Added NaCl crypto library.

67 files changed, 70260 insertions, 0 deletions

diff --git a/nacl/crypto_stream/aes128ctr/checksum b/nacl/crypto_stream/aes128ctr/checksum
new file mode 100644
index 00000000..92865436
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/checksum
@@ -0,0 +1 @@
+6e9966897837aae181e93261ae88fdf0
diff --git a/nacl/crypto_stream/aes128ctr/core2/afternm.s b/nacl/crypto_stream/aes128ctr/core2/afternm.s
new file mode 100644
index 00000000..c1ba79ef
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/afternm.s
@@ -0,0 +1,12308 @@
+# Author: Emilia Käsper and Peter Schwabe
+# Date: 2009-03-19
+# +2010.01.31: minor namespace modifications
+# Public domain
+
+.data
+.p2align 6
+
+RCON: .int 0x00000000, 0x00000000, 0x00000000, 0xffffffff
+ROTB: .int 0x0c000000, 0x00000000, 0x04000000, 0x08000000
+EXPB0: .int 0x03030303, 0x07070707, 0x0b0b0b0b, 0x0f0f0f0f
+CTRINC1: .int 0x00000001, 0x00000000, 0x00000000, 0x00000000
+CTRINC2: .int 0x00000002, 0x00000000, 0x00000000, 0x00000000
+CTRINC3: .int 0x00000003, 0x00000000, 0x00000000, 0x00000000
+CTRINC4: .int 0x00000004, 0x00000000, 0x00000000, 0x00000000
+CTRINC5: .int 0x00000005, 0x00000000, 0x00000000, 0x00000000
+CTRINC6: .int 0x00000006, 0x00000000, 0x00000000, 0x00000000
+CTRINC7: .int 0x00000007, 0x00000000, 0x00000000, 0x00000000
+RCTRINC1: .int 0x00000000, 0x00000000, 0x00000000, 0x00000001
+RCTRINC2: .int 0x00000000, 0x00000000, 0x00000000, 0x00000002
+RCTRINC3: .int 0x00000000, 0x00000000, 0x00000000, 0x00000003
+RCTRINC4: .int 0x00000000, 0x00000000, 0x00000000, 0x00000004
+RCTRINC5: .int 0x00000000, 0x00000000, 0x00000000, 0x00000005
+RCTRINC6: .int 0x00000000, 0x00000000, 0x00000000, 0x00000006
+RCTRINC7: .int 0x00000000, 0x00000000, 0x00000000, 0x00000007
+
+SWAP32: .int 0x00010203, 0x04050607, 0x08090a0b, 0x0c0d0e0f
+M0SWAP: .quad 0x0105090d0004080c , 0x03070b0f02060a0e
+
+BS0: .quad 0x5555555555555555, 0x5555555555555555
+BS1: .quad 0x3333333333333333, 0x3333333333333333
+BS2: .quad 0x0f0f0f0f0f0f0f0f, 0x0f0f0f0f0f0f0f0f
+ONE: .quad 0xffffffffffffffff, 0xffffffffffffffff
+M0:  .quad 0x02060a0e03070b0f, 0x0004080c0105090d
+SRM0:   .quad 0x0304090e00050a0f, 0x01060b0c0207080d
+SR: .quad 0x0504070600030201, 0x0f0e0d0c0a09080b
+
+# qhasm: int64 outp
+
+# qhasm: int64 len
+
+# qhasm: int64 np
+
+# qhasm: int64 c
+
+# qhasm: input outp
+
+# qhasm: input len
+
+# qhasm: input np
+
+# qhasm: input c
+
+# qhasm: int64 lensav
+
+# qhasm: int6464 xmm0
+
+# qhasm: int6464 xmm1
+
+# qhasm: int6464 xmm2
+
+# qhasm: int6464 xmm3
+
+# qhasm: int6464 xmm4
+
+# qhasm: int6464 xmm5
+
+# qhasm: int6464 xmm6
+
+# qhasm: int6464 xmm7
+
+# qhasm: int6464 xmm8
+
+# qhasm: int6464 xmm9
+
+# qhasm: int6464 xmm10
+
+# qhasm: int6464 xmm11
+
+# qhasm: int6464 xmm12
+
+# qhasm: int6464 xmm13
+
+# qhasm: int6464 xmm14
+
+# qhasm: int6464 xmm15
+
+# qhasm: int6464 t
+
+# qhasm: stack1024 bl
+
+# qhasm: stack128 nonce_stack
+
+# qhasm: int64 blp
+
+# qhasm: int64 b
+
+# qhasm: int64 tmp
+
+# qhasm: enter crypto_stream_aes128ctr_core2_afternm
+.text
+.p2align 5
+.globl _crypto_stream_aes128ctr_core2_afternm
+.globl crypto_stream_aes128ctr_core2_afternm
+_crypto_stream_aes128ctr_core2_afternm:
+crypto_stream_aes128ctr_core2_afternm:
+mov %rsp,%r11
+and $31,%r11
+add $160,%r11
+sub %r11,%rsp
+
+# qhasm: xmm0 = *(int128 *) (np + 0)
+# asm 1: movdqa 0(<np=int64#3),>xmm0=int6464#1
+# asm 2: movdqa 0(<np=%rdx),>xmm0=%xmm0
+movdqa 0(%rdx),%xmm0
+
+# qhasm: nonce_stack = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>nonce_stack=stack128#1
+# asm 2: movdqa <xmm0=%xmm0,>nonce_stack=0(%rsp)
+movdqa %xmm0,0(%rsp)
+
+# qhasm: np = &nonce_stack
+# asm 1: leaq <nonce_stack=stack128#1,>np=int64#3
+# asm 2: leaq <nonce_stack=0(%rsp),>np=%rdx
+leaq 0(%rsp),%rdx
+
+# qhasm: enc_block:
+._enc_block:
+
+# qhasm: xmm0 = *(int128 *) (np + 0)
+# asm 1: movdqa 0(<np=int64#3),>xmm0=int6464#1
+# asm 2: movdqa 0(<np=%rdx),>xmm0=%xmm0
+movdqa 0(%rdx),%xmm0
+
+# qhasm: xmm1 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm1=int6464#2
+# asm 2: movdqa <xmm0=%xmm0,>xmm1=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: shuffle bytes of xmm1 by SWAP32
+# asm 1: pshufb SWAP32,<xmm1=int6464#2
+# asm 2: pshufb SWAP32,<xmm1=%xmm1
+pshufb SWAP32,%xmm1
+
+# qhasm: xmm2 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm2=int6464#3
+# asm 2: movdqa <xmm1=%xmm1,>xmm2=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: xmm3 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm3=int6464#4
+# asm 2: movdqa <xmm1=%xmm1,>xmm3=%xmm3
+movdqa %xmm1,%xmm3
+
+# qhasm: xmm4 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm4=int6464#5
+# asm 2: movdqa <xmm1=%xmm1,>xmm4=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm5=int6464#6
+# asm 2: movdqa <xmm1=%xmm1,>xmm5=%xmm5
+movdqa %xmm1,%xmm5
+
+# qhasm: xmm6 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm6=int6464#7
+# asm 2: movdqa <xmm1=%xmm1,>xmm6=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: xmm7 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm7=int6464#8
+# asm 2: movdqa <xmm1=%xmm1,>xmm7=%xmm7
+movdqa %xmm1,%xmm7
+
+# qhasm: int32323232 xmm1 += RCTRINC1
+# asm 1: paddd  RCTRINC1,<xmm1=int6464#2
+# asm 2: paddd  RCTRINC1,<xmm1=%xmm1
+paddd  RCTRINC1,%xmm1
+
+# qhasm: int32323232 xmm2 += RCTRINC2
+# asm 1: paddd  RCTRINC2,<xmm2=int6464#3
+# asm 2: paddd  RCTRINC2,<xmm2=%xmm2
+paddd  RCTRINC2,%xmm2
+
+# qhasm: int32323232 xmm3 += RCTRINC3
+# asm 1: paddd  RCTRINC3,<xmm3=int6464#4
+# asm 2: paddd  RCTRINC3,<xmm3=%xmm3
+paddd  RCTRINC3,%xmm3
+
+# qhasm: int32323232 xmm4 += RCTRINC4
+# asm 1: paddd  RCTRINC4,<xmm4=int6464#5
+# asm 2: paddd  RCTRINC4,<xmm4=%xmm4
+paddd  RCTRINC4,%xmm4
+
+# qhasm: int32323232 xmm5 += RCTRINC5
+# asm 1: paddd  RCTRINC5,<xmm5=int6464#6
+# asm 2: paddd  RCTRINC5,<xmm5=%xmm5
+paddd  RCTRINC5,%xmm5
+
+# qhasm: int32323232 xmm6 += RCTRINC6
+# asm 1: paddd  RCTRINC6,<xmm6=int6464#7
+# asm 2: paddd  RCTRINC6,<xmm6=%xmm6
+paddd  RCTRINC6,%xmm6
+
+# qhasm: int32323232 xmm7 += RCTRINC7
+# asm 1: paddd  RCTRINC7,<xmm7=int6464#8
+# asm 2: paddd  RCTRINC7,<xmm7=%xmm7
+paddd  RCTRINC7,%xmm7
+
+# qhasm: shuffle bytes of xmm0 by M0
+# asm 1: pshufb M0,<xmm0=int6464#1
+# asm 2: pshufb M0,<xmm0=%xmm0
+pshufb M0,%xmm0
+
+# qhasm: shuffle bytes of xmm1 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm1=int6464#2
+# asm 2: pshufb M0SWAP,<xmm1=%xmm1
+pshufb M0SWAP,%xmm1
+
+# qhasm: shuffle bytes of xmm2 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm2=int6464#3
+# asm 2: pshufb M0SWAP,<xmm2=%xmm2
+pshufb M0SWAP,%xmm2
+
+# qhasm: shuffle bytes of xmm3 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm3=int6464#4
+# asm 2: pshufb M0SWAP,<xmm3=%xmm3
+pshufb M0SWAP,%xmm3
+
+# qhasm: shuffle bytes of xmm4 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm4=int6464#5
+# asm 2: pshufb M0SWAP,<xmm4=%xmm4
+pshufb M0SWAP,%xmm4
+
+# qhasm: shuffle bytes of xmm5 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm5=int6464#6
+# asm 2: pshufb M0SWAP,<xmm5=%xmm5
+pshufb M0SWAP,%xmm5
+
+# qhasm: shuffle bytes of xmm6 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm6=int6464#7
+# asm 2: pshufb M0SWAP,<xmm6=%xmm6
+pshufb M0SWAP,%xmm6
+
+# qhasm: shuffle bytes of xmm7 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm7=int6464#8
+# asm 2: pshufb M0SWAP,<xmm7=%xmm7
+pshufb M0SWAP,%xmm7
+
+# qhasm:     xmm8 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm8=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm8=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     xmm8 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm8=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm8=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     xmm8 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm8=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm8=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm8=int6464#9
+# asm 2: pxor  <xmm1=%xmm1,<xmm8=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     xmm8 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm8=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm8=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm8 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm8=int6464#9
+# asm 2: movdqa <xmm3=%xmm3,>xmm8=%xmm8
+movdqa %xmm3,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     xmm8 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm8=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm8=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 0)
+# asm 1: pxor 0(<c=int64#4),<xmm0=int6464#1
+# asm 2: pxor 0(<c=%rcx),<xmm0=%xmm0
+pxor 0(%rcx),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 16)
+# asm 1: pxor 16(<c=int64#4),<xmm1=int6464#2
+# asm 2: pxor 16(<c=%rcx),<xmm1=%xmm1
+pxor 16(%rcx),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 32)
+# asm 1: pxor 32(<c=int64#4),<xmm2=int6464#3
+# asm 2: pxor 32(<c=%rcx),<xmm2=%xmm2
+pxor 32(%rcx),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 48)
+# asm 1: pxor 48(<c=int64#4),<xmm3=int6464#4
+# asm 2: pxor 48(<c=%rcx),<xmm3=%xmm3
+pxor 48(%rcx),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 64)
+# asm 1: pxor 64(<c=int64#4),<xmm4=int6464#5
+# asm 2: pxor 64(<c=%rcx),<xmm4=%xmm4
+pxor 64(%rcx),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 80)
+# asm 1: pxor 80(<c=int64#4),<xmm5=int6464#6
+# asm 2: pxor 80(<c=%rcx),<xmm5=%xmm5
+pxor 80(%rcx),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 96)
+# asm 1: pxor 96(<c=int64#4),<xmm6=int6464#7
+# asm 2: pxor 96(<c=%rcx),<xmm6=%xmm6
+pxor 96(%rcx),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 112)
+# asm 1: pxor 112(<c=int64#4),<xmm7=int6464#8
+# asm 2: pxor 112(<c=%rcx),<xmm7=%xmm7
+pxor 112(%rcx),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 128)
+# asm 1: pxor 128(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 128(<c=%rcx),<xmm8=%xmm8
+pxor 128(%rcx),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 144)
+# asm 1: pxor 144(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 144(<c=%rcx),<xmm9=%xmm9
+pxor 144(%rcx),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 160)
+# asm 1: pxor 160(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 160(<c=%rcx),<xmm10=%xmm10
+pxor 160(%rcx),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 176)
+# asm 1: pxor 176(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 176(<c=%rcx),<xmm11=%xmm11
+pxor 176(%rcx),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 192)
+# asm 1: pxor 192(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 192(<c=%rcx),<xmm12=%xmm12
+pxor 192(%rcx),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 208)
+# asm 1: pxor 208(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 208(<c=%rcx),<xmm13=%xmm13
+pxor 208(%rcx),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 224)
+# asm 1: pxor 224(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 224(<c=%rcx),<xmm14=%xmm14
+pxor 224(%rcx),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 240)
+# asm 1: pxor 240(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 240(<c=%rcx),<xmm15=%xmm15
+pxor 240(%rcx),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 256)
+# asm 1: pxor 256(<c=int64#4),<xmm0=int6464#1
+# asm 2: pxor 256(<c=%rcx),<xmm0=%xmm0
+pxor 256(%rcx),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 272)
+# asm 1: pxor 272(<c=int64#4),<xmm1=int6464#2
+# asm 2: pxor 272(<c=%rcx),<xmm1=%xmm1
+pxor 272(%rcx),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 288)
+# asm 1: pxor 288(<c=int64#4),<xmm2=int6464#3
+# asm 2: pxor 288(<c=%rcx),<xmm2=%xmm2
+pxor 288(%rcx),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 304)
+# asm 1: pxor 304(<c=int64#4),<xmm3=int6464#4
+# asm 2: pxor 304(<c=%rcx),<xmm3=%xmm3
+pxor 304(%rcx),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 320)
+# asm 1: pxor 320(<c=int64#4),<xmm4=int6464#5
+# asm 2: pxor 320(<c=%rcx),<xmm4=%xmm4
+pxor 320(%rcx),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 336)
+# asm 1: pxor 336(<c=int64#4),<xmm5=int6464#6
+# asm 2: pxor 336(<c=%rcx),<xmm5=%xmm5
+pxor 336(%rcx),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 352)
+# asm 1: pxor 352(<c=int64#4),<xmm6=int6464#7
+# asm 2: pxor 352(<c=%rcx),<xmm6=%xmm6
+pxor 352(%rcx),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 368)
+# asm 1: pxor 368(<c=int64#4),<xmm7=int6464#8
+# asm 2: pxor 368(<c=%rcx),<xmm7=%xmm7
+pxor 368(%rcx),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 384)
+# asm 1: pxor 384(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 384(<c=%rcx),<xmm8=%xmm8
+pxor 384(%rcx),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 400)
+# asm 1: pxor 400(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 400(<c=%rcx),<xmm9=%xmm9
+pxor 400(%rcx),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 416)
+# asm 1: pxor 416(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 416(<c=%rcx),<xmm10=%xmm10
+pxor 416(%rcx),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 432)
+# asm 1: pxor 432(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 432(<c=%rcx),<xmm11=%xmm11
+pxor 432(%rcx),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 448)
+# asm 1: pxor 448(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 448(<c=%rcx),<xmm12=%xmm12
+pxor 448(%rcx),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 464)
+# asm 1: pxor 464(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 464(<c=%rcx),<xmm13=%xmm13
+pxor 464(%rcx),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 480)
+# asm 1: pxor 480(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 480(<c=%rcx),<xmm14=%xmm14
+pxor 480(%rcx),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 496)
+# asm 1: pxor 496(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 496(<c=%rcx),<xmm15=%xmm15
+pxor 496(%rcx),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 512)
+# asm 1: pxor 512(<c=int64#4),<xmm0=int6464#1
+# asm 2: pxor 512(<c=%rcx),<xmm0=%xmm0
+pxor 512(%rcx),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 528)
+# asm 1: pxor 528(<c=int64#4),<xmm1=int6464#2
+# asm 2: pxor 528(<c=%rcx),<xmm1=%xmm1
+pxor 528(%rcx),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 544)
+# asm 1: pxor 544(<c=int64#4),<xmm2=int6464#3
+# asm 2: pxor 544(<c=%rcx),<xmm2=%xmm2
+pxor 544(%rcx),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 560)
+# asm 1: pxor 560(<c=int64#4),<xmm3=int6464#4
+# asm 2: pxor 560(<c=%rcx),<xmm3=%xmm3
+pxor 560(%rcx),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 576)
+# asm 1: pxor 576(<c=int64#4),<xmm4=int6464#5
+# asm 2: pxor 576(<c=%rcx),<xmm4=%xmm4
+pxor 576(%rcx),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 592)
+# asm 1: pxor 592(<c=int64#4),<xmm5=int6464#6
+# asm 2: pxor 592(<c=%rcx),<xmm5=%xmm5
+pxor 592(%rcx),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 608)
+# asm 1: pxor 608(<c=int64#4),<xmm6=int6464#7
+# asm 2: pxor 608(<c=%rcx),<xmm6=%xmm6
+pxor 608(%rcx),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 624)
+# asm 1: pxor 624(<c=int64#4),<xmm7=int6464#8
+# asm 2: pxor 624(<c=%rcx),<xmm7=%xmm7
+pxor 624(%rcx),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 640)
+# asm 1: pxor 640(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 640(<c=%rcx),<xmm8=%xmm8
+pxor 640(%rcx),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 656)
+# asm 1: pxor 656(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 656(<c=%rcx),<xmm9=%xmm9
+pxor 656(%rcx),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 672)
+# asm 1: pxor 672(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 672(<c=%rcx),<xmm10=%xmm10
+pxor 672(%rcx),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 688)
+# asm 1: pxor 688(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 688(<c=%rcx),<xmm11=%xmm11
+pxor 688(%rcx),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 704)
+# asm 1: pxor 704(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 704(<c=%rcx),<xmm12=%xmm12
+pxor 704(%rcx),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 720)
+# asm 1: pxor 720(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 720(<c=%rcx),<xmm13=%xmm13
+pxor 720(%rcx),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 736)
+# asm 1: pxor 736(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 736(<c=%rcx),<xmm14=%xmm14
+pxor 736(%rcx),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 752)
+# asm 1: pxor 752(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 752(<c=%rcx),<xmm15=%xmm15
+pxor 752(%rcx),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 768)
+# asm 1: pxor 768(<c=int64#4),<xmm0=int6464#1
+# asm 2: pxor 768(<c=%rcx),<xmm0=%xmm0
+pxor 768(%rcx),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 784)
+# asm 1: pxor 784(<c=int64#4),<xmm1=int6464#2
+# asm 2: pxor 784(<c=%rcx),<xmm1=%xmm1
+pxor 784(%rcx),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 800)
+# asm 1: pxor 800(<c=int64#4),<xmm2=int6464#3
+# asm 2: pxor 800(<c=%rcx),<xmm2=%xmm2
+pxor 800(%rcx),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 816)
+# asm 1: pxor 816(<c=int64#4),<xmm3=int6464#4
+# asm 2: pxor 816(<c=%rcx),<xmm3=%xmm3
+pxor 816(%rcx),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 832)
+# asm 1: pxor 832(<c=int64#4),<xmm4=int6464#5
+# asm 2: pxor 832(<c=%rcx),<xmm4=%xmm4
+pxor 832(%rcx),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 848)
+# asm 1: pxor 848(<c=int64#4),<xmm5=int6464#6
+# asm 2: pxor 848(<c=%rcx),<xmm5=%xmm5
+pxor 848(%rcx),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 864)
+# asm 1: pxor 864(<c=int64#4),<xmm6=int6464#7
+# asm 2: pxor 864(<c=%rcx),<xmm6=%xmm6
+pxor 864(%rcx),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 880)
+# asm 1: pxor 880(<c=int64#4),<xmm7=int6464#8
+# asm 2: pxor 880(<c=%rcx),<xmm7=%xmm7
+pxor 880(%rcx),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 896)
+# asm 1: pxor 896(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 896(<c=%rcx),<xmm8=%xmm8
+pxor 896(%rcx),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 912)
+# asm 1: pxor 912(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 912(<c=%rcx),<xmm9=%xmm9
+pxor 912(%rcx),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 928)
+# asm 1: pxor 928(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 928(<c=%rcx),<xmm10=%xmm10
+pxor 928(%rcx),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 944)
+# asm 1: pxor 944(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 944(<c=%rcx),<xmm11=%xmm11
+pxor 944(%rcx),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 960)
+# asm 1: pxor 960(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 960(<c=%rcx),<xmm12=%xmm12
+pxor 960(%rcx),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 976)
+# asm 1: pxor 976(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 976(<c=%rcx),<xmm13=%xmm13
+pxor 976(%rcx),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 992)
+# asm 1: pxor 992(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 992(<c=%rcx),<xmm14=%xmm14
+pxor 992(%rcx),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 1008)
+# asm 1: pxor 1008(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 1008(<c=%rcx),<xmm15=%xmm15
+pxor 1008(%rcx),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 1024)
+# asm 1: pxor 1024(<c=int64#4),<xmm0=int6464#1
+# asm 2: pxor 1024(<c=%rcx),<xmm0=%xmm0
+pxor 1024(%rcx),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 1040)
+# asm 1: pxor 1040(<c=int64#4),<xmm1=int6464#2
+# asm 2: pxor 1040(<c=%rcx),<xmm1=%xmm1
+pxor 1040(%rcx),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 1056)
+# asm 1: pxor 1056(<c=int64#4),<xmm2=int6464#3
+# asm 2: pxor 1056(<c=%rcx),<xmm2=%xmm2
+pxor 1056(%rcx),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 1072)
+# asm 1: pxor 1072(<c=int64#4),<xmm3=int6464#4
+# asm 2: pxor 1072(<c=%rcx),<xmm3=%xmm3
+pxor 1072(%rcx),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 1088)
+# asm 1: pxor 1088(<c=int64#4),<xmm4=int6464#5
+# asm 2: pxor 1088(<c=%rcx),<xmm4=%xmm4
+pxor 1088(%rcx),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 1104)
+# asm 1: pxor 1104(<c=int64#4),<xmm5=int6464#6
+# asm 2: pxor 1104(<c=%rcx),<xmm5=%xmm5
+pxor 1104(%rcx),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 1120)
+# asm 1: pxor 1120(<c=int64#4),<xmm6=int6464#7
+# asm 2: pxor 1120(<c=%rcx),<xmm6=%xmm6
+pxor 1120(%rcx),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 1136)
+# asm 1: pxor 1136(<c=int64#4),<xmm7=int6464#8
+# asm 2: pxor 1136(<c=%rcx),<xmm7=%xmm7
+pxor 1136(%rcx),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 1152)
+# asm 1: pxor 1152(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 1152(<c=%rcx),<xmm8=%xmm8
+pxor 1152(%rcx),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SRM0
+# asm 1: pshufb SRM0,<xmm8=int6464#9
+# asm 2: pshufb SRM0,<xmm8=%xmm8
+pshufb SRM0,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 1168)
+# asm 1: pxor 1168(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 1168(<c=%rcx),<xmm9=%xmm9
+pxor 1168(%rcx),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SRM0
+# asm 1: pshufb SRM0,<xmm9=int6464#10
+# asm 2: pshufb SRM0,<xmm9=%xmm9
+pshufb SRM0,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 1184)
+# asm 1: pxor 1184(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 1184(<c=%rcx),<xmm10=%xmm10
+pxor 1184(%rcx),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SRM0
+# asm 1: pshufb SRM0,<xmm10=int6464#11
+# asm 2: pshufb SRM0,<xmm10=%xmm10
+pshufb SRM0,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 1200)
+# asm 1: pxor 1200(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 1200(<c=%rcx),<xmm11=%xmm11
+pxor 1200(%rcx),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SRM0
+# asm 1: pshufb SRM0,<xmm11=int6464#12
+# asm 2: pshufb SRM0,<xmm11=%xmm11
+pshufb SRM0,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 1216)
+# asm 1: pxor 1216(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 1216(<c=%rcx),<xmm12=%xmm12
+pxor 1216(%rcx),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SRM0
+# asm 1: pshufb SRM0,<xmm12=int6464#13
+# asm 2: pshufb SRM0,<xmm12=%xmm12
+pshufb SRM0,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 1232)
+# asm 1: pxor 1232(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 1232(<c=%rcx),<xmm13=%xmm13
+pxor 1232(%rcx),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SRM0
+# asm 1: pshufb SRM0,<xmm13=int6464#14
+# asm 2: pshufb SRM0,<xmm13=%xmm13
+pshufb SRM0,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 1248)
+# asm 1: pxor 1248(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 1248(<c=%rcx),<xmm14=%xmm14
+pxor 1248(%rcx),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SRM0
+# asm 1: pshufb SRM0,<xmm14=int6464#15
+# asm 2: pshufb SRM0,<xmm14=%xmm14
+pshufb SRM0,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 1264)
+# asm 1: pxor 1264(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 1264(<c=%rcx),<xmm15=%xmm15
+pxor 1264(%rcx),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SRM0
+# asm 1: pshufb SRM0,<xmm15=int6464#16
+# asm 2: pshufb SRM0,<xmm15=%xmm15
+pshufb SRM0,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:   xmm8 ^= *(int128 *)(c + 1280)
+# asm 1: pxor 1280(<c=int64#4),<xmm8=int6464#9
+# asm 2: pxor 1280(<c=%rcx),<xmm8=%xmm8
+pxor 1280(%rcx),%xmm8
+
+# qhasm:   xmm9 ^= *(int128 *)(c + 1296)
+# asm 1: pxor 1296(<c=int64#4),<xmm9=int6464#10
+# asm 2: pxor 1296(<c=%rcx),<xmm9=%xmm9
+pxor 1296(%rcx),%xmm9
+
+# qhasm:   xmm12 ^= *(int128 *)(c + 1312)
+# asm 1: pxor 1312(<c=int64#4),<xmm12=int6464#13
+# asm 2: pxor 1312(<c=%rcx),<xmm12=%xmm12
+pxor 1312(%rcx),%xmm12
+
+# qhasm:   xmm14 ^= *(int128 *)(c + 1328)
+# asm 1: pxor 1328(<c=int64#4),<xmm14=int6464#15
+# asm 2: pxor 1328(<c=%rcx),<xmm14=%xmm14
+pxor 1328(%rcx),%xmm14
+
+# qhasm:   xmm11 ^= *(int128 *)(c + 1344)
+# asm 1: pxor 1344(<c=int64#4),<xmm11=int6464#12
+# asm 2: pxor 1344(<c=%rcx),<xmm11=%xmm11
+pxor 1344(%rcx),%xmm11
+
+# qhasm:   xmm15 ^= *(int128 *)(c + 1360)
+# asm 1: pxor 1360(<c=int64#4),<xmm15=int6464#16
+# asm 2: pxor 1360(<c=%rcx),<xmm15=%xmm15
+pxor 1360(%rcx),%xmm15
+
+# qhasm:   xmm10 ^= *(int128 *)(c + 1376)
+# asm 1: pxor 1376(<c=int64#4),<xmm10=int6464#11
+# asm 2: pxor 1376(<c=%rcx),<xmm10=%xmm10
+pxor 1376(%rcx),%xmm10
+
+# qhasm:   xmm13 ^= *(int128 *)(c + 1392)
+# asm 1: pxor 1392(<c=int64#4),<xmm13=int6464#14
+# asm 2: pxor 1392(<c=%rcx),<xmm13=%xmm13
+pxor 1392(%rcx),%xmm13
+
+# qhasm:     xmm0 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm0=int6464#1
+# asm 2: movdqa <xmm10=%xmm10,>xmm0=%xmm0
+movdqa %xmm10,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     xmm0 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm0=int6464#1
+# asm 2: movdqa <xmm11=%xmm11,>xmm0=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm0=int6464#1
+# asm 2: pxor  <xmm15=%xmm15,<xmm0=%xmm0
+pxor  %xmm15,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     xmm0 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm0=int6464#1
+# asm 2: movdqa <xmm12=%xmm12,>xmm0=%xmm0
+movdqa %xmm12,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm0=int6464#1
+# asm 2: pxor  <xmm14=%xmm14,<xmm0=%xmm0
+pxor  %xmm14,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm0=int6464#1
+# asm 2: pxor  <xmm9=%xmm9,<xmm0=%xmm0
+pxor  %xmm9,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm0 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm0=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm0=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     xmm0 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm0=int6464#1
+# asm 2: movdqa <xmm11=%xmm11,>xmm0=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#1
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm0
+pxor  %xmm10,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#1
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm0=int6464#1
+# asm 2: pxor  <xmm14=%xmm14,<xmm0=%xmm0
+pxor  %xmm14,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm0=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm0 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm0=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm0=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     xmm0 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm0=int6464#1
+# asm 2: movdqa <xmm12=%xmm12,>xmm0=%xmm0
+movdqa %xmm12,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#1
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm0
+pxor  %xmm10,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#1
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm0=int6464#1
+# asm 2: pxor  <xmm15=%xmm15,<xmm0=%xmm0
+pxor  %xmm15,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#1
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm0
+pxor  %xmm11,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm: unsigned<? =? len-128
+# asm 1: cmp  $128,<len=int64#2
+# asm 2: cmp  $128,<len=%rsi
+cmp  $128,%rsi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto partial if unsigned<
+jb ._partial
+# comment:fp stack unchanged by jump
+
+# qhasm: goto full if =
+je ._full
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#3),>tmp=int64#5d
+# asm 2: movl   12(<np=%rdx),>tmp=%r8d
+movl   12(%rdx),%r8d
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#5d
+# asm 2: bswap <tmp=%r8d
+bswap %r8d
+
+# qhasm: tmp += 8
+# asm 1: add  $8,<tmp=int64#5
+# asm 2: add  $8,<tmp=%r8
+add  $8,%r8
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#5d
+# asm 2: bswap <tmp=%r8d
+bswap %r8d
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#5d,12(<np=int64#3)
+# asm 2: movl   <tmp=%r8d,12(<np=%rdx)
+movl   %r8d,12(%rdx)
+
+# qhasm: *(int128 *) (outp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<outp=int64#1)
+# asm 2: movdqa <xmm8=%xmm8,0(<outp=%rdi)
+movdqa %xmm8,0(%rdi)
+
+# qhasm: *(int128 *) (outp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<outp=int64#1)
+# asm 2: movdqa <xmm9=%xmm9,16(<outp=%rdi)
+movdqa %xmm9,16(%rdi)
+
+# qhasm: *(int128 *) (outp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<outp=int64#1)
+# asm 2: movdqa <xmm12=%xmm12,32(<outp=%rdi)
+movdqa %xmm12,32(%rdi)
+
+# qhasm: *(int128 *) (outp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<outp=int64#1)
+# asm 2: movdqa <xmm14=%xmm14,48(<outp=%rdi)
+movdqa %xmm14,48(%rdi)
+
+# qhasm: *(int128 *) (outp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<outp=int64#1)
+# asm 2: movdqa <xmm11=%xmm11,64(<outp=%rdi)
+movdqa %xmm11,64(%rdi)
+
+# qhasm: *(int128 *) (outp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<outp=int64#1)
+# asm 2: movdqa <xmm15=%xmm15,80(<outp=%rdi)
+movdqa %xmm15,80(%rdi)
+
+# qhasm: *(int128 *) (outp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<outp=int64#1)
+# asm 2: movdqa <xmm10=%xmm10,96(<outp=%rdi)
+movdqa %xmm10,96(%rdi)
+
+# qhasm: *(int128 *) (outp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<outp=int64#1)
+# asm 2: movdqa <xmm13=%xmm13,112(<outp=%rdi)
+movdqa %xmm13,112(%rdi)
+
+# qhasm: len -= 128
+# asm 1: sub  $128,<len=int64#2
+# asm 2: sub  $128,<len=%rsi
+sub  $128,%rsi
+
+# qhasm: outp += 128
+# asm 1: add  $128,<outp=int64#1
+# asm 2: add  $128,<outp=%rdi
+add  $128,%rdi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto enc_block
+jmp ._enc_block
+
+# qhasm: partial:
+._partial:
+
+# qhasm: lensav = len
+# asm 1: mov  <len=int64#2,>lensav=int64#4
+# asm 2: mov  <len=%rsi,>lensav=%rcx
+mov  %rsi,%rcx
+
+# qhasm: (uint32) len >>= 4
+# asm 1: shr  $4,<len=int64#2d
+# asm 2: shr  $4,<len=%esi
+shr  $4,%esi
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#3),>tmp=int64#5d
+# asm 2: movl   12(<np=%rdx),>tmp=%r8d
+movl   12(%rdx),%r8d
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#5d
+# asm 2: bswap <tmp=%r8d
+bswap %r8d
+
+# qhasm: tmp += len
+# asm 1: add  <len=int64#2,<tmp=int64#5
+# asm 2: add  <len=%rsi,<tmp=%r8
+add  %rsi,%r8
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#5d
+# asm 2: bswap <tmp=%r8d
+bswap %r8d
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#5d,12(<np=int64#3)
+# asm 2: movl   <tmp=%r8d,12(<np=%rdx)
+movl   %r8d,12(%rdx)
+
+# qhasm: blp = &bl
+# asm 1: leaq <bl=stack1024#1,>blp=int64#2
+# asm 2: leaq <bl=32(%rsp),>blp=%rsi
+leaq 32(%rsp),%rsi
+
+# qhasm: *(int128 *)(blp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<blp=int64#2)
+# asm 2: movdqa <xmm8=%xmm8,0(<blp=%rsi)
+movdqa %xmm8,0(%rsi)
+
+# qhasm: *(int128 *)(blp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<blp=int64#2)
+# asm 2: movdqa <xmm9=%xmm9,16(<blp=%rsi)
+movdqa %xmm9,16(%rsi)
+
+# qhasm: *(int128 *)(blp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<blp=int64#2)
+# asm 2: movdqa <xmm12=%xmm12,32(<blp=%rsi)
+movdqa %xmm12,32(%rsi)
+
+# qhasm: *(int128 *)(blp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<blp=int64#2)
+# asm 2: movdqa <xmm14=%xmm14,48(<blp=%rsi)
+movdqa %xmm14,48(%rsi)
+
+# qhasm: *(int128 *)(blp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<blp=int64#2)
+# asm 2: movdqa <xmm11=%xmm11,64(<blp=%rsi)
+movdqa %xmm11,64(%rsi)
+
+# qhasm: *(int128 *)(blp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<blp=int64#2)
+# asm 2: movdqa <xmm15=%xmm15,80(<blp=%rsi)
+movdqa %xmm15,80(%rsi)
+
+# qhasm: *(int128 *)(blp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<blp=int64#2)
+# asm 2: movdqa <xmm10=%xmm10,96(<blp=%rsi)
+movdqa %xmm10,96(%rsi)
+
+# qhasm: *(int128 *)(blp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<blp=int64#2)
+# asm 2: movdqa <xmm13=%xmm13,112(<blp=%rsi)
+movdqa %xmm13,112(%rsi)
+
+# qhasm: bytes:
+._bytes:
+
+# qhasm: =? lensav-0
+# asm 1: cmp  $0,<lensav=int64#4
+# asm 2: cmp  $0,<lensav=%rcx
+cmp  $0,%rcx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto end if =
+je ._end
+
+# qhasm: b = *(uint8 *)(blp + 0)
+# asm 1: movzbq 0(<blp=int64#2),>b=int64#3
+# asm 2: movzbq 0(<blp=%rsi),>b=%rdx
+movzbq 0(%rsi),%rdx
+
+# qhasm: *(uint8 *)(outp + 0) = b
+# asm 1: movb   <b=int64#3b,0(<outp=int64#1)
+# asm 2: movb   <b=%dl,0(<outp=%rdi)
+movb   %dl,0(%rdi)
+
+# qhasm: blp += 1
+# asm 1: add  $1,<blp=int64#2
+# asm 2: add  $1,<blp=%rsi
+add  $1,%rsi
+
+# qhasm: outp +=1
+# asm 1: add  $1,<outp=int64#1
+# asm 2: add  $1,<outp=%rdi
+add  $1,%rdi
+
+# qhasm: lensav -= 1
+# asm 1: sub  $1,<lensav=int64#4
+# asm 2: sub  $1,<lensav=%rcx
+sub  $1,%rcx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytes
+jmp ._bytes
+
+# qhasm: full:
+._full:
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#3),>tmp=int64#4d
+# asm 2: movl   12(<np=%rdx),>tmp=%ecx
+movl   12(%rdx),%ecx
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#4d
+# asm 2: bswap <tmp=%ecx
+bswap %ecx
+
+# qhasm: tmp += len
+# asm 1: add  <len=int64#2,<tmp=int64#4
+# asm 2: add  <len=%rsi,<tmp=%rcx
+add  %rsi,%rcx
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#4d
+# asm 2: bswap <tmp=%ecx
+bswap %ecx
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#4d,12(<np=int64#3)
+# asm 2: movl   <tmp=%ecx,12(<np=%rdx)
+movl   %ecx,12(%rdx)
+
+# qhasm: *(int128 *) (outp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<outp=int64#1)
+# asm 2: movdqa <xmm8=%xmm8,0(<outp=%rdi)
+movdqa %xmm8,0(%rdi)
+
+# qhasm: *(int128 *) (outp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<outp=int64#1)
+# asm 2: movdqa <xmm9=%xmm9,16(<outp=%rdi)
+movdqa %xmm9,16(%rdi)
+
+# qhasm: *(int128 *) (outp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<outp=int64#1)
+# asm 2: movdqa <xmm12=%xmm12,32(<outp=%rdi)
+movdqa %xmm12,32(%rdi)
+
+# qhasm: *(int128 *) (outp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<outp=int64#1)
+# asm 2: movdqa <xmm14=%xmm14,48(<outp=%rdi)
+movdqa %xmm14,48(%rdi)
+
+# qhasm: *(int128 *) (outp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<outp=int64#1)
+# asm 2: movdqa <xmm11=%xmm11,64(<outp=%rdi)
+movdqa %xmm11,64(%rdi)
+
+# qhasm: *(int128 *) (outp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<outp=int64#1)
+# asm 2: movdqa <xmm15=%xmm15,80(<outp=%rdi)
+movdqa %xmm15,80(%rdi)
+
+# qhasm: *(int128 *) (outp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<outp=int64#1)
+# asm 2: movdqa <xmm10=%xmm10,96(<outp=%rdi)
+movdqa %xmm10,96(%rdi)
+
+# qhasm: *(int128 *) (outp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<outp=int64#1)
+# asm 2: movdqa <xmm13=%xmm13,112(<outp=%rdi)
+movdqa %xmm13,112(%rdi)
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: end:
+._end:
+
+# qhasm: leave
+add %r11,%rsp
+mov %rdi,%rax
+mov %rsi,%rdx
+xor %rax,%rax
+ret
diff --git a/nacl/crypto_stream/aes128ctr/core2/api.h b/nacl/crypto_stream/aes128ctr/core2/api.h
new file mode 100644
index 00000000..62fc8d88
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/api.h
@@ -0,0 +1,3 @@
+#define CRYPTO_KEYBYTES 16
+#define CRYPTO_NONCEBYTES 16
+#define CRYPTO_BEFORENMBYTES 1408
diff --git a/nacl/crypto_stream/aes128ctr/core2/beforenm.s b/nacl/crypto_stream/aes128ctr/core2/beforenm.s
new file mode 100644
index 00000000..689ad8c3
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/beforenm.s
@@ -0,0 +1,13694 @@
+# Author: Emilia Käsper and Peter Schwabe
+# Date: 2009-03-19
+# +2010.01.31: minor namespace modifications
+# Public domain
+
+.data
+.p2align 6
+
+RCON: .int 0x00000000, 0x00000000, 0x00000000, 0xffffffff
+ROTB: .int 0x0c000000, 0x00000000, 0x04000000, 0x08000000
+EXPB0: .int 0x03030303, 0x07070707, 0x0b0b0b0b, 0x0f0f0f0f
+CTRINC1: .int 0x00000001, 0x00000000, 0x00000000, 0x00000000
+CTRINC2: .int 0x00000002, 0x00000000, 0x00000000, 0x00000000
+CTRINC3: .int 0x00000003, 0x00000000, 0x00000000, 0x00000000
+CTRINC4: .int 0x00000004, 0x00000000, 0x00000000, 0x00000000
+CTRINC5: .int 0x00000005, 0x00000000, 0x00000000, 0x00000000
+CTRINC6: .int 0x00000006, 0x00000000, 0x00000000, 0x00000000
+CTRINC7: .int 0x00000007, 0x00000000, 0x00000000, 0x00000000
+RCTRINC1: .int 0x00000000, 0x00000000, 0x00000000, 0x00000001
+RCTRINC2: .int 0x00000000, 0x00000000, 0x00000000, 0x00000002
+RCTRINC3: .int 0x00000000, 0x00000000, 0x00000000, 0x00000003
+RCTRINC4: .int 0x00000000, 0x00000000, 0x00000000, 0x00000004
+RCTRINC5: .int 0x00000000, 0x00000000, 0x00000000, 0x00000005
+RCTRINC6: .int 0x00000000, 0x00000000, 0x00000000, 0x00000006
+RCTRINC7: .int 0x00000000, 0x00000000, 0x00000000, 0x00000007
+
+SWAP32: .int 0x00010203, 0x04050607, 0x08090a0b, 0x0c0d0e0f
+M0SWAP: .quad 0x0105090d0004080c , 0x03070b0f02060a0e
+
+BS0: .quad 0x5555555555555555, 0x5555555555555555
+BS1: .quad 0x3333333333333333, 0x3333333333333333
+BS2: .quad 0x0f0f0f0f0f0f0f0f, 0x0f0f0f0f0f0f0f0f
+ONE: .quad 0xffffffffffffffff, 0xffffffffffffffff
+M0:  .quad 0x02060a0e03070b0f, 0x0004080c0105090d
+SRM0:   .quad 0x0304090e00050a0f, 0x01060b0c0207080d
+SR: .quad 0x0504070600030201, 0x0f0e0d0c0a09080b
+
+# qhasm: int64 arg1
+
+# qhasm: int64 arg2
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: int64 r11_caller
+
+# qhasm: int64 r12_caller
+
+# qhasm: int64 r13_caller
+
+# qhasm: int64 r14_caller
+
+# qhasm: int64 r15_caller
+
+# qhasm: int64 rbx_caller
+
+# qhasm: int64 rbp_caller
+
+# qhasm: caller r11_caller
+
+# qhasm: caller r12_caller
+
+# qhasm: caller r13_caller
+
+# qhasm: caller r14_caller
+
+# qhasm: caller r15_caller
+
+# qhasm: caller rbx_caller
+
+# qhasm: caller rbp_caller
+
+# qhasm: int64 sboxp
+
+# qhasm: int64 c
+
+# qhasm: int64 k
+
+# qhasm: int64 x0
+
+# qhasm: int64 x1
+
+# qhasm: int64 x2
+
+# qhasm: int64 x3
+
+# qhasm: int64 e
+
+# qhasm: int64 q0
+
+# qhasm: int64 q1
+
+# qhasm: int64 q2
+
+# qhasm: int64 q3
+
+# qhasm: int6464 xmm0
+
+# qhasm: int6464 xmm1
+
+# qhasm: int6464 xmm2
+
+# qhasm: int6464 xmm3
+
+# qhasm: int6464 xmm4
+
+# qhasm: int6464 xmm5
+
+# qhasm: int6464 xmm6
+
+# qhasm: int6464 xmm7
+
+# qhasm: int6464 xmm8
+
+# qhasm: int6464 xmm9
+
+# qhasm: int6464 xmm10
+
+# qhasm: int6464 xmm11
+
+# qhasm: int6464 xmm12
+
+# qhasm: int6464 xmm13
+
+# qhasm: int6464 xmm14
+
+# qhasm: int6464 xmm15
+
+# qhasm: int6464 t
+
+# qhasm: enter crypto_stream_aes128ctr_core2_beforenm
+.text
+.p2align 5
+.globl _crypto_stream_aes128ctr_core2_beforenm
+.globl crypto_stream_aes128ctr_core2_beforenm
+_crypto_stream_aes128ctr_core2_beforenm:
+crypto_stream_aes128ctr_core2_beforenm:
+mov %rsp,%r11
+and $31,%r11
+add $0,%r11
+sub %r11,%rsp
+
+# qhasm: c = arg1
+# asm 1: mov  <arg1=int64#1,>c=int64#1
+# asm 2: mov  <arg1=%rdi,>c=%rdi
+mov  %rdi,%rdi
+
+# qhasm: k = arg2
+# asm 1: mov  <arg2=int64#2,>k=int64#2
+# asm 2: mov  <arg2=%rsi,>k=%rsi
+mov  %rsi,%rsi
+
+# qhasm:   xmm0 = *(int128 *) (k + 0)
+# asm 1: movdqa 0(<k=int64#2),>xmm0=int6464#1
+# asm 2: movdqa 0(<k=%rsi),>xmm0=%xmm0
+movdqa 0(%rsi),%xmm0
+
+# qhasm:   shuffle bytes of xmm0 by M0
+# asm 1: pshufb M0,<xmm0=int6464#1
+# asm 2: pshufb M0,<xmm0=%xmm0
+pshufb M0,%xmm0
+
+# qhasm:   xmm1 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm1=int6464#2
+# asm 2: movdqa <xmm0=%xmm0,>xmm1=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm:   xmm2 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm2=int6464#3
+# asm 2: movdqa <xmm0=%xmm0,>xmm2=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:   xmm3 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm3=int6464#4
+# asm 2: movdqa <xmm0=%xmm0,>xmm3=%xmm3
+movdqa %xmm0,%xmm3
+
+# qhasm:   xmm4 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm4=int6464#5
+# asm 2: movdqa <xmm0=%xmm0,>xmm4=%xmm4
+movdqa %xmm0,%xmm4
+
+# qhasm:   xmm5 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm5=int6464#6
+# asm 2: movdqa <xmm0=%xmm0,>xmm5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:   xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm6=int6464#7
+# asm 2: movdqa <xmm0=%xmm0,>xmm6=%xmm6
+movdqa %xmm0,%xmm6
+
+# qhasm:   xmm7 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm0=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       t = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>t=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>t=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:       uint6464 t >>= 1
+# asm 1: psrlq $1,<t=int6464#9
+# asm 2: psrlq $1,<t=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:       t ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<t=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<t=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:       t &= BS0
+# asm 1: pand  BS0,<t=int6464#9
+# asm 2: pand  BS0,<t=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:       xmm7 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <t=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:       uint6464 t <<= 1
+# asm 1: psllq $1,<t=int6464#9
+# asm 2: psllq $1,<t=%xmm8
+psllq $1,%xmm8
+
+# qhasm:       xmm6 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <t=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:       t = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>t=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>t=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:       uint6464 t >>= 1
+# asm 1: psrlq $1,<t=int6464#9
+# asm 2: psrlq $1,<t=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:       t ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<t=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<t=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:       t &= BS0
+# asm 1: pand  BS0,<t=int6464#9
+# asm 2: pand  BS0,<t=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:       xmm5 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <t=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:       uint6464 t <<= 1
+# asm 1: psllq $1,<t=int6464#9
+# asm 2: psllq $1,<t=%xmm8
+psllq $1,%xmm8
+
+# qhasm:       xmm4 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <t=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       t = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>t=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>t=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:       uint6464 t >>= 1
+# asm 1: psrlq $1,<t=int6464#9
+# asm 2: psrlq $1,<t=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:       t ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<t=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<t=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:       t &= BS0
+# asm 1: pand  BS0,<t=int6464#9
+# asm 2: pand  BS0,<t=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:       xmm3 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <t=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:       uint6464 t <<= 1
+# asm 1: psllq $1,<t=int6464#9
+# asm 2: psllq $1,<t=%xmm8
+psllq $1,%xmm8
+
+# qhasm:       xmm2 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <t=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       t = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>t=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>t=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:       uint6464 t >>= 1
+# asm 1: psrlq $1,<t=int6464#9
+# asm 2: psrlq $1,<t=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:       t ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<t=int6464#9
+# asm 2: pxor  <xmm1=%xmm1,<t=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm:       t &= BS0
+# asm 1: pand  BS0,<t=int6464#9
+# asm 2: pand  BS0,<t=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:       xmm1 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <t=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:       uint6464 t <<= 1
+# asm 1: psllq $1,<t=int6464#9
+# asm 2: psllq $1,<t=%xmm8
+psllq $1,%xmm8
+
+# qhasm:       xmm0 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <t=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:       t = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>t=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>t=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       uint6464 t >>= 2
+# asm 1: psrlq $2,<t=int6464#9
+# asm 2: psrlq $2,<t=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:       t ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<t=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<t=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:       t &= BS1
+# asm 1: pand  BS1,<t=int6464#9
+# asm 2: pand  BS1,<t=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:       xmm7 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <t=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:       uint6464 t <<= 2
+# asm 1: psllq $2,<t=int6464#9
+# asm 2: psllq $2,<t=%xmm8
+psllq $2,%xmm8
+
+# qhasm:       xmm5 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <t=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:       t = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>t=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>t=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:       uint6464 t >>= 2
+# asm 1: psrlq $2,<t=int6464#9
+# asm 2: psrlq $2,<t=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:       t ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<t=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<t=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:       t &= BS1
+# asm 1: pand  BS1,<t=int6464#9
+# asm 2: pand  BS1,<t=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:       xmm6 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <t=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:       uint6464 t <<= 2
+# asm 1: psllq $2,<t=int6464#9
+# asm 2: psllq $2,<t=%xmm8
+psllq $2,%xmm8
+
+# qhasm:       xmm4 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <t=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       t = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>t=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>t=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:       uint6464 t >>= 2
+# asm 1: psrlq $2,<t=int6464#9
+# asm 2: psrlq $2,<t=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:       t ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<t=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<t=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:       t &= BS1
+# asm 1: pand  BS1,<t=int6464#9
+# asm 2: pand  BS1,<t=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:       xmm3 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <t=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:       uint6464 t <<= 2
+# asm 1: psllq $2,<t=int6464#9
+# asm 2: psllq $2,<t=%xmm8
+psllq $2,%xmm8
+
+# qhasm:       xmm1 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <t=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:       t = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>t=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>t=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:       uint6464 t >>= 2
+# asm 1: psrlq $2,<t=int6464#9
+# asm 2: psrlq $2,<t=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:       t ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<t=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<t=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:       t &= BS1
+# asm 1: pand  BS1,<t=int6464#9
+# asm 2: pand  BS1,<t=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:       xmm2 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <t=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       uint6464 t <<= 2
+# asm 1: psllq $2,<t=int6464#9
+# asm 2: psllq $2,<t=%xmm8
+psllq $2,%xmm8
+
+# qhasm:       xmm0 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <t=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:       t = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>t=int6464#9
+# asm 2: movdqa <xmm3=%xmm3,>t=%xmm8
+movdqa %xmm3,%xmm8
+
+# qhasm:       uint6464 t >>= 4
+# asm 1: psrlq $4,<t=int6464#9
+# asm 2: psrlq $4,<t=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:       t ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<t=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<t=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:       t &= BS2
+# asm 1: pand  BS2,<t=int6464#9
+# asm 2: pand  BS2,<t=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:       xmm7 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <t=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:       uint6464 t <<= 4
+# asm 1: psllq $4,<t=int6464#9
+# asm 2: psllq $4,<t=%xmm8
+psllq $4,%xmm8
+
+# qhasm:       xmm3 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <t=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:       t = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>t=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>t=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:       uint6464 t >>= 4
+# asm 1: psrlq $4,<t=int6464#9
+# asm 2: psrlq $4,<t=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:       t ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<t=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<t=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:       t &= BS2
+# asm 1: pand  BS2,<t=int6464#9
+# asm 2: pand  BS2,<t=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:       xmm6 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <t=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:       uint6464 t <<= 4
+# asm 1: psllq $4,<t=int6464#9
+# asm 2: psllq $4,<t=%xmm8
+psllq $4,%xmm8
+
+# qhasm:       xmm2 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <t=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       t = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>t=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>t=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:       uint6464 t >>= 4
+# asm 1: psrlq $4,<t=int6464#9
+# asm 2: psrlq $4,<t=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:       t ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<t=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<t=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:       t &= BS2
+# asm 1: pand  BS2,<t=int6464#9
+# asm 2: pand  BS2,<t=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:       xmm5 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <t=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:       uint6464 t <<= 4
+# asm 1: psllq $4,<t=int6464#9
+# asm 2: psllq $4,<t=%xmm8
+psllq $4,%xmm8
+
+# qhasm:       xmm1 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <t=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:       t = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>t=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>t=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:       uint6464 t >>= 4
+# asm 1: psrlq $4,<t=int6464#9
+# asm 2: psrlq $4,<t=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:       t ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<t=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<t=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       t &= BS2
+# asm 1: pand  BS2,<t=int6464#9
+# asm 2: pand  BS2,<t=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:       xmm4 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <t=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       uint6464 t <<= 4
+# asm 1: psllq $4,<t=int6464#9
+# asm 2: psllq $4,<t=%xmm8
+psllq $4,%xmm8
+
+# qhasm:       xmm0 ^= t
+# asm 1: pxor  <t=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <t=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   *(int128 *) (c + 0) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,0(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,0(<c=%rdi)
+movdqa %xmm0,0(%rdi)
+
+# qhasm:   *(int128 *) (c + 16) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,16(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,16(<c=%rdi)
+movdqa %xmm1,16(%rdi)
+
+# qhasm:   *(int128 *) (c + 32) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,32(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,32(<c=%rdi)
+movdqa %xmm2,32(%rdi)
+
+# qhasm:   *(int128 *) (c + 48) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,48(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,48(<c=%rdi)
+movdqa %xmm3,48(%rdi)
+
+# qhasm:   *(int128 *) (c + 64) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,64(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,64(<c=%rdi)
+movdqa %xmm4,64(%rdi)
+
+# qhasm:   *(int128 *) (c + 80) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,80(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,80(<c=%rdi)
+movdqa %xmm5,80(%rdi)
+
+# qhasm:   *(int128 *) (c + 96) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,96(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,96(<c=%rdi)
+movdqa %xmm6,96(%rdi)
+
+# qhasm:   *(int128 *) (c + 112) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,112(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,112(<c=%rdi)
+movdqa %xmm7,112(%rdi)
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:   xmm0 ^= RCON
+# asm 1: pxor  RCON,<xmm0=int6464#1
+# asm 2: pxor  RCON,<xmm0=%xmm0
+pxor  RCON,%xmm0
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   xmm8 = *(int128 *)(c + 0)
+# asm 1: movdqa 0(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 0(<c=%rdi),>xmm8=%xmm8
+movdqa 0(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 16)
+# asm 1: movdqa 16(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 16(<c=%rdi),>xmm9=%xmm9
+movdqa 16(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 32)
+# asm 1: movdqa 32(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 32(<c=%rdi),>xmm10=%xmm10
+movdqa 32(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 48)
+# asm 1: movdqa 48(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 48(<c=%rdi),>xmm11=%xmm11
+movdqa 48(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 64)
+# asm 1: movdqa 64(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 64(<c=%rdi),>xmm12=%xmm12
+movdqa 64(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 80)
+# asm 1: movdqa 80(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 80(<c=%rdi),>xmm13=%xmm13
+movdqa 80(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 96)
+# asm 1: movdqa 96(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 96(<c=%rdi),>xmm14=%xmm14
+movdqa 96(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 112)
+# asm 1: movdqa 112(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 112(<c=%rdi),>xmm15=%xmm15
+movdqa 112(%rdi),%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   *(int128 *)(c + 128) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,128(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,128(<c=%rdi)
+movdqa %xmm0,128(%rdi)
+
+# qhasm:   *(int128 *)(c + 144) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,144(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,144(<c=%rdi)
+movdqa %xmm1,144(%rdi)
+
+# qhasm:   *(int128 *)(c + 160) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,160(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,160(<c=%rdi)
+movdqa %xmm4,160(%rdi)
+
+# qhasm:   *(int128 *)(c + 176) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,176(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,176(<c=%rdi)
+movdqa %xmm6,176(%rdi)
+
+# qhasm:   *(int128 *)(c + 192) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,192(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,192(<c=%rdi)
+movdqa %xmm3,192(%rdi)
+
+# qhasm:   *(int128 *)(c + 208) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,208(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,208(<c=%rdi)
+movdqa %xmm7,208(%rdi)
+
+# qhasm:   *(int128 *)(c + 224) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,224(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,224(<c=%rdi)
+movdqa %xmm2,224(%rdi)
+
+# qhasm:   *(int128 *)(c + 240) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,240(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,240(<c=%rdi)
+movdqa %xmm5,240(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm7 ^= ONE
+# asm 1: pxor  ONE,<xmm7=int6464#8
+# asm 2: pxor  ONE,<xmm7=%xmm7
+pxor  ONE,%xmm7
+
+# qhasm:   xmm2 ^= ONE
+# asm 1: pxor  ONE,<xmm2=int6464#3
+# asm 2: pxor  ONE,<xmm2=%xmm2
+pxor  ONE,%xmm2
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm7=int6464#8
+# asm 2: pxor  <xmm2=%xmm2,<xmm7=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm1,<xmm4=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm2 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm4=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm5 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm5=int6464#6
+# asm 2: pxor  <xmm7=%xmm7,<xmm5=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:       xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm6=int6464#7
+# asm 2: pxor  <xmm1=%xmm1,<xmm6=%xmm6
+pxor  %xmm1,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm11 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm11=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm11=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm9=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm9=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#13
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm:       xmm11 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm11=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm11=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:       xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:       xmm9 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm9=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm9=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:       xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#12
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm11
+pxor  %xmm3,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#11
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm10
+movdqa %xmm6,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm13=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm13=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#12
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm11
+movdqa %xmm7,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#11
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm10
+movdqa %xmm4,%xmm10
+
+# qhasm:       xmm13 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm13=int6464#12
+# asm 2: movdqa <xmm3=%xmm3,>xmm13=%xmm11
+movdqa %xmm3,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: movdqa <xmm5=%xmm5,>xmm15=%xmm15
+movdqa %xmm5,%xmm15
+
+# qhasm:       xmm12 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm12=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm12=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm14=int6464#14
+# asm 2: pand  <xmm7=%xmm7,<xmm14=%xmm13
+pand  %xmm7,%xmm13
+
+# qhasm:       xmm15 |= xmm2
+# asm 1: por   <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: por   <xmm2=%xmm2,<xmm15=%xmm15
+por   %xmm2,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:         xmm8 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm8=int6464#10
+# asm 2: movdqa <xmm7=%xmm7,>xmm8=%xmm9
+movdqa %xmm7,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm10=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm11,<xmm2=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:           xmm7 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm7=int6464#8
+# asm 2: pand  <xmm15=%xmm13,<xmm7=%xmm7
+pand  %xmm13,%xmm7
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm6=int6464#7
+# asm 2: pand  <xmm13=%xmm15,<xmm6=%xmm6
+pand  %xmm15,%xmm6
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:         xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm9,<xmm7=%xmm7
+pxor  %xmm9,%xmm7
+
+# qhasm:         xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm9,<xmm6=%xmm6
+pxor  %xmm9,%xmm6
+
+# qhasm:         xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm10=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm10=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm3=int6464#4
+# asm 2: pand  <xmm9=%xmm12,<xmm3=%xmm3
+pand  %xmm12,%xmm3
+
+# qhasm:           xmm4 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm4=int6464#5
+# asm 2: pand  <xmm13=%xmm15,<xmm4=%xmm4
+pand  %xmm15,%xmm4
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm11=int6464#11
+# asm 2: pand  <xmm5=%xmm5,<xmm11=%xmm10
+pand  %xmm5,%xmm10
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm5 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm5=int6464#6
+# asm 2: pand  <xmm14=%xmm11,<xmm5=%xmm5
+pand  %xmm11,%xmm5
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm12=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:         xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm9,<xmm4=%xmm4
+pxor  %xmm9,%xmm4
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm1=int6464#2
+# asm 2: pxor  <xmm2=%xmm2,<xmm1=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm2 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm2=int6464#3
+# asm 2: pxor  <xmm0=%xmm0,<xmm2=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm7 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pxor  <xmm4=%xmm4,<xmm7=%xmm7
+pxor  %xmm4,%xmm7
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm4=int6464#5
+# asm 2: pxor  <xmm6=%xmm6,<xmm4=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:       xmm6 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm6=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm6=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:   xmm1 ^= RCON
+# asm 1: pxor  RCON,<xmm1=int6464#2
+# asm 2: pxor  RCON,<xmm1=%xmm1
+pxor  RCON,%xmm1
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   xmm8 = *(int128 *)(c + 128)
+# asm 1: movdqa 128(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 128(<c=%rdi),>xmm8=%xmm8
+movdqa 128(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 144)
+# asm 1: movdqa 144(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 144(<c=%rdi),>xmm9=%xmm9
+movdqa 144(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 160)
+# asm 1: movdqa 160(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 160(<c=%rdi),>xmm10=%xmm10
+movdqa 160(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 176)
+# asm 1: movdqa 176(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 176(<c=%rdi),>xmm11=%xmm11
+movdqa 176(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 192)
+# asm 1: movdqa 192(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 192(<c=%rdi),>xmm12=%xmm12
+movdqa 192(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 208)
+# asm 1: movdqa 208(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 208(<c=%rdi),>xmm13=%xmm13
+movdqa 208(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 224)
+# asm 1: movdqa 224(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 224(<c=%rdi),>xmm14=%xmm14
+movdqa 224(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 240)
+# asm 1: movdqa 240(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 240(<c=%rdi),>xmm15=%xmm15
+movdqa 240(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   *(int128 *)(c + 256) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,256(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,256(<c=%rdi)
+movdqa %xmm0,256(%rdi)
+
+# qhasm:   *(int128 *)(c + 272) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,272(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,272(<c=%rdi)
+movdqa %xmm1,272(%rdi)
+
+# qhasm:   *(int128 *)(c + 288) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,288(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,288(<c=%rdi)
+movdqa %xmm3,288(%rdi)
+
+# qhasm:   *(int128 *)(c + 304) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,304(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,304(<c=%rdi)
+movdqa %xmm2,304(%rdi)
+
+# qhasm:   *(int128 *)(c + 320) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,320(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,320(<c=%rdi)
+movdqa %xmm6,320(%rdi)
+
+# qhasm:   *(int128 *)(c + 336) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,336(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,336(<c=%rdi)
+movdqa %xmm5,336(%rdi)
+
+# qhasm:   *(int128 *)(c + 352) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,352(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,352(<c=%rdi)
+movdqa %xmm4,352(%rdi)
+
+# qhasm:   *(int128 *)(c + 368) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,368(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,368(<c=%rdi)
+movdqa %xmm7,368(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm5 ^= ONE
+# asm 1: pxor  ONE,<xmm5=int6464#6
+# asm 2: pxor  ONE,<xmm5=%xmm5
+pxor  ONE,%xmm5
+
+# qhasm:   xmm4 ^= ONE
+# asm 1: pxor  ONE,<xmm4=int6464#5
+# asm 2: pxor  ONE,<xmm4=%xmm4
+pxor  ONE,%xmm4
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:       xmm5 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm5=int6464#6
+# asm 2: pxor  <xmm4=%xmm4,<xmm5=%xmm5
+pxor  %xmm4,%xmm5
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm4=int6464#5
+# asm 2: pxor  <xmm3=%xmm3,<xmm4=%xmm4
+pxor  %xmm3,%xmm4
+
+# qhasm:       xmm2 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm2=int6464#3
+# asm 2: pxor  <xmm0=%xmm0,<xmm2=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm13=int6464#12
+# asm 2: movdqa <xmm3=%xmm3,>xmm13=%xmm11
+movdqa %xmm3,%xmm11
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#13
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm12
+movdqa %xmm4,%xmm12
+
+# qhasm:       xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:       xmm10 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm10=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm10=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:       xmm9 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm9=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm9=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:       xmm13 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm13=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm13=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm13 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm13=int6464#12
+# asm 2: movdqa <xmm6=%xmm6,>xmm13=%xmm11
+movdqa %xmm6,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm12=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm12=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm15=int6464#16
+# asm 2: por   <xmm4=%xmm4,<xmm15=%xmm15
+por   %xmm4,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:           xmm4 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm4=int6464#5
+# asm 2: pand  <xmm14=%xmm11,<xmm4=%xmm4
+pand  %xmm11,%xmm4
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm0=int6464#1
+# asm 2: pxor  <xmm2=%xmm2,<xmm0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm0 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm0=int6464#1
+# asm 2: pxor  <xmm2=%xmm2,<xmm0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:           xmm6 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm6=int6464#7
+# asm 2: pand  <xmm9=%xmm12,<xmm6=%xmm6
+pand  %xmm12,%xmm6
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm1=int6464#2
+# asm 2: pxor  <xmm4=%xmm4,<xmm1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:       xmm6 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm6=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm6=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm4=int6464#5
+# asm 2: pxor  <xmm0=%xmm0,<xmm4=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm5=int6464#6
+# asm 2: pxor  <xmm3=%xmm3,<xmm5=%xmm5
+pxor  %xmm3,%xmm5
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm3=int6464#4
+# asm 2: pxor  <xmm2=%xmm2,<xmm3=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm2=int6464#3
+# asm 2: pxor  <xmm5=%xmm5,<xmm2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:   xmm6 ^= RCON
+# asm 1: pxor  RCON,<xmm6=int6464#7
+# asm 2: pxor  RCON,<xmm6=%xmm6
+pxor  RCON,%xmm6
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   xmm8 = *(int128 *)(c + 256)
+# asm 1: movdqa 256(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 256(<c=%rdi),>xmm8=%xmm8
+movdqa 256(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 272)
+# asm 1: movdqa 272(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 272(<c=%rdi),>xmm9=%xmm9
+movdqa 272(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 288)
+# asm 1: movdqa 288(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 288(<c=%rdi),>xmm10=%xmm10
+movdqa 288(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 304)
+# asm 1: movdqa 304(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 304(<c=%rdi),>xmm11=%xmm11
+movdqa 304(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 320)
+# asm 1: movdqa 320(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 320(<c=%rdi),>xmm12=%xmm12
+movdqa 320(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 336)
+# asm 1: movdqa 336(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 336(<c=%rdi),>xmm13=%xmm13
+movdqa 336(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 352)
+# asm 1: movdqa 352(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 352(<c=%rdi),>xmm14=%xmm14
+movdqa 352(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 368)
+# asm 1: movdqa 368(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 368(<c=%rdi),>xmm15=%xmm15
+movdqa 368(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   *(int128 *)(c + 384) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,384(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,384(<c=%rdi)
+movdqa %xmm0,384(%rdi)
+
+# qhasm:   *(int128 *)(c + 400) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,400(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,400(<c=%rdi)
+movdqa %xmm1,400(%rdi)
+
+# qhasm:   *(int128 *)(c + 416) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,416(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,416(<c=%rdi)
+movdqa %xmm6,416(%rdi)
+
+# qhasm:   *(int128 *)(c + 432) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,432(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,432(<c=%rdi)
+movdqa %xmm4,432(%rdi)
+
+# qhasm:   *(int128 *)(c + 448) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,448(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,448(<c=%rdi)
+movdqa %xmm2,448(%rdi)
+
+# qhasm:   *(int128 *)(c + 464) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,464(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,464(<c=%rdi)
+movdqa %xmm7,464(%rdi)
+
+# qhasm:   *(int128 *)(c + 480) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,480(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,480(<c=%rdi)
+movdqa %xmm3,480(%rdi)
+
+# qhasm:   *(int128 *)(c + 496) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,496(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,496(<c=%rdi)
+movdqa %xmm5,496(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm7 ^= ONE
+# asm 1: pxor  ONE,<xmm7=int6464#8
+# asm 2: pxor  ONE,<xmm7=%xmm7
+pxor  ONE,%xmm7
+
+# qhasm:   xmm3 ^= ONE
+# asm 1: pxor  ONE,<xmm3=int6464#4
+# asm 2: pxor  ONE,<xmm3=%xmm3
+pxor  ONE,%xmm3
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:       xmm7 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm7=int6464#8
+# asm 2: pxor  <xmm3=%xmm3,<xmm7=%xmm7
+pxor  %xmm3,%xmm7
+
+# qhasm:       xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm6=int6464#7
+# asm 2: pxor  <xmm1=%xmm1,<xmm6=%xmm6
+pxor  %xmm1,%xmm6
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm3=int6464#4
+# asm 2: pxor  <xmm6=%xmm6,<xmm3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:       xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm4=int6464#5
+# asm 2: pxor  <xmm0=%xmm0,<xmm4=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm5 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm5=int6464#6
+# asm 2: pxor  <xmm7=%xmm7,<xmm5=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm1,<xmm4=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm11 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm11=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm11=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm9=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm9=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm13=int6464#12
+# asm 2: movdqa <xmm6=%xmm6,>xmm13=%xmm11
+movdqa %xmm6,%xmm11
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm12
+movdqa %xmm3,%xmm12
+
+# qhasm:       xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:       xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:       xmm9 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm9=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm9=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:       xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#11
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm10
+movdqa %xmm4,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm13=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm13=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#12
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm11
+movdqa %xmm7,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#12
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm11
+pxor  %xmm3,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#11
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm10
+movdqa %xmm6,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: movdqa <xmm5=%xmm5,>xmm15=%xmm15
+movdqa %xmm5,%xmm15
+
+# qhasm:       xmm12 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm12=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm12=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm14=int6464#14
+# asm 2: pand  <xmm7=%xmm7,<xmm14=%xmm13
+pand  %xmm7,%xmm13
+
+# qhasm:       xmm15 |= xmm3
+# asm 1: por   <xmm3=int6464#4,<xmm15=int6464#16
+# asm 2: por   <xmm3=%xmm3,<xmm15=%xmm15
+por   %xmm3,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#9
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm8
+movdqa %xmm3,%xmm8
+
+# qhasm:         xmm8 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm8=int6464#10
+# asm 2: movdqa <xmm7=%xmm7,>xmm8=%xmm9
+movdqa %xmm7,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm10=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm10=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:           xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:           xmm3 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm3=int6464#4
+# asm 2: pand  <xmm14=%xmm11,<xmm3=%xmm3
+pand  %xmm11,%xmm3
+
+# qhasm:           xmm7 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm7=int6464#8
+# asm 2: pand  <xmm15=%xmm13,<xmm7=%xmm7
+pand  %xmm13,%xmm7
+
+# qhasm:           xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:           xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm0=int6464#1
+# asm 2: pxor  <xmm4=%xmm4,<xmm0=%xmm0
+pxor  %xmm4,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm4 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm4=int6464#5
+# asm 2: pand  <xmm13=%xmm15,<xmm4=%xmm4
+pand  %xmm15,%xmm4
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm0=int6464#1
+# asm 2: pxor  <xmm4=%xmm4,<xmm0=%xmm0
+pxor  %xmm4,%xmm0
+
+# qhasm:           xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:         xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm9,<xmm7=%xmm7
+pxor  %xmm9,%xmm7
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm9,<xmm4=%xmm4
+pxor  %xmm9,%xmm4
+
+# qhasm:         xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:         xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm10=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:           xmm2 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm9=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm6=int6464#7
+# asm 2: pand  <xmm13=%xmm15,<xmm6=%xmm6
+pand  %xmm15,%xmm6
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:           xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm11=int6464#11
+# asm 2: pand  <xmm5=%xmm5,<xmm11=%xmm10
+pand  %xmm5,%xmm10
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm5 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm5=int6464#6
+# asm 2: pand  <xmm14=%xmm11,<xmm5=%xmm5
+pand  %xmm11,%xmm5
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm12=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:         xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm9,<xmm6=%xmm6
+pxor  %xmm9,%xmm6
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm1 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm1=int6464#2
+# asm 2: pxor  <xmm3=%xmm3,<xmm1=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm2=int6464#3
+# asm 2: pxor  <xmm5=%xmm5,<xmm2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm7 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm6=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm6=int6464#7
+# asm 2: pxor  <xmm4=%xmm4,<xmm6=%xmm6
+pxor  %xmm4,%xmm6
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:   xmm3 ^= RCON
+# asm 1: pxor  RCON,<xmm3=int6464#4
+# asm 2: pxor  RCON,<xmm3=%xmm3
+pxor  RCON,%xmm3
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   xmm8 = *(int128 *)(c + 384)
+# asm 1: movdqa 384(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 384(<c=%rdi),>xmm8=%xmm8
+movdqa 384(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 400)
+# asm 1: movdqa 400(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 400(<c=%rdi),>xmm9=%xmm9
+movdqa 400(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 416)
+# asm 1: movdqa 416(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 416(<c=%rdi),>xmm10=%xmm10
+movdqa 416(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 432)
+# asm 1: movdqa 432(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 432(<c=%rdi),>xmm11=%xmm11
+movdqa 432(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 448)
+# asm 1: movdqa 448(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 448(<c=%rdi),>xmm12=%xmm12
+movdqa 448(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 464)
+# asm 1: movdqa 464(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 464(<c=%rdi),>xmm13=%xmm13
+movdqa 464(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 480)
+# asm 1: movdqa 480(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 480(<c=%rdi),>xmm14=%xmm14
+movdqa 480(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 496)
+# asm 1: movdqa 496(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 496(<c=%rdi),>xmm15=%xmm15
+movdqa 496(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   *(int128 *)(c + 512) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,512(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,512(<c=%rdi)
+movdqa %xmm0,512(%rdi)
+
+# qhasm:   *(int128 *)(c + 528) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,528(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,528(<c=%rdi)
+movdqa %xmm1,528(%rdi)
+
+# qhasm:   *(int128 *)(c + 544) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,544(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,544(<c=%rdi)
+movdqa %xmm2,544(%rdi)
+
+# qhasm:   *(int128 *)(c + 560) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,560(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,560(<c=%rdi)
+movdqa %xmm3,560(%rdi)
+
+# qhasm:   *(int128 *)(c + 576) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,576(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,576(<c=%rdi)
+movdqa %xmm4,576(%rdi)
+
+# qhasm:   *(int128 *)(c + 592) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,592(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,592(<c=%rdi)
+movdqa %xmm5,592(%rdi)
+
+# qhasm:   *(int128 *)(c + 608) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,608(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,608(<c=%rdi)
+movdqa %xmm6,608(%rdi)
+
+# qhasm:   *(int128 *)(c + 624) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,624(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,624(<c=%rdi)
+movdqa %xmm7,624(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm5 ^= ONE
+# asm 1: pxor  ONE,<xmm5=int6464#6
+# asm 2: pxor  ONE,<xmm5=%xmm5
+pxor  ONE,%xmm5
+
+# qhasm:   xmm6 ^= ONE
+# asm 1: pxor  ONE,<xmm6=int6464#7
+# asm 2: pxor  ONE,<xmm6=%xmm6
+pxor  ONE,%xmm6
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:   xmm3 ^= RCON
+# asm 1: pxor  RCON,<xmm3=int6464#4
+# asm 2: pxor  RCON,<xmm3=%xmm3
+pxor  RCON,%xmm3
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   xmm8 = *(int128 *)(c + 512)
+# asm 1: movdqa 512(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 512(<c=%rdi),>xmm8=%xmm8
+movdqa 512(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 528)
+# asm 1: movdqa 528(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 528(<c=%rdi),>xmm9=%xmm9
+movdqa 528(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 544)
+# asm 1: movdqa 544(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 544(<c=%rdi),>xmm10=%xmm10
+movdqa 544(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 560)
+# asm 1: movdqa 560(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 560(<c=%rdi),>xmm11=%xmm11
+movdqa 560(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 576)
+# asm 1: movdqa 576(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 576(<c=%rdi),>xmm12=%xmm12
+movdqa 576(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 592)
+# asm 1: movdqa 592(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 592(<c=%rdi),>xmm13=%xmm13
+movdqa 592(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 608)
+# asm 1: movdqa 608(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 608(<c=%rdi),>xmm14=%xmm14
+movdqa 608(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 624)
+# asm 1: movdqa 624(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 624(<c=%rdi),>xmm15=%xmm15
+movdqa 624(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   *(int128 *)(c + 640) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,640(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,640(<c=%rdi)
+movdqa %xmm0,640(%rdi)
+
+# qhasm:   *(int128 *)(c + 656) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,656(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,656(<c=%rdi)
+movdqa %xmm1,656(%rdi)
+
+# qhasm:   *(int128 *)(c + 672) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,672(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,672(<c=%rdi)
+movdqa %xmm4,672(%rdi)
+
+# qhasm:   *(int128 *)(c + 688) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,688(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,688(<c=%rdi)
+movdqa %xmm6,688(%rdi)
+
+# qhasm:   *(int128 *)(c + 704) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,704(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,704(<c=%rdi)
+movdqa %xmm3,704(%rdi)
+
+# qhasm:   *(int128 *)(c + 720) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,720(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,720(<c=%rdi)
+movdqa %xmm7,720(%rdi)
+
+# qhasm:   *(int128 *)(c + 736) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,736(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,736(<c=%rdi)
+movdqa %xmm2,736(%rdi)
+
+# qhasm:   *(int128 *)(c + 752) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,752(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,752(<c=%rdi)
+movdqa %xmm5,752(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm7 ^= ONE
+# asm 1: pxor  ONE,<xmm7=int6464#8
+# asm 2: pxor  ONE,<xmm7=%xmm7
+pxor  ONE,%xmm7
+
+# qhasm:   xmm2 ^= ONE
+# asm 1: pxor  ONE,<xmm2=int6464#3
+# asm 2: pxor  ONE,<xmm2=%xmm2
+pxor  ONE,%xmm2
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm7=int6464#8
+# asm 2: pxor  <xmm2=%xmm2,<xmm7=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm1,<xmm4=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm2 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm4=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm5 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm5=int6464#6
+# asm 2: pxor  <xmm7=%xmm7,<xmm5=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:       xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm6=int6464#7
+# asm 2: pxor  <xmm1=%xmm1,<xmm6=%xmm6
+pxor  %xmm1,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm11 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm11=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm11=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm9=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm9=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#13
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm:       xmm11 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm11=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm11=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:       xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:       xmm9 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm9=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm9=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:       xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#12
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm11
+pxor  %xmm3,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#11
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm10
+movdqa %xmm6,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm13=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm13=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#12
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm11
+movdqa %xmm7,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#11
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm10
+movdqa %xmm4,%xmm10
+
+# qhasm:       xmm13 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm13=int6464#12
+# asm 2: movdqa <xmm3=%xmm3,>xmm13=%xmm11
+movdqa %xmm3,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: movdqa <xmm5=%xmm5,>xmm15=%xmm15
+movdqa %xmm5,%xmm15
+
+# qhasm:       xmm12 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm12=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm12=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm14=int6464#14
+# asm 2: pand  <xmm7=%xmm7,<xmm14=%xmm13
+pand  %xmm7,%xmm13
+
+# qhasm:       xmm15 |= xmm2
+# asm 1: por   <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: por   <xmm2=%xmm2,<xmm15=%xmm15
+por   %xmm2,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:         xmm8 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm8=int6464#10
+# asm 2: movdqa <xmm7=%xmm7,>xmm8=%xmm9
+movdqa %xmm7,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm10=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm11,<xmm2=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:           xmm7 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm7=int6464#8
+# asm 2: pand  <xmm15=%xmm13,<xmm7=%xmm7
+pand  %xmm13,%xmm7
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm6=int6464#7
+# asm 2: pand  <xmm13=%xmm15,<xmm6=%xmm6
+pand  %xmm15,%xmm6
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:         xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm9,<xmm7=%xmm7
+pxor  %xmm9,%xmm7
+
+# qhasm:         xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm9,<xmm6=%xmm6
+pxor  %xmm9,%xmm6
+
+# qhasm:         xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm10=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm10=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm3=int6464#4
+# asm 2: pand  <xmm9=%xmm12,<xmm3=%xmm3
+pand  %xmm12,%xmm3
+
+# qhasm:           xmm4 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm4=int6464#5
+# asm 2: pand  <xmm13=%xmm15,<xmm4=%xmm4
+pand  %xmm15,%xmm4
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm11=int6464#11
+# asm 2: pand  <xmm5=%xmm5,<xmm11=%xmm10
+pand  %xmm5,%xmm10
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm5 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm5=int6464#6
+# asm 2: pand  <xmm14=%xmm11,<xmm5=%xmm5
+pand  %xmm11,%xmm5
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm12=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:         xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm9,<xmm4=%xmm4
+pxor  %xmm9,%xmm4
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm1=int6464#2
+# asm 2: pxor  <xmm2=%xmm2,<xmm1=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm2 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm2=int6464#3
+# asm 2: pxor  <xmm0=%xmm0,<xmm2=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm7 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pxor  <xmm4=%xmm4,<xmm7=%xmm7
+pxor  %xmm4,%xmm7
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm4=int6464#5
+# asm 2: pxor  <xmm6=%xmm6,<xmm4=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:       xmm6 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm6=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm6=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:   xmm5 ^= RCON
+# asm 1: pxor  RCON,<xmm5=int6464#6
+# asm 2: pxor  RCON,<xmm5=%xmm5
+pxor  RCON,%xmm5
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   xmm8 = *(int128 *)(c + 640)
+# asm 1: movdqa 640(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 640(<c=%rdi),>xmm8=%xmm8
+movdqa 640(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 656)
+# asm 1: movdqa 656(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 656(<c=%rdi),>xmm9=%xmm9
+movdqa 656(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 672)
+# asm 1: movdqa 672(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 672(<c=%rdi),>xmm10=%xmm10
+movdqa 672(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 688)
+# asm 1: movdqa 688(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 688(<c=%rdi),>xmm11=%xmm11
+movdqa 688(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 704)
+# asm 1: movdqa 704(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 704(<c=%rdi),>xmm12=%xmm12
+movdqa 704(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 720)
+# asm 1: movdqa 720(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 720(<c=%rdi),>xmm13=%xmm13
+movdqa 720(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 736)
+# asm 1: movdqa 736(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 736(<c=%rdi),>xmm14=%xmm14
+movdqa 736(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 752)
+# asm 1: movdqa 752(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 752(<c=%rdi),>xmm15=%xmm15
+movdqa 752(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   *(int128 *)(c + 768) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,768(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,768(<c=%rdi)
+movdqa %xmm0,768(%rdi)
+
+# qhasm:   *(int128 *)(c + 784) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,784(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,784(<c=%rdi)
+movdqa %xmm1,784(%rdi)
+
+# qhasm:   *(int128 *)(c + 800) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,800(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,800(<c=%rdi)
+movdqa %xmm3,800(%rdi)
+
+# qhasm:   *(int128 *)(c + 816) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,816(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,816(<c=%rdi)
+movdqa %xmm2,816(%rdi)
+
+# qhasm:   *(int128 *)(c + 832) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,832(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,832(<c=%rdi)
+movdqa %xmm6,832(%rdi)
+
+# qhasm:   *(int128 *)(c + 848) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,848(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,848(<c=%rdi)
+movdqa %xmm5,848(%rdi)
+
+# qhasm:   *(int128 *)(c + 864) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,864(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,864(<c=%rdi)
+movdqa %xmm4,864(%rdi)
+
+# qhasm:   *(int128 *)(c + 880) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,880(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,880(<c=%rdi)
+movdqa %xmm7,880(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm5 ^= ONE
+# asm 1: pxor  ONE,<xmm5=int6464#6
+# asm 2: pxor  ONE,<xmm5=%xmm5
+pxor  ONE,%xmm5
+
+# qhasm:   xmm4 ^= ONE
+# asm 1: pxor  ONE,<xmm4=int6464#5
+# asm 2: pxor  ONE,<xmm4=%xmm4
+pxor  ONE,%xmm4
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:       xmm5 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm5=int6464#6
+# asm 2: pxor  <xmm4=%xmm4,<xmm5=%xmm5
+pxor  %xmm4,%xmm5
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm4=int6464#5
+# asm 2: pxor  <xmm3=%xmm3,<xmm4=%xmm4
+pxor  %xmm3,%xmm4
+
+# qhasm:       xmm2 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm2=int6464#3
+# asm 2: pxor  <xmm0=%xmm0,<xmm2=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm13=int6464#12
+# asm 2: movdqa <xmm3=%xmm3,>xmm13=%xmm11
+movdqa %xmm3,%xmm11
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#13
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm12
+movdqa %xmm4,%xmm12
+
+# qhasm:       xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:       xmm10 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm10=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm10=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:       xmm9 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm9=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm9=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:       xmm13 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm13=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm13=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm13 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm13=int6464#12
+# asm 2: movdqa <xmm6=%xmm6,>xmm13=%xmm11
+movdqa %xmm6,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm12=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm12=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm15=int6464#16
+# asm 2: por   <xmm4=%xmm4,<xmm15=%xmm15
+por   %xmm4,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:           xmm4 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm4=int6464#5
+# asm 2: pand  <xmm14=%xmm11,<xmm4=%xmm4
+pand  %xmm11,%xmm4
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm0=int6464#1
+# asm 2: pxor  <xmm2=%xmm2,<xmm0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm0 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm0=int6464#1
+# asm 2: pxor  <xmm2=%xmm2,<xmm0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:           xmm6 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm6=int6464#7
+# asm 2: pand  <xmm9=%xmm12,<xmm6=%xmm6
+pand  %xmm12,%xmm6
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm1=int6464#2
+# asm 2: pxor  <xmm4=%xmm4,<xmm1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:       xmm6 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm6=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm6=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm4=int6464#5
+# asm 2: pxor  <xmm0=%xmm0,<xmm4=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm5=int6464#6
+# asm 2: pxor  <xmm3=%xmm3,<xmm5=%xmm5
+pxor  %xmm3,%xmm5
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm3=int6464#4
+# asm 2: pxor  <xmm2=%xmm2,<xmm3=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm2=int6464#3
+# asm 2: pxor  <xmm5=%xmm5,<xmm2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:   xmm3 ^= RCON
+# asm 1: pxor  RCON,<xmm3=int6464#4
+# asm 2: pxor  RCON,<xmm3=%xmm3
+pxor  RCON,%xmm3
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   xmm8 = *(int128 *)(c + 768)
+# asm 1: movdqa 768(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 768(<c=%rdi),>xmm8=%xmm8
+movdqa 768(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 784)
+# asm 1: movdqa 784(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 784(<c=%rdi),>xmm9=%xmm9
+movdqa 784(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 800)
+# asm 1: movdqa 800(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 800(<c=%rdi),>xmm10=%xmm10
+movdqa 800(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 816)
+# asm 1: movdqa 816(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 816(<c=%rdi),>xmm11=%xmm11
+movdqa 816(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 832)
+# asm 1: movdqa 832(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 832(<c=%rdi),>xmm12=%xmm12
+movdqa 832(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 848)
+# asm 1: movdqa 848(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 848(<c=%rdi),>xmm13=%xmm13
+movdqa 848(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 864)
+# asm 1: movdqa 864(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 864(<c=%rdi),>xmm14=%xmm14
+movdqa 864(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 880)
+# asm 1: movdqa 880(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 880(<c=%rdi),>xmm15=%xmm15
+movdqa 880(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:   xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:   xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   *(int128 *)(c + 896) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,896(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,896(<c=%rdi)
+movdqa %xmm0,896(%rdi)
+
+# qhasm:   *(int128 *)(c + 912) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,912(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,912(<c=%rdi)
+movdqa %xmm1,912(%rdi)
+
+# qhasm:   *(int128 *)(c + 928) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,928(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,928(<c=%rdi)
+movdqa %xmm6,928(%rdi)
+
+# qhasm:   *(int128 *)(c + 944) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,944(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,944(<c=%rdi)
+movdqa %xmm4,944(%rdi)
+
+# qhasm:   *(int128 *)(c + 960) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,960(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,960(<c=%rdi)
+movdqa %xmm2,960(%rdi)
+
+# qhasm:   *(int128 *)(c + 976) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,976(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,976(<c=%rdi)
+movdqa %xmm7,976(%rdi)
+
+# qhasm:   *(int128 *)(c + 992) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,992(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,992(<c=%rdi)
+movdqa %xmm3,992(%rdi)
+
+# qhasm:   *(int128 *)(c + 1008) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,1008(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,1008(<c=%rdi)
+movdqa %xmm5,1008(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm7 ^= ONE
+# asm 1: pxor  ONE,<xmm7=int6464#8
+# asm 2: pxor  ONE,<xmm7=%xmm7
+pxor  ONE,%xmm7
+
+# qhasm:   xmm3 ^= ONE
+# asm 1: pxor  ONE,<xmm3=int6464#4
+# asm 2: pxor  ONE,<xmm3=%xmm3
+pxor  ONE,%xmm3
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:       xmm7 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm7=int6464#8
+# asm 2: pxor  <xmm3=%xmm3,<xmm7=%xmm7
+pxor  %xmm3,%xmm7
+
+# qhasm:       xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm6=int6464#7
+# asm 2: pxor  <xmm1=%xmm1,<xmm6=%xmm6
+pxor  %xmm1,%xmm6
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm3=int6464#4
+# asm 2: pxor  <xmm6=%xmm6,<xmm3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:       xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm4=int6464#5
+# asm 2: pxor  <xmm0=%xmm0,<xmm4=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm5 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm5=int6464#6
+# asm 2: pxor  <xmm7=%xmm7,<xmm5=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm1,<xmm4=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm11 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm11=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm11=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm9=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm9=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm13=int6464#12
+# asm 2: movdqa <xmm6=%xmm6,>xmm13=%xmm11
+movdqa %xmm6,%xmm11
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm12
+movdqa %xmm3,%xmm12
+
+# qhasm:       xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:       xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:       xmm9 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm9=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm9=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:       xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#11
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm10
+movdqa %xmm4,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm13=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm13=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#12
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm11
+movdqa %xmm7,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#12
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm11
+pxor  %xmm3,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#11
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm10
+movdqa %xmm6,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: movdqa <xmm5=%xmm5,>xmm15=%xmm15
+movdqa %xmm5,%xmm15
+
+# qhasm:       xmm12 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm12=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm12=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm14=int6464#14
+# asm 2: pand  <xmm7=%xmm7,<xmm14=%xmm13
+pand  %xmm7,%xmm13
+
+# qhasm:       xmm15 |= xmm3
+# asm 1: por   <xmm3=int6464#4,<xmm15=int6464#16
+# asm 2: por   <xmm3=%xmm3,<xmm15=%xmm15
+por   %xmm3,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#9
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm8
+movdqa %xmm3,%xmm8
+
+# qhasm:         xmm8 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm8=int6464#10
+# asm 2: movdqa <xmm7=%xmm7,>xmm8=%xmm9
+movdqa %xmm7,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm10=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm10=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:           xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:           xmm3 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm3=int6464#4
+# asm 2: pand  <xmm14=%xmm11,<xmm3=%xmm3
+pand  %xmm11,%xmm3
+
+# qhasm:           xmm7 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm7=int6464#8
+# asm 2: pand  <xmm15=%xmm13,<xmm7=%xmm7
+pand  %xmm13,%xmm7
+
+# qhasm:           xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:           xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm0=int6464#1
+# asm 2: pxor  <xmm4=%xmm4,<xmm0=%xmm0
+pxor  %xmm4,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm4 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm4=int6464#5
+# asm 2: pand  <xmm13=%xmm15,<xmm4=%xmm4
+pand  %xmm15,%xmm4
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm0=int6464#1
+# asm 2: pxor  <xmm4=%xmm4,<xmm0=%xmm0
+pxor  %xmm4,%xmm0
+
+# qhasm:           xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:         xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm9,<xmm7=%xmm7
+pxor  %xmm9,%xmm7
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm9,<xmm4=%xmm4
+pxor  %xmm9,%xmm4
+
+# qhasm:         xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:         xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm10=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:           xmm2 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm9=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm6=int6464#7
+# asm 2: pand  <xmm13=%xmm15,<xmm6=%xmm6
+pand  %xmm15,%xmm6
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:           xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm11=int6464#11
+# asm 2: pand  <xmm5=%xmm5,<xmm11=%xmm10
+pand  %xmm5,%xmm10
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm5 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm5=int6464#6
+# asm 2: pand  <xmm14=%xmm11,<xmm5=%xmm5
+pand  %xmm11,%xmm5
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm12=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:         xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm9,<xmm6=%xmm6
+pxor  %xmm9,%xmm6
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm1 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm1=int6464#2
+# asm 2: pxor  <xmm3=%xmm3,<xmm1=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm2=int6464#3
+# asm 2: pxor  <xmm5=%xmm5,<xmm2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm7 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm6=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm6=int6464#7
+# asm 2: pxor  <xmm4=%xmm4,<xmm6=%xmm6
+pxor  %xmm4,%xmm6
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:   xmm7 ^= RCON
+# asm 1: pxor  RCON,<xmm7=int6464#8
+# asm 2: pxor  RCON,<xmm7=%xmm7
+pxor  RCON,%xmm7
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   xmm8 = *(int128 *)(c + 896)
+# asm 1: movdqa 896(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 896(<c=%rdi),>xmm8=%xmm8
+movdqa 896(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 912)
+# asm 1: movdqa 912(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 912(<c=%rdi),>xmm9=%xmm9
+movdqa 912(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 928)
+# asm 1: movdqa 928(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 928(<c=%rdi),>xmm10=%xmm10
+movdqa 928(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 944)
+# asm 1: movdqa 944(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 944(<c=%rdi),>xmm11=%xmm11
+movdqa 944(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 960)
+# asm 1: movdqa 960(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 960(<c=%rdi),>xmm12=%xmm12
+movdqa 960(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 976)
+# asm 1: movdqa 976(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 976(<c=%rdi),>xmm13=%xmm13
+movdqa 976(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 992)
+# asm 1: movdqa 992(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 992(<c=%rdi),>xmm14=%xmm14
+movdqa 992(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 1008)
+# asm 1: movdqa 1008(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 1008(<c=%rdi),>xmm15=%xmm15
+movdqa 1008(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:   xmm3 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm3=int6464#4
+# asm 2: pxor  <xmm11=%xmm11,<xmm3=%xmm3
+pxor  %xmm11,%xmm3
+
+# qhasm:   xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm6 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm6=int6464#7
+# asm 2: pxor  <xmm14=%xmm14,<xmm6=%xmm6
+pxor  %xmm14,%xmm6
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   *(int128 *)(c + 1024) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,1024(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,1024(<c=%rdi)
+movdqa %xmm0,1024(%rdi)
+
+# qhasm:   *(int128 *)(c + 1040) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,1040(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,1040(<c=%rdi)
+movdqa %xmm1,1040(%rdi)
+
+# qhasm:   *(int128 *)(c + 1056) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,1056(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,1056(<c=%rdi)
+movdqa %xmm2,1056(%rdi)
+
+# qhasm:   *(int128 *)(c + 1072) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,1072(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,1072(<c=%rdi)
+movdqa %xmm3,1072(%rdi)
+
+# qhasm:   *(int128 *)(c + 1088) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,1088(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,1088(<c=%rdi)
+movdqa %xmm4,1088(%rdi)
+
+# qhasm:   *(int128 *)(c + 1104) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,1104(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,1104(<c=%rdi)
+movdqa %xmm5,1104(%rdi)
+
+# qhasm:   *(int128 *)(c + 1120) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,1120(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,1120(<c=%rdi)
+movdqa %xmm6,1120(%rdi)
+
+# qhasm:   *(int128 *)(c + 1136) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,1136(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,1136(<c=%rdi)
+movdqa %xmm7,1136(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm5 ^= ONE
+# asm 1: pxor  ONE,<xmm5=int6464#6
+# asm 2: pxor  ONE,<xmm5=%xmm5
+pxor  ONE,%xmm5
+
+# qhasm:   xmm6 ^= ONE
+# asm 1: pxor  ONE,<xmm6=int6464#7
+# asm 2: pxor  ONE,<xmm6=%xmm6
+pxor  ONE,%xmm6
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:   xmm0 ^= RCON
+# asm 1: pxor  RCON,<xmm0=int6464#1
+# asm 2: pxor  RCON,<xmm0=%xmm0
+pxor  RCON,%xmm0
+
+# qhasm:   xmm1 ^= RCON
+# asm 1: pxor  RCON,<xmm1=int6464#2
+# asm 2: pxor  RCON,<xmm1=%xmm1
+pxor  RCON,%xmm1
+
+# qhasm:   xmm6 ^= RCON
+# asm 1: pxor  RCON,<xmm6=int6464#7
+# asm 2: pxor  RCON,<xmm6=%xmm6
+pxor  RCON,%xmm6
+
+# qhasm:   xmm3 ^= RCON
+# asm 1: pxor  RCON,<xmm3=int6464#4
+# asm 2: pxor  RCON,<xmm3=%xmm3
+pxor  RCON,%xmm3
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   xmm8 = *(int128 *)(c + 1024)
+# asm 1: movdqa 1024(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 1024(<c=%rdi),>xmm8=%xmm8
+movdqa 1024(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 1040)
+# asm 1: movdqa 1040(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 1040(<c=%rdi),>xmm9=%xmm9
+movdqa 1040(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 1056)
+# asm 1: movdqa 1056(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 1056(<c=%rdi),>xmm10=%xmm10
+movdqa 1056(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 1072)
+# asm 1: movdqa 1072(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 1072(<c=%rdi),>xmm11=%xmm11
+movdqa 1072(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 1088)
+# asm 1: movdqa 1088(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 1088(<c=%rdi),>xmm12=%xmm12
+movdqa 1088(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 1104)
+# asm 1: movdqa 1104(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 1104(<c=%rdi),>xmm13=%xmm13
+movdqa 1104(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 1120)
+# asm 1: movdqa 1120(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 1120(<c=%rdi),>xmm14=%xmm14
+movdqa 1120(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 1136)
+# asm 1: movdqa 1136(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 1136(<c=%rdi),>xmm15=%xmm15
+movdqa 1136(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:   xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:   xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:   xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:   xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:   xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:   *(int128 *)(c + 1152) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,1152(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,1152(<c=%rdi)
+movdqa %xmm0,1152(%rdi)
+
+# qhasm:   *(int128 *)(c + 1168) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,1168(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,1168(<c=%rdi)
+movdqa %xmm1,1168(%rdi)
+
+# qhasm:   *(int128 *)(c + 1184) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,1184(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,1184(<c=%rdi)
+movdqa %xmm4,1184(%rdi)
+
+# qhasm:   *(int128 *)(c + 1200) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,1200(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,1200(<c=%rdi)
+movdqa %xmm6,1200(%rdi)
+
+# qhasm:   *(int128 *)(c + 1216) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,1216(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,1216(<c=%rdi)
+movdqa %xmm3,1216(%rdi)
+
+# qhasm:   *(int128 *)(c + 1232) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,1232(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,1232(<c=%rdi)
+movdqa %xmm7,1232(%rdi)
+
+# qhasm:   *(int128 *)(c + 1248) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,1248(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,1248(<c=%rdi)
+movdqa %xmm2,1248(%rdi)
+
+# qhasm:   *(int128 *)(c + 1264) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,1264(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,1264(<c=%rdi)
+movdqa %xmm5,1264(%rdi)
+
+# qhasm:   xmm0 ^= ONE
+# asm 1: pxor  ONE,<xmm0=int6464#1
+# asm 2: pxor  ONE,<xmm0=%xmm0
+pxor  ONE,%xmm0
+
+# qhasm:   xmm1 ^= ONE
+# asm 1: pxor  ONE,<xmm1=int6464#2
+# asm 2: pxor  ONE,<xmm1=%xmm1
+pxor  ONE,%xmm1
+
+# qhasm:   xmm7 ^= ONE
+# asm 1: pxor  ONE,<xmm7=int6464#8
+# asm 2: pxor  ONE,<xmm7=%xmm7
+pxor  ONE,%xmm7
+
+# qhasm:   xmm2 ^= ONE
+# asm 1: pxor  ONE,<xmm2=int6464#3
+# asm 2: pxor  ONE,<xmm2=%xmm2
+pxor  ONE,%xmm2
+
+# qhasm:     shuffle bytes of xmm0 by ROTB
+# asm 1: pshufb ROTB,<xmm0=int6464#1
+# asm 2: pshufb ROTB,<xmm0=%xmm0
+pshufb ROTB,%xmm0
+
+# qhasm:     shuffle bytes of xmm1 by ROTB
+# asm 1: pshufb ROTB,<xmm1=int6464#2
+# asm 2: pshufb ROTB,<xmm1=%xmm1
+pshufb ROTB,%xmm1
+
+# qhasm:     shuffle bytes of xmm4 by ROTB
+# asm 1: pshufb ROTB,<xmm4=int6464#5
+# asm 2: pshufb ROTB,<xmm4=%xmm4
+pshufb ROTB,%xmm4
+
+# qhasm:     shuffle bytes of xmm6 by ROTB
+# asm 1: pshufb ROTB,<xmm6=int6464#7
+# asm 2: pshufb ROTB,<xmm6=%xmm6
+pshufb ROTB,%xmm6
+
+# qhasm:     shuffle bytes of xmm3 by ROTB
+# asm 1: pshufb ROTB,<xmm3=int6464#4
+# asm 2: pshufb ROTB,<xmm3=%xmm3
+pshufb ROTB,%xmm3
+
+# qhasm:     shuffle bytes of xmm7 by ROTB
+# asm 1: pshufb ROTB,<xmm7=int6464#8
+# asm 2: pshufb ROTB,<xmm7=%xmm7
+pshufb ROTB,%xmm7
+
+# qhasm:     shuffle bytes of xmm2 by ROTB
+# asm 1: pshufb ROTB,<xmm2=int6464#3
+# asm 2: pshufb ROTB,<xmm2=%xmm2
+pshufb ROTB,%xmm2
+
+# qhasm:     shuffle bytes of xmm5 by ROTB
+# asm 1: pshufb ROTB,<xmm5=int6464#6
+# asm 2: pshufb ROTB,<xmm5=%xmm5
+pshufb ROTB,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm7=int6464#8
+# asm 2: pxor  <xmm2=%xmm2,<xmm7=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm1,<xmm4=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm2 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm4=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:       xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm5 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm5=int6464#6
+# asm 2: pxor  <xmm7=%xmm7,<xmm5=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:       xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm6=int6464#7
+# asm 2: pxor  <xmm1=%xmm1,<xmm6=%xmm6
+pxor  %xmm1,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm11 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm11=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm11=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm9=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm9=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#13
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm:       xmm11 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm11=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm11=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:       xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:       xmm9 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm9=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm9=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:       xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#12
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm11
+pxor  %xmm3,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#11
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm10
+movdqa %xmm6,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm13=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm13=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#12
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm11
+movdqa %xmm7,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm12=int6464#11
+# asm 2: movdqa <xmm4=%xmm4,>xmm12=%xmm10
+movdqa %xmm4,%xmm10
+
+# qhasm:       xmm13 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm13=int6464#12
+# asm 2: movdqa <xmm3=%xmm3,>xmm13=%xmm11
+movdqa %xmm3,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: movdqa <xmm5=%xmm5,>xmm15=%xmm15
+movdqa %xmm5,%xmm15
+
+# qhasm:       xmm12 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm12=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm12=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm14=int6464#14
+# asm 2: pand  <xmm7=%xmm7,<xmm14=%xmm13
+pand  %xmm7,%xmm13
+
+# qhasm:       xmm15 |= xmm2
+# asm 1: por   <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: por   <xmm2=%xmm2,<xmm15=%xmm15
+por   %xmm2,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:         xmm8 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm8=int6464#10
+# asm 2: movdqa <xmm7=%xmm7,>xmm8=%xmm9
+movdqa %xmm7,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm2
+# asm 1: pand  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pand  <xmm2=%xmm2,<xmm10=%xmm10
+pand  %xmm2,%xmm10
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm11,<xmm2=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:           xmm7 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm7=int6464#8
+# asm 2: pand  <xmm15=%xmm13,<xmm7=%xmm7
+pand  %xmm13,%xmm7
+
+# qhasm:           xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:           xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#10
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm9
+pxor  %xmm6,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm6=int6464#7
+# asm 2: pand  <xmm13=%xmm15,<xmm6=%xmm6
+pand  %xmm15,%xmm6
+
+# qhasm:           xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm0=int6464#1
+# asm 2: pxor  <xmm6=%xmm6,<xmm0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:           xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:         xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm9,<xmm7=%xmm7
+pxor  %xmm9,%xmm7
+
+# qhasm:         xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm9,<xmm6=%xmm6
+pxor  %xmm9,%xmm6
+
+# qhasm:         xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#10
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm9
+pxor  %xmm4,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm10=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm10=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm3=int6464#4
+# asm 2: pand  <xmm9=%xmm12,<xmm3=%xmm3
+pand  %xmm12,%xmm3
+
+# qhasm:           xmm4 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm4=int6464#5
+# asm 2: pand  <xmm13=%xmm15,<xmm4=%xmm4
+pand  %xmm15,%xmm4
+
+# qhasm:           xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm11=int6464#11
+# asm 2: pand  <xmm5=%xmm5,<xmm11=%xmm10
+pand  %xmm5,%xmm10
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm5 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm5=int6464#6
+# asm 2: pand  <xmm14=%xmm11,<xmm5=%xmm5
+pand  %xmm11,%xmm5
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm5 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm5=int6464#6
+# asm 2: pxor  <xmm1=%xmm1,<xmm5=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm12=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:         xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm9,<xmm4=%xmm4
+pxor  %xmm9,%xmm4
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm1=int6464#2
+# asm 2: pxor  <xmm2=%xmm2,<xmm1=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm2 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm2=int6464#3
+# asm 2: pxor  <xmm0=%xmm0,<xmm2=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#2
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm1
+pxor  %xmm7,%xmm1
+
+# qhasm:       xmm7 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pxor  <xmm4=%xmm4,<xmm7=%xmm7
+pxor  %xmm4,%xmm7
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm4 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm4=int6464#5
+# asm 2: pxor  <xmm6=%xmm6,<xmm4=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:       xmm6 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm6=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm6=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm6,<xmm2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm:   xmm1 ^= RCON
+# asm 1: pxor  RCON,<xmm1=int6464#2
+# asm 2: pxor  RCON,<xmm1=%xmm1
+pxor  RCON,%xmm1
+
+# qhasm:   xmm3 ^= RCON
+# asm 1: pxor  RCON,<xmm3=int6464#4
+# asm 2: pxor  RCON,<xmm3=%xmm3
+pxor  RCON,%xmm3
+
+# qhasm:   xmm6 ^= RCON
+# asm 1: pxor  RCON,<xmm6=int6464#7
+# asm 2: pxor  RCON,<xmm6=%xmm6
+pxor  RCON,%xmm6
+
+# qhasm:   xmm5 ^= RCON
+# asm 1: pxor  RCON,<xmm5=int6464#6
+# asm 2: pxor  RCON,<xmm5=%xmm5
+pxor  RCON,%xmm5
+
+# qhasm:   shuffle bytes of xmm0 by EXPB0
+# asm 1: pshufb EXPB0,<xmm0=int6464#1
+# asm 2: pshufb EXPB0,<xmm0=%xmm0
+pshufb EXPB0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by EXPB0
+# asm 1: pshufb EXPB0,<xmm1=int6464#2
+# asm 2: pshufb EXPB0,<xmm1=%xmm1
+pshufb EXPB0,%xmm1
+
+# qhasm:   shuffle bytes of xmm3 by EXPB0
+# asm 1: pshufb EXPB0,<xmm3=int6464#4
+# asm 2: pshufb EXPB0,<xmm3=%xmm3
+pshufb EXPB0,%xmm3
+
+# qhasm:   shuffle bytes of xmm2 by EXPB0
+# asm 1: pshufb EXPB0,<xmm2=int6464#3
+# asm 2: pshufb EXPB0,<xmm2=%xmm2
+pshufb EXPB0,%xmm2
+
+# qhasm:   shuffle bytes of xmm6 by EXPB0
+# asm 1: pshufb EXPB0,<xmm6=int6464#7
+# asm 2: pshufb EXPB0,<xmm6=%xmm6
+pshufb EXPB0,%xmm6
+
+# qhasm:   shuffle bytes of xmm5 by EXPB0
+# asm 1: pshufb EXPB0,<xmm5=int6464#6
+# asm 2: pshufb EXPB0,<xmm5=%xmm5
+pshufb EXPB0,%xmm5
+
+# qhasm:   shuffle bytes of xmm4 by EXPB0
+# asm 1: pshufb EXPB0,<xmm4=int6464#5
+# asm 2: pshufb EXPB0,<xmm4=%xmm4
+pshufb EXPB0,%xmm4
+
+# qhasm:   shuffle bytes of xmm7 by EXPB0
+# asm 1: pshufb EXPB0,<xmm7=int6464#8
+# asm 2: pshufb EXPB0,<xmm7=%xmm7
+pshufb EXPB0,%xmm7
+
+# qhasm:   xmm8 = *(int128 *)(c + 1152)
+# asm 1: movdqa 1152(<c=int64#1),>xmm8=int6464#9
+# asm 2: movdqa 1152(<c=%rdi),>xmm8=%xmm8
+movdqa 1152(%rdi),%xmm8
+
+# qhasm:   xmm9 = *(int128 *)(c + 1168)
+# asm 1: movdqa 1168(<c=int64#1),>xmm9=int6464#10
+# asm 2: movdqa 1168(<c=%rdi),>xmm9=%xmm9
+movdqa 1168(%rdi),%xmm9
+
+# qhasm:   xmm10 = *(int128 *)(c + 1184)
+# asm 1: movdqa 1184(<c=int64#1),>xmm10=int6464#11
+# asm 2: movdqa 1184(<c=%rdi),>xmm10=%xmm10
+movdqa 1184(%rdi),%xmm10
+
+# qhasm:   xmm11 = *(int128 *)(c + 1200)
+# asm 1: movdqa 1200(<c=int64#1),>xmm11=int6464#12
+# asm 2: movdqa 1200(<c=%rdi),>xmm11=%xmm11
+movdqa 1200(%rdi),%xmm11
+
+# qhasm:   xmm12 = *(int128 *)(c + 1216)
+# asm 1: movdqa 1216(<c=int64#1),>xmm12=int6464#13
+# asm 2: movdqa 1216(<c=%rdi),>xmm12=%xmm12
+movdqa 1216(%rdi),%xmm12
+
+# qhasm:   xmm13 = *(int128 *)(c + 1232)
+# asm 1: movdqa 1232(<c=int64#1),>xmm13=int6464#14
+# asm 2: movdqa 1232(<c=%rdi),>xmm13=%xmm13
+movdqa 1232(%rdi),%xmm13
+
+# qhasm:   xmm14 = *(int128 *)(c + 1248)
+# asm 1: movdqa 1248(<c=int64#1),>xmm14=int6464#15
+# asm 2: movdqa 1248(<c=%rdi),>xmm14=%xmm14
+movdqa 1248(%rdi),%xmm14
+
+# qhasm:   xmm15 = *(int128 *)(c + 1264)
+# asm 1: movdqa 1264(<c=int64#1),>xmm15=int6464#16
+# asm 2: movdqa 1264(<c=%rdi),>xmm15=%xmm15
+movdqa 1264(%rdi),%xmm15
+
+# qhasm:   xmm8 ^= ONE
+# asm 1: pxor  ONE,<xmm8=int6464#9
+# asm 2: pxor  ONE,<xmm8=%xmm8
+pxor  ONE,%xmm8
+
+# qhasm:   xmm9 ^= ONE
+# asm 1: pxor  ONE,<xmm9=int6464#10
+# asm 2: pxor  ONE,<xmm9=%xmm9
+pxor  ONE,%xmm9
+
+# qhasm:   xmm13 ^= ONE
+# asm 1: pxor  ONE,<xmm13=int6464#14
+# asm 2: pxor  ONE,<xmm13=%xmm13
+pxor  ONE,%xmm13
+
+# qhasm:   xmm14 ^= ONE
+# asm 1: pxor  ONE,<xmm14=int6464#15
+# asm 2: pxor  ONE,<xmm14=%xmm14
+pxor  ONE,%xmm14
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   uint32323232 xmm8 >>= 8
+# asm 1: psrld $8,<xmm8=int6464#9
+# asm 2: psrld $8,<xmm8=%xmm8
+psrld $8,%xmm8
+
+# qhasm:   uint32323232 xmm9 >>= 8
+# asm 1: psrld $8,<xmm9=int6464#10
+# asm 2: psrld $8,<xmm9=%xmm9
+psrld $8,%xmm9
+
+# qhasm:   uint32323232 xmm10 >>= 8
+# asm 1: psrld $8,<xmm10=int6464#11
+# asm 2: psrld $8,<xmm10=%xmm10
+psrld $8,%xmm10
+
+# qhasm:   uint32323232 xmm11 >>= 8
+# asm 1: psrld $8,<xmm11=int6464#12
+# asm 2: psrld $8,<xmm11=%xmm11
+psrld $8,%xmm11
+
+# qhasm:   uint32323232 xmm12 >>= 8
+# asm 1: psrld $8,<xmm12=int6464#13
+# asm 2: psrld $8,<xmm12=%xmm12
+psrld $8,%xmm12
+
+# qhasm:   uint32323232 xmm13 >>= 8
+# asm 1: psrld $8,<xmm13=int6464#14
+# asm 2: psrld $8,<xmm13=%xmm13
+psrld $8,%xmm13
+
+# qhasm:   uint32323232 xmm14 >>= 8
+# asm 1: psrld $8,<xmm14=int6464#15
+# asm 2: psrld $8,<xmm14=%xmm14
+psrld $8,%xmm14
+
+# qhasm:   uint32323232 xmm15 >>= 8
+# asm 1: psrld $8,<xmm15=int6464#16
+# asm 2: psrld $8,<xmm15=%xmm15
+psrld $8,%xmm15
+
+# qhasm:   xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:   xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:   xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:   xmm2 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm2=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm2=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:   xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm12,<xmm6=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:   xmm5 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm5=int6464#6
+# asm 2: pxor  <xmm13=%xmm13,<xmm5=%xmm5
+pxor  %xmm13,%xmm5
+
+# qhasm:   xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:   xmm7 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm7=int6464#8
+# asm 2: pxor  <xmm15=%xmm15,<xmm7=%xmm7
+pxor  %xmm15,%xmm7
+
+# qhasm:   shuffle bytes of xmm0 by M0
+# asm 1: pshufb M0,<xmm0=int6464#1
+# asm 2: pshufb M0,<xmm0=%xmm0
+pshufb M0,%xmm0
+
+# qhasm:   shuffle bytes of xmm1 by M0
+# asm 1: pshufb M0,<xmm1=int6464#2
+# asm 2: pshufb M0,<xmm1=%xmm1
+pshufb M0,%xmm1
+
+# qhasm:   shuffle bytes of xmm4 by M0
+# asm 1: pshufb M0,<xmm4=int6464#5
+# asm 2: pshufb M0,<xmm4=%xmm4
+pshufb M0,%xmm4
+
+# qhasm:   shuffle bytes of xmm6 by M0
+# asm 1: pshufb M0,<xmm6=int6464#7
+# asm 2: pshufb M0,<xmm6=%xmm6
+pshufb M0,%xmm6
+
+# qhasm:   shuffle bytes of xmm3 by M0
+# asm 1: pshufb M0,<xmm3=int6464#4
+# asm 2: pshufb M0,<xmm3=%xmm3
+pshufb M0,%xmm3
+
+# qhasm:   shuffle bytes of xmm7 by M0
+# asm 1: pshufb M0,<xmm7=int6464#8
+# asm 2: pshufb M0,<xmm7=%xmm7
+pshufb M0,%xmm7
+
+# qhasm:   shuffle bytes of xmm2 by M0
+# asm 1: pshufb M0,<xmm2=int6464#3
+# asm 2: pshufb M0,<xmm2=%xmm2
+pshufb M0,%xmm2
+
+# qhasm:   shuffle bytes of xmm5 by M0
+# asm 1: pshufb M0,<xmm5=int6464#6
+# asm 2: pshufb M0,<xmm5=%xmm5
+pshufb M0,%xmm5
+
+# qhasm:   *(int128 *)(c + 1280) = xmm0
+# asm 1: movdqa <xmm0=int6464#1,1280(<c=int64#1)
+# asm 2: movdqa <xmm0=%xmm0,1280(<c=%rdi)
+movdqa %xmm0,1280(%rdi)
+
+# qhasm:   *(int128 *)(c + 1296) = xmm1
+# asm 1: movdqa <xmm1=int6464#2,1296(<c=int64#1)
+# asm 2: movdqa <xmm1=%xmm1,1296(<c=%rdi)
+movdqa %xmm1,1296(%rdi)
+
+# qhasm:   *(int128 *)(c + 1312) = xmm3
+# asm 1: movdqa <xmm3=int6464#4,1312(<c=int64#1)
+# asm 2: movdqa <xmm3=%xmm3,1312(<c=%rdi)
+movdqa %xmm3,1312(%rdi)
+
+# qhasm:   *(int128 *)(c + 1328) = xmm2
+# asm 1: movdqa <xmm2=int6464#3,1328(<c=int64#1)
+# asm 2: movdqa <xmm2=%xmm2,1328(<c=%rdi)
+movdqa %xmm2,1328(%rdi)
+
+# qhasm:   *(int128 *)(c + 1344) = xmm6
+# asm 1: movdqa <xmm6=int6464#7,1344(<c=int64#1)
+# asm 2: movdqa <xmm6=%xmm6,1344(<c=%rdi)
+movdqa %xmm6,1344(%rdi)
+
+# qhasm:   *(int128 *)(c + 1360) = xmm5
+# asm 1: movdqa <xmm5=int6464#6,1360(<c=int64#1)
+# asm 2: movdqa <xmm5=%xmm5,1360(<c=%rdi)
+movdqa %xmm5,1360(%rdi)
+
+# qhasm:   *(int128 *)(c + 1376) = xmm4
+# asm 1: movdqa <xmm4=int6464#5,1376(<c=int64#1)
+# asm 2: movdqa <xmm4=%xmm4,1376(<c=%rdi)
+movdqa %xmm4,1376(%rdi)
+
+# qhasm:   *(int128 *)(c + 1392) = xmm7
+# asm 1: movdqa <xmm7=int6464#8,1392(<c=int64#1)
+# asm 2: movdqa <xmm7=%xmm7,1392(<c=%rdi)
+movdqa %xmm7,1392(%rdi)
+
+# qhasm: leave
+add %r11,%rsp
+mov %rdi,%rax
+mov %rsi,%rdx
+xor %rax,%rax
+ret
diff --git a/nacl/crypto_stream/aes128ctr/core2/stream.c b/nacl/crypto_stream/aes128ctr/core2/stream.c
new file mode 100644
index 00000000..53524a62
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/stream.c
@@ -0,0 +1,14 @@
+#include "crypto_stream.h"
+
+int crypto_stream(
+        unsigned char *out,
+        unsigned long long outlen,
+        const unsigned char *n,
+        const unsigned char *k
+        )
+{
+    unsigned char d[crypto_stream_BEFORENMBYTES];
+    crypto_stream_beforenm(d, k);
+    crypto_stream_afternm(out, outlen, n, d);
+    return 0;
+}
diff --git a/nacl/crypto_stream/aes128ctr/core2/xor.c b/nacl/crypto_stream/aes128ctr/core2/xor.c
new file mode 100644
index 00000000..825088cc
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/xor.c
@@ -0,0 +1,15 @@
+#include "crypto_stream.h"
+
+int crypto_stream_xor(
+        unsigned char *out,
+        const unsigned char *in,
+        unsigned long long inlen,
+        const unsigned char *n,
+        const unsigned char *k
+        )
+{
+    unsigned char d[crypto_stream_BEFORENMBYTES];
+    crypto_stream_beforenm(d, k);
+    crypto_stream_xor_afternm(out, in, inlen, n, d);
+    return 0;
+}
diff --git a/nacl/crypto_stream/aes128ctr/core2/xor_afternm.s b/nacl/crypto_stream/aes128ctr/core2/xor_afternm.s
new file mode 100644
index 00000000..022691a2
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/core2/xor_afternm.s
@@ -0,0 +1,12407 @@
+# Author: Emilia Käsper and Peter Schwabe
+# Date: 2009-03-19
+# +2010.01.31: minor namespace modifications
+# Public domain
+
+.data
+.p2align 6
+
+RCON: .int 0x00000000, 0x00000000, 0x00000000, 0xffffffff
+ROTB: .int 0x0c000000, 0x00000000, 0x04000000, 0x08000000
+EXPB0: .int 0x03030303, 0x07070707, 0x0b0b0b0b, 0x0f0f0f0f
+CTRINC1: .int 0x00000001, 0x00000000, 0x00000000, 0x00000000
+CTRINC2: .int 0x00000002, 0x00000000, 0x00000000, 0x00000000
+CTRINC3: .int 0x00000003, 0x00000000, 0x00000000, 0x00000000
+CTRINC4: .int 0x00000004, 0x00000000, 0x00000000, 0x00000000
+CTRINC5: .int 0x00000005, 0x00000000, 0x00000000, 0x00000000
+CTRINC6: .int 0x00000006, 0x00000000, 0x00000000, 0x00000000
+CTRINC7: .int 0x00000007, 0x00000000, 0x00000000, 0x00000000
+RCTRINC1: .int 0x00000000, 0x00000000, 0x00000000, 0x00000001
+RCTRINC2: .int 0x00000000, 0x00000000, 0x00000000, 0x00000002
+RCTRINC3: .int 0x00000000, 0x00000000, 0x00000000, 0x00000003
+RCTRINC4: .int 0x00000000, 0x00000000, 0x00000000, 0x00000004
+RCTRINC5: .int 0x00000000, 0x00000000, 0x00000000, 0x00000005
+RCTRINC6: .int 0x00000000, 0x00000000, 0x00000000, 0x00000006
+RCTRINC7: .int 0x00000000, 0x00000000, 0x00000000, 0x00000007
+
+SWAP32: .int 0x00010203, 0x04050607, 0x08090a0b, 0x0c0d0e0f
+M0SWAP: .quad 0x0105090d0004080c , 0x03070b0f02060a0e
+
+BS0: .quad 0x5555555555555555, 0x5555555555555555
+BS1: .quad 0x3333333333333333, 0x3333333333333333
+BS2: .quad 0x0f0f0f0f0f0f0f0f, 0x0f0f0f0f0f0f0f0f
+ONE: .quad 0xffffffffffffffff, 0xffffffffffffffff
+M0:  .quad 0x02060a0e03070b0f, 0x0004080c0105090d
+SRM0:   .quad 0x0304090e00050a0f, 0x01060b0c0207080d
+SR: .quad 0x0504070600030201, 0x0f0e0d0c0a09080b
+
+# qhasm: int64 outp
+
+# qhasm: int64 inp
+
+# qhasm: int64 len
+
+# qhasm: int64 np
+
+# qhasm: int64 c
+
+# qhasm: input outp
+
+# qhasm: input inp
+
+# qhasm: input len
+
+# qhasm: input np
+
+# qhasm: input c
+
+# qhasm: int64 lensav
+
+# qhasm: int64 tmp
+
+# qhasm: int6464 xmm0
+
+# qhasm: int6464 xmm1
+
+# qhasm: int6464 xmm2
+
+# qhasm: int6464 xmm3
+
+# qhasm: int6464 xmm4
+
+# qhasm: int6464 xmm5
+
+# qhasm: int6464 xmm6
+
+# qhasm: int6464 xmm7
+
+# qhasm: int6464 xmm8
+
+# qhasm: int6464 xmm9
+
+# qhasm: int6464 xmm10
+
+# qhasm: int6464 xmm11
+
+# qhasm: int6464 xmm12
+
+# qhasm: int6464 xmm13
+
+# qhasm: int6464 xmm14
+
+# qhasm: int6464 xmm15
+
+# qhasm: int6464 t
+
+# qhasm: stack1024 bl
+
+# qhasm: stack128 nonce_stack
+
+# qhasm: int64 blp
+
+# qhasm: int64 b
+
+# qhasm: enter crypto_stream_aes128ctr_core2_xor_afternm
+.text
+.p2align 5
+.globl _crypto_stream_aes128ctr_core2_xor_afternm
+.globl crypto_stream_aes128ctr_core2_xor_afternm
+_crypto_stream_aes128ctr_core2_xor_afternm:
+crypto_stream_aes128ctr_core2_xor_afternm:
+mov %rsp,%r11
+and $31,%r11
+add $160,%r11
+sub %r11,%rsp
+
+# qhasm: xmm0 = *(int128 *) (np + 0)
+# asm 1: movdqa 0(<np=int64#4),>xmm0=int6464#1
+# asm 2: movdqa 0(<np=%rcx),>xmm0=%xmm0
+movdqa 0(%rcx),%xmm0
+
+# qhasm: nonce_stack = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>nonce_stack=stack128#1
+# asm 2: movdqa <xmm0=%xmm0,>nonce_stack=0(%rsp)
+movdqa %xmm0,0(%rsp)
+
+# qhasm: np = &nonce_stack
+# asm 1: leaq <nonce_stack=stack128#1,>np=int64#4
+# asm 2: leaq <nonce_stack=0(%rsp),>np=%rcx
+leaq 0(%rsp),%rcx
+
+# qhasm: enc_block:
+._enc_block:
+
+# qhasm: xmm0 = *(int128 *) (np + 0)
+# asm 1: movdqa 0(<np=int64#4),>xmm0=int6464#1
+# asm 2: movdqa 0(<np=%rcx),>xmm0=%xmm0
+movdqa 0(%rcx),%xmm0
+
+# qhasm: xmm1 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm1=int6464#2
+# asm 2: movdqa <xmm0=%xmm0,>xmm1=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: shuffle bytes of xmm1 by SWAP32
+# asm 1: pshufb SWAP32,<xmm1=int6464#2
+# asm 2: pshufb SWAP32,<xmm1=%xmm1
+pshufb SWAP32,%xmm1
+
+# qhasm: xmm2 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm2=int6464#3
+# asm 2: movdqa <xmm1=%xmm1,>xmm2=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: xmm3 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm3=int6464#4
+# asm 2: movdqa <xmm1=%xmm1,>xmm3=%xmm3
+movdqa %xmm1,%xmm3
+
+# qhasm: xmm4 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm4=int6464#5
+# asm 2: movdqa <xmm1=%xmm1,>xmm4=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm5=int6464#6
+# asm 2: movdqa <xmm1=%xmm1,>xmm5=%xmm5
+movdqa %xmm1,%xmm5
+
+# qhasm: xmm6 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm6=int6464#7
+# asm 2: movdqa <xmm1=%xmm1,>xmm6=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: xmm7 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm7=int6464#8
+# asm 2: movdqa <xmm1=%xmm1,>xmm7=%xmm7
+movdqa %xmm1,%xmm7
+
+# qhasm: int32323232 xmm1 += RCTRINC1
+# asm 1: paddd  RCTRINC1,<xmm1=int6464#2
+# asm 2: paddd  RCTRINC1,<xmm1=%xmm1
+paddd  RCTRINC1,%xmm1
+
+# qhasm: int32323232 xmm2 += RCTRINC2
+# asm 1: paddd  RCTRINC2,<xmm2=int6464#3
+# asm 2: paddd  RCTRINC2,<xmm2=%xmm2
+paddd  RCTRINC2,%xmm2
+
+# qhasm: int32323232 xmm3 += RCTRINC3
+# asm 1: paddd  RCTRINC3,<xmm3=int6464#4
+# asm 2: paddd  RCTRINC3,<xmm3=%xmm3
+paddd  RCTRINC3,%xmm3
+
+# qhasm: int32323232 xmm4 += RCTRINC4
+# asm 1: paddd  RCTRINC4,<xmm4=int6464#5
+# asm 2: paddd  RCTRINC4,<xmm4=%xmm4
+paddd  RCTRINC4,%xmm4
+
+# qhasm: int32323232 xmm5 += RCTRINC5
+# asm 1: paddd  RCTRINC5,<xmm5=int6464#6
+# asm 2: paddd  RCTRINC5,<xmm5=%xmm5
+paddd  RCTRINC5,%xmm5
+
+# qhasm: int32323232 xmm6 += RCTRINC6
+# asm 1: paddd  RCTRINC6,<xmm6=int6464#7
+# asm 2: paddd  RCTRINC6,<xmm6=%xmm6
+paddd  RCTRINC6,%xmm6
+
+# qhasm: int32323232 xmm7 += RCTRINC7
+# asm 1: paddd  RCTRINC7,<xmm7=int6464#8
+# asm 2: paddd  RCTRINC7,<xmm7=%xmm7
+paddd  RCTRINC7,%xmm7
+
+# qhasm: shuffle bytes of xmm0 by M0
+# asm 1: pshufb M0,<xmm0=int6464#1
+# asm 2: pshufb M0,<xmm0=%xmm0
+pshufb M0,%xmm0
+
+# qhasm: shuffle bytes of xmm1 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm1=int6464#2
+# asm 2: pshufb M0SWAP,<xmm1=%xmm1
+pshufb M0SWAP,%xmm1
+
+# qhasm: shuffle bytes of xmm2 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm2=int6464#3
+# asm 2: pshufb M0SWAP,<xmm2=%xmm2
+pshufb M0SWAP,%xmm2
+
+# qhasm: shuffle bytes of xmm3 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm3=int6464#4
+# asm 2: pshufb M0SWAP,<xmm3=%xmm3
+pshufb M0SWAP,%xmm3
+
+# qhasm: shuffle bytes of xmm4 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm4=int6464#5
+# asm 2: pshufb M0SWAP,<xmm4=%xmm4
+pshufb M0SWAP,%xmm4
+
+# qhasm: shuffle bytes of xmm5 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm5=int6464#6
+# asm 2: pshufb M0SWAP,<xmm5=%xmm5
+pshufb M0SWAP,%xmm5
+
+# qhasm: shuffle bytes of xmm6 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm6=int6464#7
+# asm 2: pshufb M0SWAP,<xmm6=%xmm6
+pshufb M0SWAP,%xmm6
+
+# qhasm: shuffle bytes of xmm7 by M0SWAP
+# asm 1: pshufb M0SWAP,<xmm7=int6464#8
+# asm 2: pshufb M0SWAP,<xmm7=%xmm7
+pshufb M0SWAP,%xmm7
+
+# qhasm:     xmm8 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm8=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm8=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     xmm8 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm8=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm8=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     xmm8 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm8=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm8=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 1
+# asm 1: psrlq $1,<xmm8=int6464#9
+# asm 2: psrlq $1,<xmm8=%xmm8
+psrlq $1,%xmm8
+
+# qhasm:     xmm8 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm8=int6464#9
+# asm 2: pxor  <xmm1=%xmm1,<xmm8=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm:     xmm8 &= BS0
+# asm 1: pand  BS0,<xmm8=int6464#9
+# asm 2: pand  BS0,<xmm8=%xmm8
+pand  BS0,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     uint6464 xmm8 <<= 1
+# asm 1: psllq $1,<xmm8=int6464#9
+# asm 2: psllq $1,<xmm8=%xmm8
+psllq $1,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#9
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm8
+movdqa %xmm5,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     xmm8 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm8=int6464#9
+# asm 2: movdqa <xmm4=%xmm4,>xmm8=%xmm8
+movdqa %xmm4,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#9
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm8
+pxor  %xmm3,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 2
+# asm 1: psrlq $2,<xmm8=int6464#9
+# asm 2: psrlq $2,<xmm8=%xmm8
+psrlq $2,%xmm8
+
+# qhasm:     xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#9
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:     xmm8 &= BS1
+# asm 1: pand  BS1,<xmm8=int6464#9
+# asm 2: pand  BS1,<xmm8=%xmm8
+pand  BS1,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     uint6464 xmm8 <<= 2
+# asm 1: psllq $2,<xmm8=int6464#9
+# asm 2: psllq $2,<xmm8=%xmm8
+psllq $2,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm8 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm8=int6464#9
+# asm 2: movdqa <xmm3=%xmm3,>xmm8=%xmm8
+movdqa %xmm3,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm8=int6464#9
+# asm 2: pxor  <xmm7=%xmm7,<xmm8=%xmm8
+pxor  %xmm7,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm7 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm8=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm8,<xmm3=%xmm3
+pxor  %xmm8,%xmm3
+
+# qhasm:     xmm8 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm8=int6464#9
+# asm 2: movdqa <xmm2=%xmm2,>xmm8=%xmm8
+movdqa %xmm2,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm8=int6464#9
+# asm 2: pxor  <xmm6=%xmm6,<xmm8=%xmm8
+pxor  %xmm6,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm6 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm8=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm2=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:     xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#9
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm8
+movdqa %xmm1,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm8,<xmm5=%xmm5
+pxor  %xmm8,%xmm5
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm8 = xmm0
+# asm 1: movdqa <xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: movdqa <xmm0=%xmm0,>xmm8=%xmm8
+movdqa %xmm0,%xmm8
+
+# qhasm:     uint6464 xmm8 >>= 4
+# asm 1: psrlq $4,<xmm8=int6464#9
+# asm 2: psrlq $4,<xmm8=%xmm8
+psrlq $4,%xmm8
+
+# qhasm:     xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm8=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:     xmm8 &= BS2
+# asm 1: pand  BS2,<xmm8=int6464#9
+# asm 2: pand  BS2,<xmm8=%xmm8
+pand  BS2,%xmm8
+
+# qhasm:     xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:     uint6464 xmm8 <<= 4
+# asm 1: psllq $4,<xmm8=int6464#9
+# asm 2: psllq $4,<xmm8=%xmm8
+psllq $4,%xmm8
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 0)
+# asm 1: pxor 0(<c=int64#5),<xmm0=int6464#1
+# asm 2: pxor 0(<c=%r8),<xmm0=%xmm0
+pxor 0(%r8),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 16)
+# asm 1: pxor 16(<c=int64#5),<xmm1=int6464#2
+# asm 2: pxor 16(<c=%r8),<xmm1=%xmm1
+pxor 16(%r8),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 32)
+# asm 1: pxor 32(<c=int64#5),<xmm2=int6464#3
+# asm 2: pxor 32(<c=%r8),<xmm2=%xmm2
+pxor 32(%r8),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 48)
+# asm 1: pxor 48(<c=int64#5),<xmm3=int6464#4
+# asm 2: pxor 48(<c=%r8),<xmm3=%xmm3
+pxor 48(%r8),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 64)
+# asm 1: pxor 64(<c=int64#5),<xmm4=int6464#5
+# asm 2: pxor 64(<c=%r8),<xmm4=%xmm4
+pxor 64(%r8),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 80)
+# asm 1: pxor 80(<c=int64#5),<xmm5=int6464#6
+# asm 2: pxor 80(<c=%r8),<xmm5=%xmm5
+pxor 80(%r8),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 96)
+# asm 1: pxor 96(<c=int64#5),<xmm6=int6464#7
+# asm 2: pxor 96(<c=%r8),<xmm6=%xmm6
+pxor 96(%r8),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 112)
+# asm 1: pxor 112(<c=int64#5),<xmm7=int6464#8
+# asm 2: pxor 112(<c=%r8),<xmm7=%xmm7
+pxor 112(%r8),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 128)
+# asm 1: pxor 128(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 128(<c=%r8),<xmm8=%xmm8
+pxor 128(%r8),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 144)
+# asm 1: pxor 144(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 144(<c=%r8),<xmm9=%xmm9
+pxor 144(%r8),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 160)
+# asm 1: pxor 160(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 160(<c=%r8),<xmm10=%xmm10
+pxor 160(%r8),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 176)
+# asm 1: pxor 176(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 176(<c=%r8),<xmm11=%xmm11
+pxor 176(%r8),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 192)
+# asm 1: pxor 192(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 192(<c=%r8),<xmm12=%xmm12
+pxor 192(%r8),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 208)
+# asm 1: pxor 208(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 208(<c=%r8),<xmm13=%xmm13
+pxor 208(%r8),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 224)
+# asm 1: pxor 224(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 224(<c=%r8),<xmm14=%xmm14
+pxor 224(%r8),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 240)
+# asm 1: pxor 240(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 240(<c=%r8),<xmm15=%xmm15
+pxor 240(%r8),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 256)
+# asm 1: pxor 256(<c=int64#5),<xmm0=int6464#1
+# asm 2: pxor 256(<c=%r8),<xmm0=%xmm0
+pxor 256(%r8),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 272)
+# asm 1: pxor 272(<c=int64#5),<xmm1=int6464#2
+# asm 2: pxor 272(<c=%r8),<xmm1=%xmm1
+pxor 272(%r8),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 288)
+# asm 1: pxor 288(<c=int64#5),<xmm2=int6464#3
+# asm 2: pxor 288(<c=%r8),<xmm2=%xmm2
+pxor 288(%r8),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 304)
+# asm 1: pxor 304(<c=int64#5),<xmm3=int6464#4
+# asm 2: pxor 304(<c=%r8),<xmm3=%xmm3
+pxor 304(%r8),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 320)
+# asm 1: pxor 320(<c=int64#5),<xmm4=int6464#5
+# asm 2: pxor 320(<c=%r8),<xmm4=%xmm4
+pxor 320(%r8),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 336)
+# asm 1: pxor 336(<c=int64#5),<xmm5=int6464#6
+# asm 2: pxor 336(<c=%r8),<xmm5=%xmm5
+pxor 336(%r8),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 352)
+# asm 1: pxor 352(<c=int64#5),<xmm6=int6464#7
+# asm 2: pxor 352(<c=%r8),<xmm6=%xmm6
+pxor 352(%r8),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 368)
+# asm 1: pxor 368(<c=int64#5),<xmm7=int6464#8
+# asm 2: pxor 368(<c=%r8),<xmm7=%xmm7
+pxor 368(%r8),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 384)
+# asm 1: pxor 384(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 384(<c=%r8),<xmm8=%xmm8
+pxor 384(%r8),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 400)
+# asm 1: pxor 400(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 400(<c=%r8),<xmm9=%xmm9
+pxor 400(%r8),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 416)
+# asm 1: pxor 416(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 416(<c=%r8),<xmm10=%xmm10
+pxor 416(%r8),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 432)
+# asm 1: pxor 432(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 432(<c=%r8),<xmm11=%xmm11
+pxor 432(%r8),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 448)
+# asm 1: pxor 448(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 448(<c=%r8),<xmm12=%xmm12
+pxor 448(%r8),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 464)
+# asm 1: pxor 464(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 464(<c=%r8),<xmm13=%xmm13
+pxor 464(%r8),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 480)
+# asm 1: pxor 480(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 480(<c=%r8),<xmm14=%xmm14
+pxor 480(%r8),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 496)
+# asm 1: pxor 496(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 496(<c=%r8),<xmm15=%xmm15
+pxor 496(%r8),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 512)
+# asm 1: pxor 512(<c=int64#5),<xmm0=int6464#1
+# asm 2: pxor 512(<c=%r8),<xmm0=%xmm0
+pxor 512(%r8),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 528)
+# asm 1: pxor 528(<c=int64#5),<xmm1=int6464#2
+# asm 2: pxor 528(<c=%r8),<xmm1=%xmm1
+pxor 528(%r8),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 544)
+# asm 1: pxor 544(<c=int64#5),<xmm2=int6464#3
+# asm 2: pxor 544(<c=%r8),<xmm2=%xmm2
+pxor 544(%r8),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 560)
+# asm 1: pxor 560(<c=int64#5),<xmm3=int6464#4
+# asm 2: pxor 560(<c=%r8),<xmm3=%xmm3
+pxor 560(%r8),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 576)
+# asm 1: pxor 576(<c=int64#5),<xmm4=int6464#5
+# asm 2: pxor 576(<c=%r8),<xmm4=%xmm4
+pxor 576(%r8),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 592)
+# asm 1: pxor 592(<c=int64#5),<xmm5=int6464#6
+# asm 2: pxor 592(<c=%r8),<xmm5=%xmm5
+pxor 592(%r8),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 608)
+# asm 1: pxor 608(<c=int64#5),<xmm6=int6464#7
+# asm 2: pxor 608(<c=%r8),<xmm6=%xmm6
+pxor 608(%r8),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 624)
+# asm 1: pxor 624(<c=int64#5),<xmm7=int6464#8
+# asm 2: pxor 624(<c=%r8),<xmm7=%xmm7
+pxor 624(%r8),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 640)
+# asm 1: pxor 640(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 640(<c=%r8),<xmm8=%xmm8
+pxor 640(%r8),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 656)
+# asm 1: pxor 656(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 656(<c=%r8),<xmm9=%xmm9
+pxor 656(%r8),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 672)
+# asm 1: pxor 672(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 672(<c=%r8),<xmm10=%xmm10
+pxor 672(%r8),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 688)
+# asm 1: pxor 688(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 688(<c=%r8),<xmm11=%xmm11
+pxor 688(%r8),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 704)
+# asm 1: pxor 704(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 704(<c=%r8),<xmm12=%xmm12
+pxor 704(%r8),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 720)
+# asm 1: pxor 720(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 720(<c=%r8),<xmm13=%xmm13
+pxor 720(%r8),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 736)
+# asm 1: pxor 736(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 736(<c=%r8),<xmm14=%xmm14
+pxor 736(%r8),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 752)
+# asm 1: pxor 752(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 752(<c=%r8),<xmm15=%xmm15
+pxor 752(%r8),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 768)
+# asm 1: pxor 768(<c=int64#5),<xmm0=int6464#1
+# asm 2: pxor 768(<c=%r8),<xmm0=%xmm0
+pxor 768(%r8),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 784)
+# asm 1: pxor 784(<c=int64#5),<xmm1=int6464#2
+# asm 2: pxor 784(<c=%r8),<xmm1=%xmm1
+pxor 784(%r8),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 800)
+# asm 1: pxor 800(<c=int64#5),<xmm2=int6464#3
+# asm 2: pxor 800(<c=%r8),<xmm2=%xmm2
+pxor 800(%r8),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 816)
+# asm 1: pxor 816(<c=int64#5),<xmm3=int6464#4
+# asm 2: pxor 816(<c=%r8),<xmm3=%xmm3
+pxor 816(%r8),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 832)
+# asm 1: pxor 832(<c=int64#5),<xmm4=int6464#5
+# asm 2: pxor 832(<c=%r8),<xmm4=%xmm4
+pxor 832(%r8),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 848)
+# asm 1: pxor 848(<c=int64#5),<xmm5=int6464#6
+# asm 2: pxor 848(<c=%r8),<xmm5=%xmm5
+pxor 848(%r8),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 864)
+# asm 1: pxor 864(<c=int64#5),<xmm6=int6464#7
+# asm 2: pxor 864(<c=%r8),<xmm6=%xmm6
+pxor 864(%r8),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 880)
+# asm 1: pxor 880(<c=int64#5),<xmm7=int6464#8
+# asm 2: pxor 880(<c=%r8),<xmm7=%xmm7
+pxor 880(%r8),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 896)
+# asm 1: pxor 896(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 896(<c=%r8),<xmm8=%xmm8
+pxor 896(%r8),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SR
+# asm 1: pshufb SR,<xmm8=int6464#9
+# asm 2: pshufb SR,<xmm8=%xmm8
+pshufb SR,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 912)
+# asm 1: pxor 912(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 912(<c=%r8),<xmm9=%xmm9
+pxor 912(%r8),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SR
+# asm 1: pshufb SR,<xmm9=int6464#10
+# asm 2: pshufb SR,<xmm9=%xmm9
+pshufb SR,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 928)
+# asm 1: pxor 928(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 928(<c=%r8),<xmm10=%xmm10
+pxor 928(%r8),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SR
+# asm 1: pshufb SR,<xmm10=int6464#11
+# asm 2: pshufb SR,<xmm10=%xmm10
+pshufb SR,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 944)
+# asm 1: pxor 944(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 944(<c=%r8),<xmm11=%xmm11
+pxor 944(%r8),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SR
+# asm 1: pshufb SR,<xmm11=int6464#12
+# asm 2: pshufb SR,<xmm11=%xmm11
+pshufb SR,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 960)
+# asm 1: pxor 960(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 960(<c=%r8),<xmm12=%xmm12
+pxor 960(%r8),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SR
+# asm 1: pshufb SR,<xmm12=int6464#13
+# asm 2: pshufb SR,<xmm12=%xmm12
+pshufb SR,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 976)
+# asm 1: pxor 976(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 976(<c=%r8),<xmm13=%xmm13
+pxor 976(%r8),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SR
+# asm 1: pshufb SR,<xmm13=int6464#14
+# asm 2: pshufb SR,<xmm13=%xmm13
+pshufb SR,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 992)
+# asm 1: pxor 992(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 992(<c=%r8),<xmm14=%xmm14
+pxor 992(%r8),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SR
+# asm 1: pshufb SR,<xmm14=int6464#15
+# asm 2: pshufb SR,<xmm14=%xmm14
+pshufb SR,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 1008)
+# asm 1: pxor 1008(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 1008(<c=%r8),<xmm15=%xmm15
+pxor 1008(%r8),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SR
+# asm 1: pshufb SR,<xmm15=int6464#16
+# asm 2: pshufb SR,<xmm15=%xmm15
+pshufb SR,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:     xmm0 = shuffle dwords of xmm8 by 0x93
+# asm 1: pshufd $0x93,<xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: pshufd $0x93,<xmm8=%xmm8,>xmm0=%xmm0
+pshufd $0x93,%xmm8,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm9 by 0x93
+# asm 1: pshufd $0x93,<xmm9=int6464#10,>xmm1=int6464#2
+# asm 2: pshufd $0x93,<xmm9=%xmm9,>xmm1=%xmm1
+pshufd $0x93,%xmm9,%xmm1
+
+# qhasm:     xmm2 = shuffle dwords of xmm12 by 0x93
+# asm 1: pshufd $0x93,<xmm12=int6464#13,>xmm2=int6464#3
+# asm 2: pshufd $0x93,<xmm12=%xmm12,>xmm2=%xmm2
+pshufd $0x93,%xmm12,%xmm2
+
+# qhasm:     xmm3 = shuffle dwords of xmm14 by 0x93
+# asm 1: pshufd $0x93,<xmm14=int6464#15,>xmm3=int6464#4
+# asm 2: pshufd $0x93,<xmm14=%xmm14,>xmm3=%xmm3
+pshufd $0x93,%xmm14,%xmm3
+
+# qhasm:     xmm4 = shuffle dwords of xmm11 by 0x93
+# asm 1: pshufd $0x93,<xmm11=int6464#12,>xmm4=int6464#5
+# asm 2: pshufd $0x93,<xmm11=%xmm11,>xmm4=%xmm4
+pshufd $0x93,%xmm11,%xmm4
+
+# qhasm:     xmm5 = shuffle dwords of xmm15 by 0x93
+# asm 1: pshufd $0x93,<xmm15=int6464#16,>xmm5=int6464#6
+# asm 2: pshufd $0x93,<xmm15=%xmm15,>xmm5=%xmm5
+pshufd $0x93,%xmm15,%xmm5
+
+# qhasm:     xmm6 = shuffle dwords of xmm10 by 0x93
+# asm 1: pshufd $0x93,<xmm10=int6464#11,>xmm6=int6464#7
+# asm 2: pshufd $0x93,<xmm10=%xmm10,>xmm6=%xmm6
+pshufd $0x93,%xmm10,%xmm6
+
+# qhasm:     xmm7 = shuffle dwords of xmm13 by 0x93
+# asm 1: pshufd $0x93,<xmm13=int6464#14,>xmm7=int6464#8
+# asm 2: pshufd $0x93,<xmm13=%xmm13,>xmm7=%xmm7
+pshufd $0x93,%xmm13,%xmm7
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm12 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm12=int6464#13
+# asm 2: pxor  <xmm2=%xmm2,<xmm12=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:     xmm14 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm14=int6464#15
+# asm 2: pxor  <xmm3=%xmm3,<xmm14=%xmm14
+pxor  %xmm3,%xmm14
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm10 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pxor  <xmm6=%xmm6,<xmm10=%xmm10
+pxor  %xmm6,%xmm10
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm8,<xmm1=%xmm1
+pxor  %xmm8,%xmm1
+
+# qhasm:     xmm2 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:     xmm1 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm1=int6464#2
+# asm 2: pxor  <xmm13=%xmm13,<xmm1=%xmm1
+pxor  %xmm13,%xmm1
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#5
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm4
+pxor  %xmm14,%xmm4
+
+# qhasm:     xmm5 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm5=int6464#6
+# asm 2: pxor  <xmm11=%xmm11,<xmm5=%xmm5
+pxor  %xmm11,%xmm5
+
+# qhasm:     xmm3 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm3=int6464#4
+# asm 2: pxor  <xmm13=%xmm13,<xmm3=%xmm3
+pxor  %xmm13,%xmm3
+
+# qhasm:     xmm6 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm6=int6464#7
+# asm 2: pxor  <xmm15=%xmm15,<xmm6=%xmm6
+pxor  %xmm15,%xmm6
+
+# qhasm:     xmm7 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm7=int6464#8
+# asm 2: pxor  <xmm10=%xmm10,<xmm7=%xmm7
+pxor  %xmm10,%xmm7
+
+# qhasm:     xmm4 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm4=int6464#5
+# asm 2: pxor  <xmm13=%xmm13,<xmm4=%xmm4
+pxor  %xmm13,%xmm4
+
+# qhasm:     xmm8 = shuffle dwords of xmm8 by 0x4E
+# asm 1: pshufd $0x4E,<xmm8=int6464#9,>xmm8=int6464#9
+# asm 2: pshufd $0x4E,<xmm8=%xmm8,>xmm8=%xmm8
+pshufd $0x4E,%xmm8,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm9 by 0x4E
+# asm 1: pshufd $0x4E,<xmm9=int6464#10,>xmm9=int6464#10
+# asm 2: pshufd $0x4E,<xmm9=%xmm9,>xmm9=%xmm9
+pshufd $0x4E,%xmm9,%xmm9
+
+# qhasm:     xmm12 = shuffle dwords of xmm12 by 0x4E
+# asm 1: pshufd $0x4E,<xmm12=int6464#13,>xmm12=int6464#13
+# asm 2: pshufd $0x4E,<xmm12=%xmm12,>xmm12=%xmm12
+pshufd $0x4E,%xmm12,%xmm12
+
+# qhasm:     xmm14 = shuffle dwords of xmm14 by 0x4E
+# asm 1: pshufd $0x4E,<xmm14=int6464#15,>xmm14=int6464#15
+# asm 2: pshufd $0x4E,<xmm14=%xmm14,>xmm14=%xmm14
+pshufd $0x4E,%xmm14,%xmm14
+
+# qhasm:     xmm11 = shuffle dwords of xmm11 by 0x4E
+# asm 1: pshufd $0x4E,<xmm11=int6464#12,>xmm11=int6464#12
+# asm 2: pshufd $0x4E,<xmm11=%xmm11,>xmm11=%xmm11
+pshufd $0x4E,%xmm11,%xmm11
+
+# qhasm:     xmm15 = shuffle dwords of xmm15 by 0x4E
+# asm 1: pshufd $0x4E,<xmm15=int6464#16,>xmm15=int6464#16
+# asm 2: pshufd $0x4E,<xmm15=%xmm15,>xmm15=%xmm15
+pshufd $0x4E,%xmm15,%xmm15
+
+# qhasm:     xmm10 = shuffle dwords of xmm10 by 0x4E
+# asm 1: pshufd $0x4E,<xmm10=int6464#11,>xmm10=int6464#11
+# asm 2: pshufd $0x4E,<xmm10=%xmm10,>xmm10=%xmm10
+pshufd $0x4E,%xmm10,%xmm10
+
+# qhasm:     xmm13 = shuffle dwords of xmm13 by 0x4E
+# asm 1: pshufd $0x4E,<xmm13=int6464#14,>xmm13=int6464#14
+# asm 2: pshufd $0x4E,<xmm13=%xmm13,>xmm13=%xmm13
+pshufd $0x4E,%xmm13,%xmm13
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm2 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pxor  <xmm12=%xmm12,<xmm2=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:     xmm3 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm3=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm3=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:     xmm4 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm4=int6464#5
+# asm 2: pxor  <xmm11=%xmm11,<xmm4=%xmm4
+pxor  %xmm11,%xmm4
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm6 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm6=int6464#7
+# asm 2: pxor  <xmm10=%xmm10,<xmm6=%xmm6
+pxor  %xmm10,%xmm6
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm0 ^= *(int128 *)(c + 1024)
+# asm 1: pxor 1024(<c=int64#5),<xmm0=int6464#1
+# asm 2: pxor 1024(<c=%r8),<xmm0=%xmm0
+pxor 1024(%r8),%xmm0
+
+# qhasm:     shuffle bytes of xmm0 by SR
+# asm 1: pshufb SR,<xmm0=int6464#1
+# asm 2: pshufb SR,<xmm0=%xmm0
+pshufb SR,%xmm0
+
+# qhasm:     xmm1 ^= *(int128 *)(c + 1040)
+# asm 1: pxor 1040(<c=int64#5),<xmm1=int6464#2
+# asm 2: pxor 1040(<c=%r8),<xmm1=%xmm1
+pxor 1040(%r8),%xmm1
+
+# qhasm:     shuffle bytes of xmm1 by SR
+# asm 1: pshufb SR,<xmm1=int6464#2
+# asm 2: pshufb SR,<xmm1=%xmm1
+pshufb SR,%xmm1
+
+# qhasm:     xmm2 ^= *(int128 *)(c + 1056)
+# asm 1: pxor 1056(<c=int64#5),<xmm2=int6464#3
+# asm 2: pxor 1056(<c=%r8),<xmm2=%xmm2
+pxor 1056(%r8),%xmm2
+
+# qhasm:     shuffle bytes of xmm2 by SR
+# asm 1: pshufb SR,<xmm2=int6464#3
+# asm 2: pshufb SR,<xmm2=%xmm2
+pshufb SR,%xmm2
+
+# qhasm:     xmm3 ^= *(int128 *)(c + 1072)
+# asm 1: pxor 1072(<c=int64#5),<xmm3=int6464#4
+# asm 2: pxor 1072(<c=%r8),<xmm3=%xmm3
+pxor 1072(%r8),%xmm3
+
+# qhasm:     shuffle bytes of xmm3 by SR
+# asm 1: pshufb SR,<xmm3=int6464#4
+# asm 2: pshufb SR,<xmm3=%xmm3
+pshufb SR,%xmm3
+
+# qhasm:     xmm4 ^= *(int128 *)(c + 1088)
+# asm 1: pxor 1088(<c=int64#5),<xmm4=int6464#5
+# asm 2: pxor 1088(<c=%r8),<xmm4=%xmm4
+pxor 1088(%r8),%xmm4
+
+# qhasm:     shuffle bytes of xmm4 by SR
+# asm 1: pshufb SR,<xmm4=int6464#5
+# asm 2: pshufb SR,<xmm4=%xmm4
+pshufb SR,%xmm4
+
+# qhasm:     xmm5 ^= *(int128 *)(c + 1104)
+# asm 1: pxor 1104(<c=int64#5),<xmm5=int6464#6
+# asm 2: pxor 1104(<c=%r8),<xmm5=%xmm5
+pxor 1104(%r8),%xmm5
+
+# qhasm:     shuffle bytes of xmm5 by SR
+# asm 1: pshufb SR,<xmm5=int6464#6
+# asm 2: pshufb SR,<xmm5=%xmm5
+pshufb SR,%xmm5
+
+# qhasm:     xmm6 ^= *(int128 *)(c + 1120)
+# asm 1: pxor 1120(<c=int64#5),<xmm6=int6464#7
+# asm 2: pxor 1120(<c=%r8),<xmm6=%xmm6
+pxor 1120(%r8),%xmm6
+
+# qhasm:     shuffle bytes of xmm6 by SR
+# asm 1: pshufb SR,<xmm6=int6464#7
+# asm 2: pshufb SR,<xmm6=%xmm6
+pshufb SR,%xmm6
+
+# qhasm:     xmm7 ^= *(int128 *)(c + 1136)
+# asm 1: pxor 1136(<c=int64#5),<xmm7=int6464#8
+# asm 2: pxor 1136(<c=%r8),<xmm7=%xmm7
+pxor 1136(%r8),%xmm7
+
+# qhasm:     shuffle bytes of xmm7 by SR
+# asm 1: pshufb SR,<xmm7=int6464#8
+# asm 2: pshufb SR,<xmm7=%xmm7
+pshufb SR,%xmm7
+
+# qhasm:       xmm5 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm5=int6464#6
+# asm 2: pxor  <xmm6=%xmm6,<xmm5=%xmm5
+pxor  %xmm6,%xmm5
+
+# qhasm:       xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm1,<xmm2=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm5=int6464#6
+# asm 2: pxor  <xmm0=%xmm0,<xmm5=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm6=int6464#7
+# asm 2: pxor  <xmm2=%xmm2,<xmm6=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm3=int6464#4
+# asm 2: pxor  <xmm0=%xmm0,<xmm3=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#4
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm3
+pxor  %xmm7,%xmm3
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm3=int6464#4
+# asm 2: pxor  <xmm4=%xmm4,<xmm3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:       xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm7=int6464#8
+# asm 2: pxor  <xmm5=%xmm5,<xmm7=%xmm7
+pxor  %xmm5,%xmm7
+
+# qhasm:       xmm3 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm3=int6464#4
+# asm 2: pxor  <xmm1=%xmm1,<xmm3=%xmm3
+pxor  %xmm1,%xmm3
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm2=int6464#3
+# asm 2: pxor  <xmm7=%xmm7,<xmm2=%xmm2
+pxor  %xmm7,%xmm2
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm11 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm11=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm11=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:       xmm10 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm10=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm10=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:       xmm9 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm9=int6464#11
+# asm 2: movdqa <xmm5=%xmm5,>xmm9=%xmm10
+movdqa %xmm5,%xmm10
+
+# qhasm:       xmm13 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm13=int6464#12
+# asm 2: movdqa <xmm2=%xmm2,>xmm13=%xmm11
+movdqa %xmm2,%xmm11
+
+# qhasm:       xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#13
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm12
+movdqa %xmm6,%xmm12
+
+# qhasm:       xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:       xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:       xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm9=int6464#11
+# asm 2: pxor  <xmm3=%xmm3,<xmm9=%xmm10
+pxor  %xmm3,%xmm10
+
+# qhasm:       xmm13 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm13=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm13=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:       xmm14 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm14=int6464#14
+# asm 2: movdqa <xmm11=%xmm8,>xmm14=%xmm13
+movdqa %xmm8,%xmm13
+
+# qhasm:       xmm8 = xmm10
+# asm 1: movdqa <xmm10=int6464#10,>xmm8=int6464#15
+# asm 2: movdqa <xmm10=%xmm9,>xmm8=%xmm14
+movdqa %xmm9,%xmm14
+
+# qhasm:       xmm15 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm15=int6464#16
+# asm 2: movdqa <xmm11=%xmm8,>xmm15=%xmm15
+movdqa %xmm8,%xmm15
+
+# qhasm:       xmm10 |= xmm9
+# asm 1: por   <xmm9=int6464#11,<xmm10=int6464#10
+# asm 2: por   <xmm9=%xmm10,<xmm10=%xmm9
+por   %xmm10,%xmm9
+
+# qhasm:       xmm11 |= xmm12
+# asm 1: por   <xmm12=int6464#13,<xmm11=int6464#9
+# asm 2: por   <xmm12=%xmm12,<xmm11=%xmm8
+por   %xmm12,%xmm8
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm15=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm14 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm14=int6464#14
+# asm 2: pand  <xmm12=%xmm12,<xmm14=%xmm13
+pand  %xmm12,%xmm13
+
+# qhasm:       xmm8 &= xmm9
+# asm 1: pand  <xmm9=int6464#11,<xmm8=int6464#15
+# asm 2: pand  <xmm9=%xmm10,<xmm8=%xmm14
+pand  %xmm10,%xmm14
+
+# qhasm:       xmm12 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm9=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:       xmm15 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm15=int6464#16
+# asm 2: pand  <xmm12=%xmm12,<xmm15=%xmm15
+pand  %xmm12,%xmm15
+
+# qhasm:       xmm12 = xmm3
+# asm 1: movdqa <xmm3=int6464#4,>xmm12=int6464#11
+# asm 2: movdqa <xmm3=%xmm3,>xmm12=%xmm10
+movdqa %xmm3,%xmm10
+
+# qhasm:       xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#11,<xmm13=int6464#12
+# asm 2: pand  <xmm12=%xmm10,<xmm13=%xmm11
+pand  %xmm10,%xmm11
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm11=int6464#9
+# asm 2: pxor  <xmm13=%xmm11,<xmm11=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm13 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm13=int6464#11
+# asm 2: movdqa <xmm7=%xmm7,>xmm13=%xmm10
+movdqa %xmm7,%xmm10
+
+# qhasm:       xmm13 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm13=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm13=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm12 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm12=int6464#12
+# asm 2: movdqa <xmm5=%xmm5,>xmm12=%xmm11
+movdqa %xmm5,%xmm11
+
+# qhasm:       xmm9 = xmm13
+# asm 1: movdqa <xmm13=int6464#11,>xmm9=int6464#13
+# asm 2: movdqa <xmm13=%xmm10,>xmm9=%xmm12
+movdqa %xmm10,%xmm12
+
+# qhasm:       xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:       xmm9 |= xmm12
+# asm 1: por   <xmm12=int6464#12,<xmm9=int6464#13
+# asm 2: por   <xmm12=%xmm11,<xmm9=%xmm12
+por   %xmm11,%xmm12
+
+# qhasm:       xmm13 &= xmm12
+# asm 1: pand  <xmm12=int6464#12,<xmm13=int6464#11
+# asm 2: pand  <xmm12=%xmm11,<xmm13=%xmm10
+pand  %xmm11,%xmm10
+
+# qhasm:       xmm8 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#11,<xmm8=int6464#15
+# asm 2: pxor  <xmm13=%xmm10,<xmm8=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#9
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm10=int6464#10
+# asm 2: pxor  <xmm14=%xmm13,<xmm10=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm9 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm8 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm8=int6464#15
+# asm 2: pxor  <xmm14=%xmm13,<xmm8=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm12 = xmm2
+# asm 1: movdqa <xmm2=int6464#3,>xmm12=int6464#11
+# asm 2: movdqa <xmm2=%xmm2,>xmm12=%xmm10
+movdqa %xmm2,%xmm10
+
+# qhasm:       xmm13 = xmm4
+# asm 1: movdqa <xmm4=int6464#5,>xmm13=int6464#12
+# asm 2: movdqa <xmm4=%xmm4,>xmm13=%xmm11
+movdqa %xmm4,%xmm11
+
+# qhasm:       xmm14 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm14=int6464#14
+# asm 2: movdqa <xmm1=%xmm1,>xmm14=%xmm13
+movdqa %xmm1,%xmm13
+
+# qhasm:       xmm15 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm15=int6464#16
+# asm 2: movdqa <xmm7=%xmm7,>xmm15=%xmm15
+movdqa %xmm7,%xmm15
+
+# qhasm:       xmm12 &= xmm3
+# asm 1: pand  <xmm3=int6464#4,<xmm12=int6464#11
+# asm 2: pand  <xmm3=%xmm3,<xmm12=%xmm10
+pand  %xmm3,%xmm10
+
+# qhasm:       xmm13 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm13=int6464#12
+# asm 2: pand  <xmm0=%xmm0,<xmm13=%xmm11
+pand  %xmm0,%xmm11
+
+# qhasm:       xmm14 &= xmm5
+# asm 1: pand  <xmm5=int6464#6,<xmm14=int6464#14
+# asm 2: pand  <xmm5=%xmm5,<xmm14=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:       xmm15 |= xmm6
+# asm 1: por   <xmm6=int6464#7,<xmm15=int6464#16
+# asm 2: por   <xmm6=%xmm6,<xmm15=%xmm15
+por   %xmm6,%xmm15
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm11=int6464#9
+# asm 2: pxor  <xmm12=%xmm10,<xmm11=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#12,<xmm10=int6464#10
+# asm 2: pxor  <xmm13=%xmm11,<xmm10=%xmm9
+pxor  %xmm11,%xmm9
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#14,<xmm9=int6464#13
+# asm 2: pxor  <xmm14=%xmm13,<xmm9=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm8 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm8=int6464#15
+# asm 2: pxor  <xmm15=%xmm15,<xmm8=%xmm14
+pxor  %xmm15,%xmm14
+
+# qhasm:       xmm12 = xmm11
+# asm 1: movdqa <xmm11=int6464#9,>xmm12=int6464#11
+# asm 2: movdqa <xmm11=%xmm8,>xmm12=%xmm10
+movdqa %xmm8,%xmm10
+
+# qhasm:       xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm12=int6464#11
+# asm 2: pxor  <xmm10=%xmm9,<xmm12=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm11 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm11=int6464#9
+# asm 2: pand  <xmm9=%xmm12,<xmm11=%xmm8
+pand  %xmm12,%xmm8
+
+# qhasm:       xmm14 = xmm8
+# asm 1: movdqa <xmm8=int6464#15,>xmm14=int6464#12
+# asm 2: movdqa <xmm8=%xmm14,>xmm14=%xmm11
+movdqa %xmm14,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm11=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm15 = xmm12
+# asm 1: movdqa <xmm12=int6464#11,>xmm15=int6464#14
+# asm 2: movdqa <xmm12=%xmm10,>xmm15=%xmm13
+movdqa %xmm10,%xmm13
+
+# qhasm:       xmm15 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm15=int6464#14
+# asm 2: pand  <xmm14=%xmm11,<xmm15=%xmm13
+pand  %xmm11,%xmm13
+
+# qhasm:       xmm15 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm15=int6464#14
+# asm 2: pxor  <xmm10=%xmm9,<xmm15=%xmm13
+pxor  %xmm9,%xmm13
+
+# qhasm:       xmm13 = xmm9
+# asm 1: movdqa <xmm9=int6464#13,>xmm13=int6464#16
+# asm 2: movdqa <xmm9=%xmm12,>xmm13=%xmm15
+movdqa %xmm12,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm11 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#10,<xmm11=int6464#9
+# asm 2: pxor  <xmm10=%xmm9,<xmm11=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm13 &= xmm11
+# asm 1: pand  <xmm11=int6464#9,<xmm13=int6464#16
+# asm 2: pand  <xmm11=%xmm8,<xmm13=%xmm15
+pand  %xmm8,%xmm15
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#15,<xmm13=int6464#16
+# asm 2: pxor  <xmm8=%xmm14,<xmm13=%xmm15
+pxor  %xmm14,%xmm15
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm9=int6464#13
+# asm 2: pxor  <xmm13=%xmm15,<xmm9=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm10 = xmm14
+# asm 1: movdqa <xmm14=int6464#12,>xmm10=int6464#9
+# asm 2: movdqa <xmm14=%xmm11,>xmm10=%xmm8
+movdqa %xmm11,%xmm8
+
+# qhasm:       xmm10 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm10=int6464#9
+# asm 2: pxor  <xmm13=%xmm15,<xmm10=%xmm8
+pxor  %xmm15,%xmm8
+
+# qhasm:       xmm10 &= xmm8
+# asm 1: pand  <xmm8=int6464#15,<xmm10=int6464#9
+# asm 2: pand  <xmm8=%xmm14,<xmm10=%xmm8
+pand  %xmm14,%xmm8
+
+# qhasm:       xmm9 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm9=int6464#13
+# asm 2: pxor  <xmm10=%xmm8,<xmm9=%xmm12
+pxor  %xmm8,%xmm12
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#9,<xmm14=int6464#12
+# asm 2: pxor  <xmm10=%xmm8,<xmm14=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm14=int6464#12
+# asm 2: pand  <xmm15=%xmm13,<xmm14=%xmm11
+pand  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#11,<xmm14=int6464#12
+# asm 2: pxor  <xmm12=%xmm10,<xmm14=%xmm11
+pxor  %xmm10,%xmm11
+
+# qhasm:         xmm12 = xmm6
+# asm 1: movdqa <xmm6=int6464#7,>xmm12=int6464#9
+# asm 2: movdqa <xmm6=%xmm6,>xmm12=%xmm8
+movdqa %xmm6,%xmm8
+
+# qhasm:         xmm8 = xmm5
+# asm 1: movdqa <xmm5=int6464#6,>xmm8=int6464#10
+# asm 2: movdqa <xmm5=%xmm5,>xmm8=%xmm9
+movdqa %xmm5,%xmm9
+
+# qhasm:           xmm10 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm10=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm10=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm10 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm10 &= xmm6
+# asm 1: pand  <xmm6=int6464#7,<xmm10=int6464#11
+# asm 2: pand  <xmm6=%xmm6,<xmm10=%xmm10
+pand  %xmm6,%xmm10
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm6 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm6=int6464#7
+# asm 2: pand  <xmm14=%xmm11,<xmm6=%xmm6
+pand  %xmm11,%xmm6
+
+# qhasm:           xmm5 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm5=int6464#6
+# asm 2: pand  <xmm15=%xmm13,<xmm5=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:           xmm6 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm6=int6464#7
+# asm 2: pxor  <xmm5=%xmm5,<xmm6=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:           xmm5 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm5=int6464#6
+# asm 2: pxor  <xmm10=%xmm10,<xmm5=%xmm5
+pxor  %xmm10,%xmm5
+
+# qhasm:         xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm8 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm8=int6464#10
+# asm 2: pxor  <xmm3=%xmm3,<xmm8=%xmm9
+pxor  %xmm3,%xmm9
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm0
+# asm 1: pand  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pand  <xmm0=%xmm0,<xmm10=%xmm10
+pand  %xmm0,%xmm10
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm0=int6464#1
+# asm 2: pand  <xmm9=%xmm12,<xmm0=%xmm0
+pand  %xmm12,%xmm0
+
+# qhasm:           xmm3 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm3=int6464#4
+# asm 2: pand  <xmm13=%xmm15,<xmm3=%xmm3
+pand  %xmm15,%xmm3
+
+# qhasm:           xmm0 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm0=int6464#1
+# asm 2: pxor  <xmm3=%xmm3,<xmm0=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:           xmm3 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm3=int6464#4
+# asm 2: pxor  <xmm10=%xmm10,<xmm3=%xmm3
+pxor  %xmm10,%xmm3
+
+# qhasm:         xmm6 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm6=int6464#7
+# asm 2: pxor  <xmm12=%xmm8,<xmm6=%xmm6
+pxor  %xmm8,%xmm6
+
+# qhasm:         xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm5 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm5=int6464#6
+# asm 2: pxor  <xmm8=%xmm9,<xmm5=%xmm5
+pxor  %xmm9,%xmm5
+
+# qhasm:         xmm3 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm3=int6464#4
+# asm 2: pxor  <xmm8=%xmm9,<xmm3=%xmm3
+pxor  %xmm9,%xmm3
+
+# qhasm:         xmm12 = xmm7
+# asm 1: movdqa <xmm7=int6464#8,>xmm12=int6464#9
+# asm 2: movdqa <xmm7=%xmm7,>xmm12=%xmm8
+movdqa %xmm7,%xmm8
+
+# qhasm:         xmm8 = xmm1
+# asm 1: movdqa <xmm1=int6464#2,>xmm8=int6464#10
+# asm 2: movdqa <xmm1=%xmm1,>xmm8=%xmm9
+movdqa %xmm1,%xmm9
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm12=int6464#9
+# asm 2: pxor  <xmm4=%xmm4,<xmm12=%xmm8
+pxor  %xmm4,%xmm8
+
+# qhasm:         xmm8 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm8=int6464#10
+# asm 2: pxor  <xmm2=%xmm2,<xmm8=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm12
+# asm 1: pand  <xmm12=int6464#9,<xmm11=int6464#11
+# asm 2: pand  <xmm12=%xmm8,<xmm11=%xmm10
+pand  %xmm8,%xmm10
+
+# qhasm:           xmm12 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm12=int6464#9
+# asm 2: pxor  <xmm8=%xmm9,<xmm12=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:           xmm12 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm12=int6464#9
+# asm 2: pand  <xmm14=%xmm11,<xmm12=%xmm8
+pand  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm8=int6464#10
+# asm 2: pand  <xmm15=%xmm13,<xmm8=%xmm9
+pand  %xmm13,%xmm9
+
+# qhasm:           xmm8 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm8=int6464#10
+# asm 2: pxor  <xmm12=%xmm8,<xmm8=%xmm9
+pxor  %xmm8,%xmm9
+
+# qhasm:           xmm12 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm12=int6464#9
+# asm 2: pxor  <xmm11=%xmm10,<xmm12=%xmm8
+pxor  %xmm10,%xmm8
+
+# qhasm:           xmm10 = xmm13
+# asm 1: movdqa <xmm13=int6464#16,>xmm10=int6464#11
+# asm 2: movdqa <xmm13=%xmm15,>xmm10=%xmm10
+movdqa %xmm15,%xmm10
+
+# qhasm:           xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm12,<xmm10=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:           xmm10 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pand  <xmm4=%xmm4,<xmm10=%xmm10
+pand  %xmm4,%xmm10
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm4 &= xmm9
+# asm 1: pand  <xmm9=int6464#13,<xmm4=int6464#5
+# asm 2: pand  <xmm9=%xmm12,<xmm4=%xmm4
+pand  %xmm12,%xmm4
+
+# qhasm:           xmm2 &= xmm13
+# asm 1: pand  <xmm13=int6464#16,<xmm2=int6464#3
+# asm 2: pand  <xmm13=%xmm15,<xmm2=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm2=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:           xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#3
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm2
+pxor  %xmm10,%xmm2
+
+# qhasm:         xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#16,<xmm15=int6464#14
+# asm 2: pxor  <xmm13=%xmm15,<xmm15=%xmm13
+pxor  %xmm15,%xmm13
+
+# qhasm:         xmm14 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#13,<xmm14=int6464#12
+# asm 2: pxor  <xmm9=%xmm12,<xmm14=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:           xmm11 = xmm15
+# asm 1: movdqa <xmm15=int6464#14,>xmm11=int6464#11
+# asm 2: movdqa <xmm15=%xmm13,>xmm11=%xmm10
+movdqa %xmm13,%xmm10
+
+# qhasm:           xmm11 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#12,<xmm11=int6464#11
+# asm 2: pxor  <xmm14=%xmm11,<xmm11=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:           xmm11 &= xmm7
+# asm 1: pand  <xmm7=int6464#8,<xmm11=int6464#11
+# asm 2: pand  <xmm7=%xmm7,<xmm11=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm7 &= xmm14
+# asm 1: pand  <xmm14=int6464#12,<xmm7=int6464#8
+# asm 2: pand  <xmm14=%xmm11,<xmm7=%xmm7
+pand  %xmm11,%xmm7
+
+# qhasm:           xmm1 &= xmm15
+# asm 1: pand  <xmm15=int6464#14,<xmm1=int6464#2
+# asm 2: pand  <xmm15=%xmm13,<xmm1=%xmm1
+pand  %xmm13,%xmm1
+
+# qhasm:           xmm7 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm7=int6464#8
+# asm 2: pxor  <xmm1=%xmm1,<xmm7=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm:           xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#11,<xmm1=int6464#2
+# asm 2: pxor  <xmm11=%xmm10,<xmm1=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:         xmm7 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm7=int6464#8
+# asm 2: pxor  <xmm12=%xmm8,<xmm7=%xmm7
+pxor  %xmm8,%xmm7
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm12=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:         xmm1 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm8=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:         xmm2 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#10,<xmm2=int6464#3
+# asm 2: pxor  <xmm8=%xmm9,<xmm2=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm0,<xmm7=%xmm7
+pxor  %xmm0,%xmm7
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm1=int6464#2
+# asm 2: pxor  <xmm6=%xmm6,<xmm1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm:       xmm4 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm4=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm4=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm6 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm6=int6464#7
+# asm 2: pxor  <xmm0=%xmm0,<xmm6=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:       xmm0 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm0=int6464#1
+# asm 2: pxor  <xmm1=%xmm1,<xmm0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm1=int6464#2
+# asm 2: pxor  <xmm5=%xmm5,<xmm1=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm5 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm5=int6464#6
+# asm 2: pxor  <xmm2=%xmm2,<xmm5=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:       xmm4 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm4=int6464#5
+# asm 2: pxor  <xmm5=%xmm5,<xmm4=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm2 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm3=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm3=int6464#4
+# asm 2: pxor  <xmm5=%xmm5,<xmm3=%xmm3
+pxor  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm6=int6464#7
+# asm 2: pxor  <xmm3=%xmm3,<xmm6=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm:     xmm8 = shuffle dwords of xmm0 by 0x93
+# asm 1: pshufd $0x93,<xmm0=int6464#1,>xmm8=int6464#9
+# asm 2: pshufd $0x93,<xmm0=%xmm0,>xmm8=%xmm8
+pshufd $0x93,%xmm0,%xmm8
+
+# qhasm:     xmm9 = shuffle dwords of xmm1 by 0x93
+# asm 1: pshufd $0x93,<xmm1=int6464#2,>xmm9=int6464#10
+# asm 2: pshufd $0x93,<xmm1=%xmm1,>xmm9=%xmm9
+pshufd $0x93,%xmm1,%xmm9
+
+# qhasm:     xmm10 = shuffle dwords of xmm4 by 0x93
+# asm 1: pshufd $0x93,<xmm4=int6464#5,>xmm10=int6464#11
+# asm 2: pshufd $0x93,<xmm4=%xmm4,>xmm10=%xmm10
+pshufd $0x93,%xmm4,%xmm10
+
+# qhasm:     xmm11 = shuffle dwords of xmm6 by 0x93
+# asm 1: pshufd $0x93,<xmm6=int6464#7,>xmm11=int6464#12
+# asm 2: pshufd $0x93,<xmm6=%xmm6,>xmm11=%xmm11
+pshufd $0x93,%xmm6,%xmm11
+
+# qhasm:     xmm12 = shuffle dwords of xmm3 by 0x93
+# asm 1: pshufd $0x93,<xmm3=int6464#4,>xmm12=int6464#13
+# asm 2: pshufd $0x93,<xmm3=%xmm3,>xmm12=%xmm12
+pshufd $0x93,%xmm3,%xmm12
+
+# qhasm:     xmm13 = shuffle dwords of xmm7 by 0x93
+# asm 1: pshufd $0x93,<xmm7=int6464#8,>xmm13=int6464#14
+# asm 2: pshufd $0x93,<xmm7=%xmm7,>xmm13=%xmm13
+pshufd $0x93,%xmm7,%xmm13
+
+# qhasm:     xmm14 = shuffle dwords of xmm2 by 0x93
+# asm 1: pshufd $0x93,<xmm2=int6464#3,>xmm14=int6464#15
+# asm 2: pshufd $0x93,<xmm2=%xmm2,>xmm14=%xmm14
+pshufd $0x93,%xmm2,%xmm14
+
+# qhasm:     xmm15 = shuffle dwords of xmm5 by 0x93
+# asm 1: pshufd $0x93,<xmm5=int6464#6,>xmm15=int6464#16
+# asm 2: pshufd $0x93,<xmm5=%xmm5,>xmm15=%xmm15
+pshufd $0x93,%xmm5,%xmm15
+
+# qhasm:     xmm0 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm0=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm0=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:     xmm1 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm1=int6464#2
+# asm 2: pxor  <xmm9=%xmm9,<xmm1=%xmm1
+pxor  %xmm9,%xmm1
+
+# qhasm:     xmm4 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm4=int6464#5
+# asm 2: pxor  <xmm10=%xmm10,<xmm4=%xmm4
+pxor  %xmm10,%xmm4
+
+# qhasm:     xmm6 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm6=int6464#7
+# asm 2: pxor  <xmm11=%xmm11,<xmm6=%xmm6
+pxor  %xmm11,%xmm6
+
+# qhasm:     xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:     xmm7 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm7=int6464#8
+# asm 2: pxor  <xmm13=%xmm13,<xmm7=%xmm7
+pxor  %xmm13,%xmm7
+
+# qhasm:     xmm2 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pxor  <xmm14=%xmm14,<xmm2=%xmm2
+pxor  %xmm14,%xmm2
+
+# qhasm:     xmm5 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm5=int6464#6
+# asm 2: pxor  <xmm15=%xmm15,<xmm5=%xmm5
+pxor  %xmm15,%xmm5
+
+# qhasm:     xmm8 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm8=int6464#9
+# asm 2: pxor  <xmm5=%xmm5,<xmm8=%xmm8
+pxor  %xmm5,%xmm8
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm10 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm1=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:     xmm9 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm9=int6464#10
+# asm 2: pxor  <xmm5=%xmm5,<xmm9=%xmm9
+pxor  %xmm5,%xmm9
+
+# qhasm:     xmm11 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm11=int6464#12
+# asm 2: pxor  <xmm4=%xmm4,<xmm11=%xmm11
+pxor  %xmm4,%xmm11
+
+# qhasm:     xmm12 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm12=int6464#13
+# asm 2: pxor  <xmm6=%xmm6,<xmm12=%xmm12
+pxor  %xmm6,%xmm12
+
+# qhasm:     xmm13 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm13=int6464#14
+# asm 2: pxor  <xmm3=%xmm3,<xmm13=%xmm13
+pxor  %xmm3,%xmm13
+
+# qhasm:     xmm11 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm11=int6464#12
+# asm 2: pxor  <xmm5=%xmm5,<xmm11=%xmm11
+pxor  %xmm5,%xmm11
+
+# qhasm:     xmm14 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm14=int6464#15
+# asm 2: pxor  <xmm7=%xmm7,<xmm14=%xmm14
+pxor  %xmm7,%xmm14
+
+# qhasm:     xmm15 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm15=int6464#16
+# asm 2: pxor  <xmm2=%xmm2,<xmm15=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:     xmm12 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm12=int6464#13
+# asm 2: pxor  <xmm5=%xmm5,<xmm12=%xmm12
+pxor  %xmm5,%xmm12
+
+# qhasm:     xmm0 = shuffle dwords of xmm0 by 0x4E
+# asm 1: pshufd $0x4E,<xmm0=int6464#1,>xmm0=int6464#1
+# asm 2: pshufd $0x4E,<xmm0=%xmm0,>xmm0=%xmm0
+pshufd $0x4E,%xmm0,%xmm0
+
+# qhasm:     xmm1 = shuffle dwords of xmm1 by 0x4E
+# asm 1: pshufd $0x4E,<xmm1=int6464#2,>xmm1=int6464#2
+# asm 2: pshufd $0x4E,<xmm1=%xmm1,>xmm1=%xmm1
+pshufd $0x4E,%xmm1,%xmm1
+
+# qhasm:     xmm4 = shuffle dwords of xmm4 by 0x4E
+# asm 1: pshufd $0x4E,<xmm4=int6464#5,>xmm4=int6464#5
+# asm 2: pshufd $0x4E,<xmm4=%xmm4,>xmm4=%xmm4
+pshufd $0x4E,%xmm4,%xmm4
+
+# qhasm:     xmm6 = shuffle dwords of xmm6 by 0x4E
+# asm 1: pshufd $0x4E,<xmm6=int6464#7,>xmm6=int6464#7
+# asm 2: pshufd $0x4E,<xmm6=%xmm6,>xmm6=%xmm6
+pshufd $0x4E,%xmm6,%xmm6
+
+# qhasm:     xmm3 = shuffle dwords of xmm3 by 0x4E
+# asm 1: pshufd $0x4E,<xmm3=int6464#4,>xmm3=int6464#4
+# asm 2: pshufd $0x4E,<xmm3=%xmm3,>xmm3=%xmm3
+pshufd $0x4E,%xmm3,%xmm3
+
+# qhasm:     xmm7 = shuffle dwords of xmm7 by 0x4E
+# asm 1: pshufd $0x4E,<xmm7=int6464#8,>xmm7=int6464#8
+# asm 2: pshufd $0x4E,<xmm7=%xmm7,>xmm7=%xmm7
+pshufd $0x4E,%xmm7,%xmm7
+
+# qhasm:     xmm2 = shuffle dwords of xmm2 by 0x4E
+# asm 1: pshufd $0x4E,<xmm2=int6464#3,>xmm2=int6464#3
+# asm 2: pshufd $0x4E,<xmm2=%xmm2,>xmm2=%xmm2
+pshufd $0x4E,%xmm2,%xmm2
+
+# qhasm:     xmm5 = shuffle dwords of xmm5 by 0x4E
+# asm 1: pshufd $0x4E,<xmm5=int6464#6,>xmm5=int6464#6
+# asm 2: pshufd $0x4E,<xmm5=%xmm5,>xmm5=%xmm5
+pshufd $0x4E,%xmm5,%xmm5
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm9 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm1=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:     xmm10 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#5,<xmm10=int6464#11
+# asm 2: pxor  <xmm4=%xmm4,<xmm10=%xmm10
+pxor  %xmm4,%xmm10
+
+# qhasm:     xmm11 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#7,<xmm11=int6464#12
+# asm 2: pxor  <xmm6=%xmm6,<xmm11=%xmm11
+pxor  %xmm6,%xmm11
+
+# qhasm:     xmm12 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#4,<xmm12=int6464#13
+# asm 2: pxor  <xmm3=%xmm3,<xmm12=%xmm12
+pxor  %xmm3,%xmm12
+
+# qhasm:     xmm13 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm13=int6464#14
+# asm 2: pxor  <xmm7=%xmm7,<xmm13=%xmm13
+pxor  %xmm7,%xmm13
+
+# qhasm:     xmm14 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm14=int6464#15
+# asm 2: pxor  <xmm2=%xmm2,<xmm14=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:     xmm15 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#6,<xmm15=int6464#16
+# asm 2: pxor  <xmm5=%xmm5,<xmm15=%xmm15
+pxor  %xmm5,%xmm15
+
+# qhasm:     xmm8 ^= *(int128 *)(c + 1152)
+# asm 1: pxor 1152(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 1152(<c=%r8),<xmm8=%xmm8
+pxor 1152(%r8),%xmm8
+
+# qhasm:     shuffle bytes of xmm8 by SRM0
+# asm 1: pshufb SRM0,<xmm8=int6464#9
+# asm 2: pshufb SRM0,<xmm8=%xmm8
+pshufb SRM0,%xmm8
+
+# qhasm:     xmm9 ^= *(int128 *)(c + 1168)
+# asm 1: pxor 1168(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 1168(<c=%r8),<xmm9=%xmm9
+pxor 1168(%r8),%xmm9
+
+# qhasm:     shuffle bytes of xmm9 by SRM0
+# asm 1: pshufb SRM0,<xmm9=int6464#10
+# asm 2: pshufb SRM0,<xmm9=%xmm9
+pshufb SRM0,%xmm9
+
+# qhasm:     xmm10 ^= *(int128 *)(c + 1184)
+# asm 1: pxor 1184(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 1184(<c=%r8),<xmm10=%xmm10
+pxor 1184(%r8),%xmm10
+
+# qhasm:     shuffle bytes of xmm10 by SRM0
+# asm 1: pshufb SRM0,<xmm10=int6464#11
+# asm 2: pshufb SRM0,<xmm10=%xmm10
+pshufb SRM0,%xmm10
+
+# qhasm:     xmm11 ^= *(int128 *)(c + 1200)
+# asm 1: pxor 1200(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 1200(<c=%r8),<xmm11=%xmm11
+pxor 1200(%r8),%xmm11
+
+# qhasm:     shuffle bytes of xmm11 by SRM0
+# asm 1: pshufb SRM0,<xmm11=int6464#12
+# asm 2: pshufb SRM0,<xmm11=%xmm11
+pshufb SRM0,%xmm11
+
+# qhasm:     xmm12 ^= *(int128 *)(c + 1216)
+# asm 1: pxor 1216(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 1216(<c=%r8),<xmm12=%xmm12
+pxor 1216(%r8),%xmm12
+
+# qhasm:     shuffle bytes of xmm12 by SRM0
+# asm 1: pshufb SRM0,<xmm12=int6464#13
+# asm 2: pshufb SRM0,<xmm12=%xmm12
+pshufb SRM0,%xmm12
+
+# qhasm:     xmm13 ^= *(int128 *)(c + 1232)
+# asm 1: pxor 1232(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 1232(<c=%r8),<xmm13=%xmm13
+pxor 1232(%r8),%xmm13
+
+# qhasm:     shuffle bytes of xmm13 by SRM0
+# asm 1: pshufb SRM0,<xmm13=int6464#14
+# asm 2: pshufb SRM0,<xmm13=%xmm13
+pshufb SRM0,%xmm13
+
+# qhasm:     xmm14 ^= *(int128 *)(c + 1248)
+# asm 1: pxor 1248(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 1248(<c=%r8),<xmm14=%xmm14
+pxor 1248(%r8),%xmm14
+
+# qhasm:     shuffle bytes of xmm14 by SRM0
+# asm 1: pshufb SRM0,<xmm14=int6464#15
+# asm 2: pshufb SRM0,<xmm14=%xmm14
+pshufb SRM0,%xmm14
+
+# qhasm:     xmm15 ^= *(int128 *)(c + 1264)
+# asm 1: pxor 1264(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 1264(<c=%r8),<xmm15=%xmm15
+pxor 1264(%r8),%xmm15
+
+# qhasm:     shuffle bytes of xmm15 by SRM0
+# asm 1: pshufb SRM0,<xmm15=int6464#16
+# asm 2: pshufb SRM0,<xmm15=%xmm15
+pshufb SRM0,%xmm15
+
+# qhasm:       xmm13 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm13=int6464#14
+# asm 2: pxor  <xmm14=%xmm14,<xmm13=%xmm13
+pxor  %xmm14,%xmm13
+
+# qhasm:       xmm10 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm10=int6464#11
+# asm 2: pxor  <xmm9=%xmm9,<xmm10=%xmm10
+pxor  %xmm9,%xmm10
+
+# qhasm:       xmm13 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm13=int6464#14
+# asm 2: pxor  <xmm8=%xmm8,<xmm13=%xmm13
+pxor  %xmm8,%xmm13
+
+# qhasm:       xmm14 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm14=int6464#15
+# asm 2: pxor  <xmm10=%xmm10,<xmm14=%xmm14
+pxor  %xmm10,%xmm14
+
+# qhasm:       xmm11 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm11=int6464#12
+# asm 2: pxor  <xmm8=%xmm8,<xmm11=%xmm11
+pxor  %xmm8,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:       xmm11 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm11=int6464#12
+# asm 2: pxor  <xmm15=%xmm15,<xmm11=%xmm11
+pxor  %xmm15,%xmm11
+
+# qhasm:       xmm11 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm11=int6464#12
+# asm 2: pxor  <xmm12=%xmm12,<xmm11=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:       xmm15 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm15=int6464#16
+# asm 2: pxor  <xmm13=%xmm13,<xmm15=%xmm15
+pxor  %xmm13,%xmm15
+
+# qhasm:       xmm11 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm11=int6464#12
+# asm 2: pxor  <xmm9=%xmm9,<xmm11=%xmm11
+pxor  %xmm9,%xmm11
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm10=int6464#11
+# asm 2: pxor  <xmm15=%xmm15,<xmm10=%xmm10
+pxor  %xmm15,%xmm10
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm3 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm3=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm3=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:       xmm2 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm2=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm2=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:       xmm1 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm1=int6464#3
+# asm 2: movdqa <xmm13=%xmm13,>xmm1=%xmm2
+movdqa %xmm13,%xmm2
+
+# qhasm:       xmm5 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm5=int6464#4
+# asm 2: movdqa <xmm10=%xmm10,>xmm5=%xmm3
+movdqa %xmm10,%xmm3
+
+# qhasm:       xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#5
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm4
+movdqa %xmm14,%xmm4
+
+# qhasm:       xmm3 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm3=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm3=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:       xmm2 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm2=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm2=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:       xmm1 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm1=int6464#3
+# asm 2: pxor  <xmm11=%xmm11,<xmm1=%xmm2
+pxor  %xmm11,%xmm2
+
+# qhasm:       xmm5 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm5=int6464#4
+# asm 2: pxor  <xmm12=%xmm12,<xmm5=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#5
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm4
+pxor  %xmm8,%xmm4
+
+# qhasm:       xmm6 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm6=int6464#6
+# asm 2: movdqa <xmm3=%xmm0,>xmm6=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:       xmm0 = xmm2
+# asm 1: movdqa <xmm2=int6464#2,>xmm0=int6464#7
+# asm 2: movdqa <xmm2=%xmm1,>xmm0=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm:       xmm7 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm7=int6464#8
+# asm 2: movdqa <xmm3=%xmm0,>xmm7=%xmm7
+movdqa %xmm0,%xmm7
+
+# qhasm:       xmm2 |= xmm1
+# asm 1: por   <xmm1=int6464#3,<xmm2=int6464#2
+# asm 2: por   <xmm1=%xmm2,<xmm2=%xmm1
+por   %xmm2,%xmm1
+
+# qhasm:       xmm3 |= xmm4
+# asm 1: por   <xmm4=int6464#5,<xmm3=int6464#1
+# asm 2: por   <xmm4=%xmm4,<xmm3=%xmm0
+por   %xmm4,%xmm0
+
+# qhasm:       xmm7 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm7=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm7=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm6 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm6=int6464#6
+# asm 2: pand  <xmm4=%xmm4,<xmm6=%xmm5
+pand  %xmm4,%xmm5
+
+# qhasm:       xmm0 &= xmm1
+# asm 1: pand  <xmm1=int6464#3,<xmm0=int6464#7
+# asm 2: pand  <xmm1=%xmm2,<xmm0=%xmm6
+pand  %xmm2,%xmm6
+
+# qhasm:       xmm4 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#3,<xmm4=int6464#5
+# asm 2: pxor  <xmm1=%xmm2,<xmm4=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:       xmm7 &= xmm4
+# asm 1: pand  <xmm4=int6464#5,<xmm7=int6464#8
+# asm 2: pand  <xmm4=%xmm4,<xmm7=%xmm7
+pand  %xmm4,%xmm7
+
+# qhasm:       xmm4 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm4=int6464#3
+# asm 2: movdqa <xmm11=%xmm11,>xmm4=%xmm2
+movdqa %xmm11,%xmm2
+
+# qhasm:       xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#3
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm2
+pxor  %xmm8,%xmm2
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#3,<xmm5=int6464#4
+# asm 2: pand  <xmm4=%xmm2,<xmm5=%xmm3
+pand  %xmm2,%xmm3
+
+# qhasm:       xmm3 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm3=int6464#1
+# asm 2: pxor  <xmm5=%xmm3,<xmm3=%xmm0
+pxor  %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm5 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm5=int6464#3
+# asm 2: movdqa <xmm15=%xmm15,>xmm5=%xmm2
+movdqa %xmm15,%xmm2
+
+# qhasm:       xmm5 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm5=int6464#3
+# asm 2: pxor  <xmm9=%xmm9,<xmm5=%xmm2
+pxor  %xmm9,%xmm2
+
+# qhasm:       xmm4 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm4=int6464#4
+# asm 2: movdqa <xmm13=%xmm13,>xmm4=%xmm3
+movdqa %xmm13,%xmm3
+
+# qhasm:       xmm1 = xmm5
+# asm 1: movdqa <xmm5=int6464#3,>xmm1=int6464#5
+# asm 2: movdqa <xmm5=%xmm2,>xmm1=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm:       xmm4 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm4=int6464#4
+# asm 2: pxor  <xmm14=%xmm14,<xmm4=%xmm3
+pxor  %xmm14,%xmm3
+
+# qhasm:       xmm1 |= xmm4
+# asm 1: por   <xmm4=int6464#4,<xmm1=int6464#5
+# asm 2: por   <xmm4=%xmm3,<xmm1=%xmm4
+por   %xmm3,%xmm4
+
+# qhasm:       xmm5 &= xmm4
+# asm 1: pand  <xmm4=int6464#4,<xmm5=int6464#3
+# asm 2: pand  <xmm4=%xmm3,<xmm5=%xmm2
+pand  %xmm3,%xmm2
+
+# qhasm:       xmm0 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#3,<xmm0=int6464#7
+# asm 2: pxor  <xmm5=%xmm2,<xmm0=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:       xmm3 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm3=int6464#1
+# asm 2: pxor  <xmm7=%xmm7,<xmm3=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm2=int6464#2
+# asm 2: pxor  <xmm6=%xmm5,<xmm2=%xmm1
+pxor  %xmm5,%xmm1
+
+# qhasm:       xmm1 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm7=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm0 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm0=int6464#7
+# asm 2: pxor  <xmm6=%xmm5,<xmm0=%xmm6
+pxor  %xmm5,%xmm6
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm4 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm4=int6464#3
+# asm 2: movdqa <xmm10=%xmm10,>xmm4=%xmm2
+movdqa %xmm10,%xmm2
+
+# qhasm:       xmm5 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm5=int6464#4
+# asm 2: movdqa <xmm12=%xmm12,>xmm5=%xmm3
+movdqa %xmm12,%xmm3
+
+# qhasm:       xmm6 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm6=int6464#6
+# asm 2: movdqa <xmm9=%xmm9,>xmm6=%xmm5
+movdqa %xmm9,%xmm5
+
+# qhasm:       xmm7 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm7=int6464#8
+# asm 2: movdqa <xmm15=%xmm15,>xmm7=%xmm7
+movdqa %xmm15,%xmm7
+
+# qhasm:       xmm4 &= xmm11
+# asm 1: pand  <xmm11=int6464#12,<xmm4=int6464#3
+# asm 2: pand  <xmm11=%xmm11,<xmm4=%xmm2
+pand  %xmm11,%xmm2
+
+# qhasm:       xmm5 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm5=int6464#4
+# asm 2: pand  <xmm8=%xmm8,<xmm5=%xmm3
+pand  %xmm8,%xmm3
+
+# qhasm:       xmm6 &= xmm13
+# asm 1: pand  <xmm13=int6464#14,<xmm6=int6464#6
+# asm 2: pand  <xmm13=%xmm13,<xmm6=%xmm5
+pand  %xmm13,%xmm5
+
+# qhasm:       xmm7 |= xmm14
+# asm 1: por   <xmm14=int6464#15,<xmm7=int6464#8
+# asm 2: por   <xmm14=%xmm14,<xmm7=%xmm7
+por   %xmm14,%xmm7
+
+# qhasm:       xmm3 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm3=int6464#1
+# asm 2: pxor  <xmm4=%xmm2,<xmm3=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#4,<xmm2=int6464#2
+# asm 2: pxor  <xmm5=%xmm3,<xmm2=%xmm1
+pxor  %xmm3,%xmm1
+
+# qhasm:       xmm1 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#6,<xmm1=int6464#5
+# asm 2: pxor  <xmm6=%xmm5,<xmm1=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm:       xmm0 ^= xmm7
+# asm 1: pxor  <xmm7=int6464#8,<xmm0=int6464#7
+# asm 2: pxor  <xmm7=%xmm7,<xmm0=%xmm6
+pxor  %xmm7,%xmm6
+
+# qhasm:       xmm4 = xmm3
+# asm 1: movdqa <xmm3=int6464#1,>xmm4=int6464#3
+# asm 2: movdqa <xmm3=%xmm0,>xmm4=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm:       xmm4 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm4=int6464#3
+# asm 2: pxor  <xmm2=%xmm1,<xmm4=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm:       xmm3 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm3=int6464#1
+# asm 2: pand  <xmm1=%xmm4,<xmm3=%xmm0
+pand  %xmm4,%xmm0
+
+# qhasm:       xmm6 = xmm0
+# asm 1: movdqa <xmm0=int6464#7,>xmm6=int6464#4
+# asm 2: movdqa <xmm0=%xmm6,>xmm6=%xmm3
+movdqa %xmm6,%xmm3
+
+# qhasm:       xmm6 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm3=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm7 = xmm4
+# asm 1: movdqa <xmm4=int6464#3,>xmm7=int6464#6
+# asm 2: movdqa <xmm4=%xmm2,>xmm7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:       xmm7 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm7=int6464#6
+# asm 2: pand  <xmm6=%xmm3,<xmm7=%xmm5
+pand  %xmm3,%xmm5
+
+# qhasm:       xmm7 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm7=int6464#6
+# asm 2: pxor  <xmm2=%xmm1,<xmm7=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:       xmm5 = xmm1
+# asm 1: movdqa <xmm1=int6464#5,>xmm5=int6464#8
+# asm 2: movdqa <xmm1=%xmm4,>xmm5=%xmm7
+movdqa %xmm4,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm3 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#2,<xmm3=int6464#1
+# asm 2: pxor  <xmm2=%xmm1,<xmm3=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:       xmm5 &= xmm3
+# asm 1: pand  <xmm3=int6464#1,<xmm5=int6464#8
+# asm 2: pand  <xmm3=%xmm0,<xmm5=%xmm7
+pand  %xmm0,%xmm7
+
+# qhasm:       xmm5 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#7,<xmm5=int6464#8
+# asm 2: pxor  <xmm0=%xmm6,<xmm5=%xmm7
+pxor  %xmm6,%xmm7
+
+# qhasm:       xmm1 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm1=int6464#5
+# asm 2: pxor  <xmm5=%xmm7,<xmm1=%xmm4
+pxor  %xmm7,%xmm4
+
+# qhasm:       xmm2 = xmm6
+# asm 1: movdqa <xmm6=int6464#4,>xmm2=int6464#1
+# asm 2: movdqa <xmm6=%xmm3,>xmm2=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm:       xmm2 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm2=int6464#1
+# asm 2: pxor  <xmm5=%xmm7,<xmm2=%xmm0
+pxor  %xmm7,%xmm0
+
+# qhasm:       xmm2 &= xmm0
+# asm 1: pand  <xmm0=int6464#7,<xmm2=int6464#1
+# asm 2: pand  <xmm0=%xmm6,<xmm2=%xmm0
+pand  %xmm6,%xmm0
+
+# qhasm:       xmm1 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm1=int6464#5
+# asm 2: pxor  <xmm2=%xmm0,<xmm1=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm:       xmm6 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#1,<xmm6=int6464#4
+# asm 2: pxor  <xmm2=%xmm0,<xmm6=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:       xmm6 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm6=int6464#4
+# asm 2: pand  <xmm7=%xmm5,<xmm6=%xmm3
+pand  %xmm5,%xmm3
+
+# qhasm:       xmm6 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#3,<xmm6=int6464#4
+# asm 2: pxor  <xmm4=%xmm2,<xmm6=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm:         xmm4 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm4=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm4=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:         xmm0 = xmm13
+# asm 1: movdqa <xmm13=int6464#14,>xmm0=int6464#2
+# asm 2: movdqa <xmm13=%xmm13,>xmm0=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm:           xmm2 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm2=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm2=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm2 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm2=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm2=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm2 &= xmm14
+# asm 1: pand  <xmm14=int6464#15,<xmm2=int6464#3
+# asm 2: pand  <xmm14=%xmm14,<xmm2=%xmm2
+pand  %xmm14,%xmm2
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm14 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm14=int6464#15
+# asm 2: pand  <xmm6=%xmm3,<xmm14=%xmm14
+pand  %xmm3,%xmm14
+
+# qhasm:           xmm13 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm13=int6464#14
+# asm 2: pand  <xmm7=%xmm5,<xmm13=%xmm13
+pand  %xmm5,%xmm13
+
+# qhasm:           xmm14 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm14=int6464#15
+# asm 2: pxor  <xmm13=%xmm13,<xmm14=%xmm14
+pxor  %xmm13,%xmm14
+
+# qhasm:           xmm13 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm13=int6464#14
+# asm 2: pxor  <xmm2=%xmm2,<xmm13=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:         xmm4 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm4=int6464#1
+# asm 2: pxor  <xmm8=%xmm8,<xmm4=%xmm0
+pxor  %xmm8,%xmm0
+
+# qhasm:         xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#2
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm1
+pxor  %xmm11,%xmm1
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm8
+# asm 1: pand  <xmm8=int6464#9,<xmm2=int6464#3
+# asm 2: pand  <xmm8=%xmm8,<xmm2=%xmm2
+pand  %xmm8,%xmm2
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm8 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm8=int6464#9
+# asm 2: pand  <xmm1=%xmm4,<xmm8=%xmm8
+pand  %xmm4,%xmm8
+
+# qhasm:           xmm11 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm11=int6464#12
+# asm 2: pand  <xmm5=%xmm7,<xmm11=%xmm11
+pand  %xmm7,%xmm11
+
+# qhasm:           xmm8 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm8=int6464#9
+# asm 2: pxor  <xmm11=%xmm11,<xmm8=%xmm8
+pxor  %xmm11,%xmm8
+
+# qhasm:           xmm11 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm11=int6464#12
+# asm 2: pxor  <xmm2=%xmm2,<xmm11=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:         xmm14 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm4=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:         xmm8 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm4=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:         xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm1,<xmm13=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm:         xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm1,<xmm11=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm:         xmm4 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm4=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm4=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:         xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#2
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm1
+movdqa %xmm9,%xmm1
+
+# qhasm:         xmm4 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm4=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm4=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:         xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#2
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm1
+pxor  %xmm10,%xmm1
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm4
+# asm 1: pand  <xmm4=int6464#1,<xmm3=int6464#3
+# asm 2: pand  <xmm4=%xmm0,<xmm3=%xmm2
+pand  %xmm0,%xmm2
+
+# qhasm:           xmm4 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm4=int6464#1
+# asm 2: pxor  <xmm0=%xmm1,<xmm4=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm:           xmm4 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm4=int6464#1
+# asm 2: pand  <xmm6=%xmm3,<xmm4=%xmm0
+pand  %xmm3,%xmm0
+
+# qhasm:           xmm0 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm0=int6464#2
+# asm 2: pand  <xmm7=%xmm5,<xmm0=%xmm1
+pand  %xmm5,%xmm1
+
+# qhasm:           xmm0 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm0=int6464#2
+# asm 2: pxor  <xmm4=%xmm0,<xmm0=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm:           xmm4 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm4=int6464#1
+# asm 2: pxor  <xmm3=%xmm2,<xmm4=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:           xmm2 = xmm5
+# asm 1: movdqa <xmm5=int6464#8,>xmm2=int6464#3
+# asm 2: movdqa <xmm5=%xmm7,>xmm2=%xmm2
+movdqa %xmm7,%xmm2
+
+# qhasm:           xmm2 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm2=int6464#3
+# asm 2: pxor  <xmm1=%xmm4,<xmm2=%xmm2
+pxor  %xmm4,%xmm2
+
+# qhasm:           xmm2 &= xmm12
+# asm 1: pand  <xmm12=int6464#13,<xmm2=int6464#3
+# asm 2: pand  <xmm12=%xmm12,<xmm2=%xmm2
+pand  %xmm12,%xmm2
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm12 &= xmm1
+# asm 1: pand  <xmm1=int6464#5,<xmm12=int6464#13
+# asm 2: pand  <xmm1=%xmm4,<xmm12=%xmm12
+pand  %xmm4,%xmm12
+
+# qhasm:           xmm10 &= xmm5
+# asm 1: pand  <xmm5=int6464#8,<xmm10=int6464#11
+# asm 2: pand  <xmm5=%xmm7,<xmm10=%xmm10
+pand  %xmm7,%xmm10
+
+# qhasm:           xmm12 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm12=int6464#13
+# asm 2: pxor  <xmm10=%xmm10,<xmm12=%xmm12
+pxor  %xmm10,%xmm12
+
+# qhasm:           xmm10 ^= xmm2
+# asm 1: pxor  <xmm2=int6464#3,<xmm10=int6464#11
+# asm 2: pxor  <xmm2=%xmm2,<xmm10=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:         xmm7 ^= xmm5
+# asm 1: pxor  <xmm5=int6464#8,<xmm7=int6464#6
+# asm 2: pxor  <xmm5=%xmm7,<xmm7=%xmm5
+pxor  %xmm7,%xmm5
+
+# qhasm:         xmm6 ^= xmm1
+# asm 1: pxor  <xmm1=int6464#5,<xmm6=int6464#4
+# asm 2: pxor  <xmm1=%xmm4,<xmm6=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:           xmm3 = xmm7
+# asm 1: movdqa <xmm7=int6464#6,>xmm3=int6464#3
+# asm 2: movdqa <xmm7=%xmm5,>xmm3=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm:           xmm3 ^= xmm6
+# asm 1: pxor  <xmm6=int6464#4,<xmm3=int6464#3
+# asm 2: pxor  <xmm6=%xmm3,<xmm3=%xmm2
+pxor  %xmm3,%xmm2
+
+# qhasm:           xmm3 &= xmm15
+# asm 1: pand  <xmm15=int6464#16,<xmm3=int6464#3
+# asm 2: pand  <xmm15=%xmm15,<xmm3=%xmm2
+pand  %xmm15,%xmm2
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm15 &= xmm6
+# asm 1: pand  <xmm6=int6464#4,<xmm15=int6464#16
+# asm 2: pand  <xmm6=%xmm3,<xmm15=%xmm15
+pand  %xmm3,%xmm15
+
+# qhasm:           xmm9 &= xmm7
+# asm 1: pand  <xmm7=int6464#6,<xmm9=int6464#10
+# asm 2: pand  <xmm7=%xmm5,<xmm9=%xmm9
+pand  %xmm5,%xmm9
+
+# qhasm:           xmm15 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm15=int6464#16
+# asm 2: pxor  <xmm9=%xmm9,<xmm15=%xmm15
+pxor  %xmm9,%xmm15
+
+# qhasm:           xmm9 ^= xmm3
+# asm 1: pxor  <xmm3=int6464#3,<xmm9=int6464#10
+# asm 2: pxor  <xmm3=%xmm2,<xmm9=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:         xmm15 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm4=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:         xmm12 ^= xmm4
+# asm 1: pxor  <xmm4=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm4=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:         xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm1,<xmm9=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm:         xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#2,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm1,<xmm10=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm:       xmm15 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm15=int6464#16
+# asm 2: pxor  <xmm8=%xmm8,<xmm15=%xmm15
+pxor  %xmm8,%xmm15
+
+# qhasm:       xmm9 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm9=int6464#10
+# asm 2: pxor  <xmm14=%xmm14,<xmm9=%xmm9
+pxor  %xmm14,%xmm9
+
+# qhasm:       xmm12 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm12=int6464#13
+# asm 2: pxor  <xmm15=%xmm15,<xmm12=%xmm12
+pxor  %xmm15,%xmm12
+
+# qhasm:       xmm14 ^= xmm8
+# asm 1: pxor  <xmm8=int6464#9,<xmm14=int6464#15
+# asm 2: pxor  <xmm8=%xmm8,<xmm14=%xmm14
+pxor  %xmm8,%xmm14
+
+# qhasm:       xmm8 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm8=int6464#9
+# asm 2: pxor  <xmm9=%xmm9,<xmm8=%xmm8
+pxor  %xmm9,%xmm8
+
+# qhasm:       xmm9 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm9=int6464#10
+# asm 2: pxor  <xmm13=%xmm13,<xmm9=%xmm9
+pxor  %xmm13,%xmm9
+
+# qhasm:       xmm13 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm13=int6464#14
+# asm 2: pxor  <xmm10=%xmm10,<xmm13=%xmm13
+pxor  %xmm10,%xmm13
+
+# qhasm:       xmm12 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm12=int6464#13
+# asm 2: pxor  <xmm13=%xmm13,<xmm12=%xmm12
+pxor  %xmm13,%xmm12
+
+# qhasm:       xmm10 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm10=int6464#11
+# asm 2: pxor  <xmm11=%xmm11,<xmm10=%xmm10
+pxor  %xmm11,%xmm10
+
+# qhasm:       xmm11 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm11=int6464#12
+# asm 2: pxor  <xmm13=%xmm13,<xmm11=%xmm11
+pxor  %xmm13,%xmm11
+
+# qhasm:       xmm14 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm14=int6464#15
+# asm 2: pxor  <xmm11=%xmm11,<xmm14=%xmm14
+pxor  %xmm11,%xmm14
+
+# qhasm:   xmm8 ^= *(int128 *)(c + 1280)
+# asm 1: pxor 1280(<c=int64#5),<xmm8=int6464#9
+# asm 2: pxor 1280(<c=%r8),<xmm8=%xmm8
+pxor 1280(%r8),%xmm8
+
+# qhasm:   xmm9 ^= *(int128 *)(c + 1296)
+# asm 1: pxor 1296(<c=int64#5),<xmm9=int6464#10
+# asm 2: pxor 1296(<c=%r8),<xmm9=%xmm9
+pxor 1296(%r8),%xmm9
+
+# qhasm:   xmm12 ^= *(int128 *)(c + 1312)
+# asm 1: pxor 1312(<c=int64#5),<xmm12=int6464#13
+# asm 2: pxor 1312(<c=%r8),<xmm12=%xmm12
+pxor 1312(%r8),%xmm12
+
+# qhasm:   xmm14 ^= *(int128 *)(c + 1328)
+# asm 1: pxor 1328(<c=int64#5),<xmm14=int6464#15
+# asm 2: pxor 1328(<c=%r8),<xmm14=%xmm14
+pxor 1328(%r8),%xmm14
+
+# qhasm:   xmm11 ^= *(int128 *)(c + 1344)
+# asm 1: pxor 1344(<c=int64#5),<xmm11=int6464#12
+# asm 2: pxor 1344(<c=%r8),<xmm11=%xmm11
+pxor 1344(%r8),%xmm11
+
+# qhasm:   xmm15 ^= *(int128 *)(c + 1360)
+# asm 1: pxor 1360(<c=int64#5),<xmm15=int6464#16
+# asm 2: pxor 1360(<c=%r8),<xmm15=%xmm15
+pxor 1360(%r8),%xmm15
+
+# qhasm:   xmm10 ^= *(int128 *)(c + 1376)
+# asm 1: pxor 1376(<c=int64#5),<xmm10=int6464#11
+# asm 2: pxor 1376(<c=%r8),<xmm10=%xmm10
+pxor 1376(%r8),%xmm10
+
+# qhasm:   xmm13 ^= *(int128 *)(c + 1392)
+# asm 1: pxor 1392(<c=int64#5),<xmm13=int6464#14
+# asm 2: pxor 1392(<c=%r8),<xmm13=%xmm13
+pxor 1392(%r8),%xmm13
+
+# qhasm:     xmm0 = xmm10
+# asm 1: movdqa <xmm10=int6464#11,>xmm0=int6464#1
+# asm 2: movdqa <xmm10=%xmm10,>xmm0=%xmm0
+movdqa %xmm10,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     xmm0 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm0=int6464#1
+# asm 2: movdqa <xmm11=%xmm11,>xmm0=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm0=int6464#1
+# asm 2: pxor  <xmm15=%xmm15,<xmm0=%xmm0
+pxor  %xmm15,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     xmm0 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm0=int6464#1
+# asm 2: movdqa <xmm12=%xmm12,>xmm0=%xmm0
+movdqa %xmm12,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm0=int6464#1
+# asm 2: pxor  <xmm14=%xmm14,<xmm0=%xmm0
+pxor  %xmm14,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 1
+# asm 1: psrlq $1,<xmm0=int6464#1
+# asm 2: psrlq $1,<xmm0=%xmm0
+psrlq $1,%xmm0
+
+# qhasm:     xmm0 ^= xmm9
+# asm 1: pxor  <xmm9=int6464#10,<xmm0=int6464#1
+# asm 2: pxor  <xmm9=%xmm9,<xmm0=%xmm0
+pxor  %xmm9,%xmm0
+
+# qhasm:     xmm0 &= BS0
+# asm 1: pand  BS0,<xmm0=int6464#1
+# asm 2: pand  BS0,<xmm0=%xmm0
+pand  BS0,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     uint6464 xmm0 <<= 1
+# asm 1: psllq $1,<xmm0=int6464#1
+# asm 2: psllq $1,<xmm0=%xmm0
+psllq $1,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm0 = xmm15
+# asm 1: movdqa <xmm15=int6464#16,>xmm0=int6464#1
+# asm 2: movdqa <xmm15=%xmm15,>xmm0=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     xmm0 = xmm11
+# asm 1: movdqa <xmm11=int6464#12,>xmm0=int6464#1
+# asm 2: movdqa <xmm11=%xmm11,>xmm0=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#1
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm0
+pxor  %xmm10,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#1
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm14
+# asm 1: pxor  <xmm14=int6464#15,<xmm0=int6464#1
+# asm 2: pxor  <xmm14=%xmm14,<xmm0=%xmm0
+pxor  %xmm14,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 2
+# asm 1: psrlq $2,<xmm0=int6464#1
+# asm 2: psrlq $2,<xmm0=%xmm0
+psrlq $2,%xmm0
+
+# qhasm:     xmm0 ^= xmm12
+# asm 1: pxor  <xmm12=int6464#13,<xmm0=int6464#1
+# asm 2: pxor  <xmm12=%xmm12,<xmm0=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:     xmm0 &= BS1
+# asm 1: pand  BS1,<xmm0=int6464#1
+# asm 2: pand  BS1,<xmm0=%xmm0
+pand  BS1,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     uint6464 xmm0 <<= 2
+# asm 1: psllq $2,<xmm0=int6464#1
+# asm 2: psllq $2,<xmm0=%xmm0
+psllq $2,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm:     xmm0 = xmm14
+# asm 1: movdqa <xmm14=int6464#15,>xmm0=int6464#1
+# asm 2: movdqa <xmm14=%xmm14,>xmm0=%xmm0
+movdqa %xmm14,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm13
+# asm 1: pxor  <xmm13=int6464#14,<xmm0=int6464#1
+# asm 2: pxor  <xmm13=%xmm13,<xmm0=%xmm0
+pxor  %xmm13,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm13 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm13=int6464#14
+# asm 2: pxor  <xmm0=%xmm0,<xmm13=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm14 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm14=int6464#15
+# asm 2: pxor  <xmm0=%xmm0,<xmm14=%xmm14
+pxor  %xmm0,%xmm14
+
+# qhasm:     xmm0 = xmm12
+# asm 1: movdqa <xmm12=int6464#13,>xmm0=int6464#1
+# asm 2: movdqa <xmm12=%xmm12,>xmm0=%xmm0
+movdqa %xmm12,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm10
+# asm 1: pxor  <xmm10=int6464#11,<xmm0=int6464#1
+# asm 2: pxor  <xmm10=%xmm10,<xmm0=%xmm0
+pxor  %xmm10,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm10 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm10=int6464#11
+# asm 2: pxor  <xmm0=%xmm0,<xmm10=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm12 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm12=int6464#13
+# asm 2: pxor  <xmm0=%xmm0,<xmm12=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm:     xmm0 = xmm9
+# asm 1: movdqa <xmm9=int6464#10,>xmm0=int6464#1
+# asm 2: movdqa <xmm9=%xmm9,>xmm0=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm15
+# asm 1: pxor  <xmm15=int6464#16,<xmm0=int6464#1
+# asm 2: pxor  <xmm15=%xmm15,<xmm0=%xmm0
+pxor  %xmm15,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm15 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm15=int6464#16
+# asm 2: pxor  <xmm0=%xmm0,<xmm15=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm9 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm9=int6464#10
+# asm 2: pxor  <xmm0=%xmm0,<xmm9=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm:     xmm0 = xmm8
+# asm 1: movdqa <xmm8=int6464#9,>xmm0=int6464#1
+# asm 2: movdqa <xmm8=%xmm8,>xmm0=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm:     uint6464 xmm0 >>= 4
+# asm 1: psrlq $4,<xmm0=int6464#1
+# asm 2: psrlq $4,<xmm0=%xmm0
+psrlq $4,%xmm0
+
+# qhasm:     xmm0 ^= xmm11
+# asm 1: pxor  <xmm11=int6464#12,<xmm0=int6464#1
+# asm 2: pxor  <xmm11=%xmm11,<xmm0=%xmm0
+pxor  %xmm11,%xmm0
+
+# qhasm:     xmm0 &= BS2
+# asm 1: pand  BS2,<xmm0=int6464#1
+# asm 2: pand  BS2,<xmm0=%xmm0
+pand  BS2,%xmm0
+
+# qhasm:     xmm11 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm11=int6464#12
+# asm 2: pxor  <xmm0=%xmm0,<xmm11=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm:     uint6464 xmm0 <<= 4
+# asm 1: psllq $4,<xmm0=int6464#1
+# asm 2: psllq $4,<xmm0=%xmm0
+psllq $4,%xmm0
+
+# qhasm:     xmm8 ^= xmm0
+# asm 1: pxor  <xmm0=int6464#1,<xmm8=int6464#9
+# asm 2: pxor  <xmm0=%xmm0,<xmm8=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm: unsigned<? =? len-128
+# asm 1: cmp  $128,<len=int64#3
+# asm 2: cmp  $128,<len=%rdx
+cmp  $128,%rdx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto partial if unsigned<
+jb ._partial
+# comment:fp stack unchanged by jump
+
+# qhasm: goto full if =
+je ._full
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#4),>tmp=int64#6d
+# asm 2: movl   12(<np=%rcx),>tmp=%r9d
+movl   12(%rcx),%r9d
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#6d
+# asm 2: bswap <tmp=%r9d
+bswap %r9d
+
+# qhasm: tmp += 8
+# asm 1: add  $8,<tmp=int64#6
+# asm 2: add  $8,<tmp=%r9
+add  $8,%r9
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#6d
+# asm 2: bswap <tmp=%r9d
+bswap %r9d
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#6d,12(<np=int64#4)
+# asm 2: movl   <tmp=%r9d,12(<np=%rcx)
+movl   %r9d,12(%rcx)
+
+# qhasm: xmm8 ^= *(int128 *)(inp + 0)
+# asm 1: pxor 0(<inp=int64#2),<xmm8=int6464#9
+# asm 2: pxor 0(<inp=%rsi),<xmm8=%xmm8
+pxor 0(%rsi),%xmm8
+
+# qhasm: xmm9 ^= *(int128 *)(inp + 16)
+# asm 1: pxor 16(<inp=int64#2),<xmm9=int6464#10
+# asm 2: pxor 16(<inp=%rsi),<xmm9=%xmm9
+pxor 16(%rsi),%xmm9
+
+# qhasm: xmm12 ^= *(int128 *)(inp + 32)
+# asm 1: pxor 32(<inp=int64#2),<xmm12=int6464#13
+# asm 2: pxor 32(<inp=%rsi),<xmm12=%xmm12
+pxor 32(%rsi),%xmm12
+
+# qhasm: xmm14 ^= *(int128 *)(inp + 48)
+# asm 1: pxor 48(<inp=int64#2),<xmm14=int6464#15
+# asm 2: pxor 48(<inp=%rsi),<xmm14=%xmm14
+pxor 48(%rsi),%xmm14
+
+# qhasm: xmm11 ^= *(int128 *)(inp + 64)
+# asm 1: pxor 64(<inp=int64#2),<xmm11=int6464#12
+# asm 2: pxor 64(<inp=%rsi),<xmm11=%xmm11
+pxor 64(%rsi),%xmm11
+
+# qhasm: xmm15 ^= *(int128 *)(inp + 80)
+# asm 1: pxor 80(<inp=int64#2),<xmm15=int6464#16
+# asm 2: pxor 80(<inp=%rsi),<xmm15=%xmm15
+pxor 80(%rsi),%xmm15
+
+# qhasm: xmm10 ^= *(int128 *)(inp + 96)
+# asm 1: pxor 96(<inp=int64#2),<xmm10=int6464#11
+# asm 2: pxor 96(<inp=%rsi),<xmm10=%xmm10
+pxor 96(%rsi),%xmm10
+
+# qhasm: xmm13 ^= *(int128 *)(inp + 112)
+# asm 1: pxor 112(<inp=int64#2),<xmm13=int6464#14
+# asm 2: pxor 112(<inp=%rsi),<xmm13=%xmm13
+pxor 112(%rsi),%xmm13
+
+# qhasm: *(int128 *) (outp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<outp=int64#1)
+# asm 2: movdqa <xmm8=%xmm8,0(<outp=%rdi)
+movdqa %xmm8,0(%rdi)
+
+# qhasm: *(int128 *) (outp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<outp=int64#1)
+# asm 2: movdqa <xmm9=%xmm9,16(<outp=%rdi)
+movdqa %xmm9,16(%rdi)
+
+# qhasm: *(int128 *) (outp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<outp=int64#1)
+# asm 2: movdqa <xmm12=%xmm12,32(<outp=%rdi)
+movdqa %xmm12,32(%rdi)
+
+# qhasm: *(int128 *) (outp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<outp=int64#1)
+# asm 2: movdqa <xmm14=%xmm14,48(<outp=%rdi)
+movdqa %xmm14,48(%rdi)
+
+# qhasm: *(int128 *) (outp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<outp=int64#1)
+# asm 2: movdqa <xmm11=%xmm11,64(<outp=%rdi)
+movdqa %xmm11,64(%rdi)
+
+# qhasm: *(int128 *) (outp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<outp=int64#1)
+# asm 2: movdqa <xmm15=%xmm15,80(<outp=%rdi)
+movdqa %xmm15,80(%rdi)
+
+# qhasm: *(int128 *) (outp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<outp=int64#1)
+# asm 2: movdqa <xmm10=%xmm10,96(<outp=%rdi)
+movdqa %xmm10,96(%rdi)
+
+# qhasm: *(int128 *) (outp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<outp=int64#1)
+# asm 2: movdqa <xmm13=%xmm13,112(<outp=%rdi)
+movdqa %xmm13,112(%rdi)
+
+# qhasm: len -= 128
+# asm 1: sub  $128,<len=int64#3
+# asm 2: sub  $128,<len=%rdx
+sub  $128,%rdx
+
+# qhasm: inp += 128
+# asm 1: add  $128,<inp=int64#2
+# asm 2: add  $128,<inp=%rsi
+add  $128,%rsi
+
+# qhasm: outp += 128
+# asm 1: add  $128,<outp=int64#1
+# asm 2: add  $128,<outp=%rdi
+add  $128,%rdi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto enc_block
+jmp ._enc_block
+
+# qhasm: partial:
+._partial:
+
+# qhasm: lensav = len
+# asm 1: mov  <len=int64#3,>lensav=int64#5
+# asm 2: mov  <len=%rdx,>lensav=%r8
+mov  %rdx,%r8
+
+# qhasm: (uint32) len >>= 4
+# asm 1: shr  $4,<len=int64#3d
+# asm 2: shr  $4,<len=%edx
+shr  $4,%edx
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#4),>tmp=int64#6d
+# asm 2: movl   12(<np=%rcx),>tmp=%r9d
+movl   12(%rcx),%r9d
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#6d
+# asm 2: bswap <tmp=%r9d
+bswap %r9d
+
+# qhasm: tmp += len
+# asm 1: add  <len=int64#3,<tmp=int64#6
+# asm 2: add  <len=%rdx,<tmp=%r9
+add  %rdx,%r9
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#6d
+# asm 2: bswap <tmp=%r9d
+bswap %r9d
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#6d,12(<np=int64#4)
+# asm 2: movl   <tmp=%r9d,12(<np=%rcx)
+movl   %r9d,12(%rcx)
+
+# qhasm: blp = &bl
+# asm 1: leaq <bl=stack1024#1,>blp=int64#3
+# asm 2: leaq <bl=32(%rsp),>blp=%rdx
+leaq 32(%rsp),%rdx
+
+# qhasm: *(int128 *)(blp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<blp=int64#3)
+# asm 2: movdqa <xmm8=%xmm8,0(<blp=%rdx)
+movdqa %xmm8,0(%rdx)
+
+# qhasm: *(int128 *)(blp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<blp=int64#3)
+# asm 2: movdqa <xmm9=%xmm9,16(<blp=%rdx)
+movdqa %xmm9,16(%rdx)
+
+# qhasm: *(int128 *)(blp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<blp=int64#3)
+# asm 2: movdqa <xmm12=%xmm12,32(<blp=%rdx)
+movdqa %xmm12,32(%rdx)
+
+# qhasm: *(int128 *)(blp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<blp=int64#3)
+# asm 2: movdqa <xmm14=%xmm14,48(<blp=%rdx)
+movdqa %xmm14,48(%rdx)
+
+# qhasm: *(int128 *)(blp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<blp=int64#3)
+# asm 2: movdqa <xmm11=%xmm11,64(<blp=%rdx)
+movdqa %xmm11,64(%rdx)
+
+# qhasm: *(int128 *)(blp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<blp=int64#3)
+# asm 2: movdqa <xmm15=%xmm15,80(<blp=%rdx)
+movdqa %xmm15,80(%rdx)
+
+# qhasm: *(int128 *)(blp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<blp=int64#3)
+# asm 2: movdqa <xmm10=%xmm10,96(<blp=%rdx)
+movdqa %xmm10,96(%rdx)
+
+# qhasm: *(int128 *)(blp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<blp=int64#3)
+# asm 2: movdqa <xmm13=%xmm13,112(<blp=%rdx)
+movdqa %xmm13,112(%rdx)
+
+# qhasm: bytes:
+._bytes:
+
+# qhasm: =? lensav-0
+# asm 1: cmp  $0,<lensav=int64#5
+# asm 2: cmp  $0,<lensav=%r8
+cmp  $0,%r8
+# comment:fp stack unchanged by jump
+
+# qhasm: goto end if =
+je ._end
+
+# qhasm: b = *(uint8 *)(blp + 0)
+# asm 1: movzbq 0(<blp=int64#3),>b=int64#4
+# asm 2: movzbq 0(<blp=%rdx),>b=%rcx
+movzbq 0(%rdx),%rcx
+
+# qhasm: (uint8) b ^= *(uint8 *)(inp + 0)
+# asm 1: xorb 0(<inp=int64#2),<b=int64#4b
+# asm 2: xorb 0(<inp=%rsi),<b=%cl
+xorb 0(%rsi),%cl
+
+# qhasm: *(uint8 *)(outp + 0) = b
+# asm 1: movb   <b=int64#4b,0(<outp=int64#1)
+# asm 2: movb   <b=%cl,0(<outp=%rdi)
+movb   %cl,0(%rdi)
+
+# qhasm: blp += 1
+# asm 1: add  $1,<blp=int64#3
+# asm 2: add  $1,<blp=%rdx
+add  $1,%rdx
+
+# qhasm: inp +=1
+# asm 1: add  $1,<inp=int64#2
+# asm 2: add  $1,<inp=%rsi
+add  $1,%rsi
+
+# qhasm: outp +=1
+# asm 1: add  $1,<outp=int64#1
+# asm 2: add  $1,<outp=%rdi
+add  $1,%rdi
+
+# qhasm: lensav -= 1
+# asm 1: sub  $1,<lensav=int64#5
+# asm 2: sub  $1,<lensav=%r8
+sub  $1,%r8
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytes
+jmp ._bytes
+
+# qhasm: full:
+._full:
+
+# qhasm: tmp = *(uint32 *)(np + 12)
+# asm 1: movl   12(<np=int64#4),>tmp=int64#3d
+# asm 2: movl   12(<np=%rcx),>tmp=%edx
+movl   12(%rcx),%edx
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#3d
+# asm 2: bswap <tmp=%edx
+bswap %edx
+
+# qhasm: tmp += 8
+# asm 1: add  $8,<tmp=int64#3
+# asm 2: add  $8,<tmp=%rdx
+add  $8,%rdx
+
+# qhasm: (uint32) bswap tmp
+# asm 1: bswap <tmp=int64#3d
+# asm 2: bswap <tmp=%edx
+bswap %edx
+
+# qhasm: *(uint32 *)(np + 12) = tmp
+# asm 1: movl   <tmp=int64#3d,12(<np=int64#4)
+# asm 2: movl   <tmp=%edx,12(<np=%rcx)
+movl   %edx,12(%rcx)
+
+# qhasm: xmm8 ^= *(int128 *)(inp + 0)
+# asm 1: pxor 0(<inp=int64#2),<xmm8=int6464#9
+# asm 2: pxor 0(<inp=%rsi),<xmm8=%xmm8
+pxor 0(%rsi),%xmm8
+
+# qhasm: xmm9 ^= *(int128 *)(inp + 16)
+# asm 1: pxor 16(<inp=int64#2),<xmm9=int6464#10
+# asm 2: pxor 16(<inp=%rsi),<xmm9=%xmm9
+pxor 16(%rsi),%xmm9
+
+# qhasm: xmm12 ^= *(int128 *)(inp + 32)
+# asm 1: pxor 32(<inp=int64#2),<xmm12=int6464#13
+# asm 2: pxor 32(<inp=%rsi),<xmm12=%xmm12
+pxor 32(%rsi),%xmm12
+
+# qhasm: xmm14 ^= *(int128 *)(inp + 48)
+# asm 1: pxor 48(<inp=int64#2),<xmm14=int6464#15
+# asm 2: pxor 48(<inp=%rsi),<xmm14=%xmm14
+pxor 48(%rsi),%xmm14
+
+# qhasm: xmm11 ^= *(int128 *)(inp + 64)
+# asm 1: pxor 64(<inp=int64#2),<xmm11=int6464#12
+# asm 2: pxor 64(<inp=%rsi),<xmm11=%xmm11
+pxor 64(%rsi),%xmm11
+
+# qhasm: xmm15 ^= *(int128 *)(inp + 80)
+# asm 1: pxor 80(<inp=int64#2),<xmm15=int6464#16
+# asm 2: pxor 80(<inp=%rsi),<xmm15=%xmm15
+pxor 80(%rsi),%xmm15
+
+# qhasm: xmm10 ^= *(int128 *)(inp + 96)
+# asm 1: pxor 96(<inp=int64#2),<xmm10=int6464#11
+# asm 2: pxor 96(<inp=%rsi),<xmm10=%xmm10
+pxor 96(%rsi),%xmm10
+
+# qhasm: xmm13 ^= *(int128 *)(inp + 112)
+# asm 1: pxor 112(<inp=int64#2),<xmm13=int6464#14
+# asm 2: pxor 112(<inp=%rsi),<xmm13=%xmm13
+pxor 112(%rsi),%xmm13
+
+# qhasm: *(int128 *) (outp + 0) = xmm8
+# asm 1: movdqa <xmm8=int6464#9,0(<outp=int64#1)
+# asm 2: movdqa <xmm8=%xmm8,0(<outp=%rdi)
+movdqa %xmm8,0(%rdi)
+
+# qhasm: *(int128 *) (outp + 16) = xmm9
+# asm 1: movdqa <xmm9=int6464#10,16(<outp=int64#1)
+# asm 2: movdqa <xmm9=%xmm9,16(<outp=%rdi)
+movdqa %xmm9,16(%rdi)
+
+# qhasm: *(int128 *) (outp + 32) = xmm12
+# asm 1: movdqa <xmm12=int6464#13,32(<outp=int64#1)
+# asm 2: movdqa <xmm12=%xmm12,32(<outp=%rdi)
+movdqa %xmm12,32(%rdi)
+
+# qhasm: *(int128 *) (outp + 48) = xmm14
+# asm 1: movdqa <xmm14=int6464#15,48(<outp=int64#1)
+# asm 2: movdqa <xmm14=%xmm14,48(<outp=%rdi)
+movdqa %xmm14,48(%rdi)
+
+# qhasm: *(int128 *) (outp + 64) = xmm11
+# asm 1: movdqa <xmm11=int6464#12,64(<outp=int64#1)
+# asm 2: movdqa <xmm11=%xmm11,64(<outp=%rdi)
+movdqa %xmm11,64(%rdi)
+
+# qhasm: *(int128 *) (outp + 80) = xmm15
+# asm 1: movdqa <xmm15=int6464#16,80(<outp=int64#1)
+# asm 2: movdqa <xmm15=%xmm15,80(<outp=%rdi)
+movdqa %xmm15,80(%rdi)
+
+# qhasm: *(int128 *) (outp + 96) = xmm10
+# asm 1: movdqa <xmm10=int6464#11,96(<outp=int64#1)
+# asm 2: movdqa <xmm10=%xmm10,96(<outp=%rdi)
+movdqa %xmm10,96(%rdi)
+
+# qhasm: *(int128 *) (outp + 112) = xmm13
+# asm 1: movdqa <xmm13=int6464#14,112(<outp=int64#1)
+# asm 2: movdqa <xmm13=%xmm13,112(<outp=%rdi)
+movdqa %xmm13,112(%rdi)
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: end:
+._end:
+
+# qhasm: leave
+add %r11,%rsp
+mov %rdi,%rax
+mov %rsi,%rdx
+xor %rax,%rax
+ret
diff --git a/nacl/crypto_stream/aes128ctr/portable/afternm.c b/nacl/crypto_stream/aes128ctr/portable/afternm.c
new file mode 100644
index 00000000..93c96e42
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/afternm.c
@@ -0,0 +1,158 @@
+/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper
+ * Date: 2009-03-19
+ * Public domain */
+
+#include "int128.h"
+#include "common.h"
+#include "consts.h"
+#include "crypto_stream.h"
+
+int crypto_stream_afternm(unsigned char *outp, unsigned long long len, const unsigned char *noncep, const unsigned char *c)
+{
+
+  int128 xmm0;
+  int128 xmm1;
+  int128 xmm2;
+  int128 xmm3;
+  int128 xmm4;
+  int128 xmm5;
+  int128 xmm6;
+  int128 xmm7;
+
+  int128 xmm8;
+  int128 xmm9;
+  int128 xmm10;
+  int128 xmm11;
+  int128 xmm12;
+  int128 xmm13;
+  int128 xmm14;
+  int128 xmm15;
+
+  int128 nonce_stack;
+  unsigned long long lensav;
+  unsigned char bl[128];
+  unsigned char *blp;
+  unsigned char b;
+
+  uint32 tmp;
+
+  /* Copy nonce on the stack */
+  copy2(&nonce_stack, (int128 *) (noncep + 0));
+  unsigned char *np = (unsigned char *)&nonce_stack;
+
+    enc_block:
+
+    xmm0 = *(int128 *) (np + 0);
+    copy2(&xmm1, &xmm0);
+    shufb(&xmm1, SWAP32);
+    copy2(&xmm2, &xmm1);
+    copy2(&xmm3, &xmm1);
+    copy2(&xmm4, &xmm1);
+    copy2(&xmm5, &xmm1);
+    copy2(&xmm6, &xmm1);
+    copy2(&xmm7, &xmm1);
+
+    add_uint32_big(&xmm1, 1);
+    add_uint32_big(&xmm2, 2);
+    add_uint32_big(&xmm3, 3);
+    add_uint32_big(&xmm4, 4);
+    add_uint32_big(&xmm5, 5);
+    add_uint32_big(&xmm6, 6);
+    add_uint32_big(&xmm7, 7);
+
+    shufb(&xmm0, M0);
+    shufb(&xmm1, M0SWAP);
+    shufb(&xmm2, M0SWAP);
+    shufb(&xmm3, M0SWAP);
+    shufb(&xmm4, M0SWAP);
+    shufb(&xmm5, M0SWAP);
+    shufb(&xmm6, M0SWAP);
+    shufb(&xmm7, M0SWAP);
+
+    bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8)
+
+    aesround( 1, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 2, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 3, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 4, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 5, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 6, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 8, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 9, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    lastround(xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+
+    bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0)
+
+    if(len < 128) goto partial;
+    if(len == 128) goto full;
+
+    tmp = load32_bigendian(np + 12);
+    tmp += 8;
+    store32_bigendian(np + 12, tmp);
+
+    *(int128 *) (outp + 0) = xmm8;
+    *(int128 *) (outp + 16) = xmm9;
+    *(int128 *) (outp + 32) = xmm12;
+    *(int128 *) (outp + 48) = xmm14;
+    *(int128 *) (outp + 64) = xmm11;
+    *(int128 *) (outp + 80) = xmm15;
+    *(int128 *) (outp + 96) = xmm10;
+    *(int128 *) (outp + 112) = xmm13;
+
+    len -= 128;
+    outp += 128;
+
+    goto enc_block;
+
+    partial:
+
+    lensav = len;
+    len >>= 4;
+
+    tmp = load32_bigendian(np + 12);
+    tmp += len;
+    store32_bigendian(np + 12, tmp);
+
+    blp = bl;
+    *(int128 *)(blp + 0) = xmm8;
+    *(int128 *)(blp + 16) = xmm9;
+    *(int128 *)(blp + 32) = xmm12;
+    *(int128 *)(blp + 48) = xmm14;
+    *(int128 *)(blp + 64) = xmm11;
+    *(int128 *)(blp + 80) = xmm15;
+    *(int128 *)(blp + 96) = xmm10;
+    *(int128 *)(blp + 112) = xmm13;
+
+    bytes:
+
+    if(lensav == 0) goto end;
+
+    b = blp[0];
+    *(unsigned char *)(outp + 0) = b;
+
+    blp += 1;
+    outp +=1;
+    lensav -= 1;
+
+    goto bytes;
+
+    full:
+
+    tmp = load32_bigendian(np + 12);
+    tmp += 8;
+    store32_bigendian(np + 12, tmp);
+
+    *(int128 *) (outp + 0) = xmm8;
+    *(int128 *) (outp + 16) = xmm9;
+    *(int128 *) (outp + 32) = xmm12;
+    *(int128 *) (outp + 48) = xmm14;
+    *(int128 *) (outp + 64) = xmm11;
+    *(int128 *) (outp + 80) = xmm15;
+    *(int128 *) (outp + 96) = xmm10;
+    *(int128 *) (outp + 112) = xmm13;
+
+    end:
+    return 0;
+
+}
diff --git a/nacl/crypto_stream/aes128ctr/portable/api.h b/nacl/crypto_stream/aes128ctr/portable/api.h
new file mode 100644
index 00000000..62fc8d88
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/api.h
@@ -0,0 +1,3 @@
+#define CRYPTO_KEYBYTES 16
+#define CRYPTO_NONCEBYTES 16
+#define CRYPTO_BEFORENMBYTES 1408
diff --git a/nacl/crypto_stream/aes128ctr/portable/beforenm.c b/nacl/crypto_stream/aes128ctr/portable/beforenm.c
new file mode 100644
index 00000000..8fa2673d
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/beforenm.c
@@ -0,0 +1,59 @@
+/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper
+ * Date: 2009-03-19
+ * Public domain */
+
+#include "consts.h"
+#include "int128.h"
+#include "common.h"
+#include "crypto_stream.h"
+
+int crypto_stream_beforenm(unsigned char *c, const unsigned char *k)
+{
+
+  /*
+     int64 x0;
+     int64 x1;
+     int64 x2;
+     int64 x3;
+     int64 e;
+     int64 q0;
+     int64 q1;
+     int64 q2;
+     int64 q3;
+     */
+
+  int128 xmm0;
+  int128 xmm1;
+  int128 xmm2;
+  int128 xmm3;
+  int128 xmm4;
+  int128 xmm5;
+  int128 xmm6;
+  int128 xmm7;
+  int128 xmm8;
+  int128 xmm9;
+  int128 xmm10;
+  int128 xmm11;
+  int128 xmm12;
+  int128 xmm13;
+  int128 xmm14;
+  int128 xmm15;
+  int128 t;
+
+  bitslicekey0(k, c)
+
+    keyexpbs1(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    keyexpbs(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm1);, 2,c)
+    keyexpbs(xmm0, xmm1, xmm3, xmm2, xmm6, xmm5, xmm4, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm6);, 3,c)
+    keyexpbs(xmm0, xmm1, xmm6, xmm4, xmm2, xmm7, xmm3, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 4,c)
+
+    keyexpbs(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 5,c)
+    keyexpbs(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm5);, 6,c)
+    keyexpbs(xmm0, xmm1, xmm3, xmm2, xmm6, xmm5, xmm4, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm3);, 7,c)
+    keyexpbs(xmm0, xmm1, xmm6, xmm4, xmm2, xmm7, xmm3, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm7);, 8,c)
+
+    keyexpbs(xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xor_rcon(&xmm0); xor_rcon(&xmm1); xor_rcon(&xmm6); xor_rcon(&xmm3);, 9,c)
+    keyexpbs10(xmm0, xmm1, xmm4, xmm6, xmm3, xmm7, xmm2, xmm5, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+
+    return 0;
+}
diff --git a/nacl/crypto_stream/aes128ctr/portable/common.c b/nacl/crypto_stream/aes128ctr/portable/common.c
new file mode 100644
index 00000000..14a28cc6
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/common.c
@@ -0,0 +1,64 @@
+#include "common.h"
+
+uint32 load32_bigendian(const unsigned char *x)
+{
+  return
+      (uint32) (x[3]) \
+  | (((uint32) (x[2])) << 8) \
+  | (((uint32) (x[1])) << 16) \
+  | (((uint32) (x[0])) << 24)
+  ;
+}
+
+void store32_bigendian(unsigned char *x,uint32 u)
+{
+  x[3] = u; u >>= 8;
+  x[2] = u; u >>= 8;
+  x[1] = u; u >>= 8;
+  x[0] = u;
+}
+
+uint32 load32_littleendian(const unsigned char *x)
+{
+  return
+      (uint32) (x[0]) \
+  | (((uint32) (x[1])) << 8) \
+  | (((uint32) (x[2])) << 16) \
+  | (((uint32) (x[3])) << 24)
+  ;
+}
+
+void store32_littleendian(unsigned char *x,uint32 u)
+{
+  x[0] = u; u >>= 8;
+  x[1] = u; u >>= 8;
+  x[2] = u; u >>= 8;
+  x[3] = u;
+}
+
+
+uint64 load64_littleendian(const unsigned char *x)
+{
+  return
+      (uint64) (x[0]) \
+  | (((uint64) (x[1])) << 8) \
+  | (((uint64) (x[2])) << 16) \
+  | (((uint64) (x[3])) << 24)
+  | (((uint64) (x[4])) << 32)
+  | (((uint64) (x[5])) << 40)
+  | (((uint64) (x[6])) << 48)
+  | (((uint64) (x[7])) << 56)
+  ;
+}
+
+void store64_littleendian(unsigned char *x,uint64 u)
+{
+  x[0] = u; u >>= 8;
+  x[1] = u; u >>= 8;
+  x[2] = u; u >>= 8;
+  x[3] = u; u >>= 8;
+  x[4] = u; u >>= 8;
+  x[5] = u; u >>= 8;
+  x[6] = u; u >>= 8;
+  x[7] = u;
+}
diff --git a/nacl/crypto_stream/aes128ctr/portable/common.h b/nacl/crypto_stream/aes128ctr/portable/common.h
new file mode 100644
index 00000000..0f723332
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/common.h
@@ -0,0 +1,788 @@
+/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper
+ Date: 2009-03-19
+ Public domain */
+#ifndef COMMON_H
+#define COMMON_H
+
+#include "types.h"
+
+#define load32_bigendian crypto_stream_aes128ctr_portable_load32_bigendian
+uint32 load32_bigendian(const unsigned char *x);
+
+#define store32_bigendian crypto_stream_aes128ctr_portable_store32_bigendian
+void store32_bigendian(unsigned char *x,uint32 u);
+
+#define load32_littleendian crypto_stream_aes128ctr_portable_load32_littleendian
+uint32 load32_littleendian(const unsigned char *x);
+
+#define store32_littleendian crypto_stream_aes128ctr_portable_store32_littleendian
+void store32_littleendian(unsigned char *x,uint32 u);
+
+#define load64_littleendian crypto_stream_aes128ctr_portable_load64_littleendian
+uint64 load64_littleendian(const unsigned char *x);
+
+#define store64_littleendian crypto_stream_aes128ctr_portable_store64_littleendian
+void store64_littleendian(unsigned char *x,uint64 u);
+
+/* Macros required only for key expansion */
+
+#define keyexpbs1(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \
+  rotbyte(&b0);\
+  rotbyte(&b1);\
+  rotbyte(&b2);\
+  rotbyte(&b3);\
+  rotbyte(&b4);\
+  rotbyte(&b5);\
+  rotbyte(&b6);\
+  rotbyte(&b7);\
+  ;\
+  sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\
+  ;\
+  xor_rcon(&b0);\
+  shufb(&b0, EXPB0);\
+  shufb(&b1, EXPB0);\
+  shufb(&b4, EXPB0);\
+  shufb(&b6, EXPB0);\
+  shufb(&b3, EXPB0);\
+  shufb(&b7, EXPB0);\
+  shufb(&b2, EXPB0);\
+  shufb(&b5, EXPB0);\
+  shufb(&b0, EXPB0);\
+  ;\
+  t0 = *(int128 *)(bskey + 0);\
+  t1 = *(int128 *)(bskey + 16);\
+  t2 = *(int128 *)(bskey + 32);\
+  t3 = *(int128 *)(bskey + 48);\
+  t4 = *(int128 *)(bskey + 64);\
+  t5 = *(int128 *)(bskey + 80);\
+  t6 = *(int128 *)(bskey + 96);\
+  t7 = *(int128 *)(bskey + 112);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  *(int128 *)(bskey + 128) = b0;\
+  *(int128 *)(bskey + 144) = b1;\
+  *(int128 *)(bskey + 160) = b4;\
+  *(int128 *)(bskey + 176) = b6;\
+  *(int128 *)(bskey + 192) = b3;\
+  *(int128 *)(bskey + 208) = b7;\
+  *(int128 *)(bskey + 224) = b2;\
+  *(int128 *)(bskey + 240) = b5;\
+
+#define keyexpbs10(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) ;\
+  toggle(&b0);\
+  toggle(&b1);\
+  toggle(&b5);\
+  toggle(&b6);\
+  rotbyte(&b0);\
+  rotbyte(&b1);\
+  rotbyte(&b2);\
+  rotbyte(&b3);\
+  rotbyte(&b4);\
+  rotbyte(&b5);\
+  rotbyte(&b6);\
+  rotbyte(&b7);\
+  ;\
+  sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\
+  ;\
+  xor_rcon(&b1);\
+  xor_rcon(&b4);\
+  xor_rcon(&b3);\
+  xor_rcon(&b7);\
+  shufb(&b0, EXPB0);\
+  shufb(&b1, EXPB0);\
+  shufb(&b4, EXPB0);\
+  shufb(&b6, EXPB0);\
+  shufb(&b3, EXPB0);\
+  shufb(&b7, EXPB0);\
+  shufb(&b2, EXPB0);\
+  shufb(&b5, EXPB0);\
+  ;\
+  t0 = *(int128 *)(bskey + 9 * 128 +   0);\
+  t1 = *(int128 *)(bskey + 9 * 128 +  16);\
+  t2 = *(int128 *)(bskey + 9 * 128 +  32);\
+  t3 = *(int128 *)(bskey + 9 * 128 +  48);\
+  t4 = *(int128 *)(bskey + 9 * 128 +  64);\
+  t5 = *(int128 *)(bskey + 9 * 128 +  80);\
+  t6 = *(int128 *)(bskey + 9 * 128 +  96);\
+  t7 = *(int128 *)(bskey + 9 * 128 + 112);\
+  ;\
+  toggle(&t0);\
+  toggle(&t1);\
+  toggle(&t5);\
+  toggle(&t6);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  shufb(&b0, M0);\
+  shufb(&b1, M0);\
+  shufb(&b2, M0);\
+  shufb(&b3, M0);\
+  shufb(&b4, M0);\
+  shufb(&b5, M0);\
+  shufb(&b6, M0);\
+  shufb(&b7, M0);\
+  ;\
+  *(int128 *)(bskey + 1280) = b0;\
+  *(int128 *)(bskey + 1296) = b1;\
+  *(int128 *)(bskey + 1312) = b4;\
+  *(int128 *)(bskey + 1328) = b6;\
+  *(int128 *)(bskey + 1344) = b3;\
+  *(int128 *)(bskey + 1360) = b7;\
+  *(int128 *)(bskey + 1376) = b2;\
+  *(int128 *)(bskey + 1392) = b5;\
+
+
+#define keyexpbs(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, rcon, i, bskey) \
+  toggle(&b0);\
+  toggle(&b1);\
+  toggle(&b5);\
+  toggle(&b6);\
+  rotbyte(&b0);\
+  rotbyte(&b1);\
+  rotbyte(&b2);\
+  rotbyte(&b3);\
+  rotbyte(&b4);\
+  rotbyte(&b5);\
+  rotbyte(&b6);\
+  rotbyte(&b7);\
+  ;\
+  sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\
+  ;\
+  rcon;\
+  shufb(&b0, EXPB0);\
+  shufb(&b1, EXPB0);\
+  shufb(&b4, EXPB0);\
+  shufb(&b6, EXPB0);\
+  shufb(&b3, EXPB0);\
+  shufb(&b7, EXPB0);\
+  shufb(&b2, EXPB0);\
+  shufb(&b5, EXPB0);\
+  ;\
+  t0 = *(int128 *)(bskey + (i-1) * 128 +   0);\
+  t1 = *(int128 *)(bskey + (i-1) * 128 +  16);\
+  t2 = *(int128 *)(bskey + (i-1) * 128 +  32);\
+  t3 = *(int128 *)(bskey + (i-1) * 128 +  48);\
+  t4 = *(int128 *)(bskey + (i-1) * 128 +  64);\
+  t5 = *(int128 *)(bskey + (i-1) * 128 +  80);\
+  t6 = *(int128 *)(bskey + (i-1) * 128 +  96);\
+  t7 = *(int128 *)(bskey + (i-1) * 128 + 112);\
+  ;\
+  toggle(&t0);\
+  toggle(&t1);\
+  toggle(&t5);\
+  toggle(&t6);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  rshift32_littleendian(&t0, 8);\
+  rshift32_littleendian(&t1, 8);\
+  rshift32_littleendian(&t2, 8);\
+  rshift32_littleendian(&t3, 8);\
+  rshift32_littleendian(&t4, 8);\
+  rshift32_littleendian(&t5, 8);\
+  rshift32_littleendian(&t6, 8);\
+  rshift32_littleendian(&t7, 8);\
+  ;\
+  xor2(&b0, &t0);\
+  xor2(&b1, &t1);\
+  xor2(&b4, &t2);\
+  xor2(&b6, &t3);\
+  xor2(&b3, &t4);\
+  xor2(&b7, &t5);\
+  xor2(&b2, &t6);\
+  xor2(&b5, &t7);\
+  ;\
+  *(int128 *)(bskey + i*128 +   0) = b0;\
+  *(int128 *)(bskey + i*128 +  16) = b1;\
+  *(int128 *)(bskey + i*128 +  32) = b4;\
+  *(int128 *)(bskey + i*128 +  48) = b6;\
+  *(int128 *)(bskey + i*128 +  64) = b3;\
+  *(int128 *)(bskey + i*128 +  80) = b7;\
+  *(int128 *)(bskey + i*128 +  96) = b2;\
+  *(int128 *)(bskey + i*128 + 112) = b5;\
+
+/* Macros used in multiple contexts */
+
+#define bitslicekey0(key, bskey) \
+  xmm0 = *(int128 *) (key + 0);\
+  shufb(&xmm0, M0);\
+  copy2(&xmm1, &xmm0);\
+  copy2(&xmm2, &xmm0);\
+  copy2(&xmm3, &xmm0);\
+  copy2(&xmm4, &xmm0);\
+  copy2(&xmm5, &xmm0);\
+  copy2(&xmm6, &xmm0);\
+  copy2(&xmm7, &xmm0);\
+  ;\
+  bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\
+  ;\
+  *(int128 *) (bskey + 0) = xmm0;\
+  *(int128 *) (bskey + 16) = xmm1;\
+  *(int128 *) (bskey + 32) = xmm2;\
+  *(int128 *) (bskey + 48) = xmm3;\
+  *(int128 *) (bskey + 64) = xmm4;\
+  *(int128 *) (bskey + 80) = xmm5;\
+  *(int128 *) (bskey + 96) = xmm6;\
+  *(int128 *) (bskey + 112) = xmm7;\
+
+
+#define bitslicekey10(key, bskey) \
+  xmm0 = *(int128 *) (key + 0);\
+  copy2(xmm1, xmm0);\
+  copy2(xmm2, xmm0);\
+  copy2(xmm3, xmm0);\
+  copy2(xmm4, xmm0);\
+  copy2(xmm5, xmm0);\
+  copy2(xmm6, xmm0);\
+  copy2(xmm7, xmm0);\
+  ;\
+  bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\
+  ;\
+  toggle(&xmm6);\
+  toggle(&xmm5);\
+  toggle(&xmm1);\
+  toggle(&xmm0);\
+  ;\
+  *(int128 *) (bskey +   0 + 1280) = xmm0;\
+  *(int128 *) (bskey +  16 + 1280) = xmm1;\
+  *(int128 *) (bskey +  32 + 1280) = xmm2;\
+  *(int128 *) (bskey +  48 + 1280) = xmm3;\
+  *(int128 *) (bskey +  64 + 1280) = xmm4;\
+  *(int128 *) (bskey +  80 + 1280) = xmm5;\
+  *(int128 *) (bskey +  96 + 1280) = xmm6;\
+  *(int128 *) (bskey + 112 + 1280) = xmm7;\
+  
+
+#define bitslicekey(i,key,bskey) \
+  xmm0 = *(int128 *) (key + 0);\
+  shufb(&xmm0, M0);\
+  copy2(&xmm1, &xmm0);\
+  copy2(&xmm2, &xmm0);\
+  copy2(&xmm3, &xmm0);\
+  copy2(&xmm4, &xmm0);\
+  copy2(&xmm5, &xmm0);\
+  copy2(&xmm6, &xmm0);\
+  copy2(&xmm7, &xmm0);\
+  ;\
+  bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, t);\
+  ;\
+  toggle(&xmm6);\
+  toggle(&xmm5);\
+  toggle(&xmm1);\
+  toggle(&xmm0);\
+  ;\
+  *(int128 *) (bskey +   0 + 128*i) = xmm0;\
+  *(int128 *) (bskey +  16 + 128*i) = xmm1;\
+  *(int128 *) (bskey +  32 + 128*i) = xmm2;\
+  *(int128 *) (bskey +  48 + 128*i) = xmm3;\
+  *(int128 *) (bskey +  64 + 128*i) = xmm4;\
+  *(int128 *) (bskey +  80 + 128*i) = xmm5;\
+  *(int128 *) (bskey +  96 + 128*i) = xmm6;\
+  *(int128 *) (bskey + 112 + 128*i) = xmm7;\
+
+
+#define bitslice(x0, x1, x2, x3, x4, x5, x6, x7, t) \
+        swapmove(x0, x1, 1, BS0, t);\
+        swapmove(x2, x3, 1, BS0, t);\
+        swapmove(x4, x5, 1, BS0, t);\
+        swapmove(x6, x7, 1, BS0, t);\
+        ;\
+        swapmove(x0, x2, 2, BS1, t);\
+        swapmove(x1, x3, 2, BS1, t);\
+        swapmove(x4, x6, 2, BS1, t);\
+        swapmove(x5, x7, 2, BS1, t);\
+        ;\
+        swapmove(x0, x4, 4, BS2, t);\
+        swapmove(x1, x5, 4, BS2, t);\
+        swapmove(x2, x6, 4, BS2, t);\
+        swapmove(x3, x7, 4, BS2, t);\
+
+
+#define swapmove(a, b, n, m, t) \
+        copy2(&t, &b);\
+  rshift64_littleendian(&t, n);\
+        xor2(&t, &a);\
+  and2(&t, &m);\
+  xor2(&a, &t);\
+  lshift64_littleendian(&t, n);\
+  xor2(&b, &t);
+
+#define rotbyte(x) \
+  shufb(x, ROTB) /* TODO: Make faster */
+
+
+/* Macros used for encryption (and decryption) */
+
+#define shiftrows(x0, x1, x2, x3, x4, x5, x6, x7, i, M, bskey) \
+        xor2(&x0, (int128 *)(bskey + 128*(i-1) + 0));\
+  shufb(&x0, M);\
+        xor2(&x1, (int128 *)(bskey + 128*(i-1) + 16));\
+  shufb(&x1, M);\
+        xor2(&x2, (int128 *)(bskey + 128*(i-1) + 32));\
+  shufb(&x2, M);\
+        xor2(&x3, (int128 *)(bskey + 128*(i-1) + 48));\
+  shufb(&x3, M);\
+        xor2(&x4, (int128 *)(bskey + 128*(i-1) + 64));\
+  shufb(&x4, M);\
+        xor2(&x5, (int128 *)(bskey + 128*(i-1) + 80));\
+  shufb(&x5, M);\
+        xor2(&x6, (int128 *)(bskey + 128*(i-1) + 96));\
+  shufb(&x6, M);\
+        xor2(&x7, (int128 *)(bskey + 128*(i-1) + 112));\
+  shufb(&x7, M);\
+
+
+#define mixcolumns(x0, x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, t4, t5, t6, t7) \
+  shufd(&t0, &x0, 0x93);\
+  shufd(&t1, &x1, 0x93);\
+  shufd(&t2, &x2, 0x93);\
+  shufd(&t3, &x3, 0x93);\
+  shufd(&t4, &x4, 0x93);\
+  shufd(&t5, &x5, 0x93);\
+  shufd(&t6, &x6, 0x93);\
+  shufd(&t7, &x7, 0x93);\
+        ;\
+        xor2(&x0, &t0);\
+        xor2(&x1, &t1);\
+        xor2(&x2, &t2);\
+        xor2(&x3, &t3);\
+        xor2(&x4, &t4);\
+        xor2(&x5, &t5);\
+        xor2(&x6, &t6);\
+        xor2(&x7, &t7);\
+        ;\
+        xor2(&t0, &x7);\
+        xor2(&t1, &x0);\
+        xor2(&t2, &x1);\
+        xor2(&t1, &x7);\
+        xor2(&t3, &x2);\
+        xor2(&t4, &x3);\
+        xor2(&t5, &x4);\
+        xor2(&t3, &x7);\
+        xor2(&t6, &x5);\
+        xor2(&t7, &x6);\
+        xor2(&t4, &x7);\
+        ;\
+  shufd(&x0, &x0, 0x4e);\
+  shufd(&x1, &x1, 0x4e);\
+  shufd(&x2, &x2, 0x4e);\
+  shufd(&x3, &x3, 0x4e);\
+  shufd(&x4, &x4, 0x4e);\
+  shufd(&x5, &x5, 0x4e);\
+  shufd(&x6, &x6, 0x4e);\
+  shufd(&x7, &x7, 0x4e);\
+        ;\
+        xor2(&t0, &x0);\
+        xor2(&t1, &x1);\
+        xor2(&t2, &x2);\
+        xor2(&t3, &x3);\
+        xor2(&t4, &x4);\
+        xor2(&t5, &x5);\
+        xor2(&t6, &x6);\
+        xor2(&t7, &x7);\
+        
+
+#define aesround(i, b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \
+        shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, i, SR, bskey);\
+        sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\
+        mixcolumns(b0, b1, b4, b6, b3, b7, b2, b5, t0, t1, t2, t3, t4, t5, t6, t7);\
+
+
+#define lastround(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7, bskey) \
+        shiftrows(b0, b1, b2, b3, b4, b5, b6, b7, 10, SRM0, bskey);\
+        sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, t4, t5, t6, t7);\
+        xor2(&b0,(int128 *)(bskey + 128*10));\
+        xor2(&b1,(int128 *)(bskey + 128*10+16));\
+        xor2(&b4,(int128 *)(bskey + 128*10+32));\
+        xor2(&b6,(int128 *)(bskey + 128*10+48));\
+        xor2(&b3,(int128 *)(bskey + 128*10+64));\
+        xor2(&b7,(int128 *)(bskey + 128*10+80));\
+        xor2(&b2,(int128 *)(bskey + 128*10+96));\
+        xor2(&b5,(int128 *)(bskey + 128*10+112));\
+
+
+#define sbox(b0, b1, b2, b3, b4, b5, b6, b7, t0, t1, t2, t3, s0, s1, s2, s3) \
+        InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7); \
+        Inv_GF256(b6, b5, b0, b3, b7, b1, b4, b2, t0, t1, t2, t3, s0, s1, s2, s3); \
+        OutBasisChange(b7, b1, b4, b2, b6, b5, b0, b3); \
+
+
+#define InBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \
+        xor2(&b5, &b6);\
+        xor2(&b2, &b1);\
+        xor2(&b5, &b0);\
+        xor2(&b6, &b2);\
+        xor2(&b3, &b0);\
+        ;\
+        xor2(&b6, &b3);\
+        xor2(&b3, &b7);\
+        xor2(&b3, &b4);\
+        xor2(&b7, &b5);\
+        xor2(&b3, &b1);\
+        ;\
+        xor2(&b4, &b5);\
+        xor2(&b2, &b7);\
+        xor2(&b1, &b5);\
+ 
+#define OutBasisChange(b0, b1, b2, b3, b4, b5, b6, b7) \
+        xor2(&b0, &b6);\
+        xor2(&b1, &b4);\
+        xor2(&b2, &b0);\
+        xor2(&b4, &b6);\
+        xor2(&b6, &b1);\
+        ;\
+        xor2(&b1, &b5);\
+        xor2(&b5, &b3);\
+        xor2(&b2, &b5);\
+        xor2(&b3, &b7);\
+        xor2(&b7, &b5);\
+        ;\
+        xor2(&b4, &b7);\
+        
+#define Mul_GF4(x0, x1, y0, y1, t0) \
+        copy2(&t0, &y0);\
+        xor2(&t0, &y1);\
+        and2(&t0, &x0);\
+        xor2(&x0, &x1);\
+        and2(&x0, &y1);\
+        and2(&x1, &y0);\
+        xor2(&x0, &x1);\
+        xor2(&x1, &t0);\
+        
+#define Mul_GF4_N(x0, x1, y0, y1, t0) \
+        copy2(&t0, &y0);\
+        xor2(&t0, &y1);\
+        and2(&t0, &x0);\
+        xor2(&x0, &x1);\
+        and2(&x0, &y1);\
+        and2(&x1, &y0);\
+        xor2(&x1, &x0);\
+        xor2(&x0, &t0);\
+        
+#define Mul_GF4_2(x0, x1, x2, x3, y0, y1, t0, t1) \
+        copy2(&t0, = y0);\
+        xor2(&t0, &y1);\
+        copy2(&t1, &t0);\
+        and2(&t0, &x0);\
+        and2(&t1, &x2);\
+        xor2(&x0, &x1);\
+        xor2(&x2, &x3);\
+        and2(&x0, &y1);\
+        and2(&x2, &y1);\
+        and2(&x1, &y0);\
+        and2(&x3, &y0);\
+        xor2(&x0, &x1);\
+        xor2(&x2, &x3);\
+        xor2(&x1, &t0);\
+        xor2(&x3, &t1);\
+        
+#define Mul_GF16(x0, x1, x2, x3, y0, y1, y2, y3, t0, t1, t2, t3) \
+        copy2(&t0, &x0);\
+        copy2(&t1, &x1);\
+        Mul_GF4(x0, x1, y0, y1, t2);\
+        xor2(&t0, &x2);\
+        xor2(&t1, &x3);\
+        xor2(&y0, &y2);\
+        xor2(&y1, &y3);\
+        Mul_GF4_N(t0, t1, y0, y1, t2);\
+        Mul_GF4(x2, x3, y2, y3, t3);\
+        ;\
+        xor2(&x0, &t0);\
+        xor2(&x2, &t0);\
+        xor2(&x1, &t1);\
+        xor2(&x3, &t1);\
+                        
+#define Mul_GF16_2(x0, x1, x2, x3, x4, x5, x6, x7, y0, y1, y2, y3, t0, t1, t2, t3) \
+        copy2(&t0, &x0);\
+        copy2(&t1, &x1);\
+        Mul_GF4(x0, x1, y0, y1, t2);\
+        xor2(&t0, &x2);\
+        xor2(&t1, &x3);\
+        xor2(&y0, &y2);\
+        xor2(&y1, &y3);\
+        Mul_GF4_N(t0, t1, y0, y1, t3);\
+        Mul_GF4(x2, x3, y2, y3, t2);\
+        ;\
+        xor2(&x0, &t0);\
+        xor2(&x2, &t0);\
+        xor2(&x1, &t1);\
+        xor2(&x3, &t1);\
+        ;\
+        copy2(&t0, &x4);\
+        copy2(&t1, &x5);\
+        xor2(&t0, &x6);\
+        xor2(&t1, &x7);\
+        Mul_GF4_N(t0, t1, y0, y1, t3);\
+        Mul_GF4(x6, x7, y2, y3, t2);\
+        xor2(&y0, &y2);\
+        xor2(&y1, &y3);\
+        Mul_GF4(x4, x5, y0, y1, t3);\
+        ;\
+        xor2(&x4, &t0);\
+        xor2(&x6, &t0);\
+        xor2(&x5, &t1);\
+        xor2(&x7, &t1);\
+        
+#define Inv_GF16(x0, x1, x2, x3, t0, t1, t2, t3) \
+        copy2(&t0, &x1);\
+        copy2(&t1, &x0);\
+        and2(&t0, &x3);\
+        or2(&t1, &x2);\
+        copy2(&t2, &x1);\
+        copy2(&t3, &x0);\
+        or2(&t2, &x2);\
+        or2(&t3, &x3);\
+        xor2(&t2, &t3);\
+        ;\
+        xor2(&t0, &t2);\
+        xor2(&t1, &t2);\
+        ;\
+        Mul_GF4_2(x0, x1, x2, x3, t1, t0, t2, t3);\
+
+        
+#define Inv_GF256(x0,  x1, x2, x3, x4, x5, x6, x7, t0, t1, t2, t3, s0, s1, s2, s3) \
+        copy2(&t3, &x4);\
+        copy2(&t2, &x5);\
+        copy2(&t1, &x1);\
+        copy2(&s1, &x7);\
+        copy2(&s0, &x0);\
+        ;\
+        xor2(&t3, &x6);\
+        xor2(&t2, &x7);\
+        xor2(&t1, &x3);\
+        xor2(&s1, &x6);\
+        xor2(&s0, &x2);\
+        ;\
+        copy2(&s2, &t3);\
+        copy2(&t0, &t2);\
+        copy2(&s3, &t3);\
+        ;\
+        or2(&t2, &t1);\
+        or2(&t3, &s0);\
+        xor2(&s3, &t0);\
+        and2(&s2, &s0);\
+        and2(&t0, &t1);\
+        xor2(&s0, &t1);\
+        and2(&s3, &s0);\
+        copy2(&s0, &x3);\
+        xor2(&s0, &x2);\
+        and2(&s1, &s0);\
+        xor2(&t3, &s1);\
+        xor2(&t2, &s1);\
+        copy2(&s1, &x4);\
+        xor2(&s1, &x5);\
+        copy2(&s0, &x1);\
+        copy2(&t1, &s1);\
+        xor2(&s0, &x0);\
+        or2(&t1, &s0);\
+        and2(&s1, &s0);\
+        xor2(&t0, &s1);\
+        xor2(&t3, &s3);\
+        xor2(&t2, &s2);\
+        xor2(&t1, &s3);\
+        xor2(&t0, &s2);\
+        xor2(&t1, &s2);\
+        copy2(&s0, &x7);\
+        copy2(&s1, &x6);\
+        copy2(&s2, &x5);\
+        copy2(&s3, &x4);\
+        and2(&s0, &x3);\
+        and2(&s1, &x2);\
+        and2(&s2, &x1);\
+        or2(&s3, &x0);\
+        xor2(&t3, &s0);\
+        xor2(&t2, &s1);\
+        xor2(&t1, &s2);\
+        xor2(&t0, &s3);\
+  ;\
+  copy2(&s0, &t3);\
+  xor2(&s0, &t2);\
+  and2(&t3, &t1);\
+  copy2(&s2, &t0);\
+  xor2(&s2, &t3);\
+  copy2(&s3, &s0);\
+  and2(&s3, &s2);\
+  xor2(&s3, &t2);\
+  copy2(&s1, &t1);\
+  xor2(&s1, &t0);\
+  xor2(&t3, &t2);\
+  and2(&s1, &t3);\
+  xor2(&s1, &t0);\
+  xor2(&t1, &s1);\
+  copy2(&t2, &s2);\
+  xor2(&t2, &s1);\
+  and2(&t2, &t0);\
+  xor2(&t1, &t2);\
+  xor2(&s2, &t2);\
+  and2(&s2, &s3);\
+  xor2(&s2, &s0);\
+  ;\
+  Mul_GF16_2(x0, x1, x2, x3, x4, x5, x6, x7, s3, s2, s1, t1, s0, t0, t2, t3);\
+        
+#endif
diff --git a/nacl/crypto_stream/aes128ctr/portable/consts.c b/nacl/crypto_stream/aes128ctr/portable/consts.c
new file mode 100644
index 00000000..ed2835db
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/consts.c
@@ -0,0 +1,14 @@
+#include "consts.h"
+
+const unsigned char ROTB[16] = {0x00, 0x00, 0x00, 0x0c, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x04, 0x00, 0x00, 0x00, 0x08};
+const unsigned char M0[16]   = {0x0f, 0x0b, 0x07, 0x03, 0x0e, 0x0a, 0x06, 0x02, 0x0d, 0x09, 0x05, 0x01, 0x0c, 0x08, 0x04, 0x00};
+const unsigned char EXPB0[16] = {0x03, 0x03, 0x03, 0x03, 0x07, 0x07, 0x07, 0x07, 0x0b, 0x0b, 0x0b, 0x0b, 0x0f, 0x0f, 0x0f, 0x0f};
+
+const unsigned char SWAP32[16] = {0x03, 0x02, 0x01, 0x00, 0x07, 0x06, 0x05, 0x04, 0x0b, 0x0a, 0x09, 0x08, 0x0f, 0x0e, 0x0d, 0x0c};
+const unsigned char M0SWAP[16] = {0x0c, 0x08, 0x04, 0x00, 0x0d, 0x09, 0x05, 0x01, 0x0e, 0x0a, 0x06, 0x02, 0x0f, 0x0b, 0x07, 0x03}; 
+const unsigned char SR[16] = {0x01, 0x02, 0x03, 0x00, 0x06, 0x07, 0x04, 0x05, 0x0b, 0x08, 0x09, 0x0a, 0x0c, 0x0d, 0x0e, 0x0f};
+const unsigned char SRM0[16] = {0x0f, 0x0a, 0x05, 0x00, 0x0e, 0x09, 0x04, 0x03, 0x0d, 0x08, 0x07, 0x02, 0x0c, 0x0b, 0x06, 0x01};
+
+const int128 BS0 = {0x5555555555555555ULL, 0x5555555555555555ULL};
+const int128 BS1 = {0x3333333333333333ULL, 0x3333333333333333ULL};
+const int128 BS2 = {0x0f0f0f0f0f0f0f0fULL, 0x0f0f0f0f0f0f0f0fULL};
diff --git a/nacl/crypto_stream/aes128ctr/portable/consts.h b/nacl/crypto_stream/aes128ctr/portable/consts.h
new file mode 100644
index 00000000..4c50360b
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/consts.h
@@ -0,0 +1,28 @@
+#ifndef CONSTS_H
+#define CONSTS_H
+
+#include "int128.h"
+
+#define ROTB crypto_stream_aes128ctr_portable_ROTB
+#define M0 crypto_stream_aes128ctr_portable_M0
+#define EXPB0 crypto_stream_aes128ctr_portable_EXPB0
+#define SWAP32 crypto_stream_aes128ctr_portable_SWAP32
+#define M0SWAP crypto_stream_aes128ctr_portable_M0SWAP
+#define SR crypto_stream_aes128ctr_portable_SR
+#define SRM0 crypto_stream_aes128ctr_portable_SRM0
+#define BS0 crypto_stream_aes128ctr_portable_BS0
+#define BS1 crypto_stream_aes128ctr_portable_BS1
+#define BS2 crypto_stream_aes128ctr_portable_BS2
+
+extern const unsigned char ROTB[16];
+extern const unsigned char M0[16];
+extern const unsigned char EXPB0[16];
+extern const unsigned char SWAP32[16];
+extern const unsigned char M0SWAP[16];
+extern const unsigned char SR[16];
+extern const unsigned char SRM0[16];
+extern const int128 BS0;
+extern const int128 BS1;
+extern const int128 BS2;
+
+#endif
diff --git a/nacl/crypto_stream/aes128ctr/portable/int128.c b/nacl/crypto_stream/aes128ctr/portable/int128.c
new file mode 100644
index 00000000..25894d42
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/int128.c
@@ -0,0 +1,128 @@
+#include "int128.h"
+#include "common.h"
+
+void xor2(int128 *r, const int128 *x)
+{
+  r->a ^= x->a;
+  r->b ^= x->b;
+}
+
+void and2(int128 *r, const int128 *x)
+{
+  r->a &= x->a;
+  r->b &= x->b;
+}
+
+void or2(int128 *r, const int128 *x)
+{
+  r->a |= x->a;
+  r->b |= x->b;
+}
+
+void copy2(int128 *r, const int128 *x)
+{
+  r->a = x->a;
+  r->b = x->b;
+}
+
+void shufb(int128 *r, const unsigned char *l)
+{
+  int128 t;
+  copy2(&t,r);
+  unsigned char *cr = (unsigned char *)r;
+  unsigned char *ct = (unsigned char *)&t;
+  cr[0] = ct[l[0]];
+  cr[1] = ct[l[1]];
+  cr[2] = ct[l[2]];
+  cr[3] = ct[l[3]];
+  cr[4] = ct[l[4]];
+  cr[5] = ct[l[5]];
+  cr[6] = ct[l[6]];
+  cr[7] = ct[l[7]];
+  cr[8] = ct[l[8]];
+  cr[9] = ct[l[9]];
+  cr[10] = ct[l[10]];
+  cr[11] = ct[l[11]];
+  cr[12] = ct[l[12]];
+  cr[13] = ct[l[13]];
+  cr[14] = ct[l[14]];
+  cr[15] = ct[l[15]];
+}
+
+void shufd(int128 *r, const int128 *x, const unsigned int c)
+{
+  int128 t;
+  uint32 *tp = (uint32 *)&t;
+  uint32 *xp = (uint32 *)x;
+  tp[0] = xp[c&3];
+  tp[1] = xp[(c>>2)&3];
+  tp[2] = xp[(c>>4)&3];
+  tp[3] = xp[(c>>6)&3];
+  copy2(r,&t);
+}
+
+void rshift32_littleendian(int128 *r, const unsigned int n)
+{
+  unsigned char *rp = (unsigned char *)r;
+  uint32 t;
+  t = load32_littleendian(rp);
+  t >>= n;
+  store32_littleendian(rp, t);
+  t = load32_littleendian(rp+4);
+  t >>= n;
+  store32_littleendian(rp+4, t);
+  t = load32_littleendian(rp+8);
+  t >>= n;
+  store32_littleendian(rp+8, t);
+  t = load32_littleendian(rp+12);
+  t >>= n;
+  store32_littleendian(rp+12, t);
+}
+
+void rshift64_littleendian(int128 *r, const unsigned int n)
+{
+  unsigned char *rp = (unsigned char *)r;
+  uint64 t;
+  t = load64_littleendian(rp);
+  t >>= n;
+  store64_littleendian(rp, t);
+  t = load64_littleendian(rp+8);
+  t >>= n;
+  store64_littleendian(rp+8, t);
+}
+
+void lshift64_littleendian(int128 *r, const unsigned int n)
+{
+  unsigned char *rp = (unsigned char *)r;
+  uint64 t;
+  t = load64_littleendian(rp);
+  t <<= n;
+  store64_littleendian(rp, t);
+  t = load64_littleendian(rp+8);
+  t <<= n;
+  store64_littleendian(rp+8, t);
+}
+
+void toggle(int128 *r)
+{
+  r->a ^= 0xffffffffffffffffULL;
+  r->b ^= 0xffffffffffffffffULL;
+}
+
+void xor_rcon(int128 *r)
+{
+  unsigned char *rp = (unsigned char *)r;
+  uint32 t;
+  t = load32_littleendian(rp+12);
+  t ^= 0xffffffff;
+  store32_littleendian(rp+12, t);
+}
+
+void add_uint32_big(int128 *r, uint32 x)
+{
+  unsigned char *rp = (unsigned char *)r;
+  uint32 t;
+  t = load32_littleendian(rp+12);
+  t += x;
+  store32_littleendian(rp+12, t);
+}
diff --git a/nacl/crypto_stream/aes128ctr/portable/int128.h b/nacl/crypto_stream/aes128ctr/portable/int128.h
new file mode 100644
index 00000000..7099e5b1
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/int128.h
@@ -0,0 +1,47 @@
+#ifndef INT128_H
+#define INT128_H
+
+#include "common.h"
+
+typedef struct{
+  unsigned long long a;
+  unsigned long long b;
+} int128;
+
+#define xor2 crypto_stream_aes128ctr_portable_xor2
+void xor2(int128 *r, const int128 *x);
+
+#define and2 crypto_stream_aes128ctr_portable_and2
+void and2(int128 *r, const int128 *x);
+
+#define or2 crypto_stream_aes128ctr_portable_or2
+void or2(int128 *r, const int128 *x);
+
+#define copy2 crypto_stream_aes128ctr_portable_copy2
+void copy2(int128 *r, const int128 *x);
+
+#define shufb crypto_stream_aes128ctr_portable_shufb
+void shufb(int128 *r, const unsigned char *l);
+
+#define shufd crypto_stream_aes128ctr_portable_shufd
+void shufd(int128 *r, const int128 *x, const unsigned int c);
+
+#define rshift32_littleendian crypto_stream_aes128ctr_portable_rshift32_littleendian
+void rshift32_littleendian(int128 *r, const unsigned int n);
+
+#define rshift64_littleendian crypto_stream_aes128ctr_portable_rshift64_littleendian
+void rshift64_littleendian(int128 *r, const unsigned int n);
+
+#define lshift64_littleendian crypto_stream_aes128ctr_portable_lshift64_littleendian
+void lshift64_littleendian(int128 *r, const unsigned int n);
+
+#define toggle crypto_stream_aes128ctr_portable_toggle
+void toggle(int128 *r);
+
+#define xor_rcon crypto_stream_aes128ctr_portable_xor_rcon
+void xor_rcon(int128 *r);
+
+#define add_uint32_big crypto_stream_aes128ctr_portable_add_uint32_big
+void add_uint32_big(int128 *r, uint32 x);
+
+#endif
diff --git a/nacl/crypto_stream/aes128ctr/portable/stream.c b/nacl/crypto_stream/aes128ctr/portable/stream.c
new file mode 100644
index 00000000..963fa8c1
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/stream.c
@@ -0,0 +1,28 @@
+#include "crypto_stream.h"
+
+int crypto_stream(
+        unsigned char *out,
+        unsigned long long outlen,
+        const unsigned char *n,
+        const unsigned char *k
+        )
+{
+    unsigned char d[crypto_stream_BEFORENMBYTES];
+    crypto_stream_beforenm(d, k);
+    crypto_stream_afternm(out, outlen, n, d);
+    return 0;
+}
+
+int crypto_stream_xor(
+        unsigned char *out,
+        const unsigned char *in,
+        unsigned long long inlen,
+        const unsigned char *n,
+        const unsigned char *k
+        )
+{
+    unsigned char d[crypto_stream_BEFORENMBYTES];
+    crypto_stream_beforenm(d, k);
+    crypto_stream_xor_afternm(out, in, inlen, n, d);
+    return 0;
+}
diff --git a/nacl/crypto_stream/aes128ctr/portable/types.h b/nacl/crypto_stream/aes128ctr/portable/types.h
new file mode 100644
index 00000000..6aa502fc
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/types.h
@@ -0,0 +1,10 @@
+#ifndef TYPES_H
+#define TYPES_H
+
+#include "crypto_uint32.h"
+typedef crypto_uint32 uint32;
+
+#include "crypto_uint64.h"
+typedef crypto_uint64 uint64;
+
+#endif
diff --git a/nacl/crypto_stream/aes128ctr/portable/xor_afternm.c b/nacl/crypto_stream/aes128ctr/portable/xor_afternm.c
new file mode 100644
index 00000000..f2ff8ff6
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/portable/xor_afternm.c
@@ -0,0 +1,180 @@
+/* Author: Peter Schwabe, ported from an assembly implementation by Emilia Käsper
+ * Date: 2009-03-19
+ * Public domain */
+
+#include <stdio.h>
+#include "int128.h"
+#include "common.h"
+#include "consts.h"
+#include "crypto_stream.h"
+
+int crypto_stream_xor_afternm(unsigned char *outp, const unsigned char *inp, unsigned long long len, const unsigned char *noncep, const unsigned char *c)
+{
+
+  int128 xmm0;
+  int128 xmm1;
+  int128 xmm2;
+  int128 xmm3;
+  int128 xmm4;
+  int128 xmm5;
+  int128 xmm6;
+  int128 xmm7;
+
+  int128 xmm8;
+  int128 xmm9;
+  int128 xmm10;
+  int128 xmm11;
+  int128 xmm12;
+  int128 xmm13;
+  int128 xmm14;
+  int128 xmm15;
+
+  int128 nonce_stack;
+  unsigned long long lensav;
+  unsigned char bl[128];
+  unsigned char *blp;
+  unsigned char b;
+
+  uint32 tmp;
+
+  /* Copy nonce on the stack */
+  copy2(&nonce_stack, (int128 *) (noncep + 0));
+  unsigned char *np = (unsigned char *)&nonce_stack;
+
+    enc_block:
+
+    xmm0 = *(int128 *) (np + 0);
+    copy2(&xmm1, &xmm0);
+    shufb(&xmm1, SWAP32);
+    copy2(&xmm2, &xmm1);
+    copy2(&xmm3, &xmm1);
+    copy2(&xmm4, &xmm1);
+    copy2(&xmm5, &xmm1);
+    copy2(&xmm6, &xmm1);
+    copy2(&xmm7, &xmm1);
+
+    add_uint32_big(&xmm1, 1);
+    add_uint32_big(&xmm2, 2);
+    add_uint32_big(&xmm3, 3);
+    add_uint32_big(&xmm4, 4);
+    add_uint32_big(&xmm5, 5);
+    add_uint32_big(&xmm6, 6);
+    add_uint32_big(&xmm7, 7);
+
+    shufb(&xmm0, M0);
+    shufb(&xmm1, M0SWAP);
+    shufb(&xmm2, M0SWAP);
+    shufb(&xmm3, M0SWAP);
+    shufb(&xmm4, M0SWAP);
+    shufb(&xmm5, M0SWAP);
+    shufb(&xmm6, M0SWAP);
+    shufb(&xmm7, M0SWAP);
+
+    bitslice(xmm7, xmm6, xmm5, xmm4, xmm3, xmm2, xmm1, xmm0, xmm8)
+
+    aesround( 1, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 2, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 3, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 4, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 5, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 6, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 7, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    aesround( 8, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+    aesround( 9, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7, xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15,c)
+    lastround(xmm8, xmm9, xmm10, xmm11, xmm12, xmm13, xmm14, xmm15, xmm0, xmm1, xmm2, xmm3, xmm4, xmm5, xmm6, xmm7,c)
+
+    bitslice(xmm13, xmm10, xmm15, xmm11, xmm14, xmm12, xmm9, xmm8, xmm0)
+
+    if(len < 128) goto partial;
+    if(len == 128) goto full;
+
+    tmp = load32_bigendian(np + 12);
+    tmp += 8;
+    store32_bigendian(np + 12, tmp);
+
+    xor2(&xmm8, (int128 *)(inp + 0));
+    xor2(&xmm9, (int128 *)(inp + 16));
+    xor2(&xmm12, (int128 *)(inp + 32));
+    xor2(&xmm14, (int128 *)(inp + 48));
+    xor2(&xmm11, (int128 *)(inp + 64));
+    xor2(&xmm15, (int128 *)(inp + 80));
+    xor2(&xmm10, (int128 *)(inp + 96));
+    xor2(&xmm13, (int128 *)(inp + 112));
+
+    *(int128 *) (outp + 0) = xmm8;
+    *(int128 *) (outp + 16) = xmm9;
+    *(int128 *) (outp + 32) = xmm12;
+    *(int128 *) (outp + 48) = xmm14;
+    *(int128 *) (outp + 64) = xmm11;
+    *(int128 *) (outp + 80) = xmm15;
+    *(int128 *) (outp + 96) = xmm10;
+    *(int128 *) (outp + 112) = xmm13;
+
+    len -= 128;
+    inp += 128;
+    outp += 128;
+
+    goto enc_block;
+
+    partial:
+
+    lensav = len;
+    len >>= 4;
+
+    tmp = load32_bigendian(np + 12);
+    tmp += len;
+    store32_bigendian(np + 12, tmp);
+
+    blp = bl;
+    *(int128 *)(blp + 0) = xmm8;
+    *(int128 *)(blp + 16) = xmm9;
+    *(int128 *)(blp + 32) = xmm12;
+    *(int128 *)(blp + 48) = xmm14;
+    *(int128 *)(blp + 64) = xmm11;
+    *(int128 *)(blp + 80) = xmm15;
+    *(int128 *)(blp + 96) = xmm10;
+    *(int128 *)(blp + 112) = xmm13;
+
+    bytes:
+
+    if(lensav == 0) goto end;
+
+    b = blp[0];
+    b ^= *(unsigned char *)(inp + 0);
+    *(unsigned char *)(outp + 0) = b;
+
+    blp += 1;
+    inp +=1;
+    outp +=1;
+    lensav -= 1;
+
+    goto bytes;
+
+    full:
+
+    tmp = load32_bigendian(np + 12);
+    tmp += 8;
+    store32_bigendian(np + 12, tmp);
+
+    xor2(&xmm8, (int128 *)(inp + 0));
+    xor2(&xmm9, (int128 *)(inp + 16));
+    xor2(&xmm12, (int128 *)(inp + 32));
+    xor2(&xmm14, (int128 *)(inp + 48));
+    xor2(&xmm11, (int128 *)(inp + 64));
+    xor2(&xmm15, (int128 *)(inp + 80));
+    xor2(&xmm10, (int128 *)(inp + 96));
+    xor2(&xmm13, (int128 *)(inp + 112));
+
+    *(int128 *) (outp + 0) = xmm8;
+    *(int128 *) (outp + 16) = xmm9;
+    *(int128 *) (outp + 32) = xmm12;
+    *(int128 *) (outp + 48) = xmm14;
+    *(int128 *) (outp + 64) = xmm11;
+    *(int128 *) (outp + 80) = xmm15;
+    *(int128 *) (outp + 96) = xmm10;
+    *(int128 *) (outp + 112) = xmm13;
+
+    end:
+    return 0;
+
+}
diff --git a/nacl/crypto_stream/aes128ctr/used b/nacl/crypto_stream/aes128ctr/used
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/aes128ctr/used
diff --git a/nacl/crypto_stream/measure.c b/nacl/crypto_stream/measure.c
new file mode 100644
index 00000000..ff3ab610
--- /dev/null
+++ b/nacl/crypto_stream/measure.c
@@ -0,0 +1,73 @@
+#include <stdlib.h>
+#include "randombytes.h"
+#include "cpucycles.h"
+#include "crypto_stream.h"
+
+extern void printentry(long long,const char *,long long *,long long);
+extern unsigned char *alignedcalloc(unsigned long long);
+extern const char *primitiveimplementation;
+extern const char *implementationversion;
+extern const char *sizenames[];
+extern const long long sizes[];
+extern void allocate(void);
+extern void measure(void);
+
+const char *primitiveimplementation = crypto_stream_IMPLEMENTATION;
+const char *implementationversion = crypto_stream_VERSION;
+const char *sizenames[] = { "keybytes", "noncebytes", 0 };
+const long long sizes[] = { crypto_stream_KEYBYTES, crypto_stream_NONCEBYTES };
+
+#define MAXTEST_BYTES 4096
+#ifdef SUPERCOP
+#define MGAP 8192
+#else
+#define MGAP 8
+#endif
+
+static unsigned char *k;
+static unsigned char *n;
+static unsigned char *m;
+static unsigned char *c;
+
+void preallocate(void)
+{
+}
+
+void allocate(void)
+{
+  k = alignedcalloc(crypto_stream_KEYBYTES);
+  n = alignedcalloc(crypto_stream_NONCEBYTES);
+  m = alignedcalloc(MAXTEST_BYTES);
+  c = alignedcalloc(MAXTEST_BYTES);
+}
+
+#define TIMINGS 15
+static long long cycles[TIMINGS + 1];
+
+void measure(void)
+{
+  int i;
+  int loop;
+  int mlen;
+
+  for (loop = 0;loop < LOOPS;++loop) {
+    for (mlen = 0;mlen <= MAXTEST_BYTES;mlen += 1 + mlen / MGAP) {
+      randombytes(k,crypto_stream_KEYBYTES);
+      randombytes(n,crypto_stream_NONCEBYTES);
+      randombytes(m,mlen);
+      randombytes(c,mlen);
+      for (i = 0;i <= TIMINGS;++i) {
+        cycles[i] = cpucycles();
+        crypto_stream(c,mlen,n,k);
+      }
+      for (i = 0;i < TIMINGS;++i) cycles[i] = cycles[i + 1] - cycles[i];
+      printentry(mlen,"cycles",cycles,TIMINGS);
+      for (i = 0;i <= TIMINGS;++i) {
+        cycles[i] = cpucycles();
+        crypto_stream_xor(c,m,mlen,n,k);
+      }
+      for (i = 0;i < TIMINGS;++i) cycles[i] = cycles[i + 1] - cycles[i];
+      printentry(mlen,"xor_cycles",cycles,TIMINGS);
+    }
+  }
+}
diff --git a/nacl/crypto_stream/salsa20/amd64_xmm6/api.h b/nacl/crypto_stream/salsa20/amd64_xmm6/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/amd64_xmm6/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa20/amd64_xmm6/implementors b/nacl/crypto_stream/salsa20/amd64_xmm6/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/amd64_xmm6/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa20/amd64_xmm6/stream.s b/nacl/crypto_stream/salsa20/amd64_xmm6/stream.s
new file mode 100644
index 00000000..82a897f7
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/amd64_xmm6/stream.s
@@ -0,0 +1,4823 @@
+
+# qhasm: int64 r11_caller
+
+# qhasm: int64 r12_caller
+
+# qhasm: int64 r13_caller
+
+# qhasm: int64 r14_caller
+
+# qhasm: int64 r15_caller
+
+# qhasm: int64 rbx_caller
+
+# qhasm: int64 rbp_caller
+
+# qhasm: caller r11_caller
+
+# qhasm: caller r12_caller
+
+# qhasm: caller r13_caller
+
+# qhasm: caller r14_caller
+
+# qhasm: caller r15_caller
+
+# qhasm: caller rbx_caller
+
+# qhasm: caller rbp_caller
+
+# qhasm: stack64 r11_stack
+
+# qhasm: stack64 r12_stack
+
+# qhasm: stack64 r13_stack
+
+# qhasm: stack64 r14_stack
+
+# qhasm: stack64 r15_stack
+
+# qhasm: stack64 rbx_stack
+
+# qhasm: stack64 rbp_stack
+
+# qhasm: int64 a
+
+# qhasm: int64 arg1
+
+# qhasm: int64 arg2
+
+# qhasm: int64 arg3
+
+# qhasm: int64 arg4
+
+# qhasm: int64 arg5
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: int64 k
+
+# qhasm: int64 kbits
+
+# qhasm: int64 iv
+
+# qhasm: int64 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int64 m
+
+# qhasm: int64 out
+
+# qhasm: int64 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: int6464 y0
+
+# qhasm: int6464 y1
+
+# qhasm: int6464 y2
+
+# qhasm: int6464 y3
+
+# qhasm: int6464 y4
+
+# qhasm: int6464 y5
+
+# qhasm: int6464 y6
+
+# qhasm: int6464 y7
+
+# qhasm: int6464 y8
+
+# qhasm: int6464 y9
+
+# qhasm: int6464 y10
+
+# qhasm: int6464 y11
+
+# qhasm: int6464 y12
+
+# qhasm: int6464 y13
+
+# qhasm: int6464 y14
+
+# qhasm: int6464 y15
+
+# qhasm: int6464 r0
+
+# qhasm: int6464 r1
+
+# qhasm: int6464 r2
+
+# qhasm: int6464 r3
+
+# qhasm: int6464 r4
+
+# qhasm: int6464 r5
+
+# qhasm: int6464 r6
+
+# qhasm: int6464 r7
+
+# qhasm: int6464 r8
+
+# qhasm: int6464 r9
+
+# qhasm: int6464 r10
+
+# qhasm: int6464 r11
+
+# qhasm: int6464 r12
+
+# qhasm: int6464 r13
+
+# qhasm: int6464 r14
+
+# qhasm: int6464 r15
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int64 in0
+
+# qhasm: int64 in1
+
+# qhasm: int64 in2
+
+# qhasm: int64 in3
+
+# qhasm: int64 in4
+
+# qhasm: int64 in5
+
+# qhasm: int64 in6
+
+# qhasm: int64 in7
+
+# qhasm: int64 in8
+
+# qhasm: int64 in9
+
+# qhasm: int64 in10
+
+# qhasm: int64 in11
+
+# qhasm: int64 in12
+
+# qhasm: int64 in13
+
+# qhasm: int64 in14
+
+# qhasm: int64 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: int64 ctarget
+
+# qhasm: stack64 bytes_backup
+
+# qhasm: enter crypto_stream_salsa20_amd64_xmm6
+.text
+.p2align 5
+.globl _crypto_stream_salsa20_amd64_xmm6
+.globl crypto_stream_salsa20_amd64_xmm6
+_crypto_stream_salsa20_amd64_xmm6:
+crypto_stream_salsa20_amd64_xmm6:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: bytes = arg2
+# asm 1: mov  <arg2=int64#2,>bytes=int64#6
+# asm 2: mov  <arg2=%rsi,>bytes=%r9
+mov  %rsi,%r9
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm: iv = arg3
+# asm 1: mov  <arg3=int64#3,>iv=int64#3
+# asm 2: mov  <arg3=%rdx,>iv=%rdx
+mov  %rdx,%rdx
+
+# qhasm: k = arg4
+# asm 1: mov  <arg4=int64#4,>k=int64#8
+# asm 2: mov  <arg4=%rcx,>k=%r10
+mov  %rcx,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int64#7
+# asm 2: mov  $0,>a=%rax
+mov  $0,%rax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: sub  <bytes=int64#6,<out=int64#1
+# asm 2: sub  <bytes=%r9,<out=%rdi
+sub  %r9,%rdi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa20_amd64_xmm6_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa20_amd64_xmm6_xor
+.globl crypto_stream_salsa20_amd64_xmm6_xor
+_crypto_stream_salsa20_amd64_xmm6_xor:
+crypto_stream_salsa20_amd64_xmm6_xor:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = arg2
+# asm 1: mov  <arg2=int64#2,>m=int64#2
+# asm 2: mov  <arg2=%rsi,>m=%rsi
+mov  %rsi,%rsi
+
+# qhasm: bytes = arg3
+# asm 1: mov  <arg3=int64#3,>bytes=int64#6
+# asm 2: mov  <arg3=%rdx,>bytes=%r9
+mov  %rdx,%r9
+
+# qhasm: iv = arg4
+# asm 1: mov  <arg4=int64#4,>iv=int64#3
+# asm 2: mov  <arg4=%rcx,>iv=%rdx
+mov  %rcx,%rdx
+
+# qhasm: k = arg5
+# asm 1: mov  <arg5=int64#5,>k=int64#8
+# asm 2: mov  <arg5=%r8,>k=%r10
+mov  %r8,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl   20(<k=int64#8),>in12=int64#4d
+# asm 2: movl   20(<k=%r10),>in12=%ecx
+movl   20(%r10),%ecx
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl   0(<k=int64#8),>in1=int64#5d
+# asm 2: movl   0(<k=%r10),>in1=%r8d
+movl   0(%r10),%r8d
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl   0(<iv=int64#3),>in6=int64#7d
+# asm 2: movl   0(<iv=%rdx),>in6=%eax
+movl   0(%rdx),%eax
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl   16(<k=int64#8),>in11=int64#9d
+# asm 2: movl   16(<k=%r10),>in11=%r11d
+movl   16(%r10),%r11d
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int64#4d,>x1=stack128#1
+# asm 2: movl <in12=%ecx,>x1=0(%rsp)
+movl %ecx,0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int64#5d,4+<x1=stack128#1
+# asm 2: movl <in1=%r8d,4+<x1=0(%rsp)
+movl %r8d,4+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int64#7d,8+<x1=stack128#1
+# asm 2: movl <in6=%eax,8+<x1=0(%rsp)
+movl %eax,8+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int64#9d,12+<x1=stack128#1
+# asm 2: movl <in11=%r11d,12+<x1=0(%rsp)
+movl %r11d,12+0(%rsp)
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int64#4
+# asm 2: mov  $0,>in8=%rcx
+mov  $0,%rcx
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl   24(<k=int64#8),>in13=int64#5d
+# asm 2: movl   24(<k=%r10),>in13=%r8d
+movl   24(%r10),%r8d
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl   4(<k=int64#8),>in2=int64#7d
+# asm 2: movl   4(<k=%r10),>in2=%eax
+movl   4(%r10),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl   4(<iv=int64#3),>in7=int64#3d
+# asm 2: movl   4(<iv=%rdx),>in7=%edx
+movl   4(%rdx),%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int64#5d,4+<x2=stack128#2
+# asm 2: movl <in13=%r8d,4+<x2=16(%rsp)
+movl %r8d,4+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int64#7d,8+<x2=stack128#2
+# asm 2: movl <in2=%eax,8+<x2=16(%rsp)
+movl %eax,8+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int64#3d,12+<x2=stack128#2
+# asm 2: movl <in7=%edx,12+<x2=16(%rsp)
+movl %edx,12+16(%rsp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl   12(<k=int64#8),>in4=int64#3d
+# asm 2: movl   12(<k=%r10),>in4=%edx
+movl   12(%r10),%edx
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int64#4
+# asm 2: mov  $0,>in9=%rcx
+mov  $0,%rcx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl   28(<k=int64#8),>in14=int64#5d
+# asm 2: movl   28(<k=%r10),>in14=%r8d
+movl   28(%r10),%r8d
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl   8(<k=int64#8),>in3=int64#7d
+# asm 2: movl   8(<k=%r10),>in3=%eax
+movl   8(%r10),%eax
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int64#3d,>x3=stack128#3
+# asm 2: movl <in4=%edx,>x3=32(%rsp)
+movl %edx,32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int64#5d,8+<x3=stack128#3
+# asm 2: movl <in14=%r8d,8+<x3=32(%rsp)
+movl %r8d,8+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int64#7d,12+<x3=stack128#3
+# asm 2: movl <in3=%eax,12+<x3=32(%rsp)
+movl %eax,12+32(%rsp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int64#3
+# asm 2: mov  $1634760805,>in0=%rdx
+mov  $1634760805,%rdx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int64#4
+# asm 2: mov  $857760878,>in5=%rcx
+mov  $857760878,%rcx
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int64#5
+# asm 2: mov  $2036477234,>in10=%r8
+mov  $2036477234,%r8
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int64#7
+# asm 2: mov  $1797285236,>in15=%rax
+mov  $1797285236,%rax
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int64#3d,>x0=stack128#4
+# asm 2: movl <in0=%edx,>x0=48(%rsp)
+movl %edx,48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int64#4d,4+<x0=stack128#4
+# asm 2: movl <in5=%ecx,4+<x0=48(%rsp)
+movl %ecx,4+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int64#5d,8+<x0=stack128#4
+# asm 2: movl <in10=%r8d,8+<x0=48(%rsp)
+movl %r8d,8+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int64#7d,12+<x0=stack128#4
+# asm 2: movl <in15=%eax,12+<x0=48(%rsp)
+movl %eax,12+48(%rsp)
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#4,>z0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>z0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=64(%rsp)
+movdqa %xmm1,64(%rsp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=80(%rsp)
+movdqa %xmm2,80(%rsp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=96(%rsp)
+movdqa %xmm3,96(%rsp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=112(%rsp)
+movdqa %xmm0,112(%rsp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#1,>z1=int6464#1
+# asm 2: movdqa <x1=0(%rsp),>z1=%xmm0
+movdqa 0(%rsp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=128(%rsp)
+movdqa %xmm1,128(%rsp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=144(%rsp)
+movdqa %xmm2,144(%rsp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=160(%rsp)
+movdqa %xmm3,160(%rsp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=176(%rsp)
+movdqa %xmm0,176(%rsp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#2,>z2=int6464#1
+# asm 2: movdqa <x2=16(%rsp),>z2=%xmm0
+movdqa 16(%rsp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=192(%rsp)
+movdqa %xmm1,192(%rsp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=208(%rsp)
+movdqa %xmm2,208(%rsp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=224(%rsp)
+movdqa %xmm0,224(%rsp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#3,>z3=int6464#1
+# asm 2: movdqa <x3=32(%rsp),>z3=%xmm0
+movdqa 32(%rsp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=240(%rsp)
+movdqa %xmm1,240(%rsp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=256(%rsp)
+movdqa %xmm2,256(%rsp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=272(%rsp)
+movdqa %xmm0,272(%rsp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#3d
+# asm 2: movl <x2=16(%rsp),>in8=%edx
+movl 16(%rsp),%edx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#4d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%ecx
+movl 4+32(%rsp),%ecx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int64#3d,>orig8=stack128#19
+# asm 2: movl <in8=%edx,>orig8=288(%rsp)
+movl %edx,288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int64#4d,>orig9=stack128#20
+# asm 2: movl <in9=%ecx,>orig9=304(%rsp)
+movl %ecx,304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int64#3d,4+<orig8=stack128#19
+# asm 2: movl <in8=%edx,4+<orig8=288(%rsp)
+movl %edx,4+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,4+<orig9=304(%rsp)
+movl %ecx,4+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int64#3d,8+<orig8=stack128#19
+# asm 2: movl <in8=%edx,8+<orig8=288(%rsp)
+movl %edx,8+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int64#4d,8+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,8+<orig9=304(%rsp)
+movl %ecx,8+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int64#3d,12+<orig8=stack128#19
+# asm 2: movl <in8=%edx,12+<orig8=288(%rsp)
+movl %edx,12+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int64#4d,12+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,12+<orig9=304(%rsp)
+movl %ecx,12+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#3d,>x2=stack128#2
+# asm 2: movl <in8=%edx,>x2=16(%rsp)
+movl %edx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: i = 20
+# asm 1: mov  $20,>i=int64#3
+# asm 2: mov  $20,>i=%rdx
+mov  $20,%rdx
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=64(%rsp),>z5=%xmm0
+movdqa 64(%rsp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=80(%rsp),>z10=%xmm1
+movdqa 80(%rsp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=96(%rsp),>z15=%xmm2
+movdqa 96(%rsp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=256(%rsp),>z14=%xmm3
+movdqa 256(%rsp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=272(%rsp),>z3=%xmm4
+movdqa 272(%rsp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=128(%rsp),>z6=%xmm5
+movdqa 128(%rsp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=144(%rsp),>z11=%xmm6
+movdqa 144(%rsp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=176(%rsp),>z1=%xmm7
+movdqa 176(%rsp),%xmm7
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#9
+# asm 2: movdqa <orig7=192(%rsp),>z7=%xmm8
+movdqa 192(%rsp),%xmm8
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#10
+# asm 2: movdqa <orig13=208(%rsp),>z13=%xmm9
+movdqa 208(%rsp),%xmm9
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#11
+# asm 2: movdqa <orig2=224(%rsp),>z2=%xmm10
+movdqa 224(%rsp),%xmm10
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#12
+# asm 2: movdqa <orig9=304(%rsp),>z9=%xmm11
+movdqa 304(%rsp),%xmm11
+
+# qhasm:   z0 = orig0
+# asm 1: movdqa <orig0=stack128#8,>z0=int6464#13
+# asm 2: movdqa <orig0=112(%rsp),>z0=%xmm12
+movdqa 112(%rsp),%xmm12
+
+# qhasm:   z12 = orig12
+# asm 1: movdqa <orig12=stack128#11,>z12=int6464#14
+# asm 2: movdqa <orig12=160(%rsp),>z12=%xmm13
+movdqa 160(%rsp),%xmm13
+
+# qhasm:   z4 = orig4
+# asm 1: movdqa <orig4=stack128#16,>z4=int6464#15
+# asm 2: movdqa <orig4=240(%rsp),>z4=%xmm14
+movdqa 240(%rsp),%xmm14
+
+# qhasm:   z8 = orig8
+# asm 1: movdqa <orig8=stack128#19,>z8=int6464#16
+# asm 2: movdqa <orig8=288(%rsp),>z8=%xmm15
+movdqa 288(%rsp),%xmm15
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                y4 = z12
+# asm 1: movdqa <z12=int6464#14,>y4=int6464#2
+# asm 2: movdqa <z12=%xmm13,>y4=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm: uint32323232   y4 += z0
+# asm 1: paddd <z0=int6464#13,<y4=int6464#2
+# asm 2: paddd <z0=%xmm12,<y4=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y4 <<= 7
+# asm 1: pslld $7,<y4=int6464#2
+# asm 2: pslld $7,<y4=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232   r4 >>= 25
+# asm 1: psrld $25,<r4=int6464#3
+# asm 2: psrld $25,<r4=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                                y9 = z1
+# asm 1: movdqa <z1=int6464#8,>y9=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y9=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232                   y9 += z5
+# asm 1: paddd <z5=int6464#1,<y9=int6464#2
+# asm 2: paddd <z5=%xmm0,<y9=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                                r9 = y9
+# asm 1: movdqa <y9=int6464#2,>r9=int6464#3
+# asm 2: movdqa <y9=%xmm1,>r9=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y9 <<= 7
+# asm 1: pslld $7,<y9=int6464#2
+# asm 2: pslld $7,<y9=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#2,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm1,<z9=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm: uint32323232                   r9 >>= 25
+# asm 1: psrld $25,<r9=int6464#3
+# asm 2: psrld $25,<r9=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#3,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm2,<z9=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:                y8 = z0
+# asm 1: movdqa <z0=int6464#13,>y8=int6464#2
+# asm 2: movdqa <z0=%xmm12,>y8=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232   y8 += z4
+# asm 1: paddd <z4=int6464#15,<y8=int6464#2
+# asm 2: paddd <z4=%xmm14,<y8=%xmm1
+paddd %xmm14,%xmm1
+
+# qhasm:                r8 = y8
+# asm 1: movdqa <y8=int6464#2,>r8=int6464#3
+# asm 2: movdqa <y8=%xmm1,>r8=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#2
+# asm 2: pslld $9,<y8=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z8 ^= y8
+# asm 1: pxor  <y8=int6464#2,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm1,<z8=%xmm15
+pxor  %xmm1,%xmm15
+
+# qhasm: uint32323232   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#3
+# asm 2: psrld $23,<r8=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z8 ^= r8
+# asm 1: pxor  <r8=int6464#3,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm2,<z8=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:                                y13 = z5
+# asm 1: movdqa <z5=int6464#1,>y13=int6464#2
+# asm 2: movdqa <z5=%xmm0,>y13=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                   y13 += z9
+# asm 1: paddd <z9=int6464#12,<y13=int6464#2
+# asm 2: paddd <z9=%xmm11,<y13=%xmm1
+paddd %xmm11,%xmm1
+
+# qhasm:                                r13 = y13
+# asm 1: movdqa <y13=int6464#2,>r13=int6464#3
+# asm 2: movdqa <y13=%xmm1,>r13=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#2
+# asm 2: pslld $9,<y13=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#2,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm1,<z13=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm: uint32323232                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#3
+# asm 2: psrld $23,<r13=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#3,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm2,<z13=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:                y12 = z4
+# asm 1: movdqa <z4=int6464#15,>y12=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y12=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232   y12 += z8
+# asm 1: paddd <z8=int6464#16,<y12=int6464#2
+# asm 2: paddd <z8=%xmm15,<y12=%xmm1
+paddd %xmm15,%xmm1
+
+# qhasm:                r12 = y12
+# asm 1: movdqa <y12=int6464#2,>r12=int6464#3
+# asm 2: movdqa <y12=%xmm1,>r12=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y12 <<= 13
+# asm 1: pslld $13,<y12=int6464#2
+# asm 2: pslld $13,<y12=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z12 ^= y12
+# asm 1: pxor  <y12=int6464#2,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm1,<z12=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm: uint32323232   r12 >>= 19
+# asm 1: psrld $19,<r12=int6464#3
+# asm 2: psrld $19,<r12=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z12 ^= r12
+# asm 1: pxor  <r12=int6464#3,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm2,<z12=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:                                y1 = z9
+# asm 1: movdqa <z9=int6464#12,>y1=int6464#2
+# asm 2: movdqa <z9=%xmm11,>y1=%xmm1
+movdqa %xmm11,%xmm1
+
+# qhasm: uint32323232                   y1 += z13
+# asm 1: paddd <z13=int6464#10,<y1=int6464#2
+# asm 2: paddd <z13=%xmm9,<y1=%xmm1
+paddd %xmm9,%xmm1
+
+# qhasm:                                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#3
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y1 <<= 13
+# asm 1: pslld $13,<y1=int6464#2
+# asm 2: pslld $13,<y1=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232                   r1 >>= 19
+# asm 1: psrld $19,<r1=int6464#3
+# asm 2: psrld $19,<r1=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z1 ^= r1
+# asm 1: pxor  <r1=int6464#3,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm2,<z1=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:                y0 = z8
+# asm 1: movdqa <z8=int6464#16,>y0=int6464#2
+# asm 2: movdqa <z8=%xmm15,>y0=%xmm1
+movdqa %xmm15,%xmm1
+
+# qhasm: uint32323232   y0 += z12
+# asm 1: paddd <z12=int6464#14,<y0=int6464#2
+# asm 2: paddd <z12=%xmm13,<y0=%xmm1
+paddd %xmm13,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#13
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm12
+pxor  %xmm1,%xmm12
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#13
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#13,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm12,>z0_stack=320(%rsp)
+movdqa %xmm12,320(%rsp)
+
+# qhasm:                                y5 = z13
+# asm 1: movdqa <z13=int6464#10,>y5=int6464#3
+# asm 2: movdqa <z13=%xmm9,>y5=%xmm2
+movdqa %xmm9,%xmm2
+
+# qhasm: uint32323232                   y5 += z1
+# asm 1: paddd <z1=int6464#8,<y5=int6464#3
+# asm 2: paddd <z1=%xmm7,<y5=%xmm2
+paddd %xmm7,%xmm2
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#3,>r5=int6464#13
+# asm 2: movdqa <y5=%xmm2,>r5=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#3
+# asm 2: pslld $18,<y5=%xmm2
+pslld $18,%xmm2
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#3,<z5=int6464#1
+# asm 2: pxor  <y5=%xmm2,<z5=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#13
+# asm 2: psrld $14,<r5=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#13,<z5=int6464#1
+# asm 2: pxor  <r5=%xmm12,<z5=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:                                                y14 = z6
+# asm 1: movdqa <z6=int6464#6,>y14=int6464#3
+# asm 2: movdqa <z6=%xmm5,>y14=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm: uint32323232                                   y14 += z10
+# asm 1: paddd <z10=int6464#2,<y14=int6464#3
+# asm 2: paddd <z10=%xmm1,<y14=%xmm2
+paddd %xmm1,%xmm2
+
+# qhasm:                                                r14 = y14
+# asm 1: movdqa <y14=int6464#3,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm2,>r14=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                                   y14 <<= 7
+# asm 1: pslld $7,<y14=int6464#3
+# asm 2: pslld $7,<y14=%xmm2
+pslld $7,%xmm2
+
+# qhasm:                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#3,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm2,<z14=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                   r14 >>= 25
+# asm 1: psrld $25,<r14=int6464#13
+# asm 2: psrld $25,<r14=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm0,>z5_stack=336(%rsp)
+movdqa %xmm0,336(%rsp)
+
+# qhasm:                                                                y3 = z11
+# asm 1: movdqa <z11=int6464#7,>y3=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y3=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                                   y3 += z15
+# asm 1: paddd <z15=int6464#3,<y3=int6464#1
+# asm 2: paddd <z15=%xmm2,<y3=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r3 = y3
+# asm 1: movdqa <y3=int6464#1,>r3=int6464#13
+# asm 2: movdqa <y3=%xmm0,>r3=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y3 <<= 7
+# asm 1: pslld $7,<y3=int6464#1
+# asm 2: pslld $7,<y3=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z3 ^= y3
+# asm 1: pxor  <y3=int6464#1,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm0,<z3=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm: uint32323232                                                   r3 >>= 25
+# asm 1: psrld $25,<r3=int6464#13
+# asm 2: psrld $25,<r3=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z3 ^= r3
+# asm 1: pxor  <r3=int6464#13,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm12,<z3=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:                                                y2 = z10
+# asm 1: movdqa <z10=int6464#2,>y2=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y2=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y2 += z14
+# asm 1: paddd <z14=int6464#4,<y2=int6464#1
+# asm 2: paddd <z14=%xmm3,<y2=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                r2 = y2
+# asm 1: movdqa <y2=int6464#1,>r2=int6464#13
+# asm 2: movdqa <y2=%xmm0,>r2=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#1
+# asm 2: pslld $9,<y2=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z2 ^= y2
+# asm 1: pxor  <y2=int6464#1,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm0,<z2=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm: uint32323232                                   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#13
+# asm 2: psrld $23,<r2=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z2 ^= r2
+# asm 1: pxor  <r2=int6464#13,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm12,<z2=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:                                                                y7 = z15
+# asm 1: movdqa <z15=int6464#3,>y7=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y7=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y7 += z3
+# asm 1: paddd <z3=int6464#5,<y7=int6464#1
+# asm 2: paddd <z3=%xmm4,<y7=%xmm0
+paddd %xmm4,%xmm0
+
+# qhasm:                                                                r7 = y7
+# asm 1: movdqa <y7=int6464#1,>r7=int6464#13
+# asm 2: movdqa <y7=%xmm0,>r7=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#1
+# asm 2: pslld $9,<y7=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#1,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm0,<z7=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm: uint32323232                                                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#13
+# asm 2: psrld $23,<r7=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#13,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm12,<z7=%xmm8
+pxor  %xmm12,%xmm8
+
+# qhasm:                                                y6 = z14
+# asm 1: movdqa <z14=int6464#4,>y6=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y6=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                   y6 += z2
+# asm 1: paddd <z2=int6464#11,<y6=int6464#1
+# asm 2: paddd <z2=%xmm10,<y6=%xmm0
+paddd %xmm10,%xmm0
+
+# qhasm:                                                r6 = y6
+# asm 1: movdqa <y6=int6464#1,>r6=int6464#13
+# asm 2: movdqa <y6=%xmm0,>r6=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y6 <<= 13
+# asm 1: pslld $13,<y6=int6464#1
+# asm 2: pslld $13,<y6=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#1,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm0,<z6=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                   r6 >>= 19
+# asm 1: psrld $19,<r6=int6464#13
+# asm 2: psrld $19,<r6=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#13,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm12,<z6=%xmm5
+pxor  %xmm12,%xmm5
+
+# qhasm:                                                                y11 = z3
+# asm 1: movdqa <z3=int6464#5,>y11=int6464#1
+# asm 2: movdqa <z3=%xmm4,>y11=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                   y11 += z7
+# asm 1: paddd <z7=int6464#9,<y11=int6464#1
+# asm 2: paddd <z7=%xmm8,<y11=%xmm0
+paddd %xmm8,%xmm0
+
+# qhasm:                                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#13
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y11 <<= 13
+# asm 1: pslld $13,<y11=int6464#1
+# asm 2: pslld $13,<y11=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                                   r11 >>= 19
+# asm 1: psrld $19,<r11=int6464#13
+# asm 2: psrld $19,<r11=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#13,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm12,<z11=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:                                                y10 = z2
+# asm 1: movdqa <z2=int6464#11,>y10=int6464#1
+# asm 2: movdqa <z2=%xmm10,>y10=%xmm0
+movdqa %xmm10,%xmm0
+
+# qhasm: uint32323232                                   y10 += z6
+# asm 1: paddd <z6=int6464#6,<y10=int6464#1
+# asm 2: paddd <z6=%xmm5,<y10=%xmm0
+paddd %xmm5,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#1
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm0
+movdqa 320(%rsp),%xmm0
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                y1 = z3
+# asm 1: movdqa <z3=int6464#5,>y1=int6464#2
+# asm 2: movdqa <z3=%xmm4,>y1=%xmm1
+movdqa %xmm4,%xmm1
+
+# qhasm: uint32323232   y1 += z0
+# asm 1: paddd <z0=int6464#1,<y1=int6464#2
+# asm 2: paddd <z0=%xmm0,<y1=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#13
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232   y1 <<= 7
+# asm 1: pslld $7,<y1=int6464#2
+# asm 2: pslld $7,<y1=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232   r1 >>= 25
+# asm 1: psrld $25,<r1=int6464#13
+# asm 2: psrld $25,<r1=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                z1 ^= r1
+# asm 1: pxor  <r1=int6464#13,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm12,<z1=%xmm7
+pxor  %xmm12,%xmm7
+
+# qhasm:                                                                y15 = z7
+# asm 1: movdqa <z7=int6464#9,>y15=int6464#2
+# asm 2: movdqa <z7=%xmm8,>y15=%xmm1
+movdqa %xmm8,%xmm1
+
+# qhasm: uint32323232                                                   y15 += z11
+# asm 1: paddd <z11=int6464#7,<y15=int6464#2
+# asm 2: paddd <z11=%xmm6,<y15=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#2,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm1,>r15=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#2
+# asm 2: pslld $18,<y15=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#2,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm1,<z15=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#13
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm12
+movdqa 336(%rsp),%xmm12
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                                y6 = z4
+# asm 1: movdqa <z4=int6464#15,>y6=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y6=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232                   y6 += z5
+# asm 1: paddd <z5=int6464#13,<y6=int6464#2
+# asm 2: paddd <z5=%xmm12,<y6=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                                r6 = y6
+# asm 1: movdqa <y6=int6464#2,>r6=int6464#3
+# asm 2: movdqa <y6=%xmm1,>r6=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y6 <<= 7
+# asm 1: pslld $7,<y6=int6464#2
+# asm 2: pslld $7,<y6=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#2,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm1,<z6=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm: uint32323232                   r6 >>= 25
+# asm 1: psrld $25,<r6=int6464#3
+# asm 2: psrld $25,<r6=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#3,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm2,<z6=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:                y2 = z0
+# asm 1: movdqa <z0=int6464#1,>y2=int6464#2
+# asm 2: movdqa <z0=%xmm0,>y2=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232   y2 += z1
+# asm 1: paddd <z1=int6464#8,<y2=int6464#2
+# asm 2: paddd <z1=%xmm7,<y2=%xmm1
+paddd %xmm7,%xmm1
+
+# qhasm:                r2 = y2
+# asm 1: movdqa <y2=int6464#2,>r2=int6464#3
+# asm 2: movdqa <y2=%xmm1,>r2=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#2
+# asm 2: pslld $9,<y2=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z2 ^= y2
+# asm 1: pxor  <y2=int6464#2,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm1,<z2=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm: uint32323232   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#3
+# asm 2: psrld $23,<r2=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z2 ^= r2
+# asm 1: pxor  <r2=int6464#3,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm2,<z2=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:                                y7 = z5
+# asm 1: movdqa <z5=int6464#13,>y7=int6464#2
+# asm 2: movdqa <z5=%xmm12,>y7=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232                   y7 += z6
+# asm 1: paddd <z6=int6464#6,<y7=int6464#2
+# asm 2: paddd <z6=%xmm5,<y7=%xmm1
+paddd %xmm5,%xmm1
+
+# qhasm:                                r7 = y7
+# asm 1: movdqa <y7=int6464#2,>r7=int6464#3
+# asm 2: movdqa <y7=%xmm1,>r7=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#2
+# asm 2: pslld $9,<y7=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#2,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm1,<z7=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm: uint32323232                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#3
+# asm 2: psrld $23,<r7=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#3,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm2,<z7=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:                y3 = z1
+# asm 1: movdqa <z1=int6464#8,>y3=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y3=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232   y3 += z2
+# asm 1: paddd <z2=int6464#11,<y3=int6464#2
+# asm 2: paddd <z2=%xmm10,<y3=%xmm1
+paddd %xmm10,%xmm1
+
+# qhasm:                r3 = y3
+# asm 1: movdqa <y3=int6464#2,>r3=int6464#3
+# asm 2: movdqa <y3=%xmm1,>r3=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y3 <<= 13
+# asm 1: pslld $13,<y3=int6464#2
+# asm 2: pslld $13,<y3=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z3 ^= y3
+# asm 1: pxor  <y3=int6464#2,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm1,<z3=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm: uint32323232   r3 >>= 19
+# asm 1: psrld $19,<r3=int6464#3
+# asm 2: psrld $19,<r3=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z3 ^= r3
+# asm 1: pxor  <r3=int6464#3,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm2,<z3=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:                                y4 = z6
+# asm 1: movdqa <z6=int6464#6,>y4=int6464#2
+# asm 2: movdqa <z6=%xmm5,>y4=%xmm1
+movdqa %xmm5,%xmm1
+
+# qhasm: uint32323232                   y4 += z7
+# asm 1: paddd <z7=int6464#9,<y4=int6464#2
+# asm 2: paddd <z7=%xmm8,<y4=%xmm1
+paddd %xmm8,%xmm1
+
+# qhasm:                                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y4 <<= 13
+# asm 1: pslld $13,<y4=int6464#2
+# asm 2: pslld $13,<y4=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232                   r4 >>= 19
+# asm 1: psrld $19,<r4=int6464#3
+# asm 2: psrld $19,<r4=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                y0 = z2
+# asm 1: movdqa <z2=int6464#11,>y0=int6464#2
+# asm 2: movdqa <z2=%xmm10,>y0=%xmm1
+movdqa %xmm10,%xmm1
+
+# qhasm: uint32323232   y0 += z3
+# asm 1: paddd <z3=int6464#5,<y0=int6464#2
+# asm 2: paddd <z3=%xmm4,<y0=%xmm1
+paddd %xmm4,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#1
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#1
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm0,>z0_stack=320(%rsp)
+movdqa %xmm0,320(%rsp)
+
+# qhasm:                                y5 = z7
+# asm 1: movdqa <z7=int6464#9,>y5=int6464#1
+# asm 2: movdqa <z7=%xmm8,>y5=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm: uint32323232                   y5 += z4
+# asm 1: paddd <z4=int6464#15,<y5=int6464#1
+# asm 2: paddd <z4=%xmm14,<y5=%xmm0
+paddd %xmm14,%xmm0
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#1,>r5=int6464#3
+# asm 2: movdqa <y5=%xmm0,>r5=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#1
+# asm 2: pslld $18,<y5=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#1,<z5=int6464#13
+# asm 2: pxor  <y5=%xmm0,<z5=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#3
+# asm 2: psrld $14,<r5=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#3,<z5=int6464#13
+# asm 2: pxor  <r5=%xmm2,<z5=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                y11 = z9
+# asm 1: movdqa <z9=int6464#12,>y11=int6464#1
+# asm 2: movdqa <z9=%xmm11,>y11=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm: uint32323232                                   y11 += z10
+# asm 1: paddd <z10=int6464#2,<y11=int6464#1
+# asm 2: paddd <z10=%xmm1,<y11=%xmm0
+paddd %xmm1,%xmm0
+
+# qhasm:                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#3
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                   y11 <<= 7
+# asm 1: pslld $7,<y11=int6464#1
+# asm 2: pslld $7,<y11=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                   r11 >>= 25
+# asm 1: psrld $25,<r11=int6464#3
+# asm 2: psrld $25,<r11=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#3,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm2,<z11=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#13,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm12,>z5_stack=336(%rsp)
+movdqa %xmm12,336(%rsp)
+
+# qhasm:                                                                y12 = z14
+# asm 1: movdqa <z14=int6464#4,>y12=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y12=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                   y12 += z15
+# asm 1: paddd <z15=int6464#3,<y12=int6464#1
+# asm 2: paddd <z15=%xmm2,<y12=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r12 = y12
+# asm 1: movdqa <y12=int6464#1,>r12=int6464#13
+# asm 2: movdqa <y12=%xmm0,>r12=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y12 <<= 7
+# asm 1: pslld $7,<y12=int6464#1
+# asm 2: pslld $7,<y12=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z12 ^= y12
+# asm 1: pxor  <y12=int6464#1,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm0,<z12=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm: uint32323232                                                   r12 >>= 25
+# asm 1: psrld $25,<r12=int6464#13
+# asm 2: psrld $25,<r12=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z12 ^= r12
+# asm 1: pxor  <r12=int6464#13,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm12,<z12=%xmm13
+pxor  %xmm12,%xmm13
+
+# qhasm:                                                y8 = z10
+# asm 1: movdqa <z10=int6464#2,>y8=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y8=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y8 += z11
+# asm 1: paddd <z11=int6464#7,<y8=int6464#1
+# asm 2: paddd <z11=%xmm6,<y8=%xmm0
+paddd %xmm6,%xmm0
+
+# qhasm:                                                r8 = y8
+# asm 1: movdqa <y8=int6464#1,>r8=int6464#13
+# asm 2: movdqa <y8=%xmm0,>r8=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#1
+# asm 2: pslld $9,<y8=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z8 ^= y8
+# asm 1: pxor  <y8=int6464#1,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm0,<z8=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm: uint32323232                                   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#13
+# asm 2: psrld $23,<r8=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z8 ^= r8
+# asm 1: pxor  <r8=int6464#13,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm12,<z8=%xmm15
+pxor  %xmm12,%xmm15
+
+# qhasm:                                                                y13 = z15
+# asm 1: movdqa <z15=int6464#3,>y13=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y13=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y13 += z12
+# asm 1: paddd <z12=int6464#14,<y13=int6464#1
+# asm 2: paddd <z12=%xmm13,<y13=%xmm0
+paddd %xmm13,%xmm0
+
+# qhasm:                                                                r13 = y13
+# asm 1: movdqa <y13=int6464#1,>r13=int6464#13
+# asm 2: movdqa <y13=%xmm0,>r13=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#1
+# asm 2: pslld $9,<y13=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#1,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm0,<z13=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm: uint32323232                                                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#13
+# asm 2: psrld $23,<r13=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#13,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm12,<z13=%xmm9
+pxor  %xmm12,%xmm9
+
+# qhasm:                                                y9 = z11
+# asm 1: movdqa <z11=int6464#7,>y9=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y9=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                   y9 += z8
+# asm 1: paddd <z8=int6464#16,<y9=int6464#1
+# asm 2: paddd <z8=%xmm15,<y9=%xmm0
+paddd %xmm15,%xmm0
+
+# qhasm:                                                r9 = y9
+# asm 1: movdqa <y9=int6464#1,>r9=int6464#13
+# asm 2: movdqa <y9=%xmm0,>r9=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y9 <<= 13
+# asm 1: pslld $13,<y9=int6464#1
+# asm 2: pslld $13,<y9=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#1,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm0,<z9=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm: uint32323232                                   r9 >>= 19
+# asm 1: psrld $19,<r9=int6464#13
+# asm 2: psrld $19,<r9=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#13,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm12,<z9=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:                                                                y14 = z12
+# asm 1: movdqa <z12=int6464#14,>y14=int6464#1
+# asm 2: movdqa <z12=%xmm13,>y14=%xmm0
+movdqa %xmm13,%xmm0
+
+# qhasm: uint32323232                                                   y14 += z13
+# asm 1: paddd <z13=int6464#10,<y14=int6464#1
+# asm 2: paddd <z13=%xmm9,<y14=%xmm0
+paddd %xmm9,%xmm0
+
+# qhasm:                                                                r14 = y14
+# asm 1: movdqa <y14=int6464#1,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm0,>r14=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y14 <<= 13
+# asm 1: pslld $13,<y14=int6464#1
+# asm 2: pslld $13,<y14=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#1,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm0,<z14=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm: uint32323232                                                   r14 >>= 19
+# asm 1: psrld $19,<r14=int6464#13
+# asm 2: psrld $19,<r14=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                y10 = z8
+# asm 1: movdqa <z8=int6464#16,>y10=int6464#1
+# asm 2: movdqa <z8=%xmm15,>y10=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm: uint32323232                                   y10 += z9
+# asm 1: paddd <z9=int6464#12,<y10=int6464#1
+# asm 2: paddd <z9=%xmm11,<y10=%xmm0
+paddd %xmm11,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                                                                y15 = z13
+# asm 1: movdqa <z13=int6464#10,>y15=int6464#1
+# asm 2: movdqa <z13=%xmm9,>y15=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm: uint32323232                                                   y15 += z14
+# asm 1: paddd <z14=int6464#4,<y15=int6464#1
+# asm 2: paddd <z14=%xmm3,<y15=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#1,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm0,>r15=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#1
+# asm 2: pslld $18,<y15=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#1,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm0,<z15=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#13
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm12
+movdqa 320(%rsp),%xmm12
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#1
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm0
+movdqa 336(%rsp),%xmm0
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int64#3
+# asm 2: sub  $2,<i=%rdx
+sub  $2,%rdx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#13
+# asm 2: paddd <orig0=112(%rsp),<z0=%xmm12
+paddd 112(%rsp),%xmm12
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#8
+# asm 2: paddd <orig1=176(%rsp),<z1=%xmm7
+paddd 176(%rsp),%xmm7
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#11
+# asm 2: paddd <orig2=224(%rsp),<z2=%xmm10
+paddd 224(%rsp),%xmm10
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#5
+# asm 2: paddd <orig3=272(%rsp),<z3=%xmm4
+paddd 272(%rsp),%xmm4
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 0(<m=%rsi),<in0=%edx
+xorl 0(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 4(<m=%rsi),<in1=%ecx
+xorl 4(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 8(<m=%rsi),<in2=%r8d
+xorl 8(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 12(<m=%rsi),<in3=%r9d
+xorl 12(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#3d,0(<out=int64#1)
+# asm 2: movl   <in0=%edx,0(<out=%rdi)
+movl   %edx,0(%rdi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#4d,4(<out=int64#1)
+# asm 2: movl   <in1=%ecx,4(<out=%rdi)
+movl   %ecx,4(%rdi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#5d,8(<out=int64#1)
+# asm 2: movl   <in2=%r8d,8(<out=%rdi)
+movl   %r8d,8(%rdi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#6d,12(<out=int64#1)
+# asm 2: movl   <in3=%r9d,12(<out=%rdi)
+movl   %r9d,12(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 64(<m=%rsi),<in0=%edx
+xorl 64(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 68(<m=%rsi),<in1=%ecx
+xorl 68(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 72(<m=%rsi),<in2=%r8d
+xorl 72(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 76(<m=%rsi),<in3=%r9d
+xorl 76(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl   <in0=int64#3d,64(<out=int64#1)
+# asm 2: movl   <in0=%edx,64(<out=%rdi)
+movl   %edx,64(%rdi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl   <in1=int64#4d,68(<out=int64#1)
+# asm 2: movl   <in1=%ecx,68(<out=%rdi)
+movl   %ecx,68(%rdi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl   <in2=int64#5d,72(<out=int64#1)
+# asm 2: movl   <in2=%r8d,72(<out=%rdi)
+movl   %r8d,72(%rdi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl   <in3=int64#6d,76(<out=int64#1)
+# asm 2: movl   <in3=%r9d,76(<out=%rdi)
+movl   %r9d,76(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 128(<m=%rsi),<in0=%edx
+xorl 128(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 132(<m=%rsi),<in1=%ecx
+xorl 132(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 136(<m=%rsi),<in2=%r8d
+xorl 136(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 140(<m=%rsi),<in3=%r9d
+xorl 140(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl   <in0=int64#3d,128(<out=int64#1)
+# asm 2: movl   <in0=%edx,128(<out=%rdi)
+movl   %edx,128(%rdi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl   <in1=int64#4d,132(<out=int64#1)
+# asm 2: movl   <in1=%ecx,132(<out=%rdi)
+movl   %ecx,132(%rdi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl   <in2=int64#5d,136(<out=int64#1)
+# asm 2: movl   <in2=%r8d,136(<out=%rdi)
+movl   %r8d,136(%rdi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl   <in3=int64#6d,140(<out=int64#1)
+# asm 2: movl   <in3=%r9d,140(<out=%rdi)
+movl   %r9d,140(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 192(<m=%rsi),<in0=%edx
+xorl 192(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 196(<m=%rsi),<in1=%ecx
+xorl 196(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 200(<m=%rsi),<in2=%r8d
+xorl 200(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 204(<m=%rsi),<in3=%r9d
+xorl 204(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl   <in0=int64#3d,192(<out=int64#1)
+# asm 2: movl   <in0=%edx,192(<out=%rdi)
+movl   %edx,192(%rdi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl   <in1=int64#4d,196(<out=int64#1)
+# asm 2: movl   <in1=%ecx,196(<out=%rdi)
+movl   %ecx,196(%rdi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl   <in2=int64#5d,200(<out=int64#1)
+# asm 2: movl   <in2=%r8d,200(<out=%rdi)
+movl   %r8d,200(%rdi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl   <in3=int64#6d,204(<out=int64#1)
+# asm 2: movl   <in3=%r9d,204(<out=%rdi)
+movl   %r9d,204(%rdi)
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#15
+# asm 2: paddd <orig4=240(%rsp),<z4=%xmm14
+paddd 240(%rsp),%xmm14
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#1
+# asm 2: paddd <orig5=64(%rsp),<z5=%xmm0
+paddd 64(%rsp),%xmm0
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#6
+# asm 2: paddd <orig6=128(%rsp),<z6=%xmm5
+paddd 128(%rsp),%xmm5
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#9
+# asm 2: paddd <orig7=192(%rsp),<z7=%xmm8
+paddd 192(%rsp),%xmm8
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 16(<m=%rsi),<in4=%edx
+xorl 16(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#3d,16(<out=int64#1)
+# asm 2: movl   <in4=%edx,16(<out=%rdi)
+movl   %edx,16(%rdi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 80(<m=%rsi),<in4=%edx
+xorl 80(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 84(<m=%rsi),<in5=%ecx
+xorl 84(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 88(<m=%rsi),<in6=%r8d
+xorl 88(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 92(<m=%rsi),<in7=%r9d
+xorl 92(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl   <in4=int64#3d,80(<out=int64#1)
+# asm 2: movl   <in4=%edx,80(<out=%rdi)
+movl   %edx,80(%rdi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl   <in5=int64#4d,84(<out=int64#1)
+# asm 2: movl   <in5=%ecx,84(<out=%rdi)
+movl   %ecx,84(%rdi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl   <in6=int64#5d,88(<out=int64#1)
+# asm 2: movl   <in6=%r8d,88(<out=%rdi)
+movl   %r8d,88(%rdi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl   <in7=int64#6d,92(<out=int64#1)
+# asm 2: movl   <in7=%r9d,92(<out=%rdi)
+movl   %r9d,92(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 144(<m=%rsi),<in4=%edx
+xorl 144(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 148(<m=%rsi),<in5=%ecx
+xorl 148(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 152(<m=%rsi),<in6=%r8d
+xorl 152(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 156(<m=%rsi),<in7=%r9d
+xorl 156(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl   <in4=int64#3d,144(<out=int64#1)
+# asm 2: movl   <in4=%edx,144(<out=%rdi)
+movl   %edx,144(%rdi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl   <in5=int64#4d,148(<out=int64#1)
+# asm 2: movl   <in5=%ecx,148(<out=%rdi)
+movl   %ecx,148(%rdi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl   <in6=int64#5d,152(<out=int64#1)
+# asm 2: movl   <in6=%r8d,152(<out=%rdi)
+movl   %r8d,152(%rdi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl   <in7=int64#6d,156(<out=int64#1)
+# asm 2: movl   <in7=%r9d,156(<out=%rdi)
+movl   %r9d,156(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 208(<m=%rsi),<in4=%edx
+xorl 208(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 212(<m=%rsi),<in5=%ecx
+xorl 212(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 216(<m=%rsi),<in6=%r8d
+xorl 216(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 220(<m=%rsi),<in7=%r9d
+xorl 220(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl   <in4=int64#3d,208(<out=int64#1)
+# asm 2: movl   <in4=%edx,208(<out=%rdi)
+movl   %edx,208(%rdi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl   <in5=int64#4d,212(<out=int64#1)
+# asm 2: movl   <in5=%ecx,212(<out=%rdi)
+movl   %ecx,212(%rdi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl   <in6=int64#5d,216(<out=int64#1)
+# asm 2: movl   <in6=%r8d,216(<out=%rdi)
+movl   %r8d,216(%rdi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl   <in7=int64#6d,220(<out=int64#1)
+# asm 2: movl   <in7=%r9d,220(<out=%rdi)
+movl   %r9d,220(%rdi)
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#16
+# asm 2: paddd <orig8=288(%rsp),<z8=%xmm15
+paddd 288(%rsp),%xmm15
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#12
+# asm 2: paddd <orig9=304(%rsp),<z9=%xmm11
+paddd 304(%rsp),%xmm11
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#2
+# asm 2: paddd <orig10=80(%rsp),<z10=%xmm1
+paddd 80(%rsp),%xmm1
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#7
+# asm 2: paddd <orig11=144(%rsp),<z11=%xmm6
+paddd 144(%rsp),%xmm6
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 32(<m=%rsi),<in8=%edx
+xorl 32(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 36(<m=%rsi),<in9=%ecx
+xorl 36(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 40(<m=%rsi),<in10=%r8d
+xorl 40(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 44(<m=%rsi),<in11=%r9d
+xorl 44(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#3d,32(<out=int64#1)
+# asm 2: movl   <in8=%edx,32(<out=%rdi)
+movl   %edx,32(%rdi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#4d,36(<out=int64#1)
+# asm 2: movl   <in9=%ecx,36(<out=%rdi)
+movl   %ecx,36(%rdi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#5d,40(<out=int64#1)
+# asm 2: movl   <in10=%r8d,40(<out=%rdi)
+movl   %r8d,40(%rdi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#6d,44(<out=int64#1)
+# asm 2: movl   <in11=%r9d,44(<out=%rdi)
+movl   %r9d,44(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 96(<m=%rsi),<in8=%edx
+xorl 96(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 100(<m=%rsi),<in9=%ecx
+xorl 100(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 104(<m=%rsi),<in10=%r8d
+xorl 104(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 108(<m=%rsi),<in11=%r9d
+xorl 108(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl   <in8=int64#3d,96(<out=int64#1)
+# asm 2: movl   <in8=%edx,96(<out=%rdi)
+movl   %edx,96(%rdi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl   <in9=int64#4d,100(<out=int64#1)
+# asm 2: movl   <in9=%ecx,100(<out=%rdi)
+movl   %ecx,100(%rdi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl   <in10=int64#5d,104(<out=int64#1)
+# asm 2: movl   <in10=%r8d,104(<out=%rdi)
+movl   %r8d,104(%rdi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl   <in11=int64#6d,108(<out=int64#1)
+# asm 2: movl   <in11=%r9d,108(<out=%rdi)
+movl   %r9d,108(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 160(<m=%rsi),<in8=%edx
+xorl 160(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 164(<m=%rsi),<in9=%ecx
+xorl 164(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 168(<m=%rsi),<in10=%r8d
+xorl 168(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 172(<m=%rsi),<in11=%r9d
+xorl 172(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl   <in8=int64#3d,160(<out=int64#1)
+# asm 2: movl   <in8=%edx,160(<out=%rdi)
+movl   %edx,160(%rdi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl   <in9=int64#4d,164(<out=int64#1)
+# asm 2: movl   <in9=%ecx,164(<out=%rdi)
+movl   %ecx,164(%rdi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl   <in10=int64#5d,168(<out=int64#1)
+# asm 2: movl   <in10=%r8d,168(<out=%rdi)
+movl   %r8d,168(%rdi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl   <in11=int64#6d,172(<out=int64#1)
+# asm 2: movl   <in11=%r9d,172(<out=%rdi)
+movl   %r9d,172(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 224(<m=%rsi),<in8=%edx
+xorl 224(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 228(<m=%rsi),<in9=%ecx
+xorl 228(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 232(<m=%rsi),<in10=%r8d
+xorl 232(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 236(<m=%rsi),<in11=%r9d
+xorl 236(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl   <in8=int64#3d,224(<out=int64#1)
+# asm 2: movl   <in8=%edx,224(<out=%rdi)
+movl   %edx,224(%rdi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl   <in9=int64#4d,228(<out=int64#1)
+# asm 2: movl   <in9=%ecx,228(<out=%rdi)
+movl   %ecx,228(%rdi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl   <in10=int64#5d,232(<out=int64#1)
+# asm 2: movl   <in10=%r8d,232(<out=%rdi)
+movl   %r8d,232(%rdi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl   <in11=int64#6d,236(<out=int64#1)
+# asm 2: movl   <in11=%r9d,236(<out=%rdi)
+movl   %r9d,236(%rdi)
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#14
+# asm 2: paddd <orig12=160(%rsp),<z12=%xmm13
+paddd 160(%rsp),%xmm13
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#10
+# asm 2: paddd <orig13=208(%rsp),<z13=%xmm9
+paddd 208(%rsp),%xmm9
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#4
+# asm 2: paddd <orig14=256(%rsp),<z14=%xmm3
+paddd 256(%rsp),%xmm3
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#3
+# asm 2: paddd <orig15=96(%rsp),<z15=%xmm2
+paddd 96(%rsp),%xmm2
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 48(<m=%rsi),<in12=%edx
+xorl 48(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 52(<m=%rsi),<in13=%ecx
+xorl 52(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 56(<m=%rsi),<in14=%r8d
+xorl 56(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 60(<m=%rsi),<in15=%r9d
+xorl 60(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#3d,48(<out=int64#1)
+# asm 2: movl   <in12=%edx,48(<out=%rdi)
+movl   %edx,48(%rdi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#4d,52(<out=int64#1)
+# asm 2: movl   <in13=%ecx,52(<out=%rdi)
+movl   %ecx,52(%rdi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#5d,56(<out=int64#1)
+# asm 2: movl   <in14=%r8d,56(<out=%rdi)
+movl   %r8d,56(%rdi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#6d,60(<out=int64#1)
+# asm 2: movl   <in15=%r9d,60(<out=%rdi)
+movl   %r9d,60(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 112(<m=%rsi),<in12=%edx
+xorl 112(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 116(<m=%rsi),<in13=%ecx
+xorl 116(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 120(<m=%rsi),<in14=%r8d
+xorl 120(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 124(<m=%rsi),<in15=%r9d
+xorl 124(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl   <in12=int64#3d,112(<out=int64#1)
+# asm 2: movl   <in12=%edx,112(<out=%rdi)
+movl   %edx,112(%rdi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl   <in13=int64#4d,116(<out=int64#1)
+# asm 2: movl   <in13=%ecx,116(<out=%rdi)
+movl   %ecx,116(%rdi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl   <in14=int64#5d,120(<out=int64#1)
+# asm 2: movl   <in14=%r8d,120(<out=%rdi)
+movl   %r8d,120(%rdi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl   <in15=int64#6d,124(<out=int64#1)
+# asm 2: movl   <in15=%r9d,124(<out=%rdi)
+movl   %r9d,124(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 176(<m=%rsi),<in12=%edx
+xorl 176(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 180(<m=%rsi),<in13=%ecx
+xorl 180(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 184(<m=%rsi),<in14=%r8d
+xorl 184(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 188(<m=%rsi),<in15=%r9d
+xorl 188(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl   <in12=int64#3d,176(<out=int64#1)
+# asm 2: movl   <in12=%edx,176(<out=%rdi)
+movl   %edx,176(%rdi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl   <in13=int64#4d,180(<out=int64#1)
+# asm 2: movl   <in13=%ecx,180(<out=%rdi)
+movl   %ecx,180(%rdi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl   <in14=int64#5d,184(<out=int64#1)
+# asm 2: movl   <in14=%r8d,184(<out=%rdi)
+movl   %r8d,184(%rdi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl   <in15=int64#6d,188(<out=int64#1)
+# asm 2: movl   <in15=%r9d,188(<out=%rdi)
+movl   %r9d,188(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 240(<m=%rsi),<in12=%edx
+xorl 240(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 244(<m=%rsi),<in13=%ecx
+xorl 244(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 248(<m=%rsi),<in14=%r8d
+xorl 248(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 252(<m=%rsi),<in15=%r9d
+xorl 252(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl   <in12=int64#3d,240(<out=int64#1)
+# asm 2: movl   <in12=%edx,240(<out=%rdi)
+movl   %edx,240(%rdi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl   <in13=int64#4d,244(<out=int64#1)
+# asm 2: movl   <in13=%ecx,244(<out=%rdi)
+movl   %ecx,244(%rdi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl   <in14=int64#5d,248(<out=int64#1)
+# asm 2: movl   <in14=%r8d,248(<out=%rdi)
+movl   %r8d,248(%rdi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl   <in15=int64#6d,252(<out=int64#1)
+# asm 2: movl   <in15=%r9d,252(<out=%rdi)
+movl   %r9d,252(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int64#6
+# asm 2: sub  $256,<bytes=%r9
+sub  $256,%r9
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int64#2
+# asm 2: add  $256,<m=%rsi
+add  $256,%rsi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int64#1
+# asm 2: add  $256,<out=%rdi
+add  $256,%rdi
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: mov  <out=int64#1,>ctarget=int64#3
+# asm 2: mov  <out=%rdi,>ctarget=%rdx
+mov  %rdi,%rdx
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     m = &tmp
+# asm 1: leaq <tmp=stack512#1,>m=int64#2
+# asm 2: leaq <tmp=416(%rsp),>m=%rsi
+leaq 416(%rsp),%rsi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#4,>diag0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>diag0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#1,>diag1=int6464#2
+# asm 2: movdqa <x1=0(%rsp),>diag1=%xmm1
+movdqa 0(%rsp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#2,>diag2=int6464#3
+# asm 2: movdqa <x2=16(%rsp),>diag2=%xmm2
+movdqa 16(%rsp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#3,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%rsp),>diag3=%xmm3
+movdqa 32(%rsp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 20
+# asm 1: mov  $20,>i=int64#4
+# asm 2: mov  $20,>i=%rcx
+mov  $20,%rcx
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int64#4
+# asm 2: sub  $4,<i=%rcx
+sub  $4,%rcx
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#4,<diag0=int6464#1
+# asm 2: paddd <x0=48(%rsp),<diag0=%xmm0
+paddd 48(%rsp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#1,<diag1=int6464#2
+# asm 2: paddd <x1=0(%rsp),<diag1=%xmm1
+paddd 0(%rsp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#2,<diag2=int6464#3
+# asm 2: paddd <x2=16(%rsp),<diag2=%xmm2
+paddd 16(%rsp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#3,<diag3=int6464#4
+# asm 2: paddd <x3=32(%rsp),<diag3=%xmm3
+paddd 32(%rsp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int64#4
+# asm 2: movd   <diag0=%xmm0,>in0=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int64#5
+# asm 2: movd   <diag1=%xmm1,>in12=%r8
+movd   %xmm1,%r8
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int64#6
+# asm 2: movd   <diag2=%xmm2,>in8=%r9
+movd   %xmm2,%r9
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int64#7
+# asm 2: movd   <diag3=%xmm3,>in4=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#4d
+# asm 2: xorl 0(<m=%rsi),<in0=%ecx
+xorl 0(%rsi),%ecx
+
+# qhasm: (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#5d
+# asm 2: xorl 48(<m=%rsi),<in12=%r8d
+xorl 48(%rsi),%r8d
+
+# qhasm: (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#6d
+# asm 2: xorl 32(<m=%rsi),<in8=%r9d
+xorl 32(%rsi),%r9d
+
+# qhasm: (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#7d
+# asm 2: xorl 16(<m=%rsi),<in4=%eax
+xorl 16(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#4d,0(<out=int64#1)
+# asm 2: movl   <in0=%ecx,0(<out=%rdi)
+movl   %ecx,0(%rdi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#5d,48(<out=int64#1)
+# asm 2: movl   <in12=%r8d,48(<out=%rdi)
+movl   %r8d,48(%rdi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#6d,32(<out=int64#1)
+# asm 2: movl   <in8=%r9d,32(<out=%rdi)
+movl   %r9d,32(%rdi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#7d,16(<out=int64#1)
+# asm 2: movl   <in4=%eax,16(<out=%rdi)
+movl   %eax,16(%rdi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int64#4
+# asm 2: movd   <diag0=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int64#5
+# asm 2: movd   <diag1=%xmm1,>in1=%r8
+movd   %xmm1,%r8
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int64#6
+# asm 2: movd   <diag2=%xmm2,>in13=%r9
+movd   %xmm2,%r9
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int64#7
+# asm 2: movd   <diag3=%xmm3,>in9=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm: (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#5d
+# asm 2: xorl 4(<m=%rsi),<in1=%r8d
+xorl 4(%rsi),%r8d
+
+# qhasm: (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#6d
+# asm 2: xorl 52(<m=%rsi),<in13=%r9d
+xorl 52(%rsi),%r9d
+
+# qhasm: (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#7d
+# asm 2: xorl 36(<m=%rsi),<in9=%eax
+xorl 36(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#5d,4(<out=int64#1)
+# asm 2: movl   <in1=%r8d,4(<out=%rdi)
+movl   %r8d,4(%rdi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#6d,52(<out=int64#1)
+# asm 2: movl   <in13=%r9d,52(<out=%rdi)
+movl   %r9d,52(%rdi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#7d,36(<out=int64#1)
+# asm 2: movl   <in9=%eax,36(<out=%rdi)
+movl   %eax,36(%rdi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int64#4
+# asm 2: movd   <diag0=%xmm0,>in10=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int64#5
+# asm 2: movd   <diag1=%xmm1,>in6=%r8
+movd   %xmm1,%r8
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int64#6
+# asm 2: movd   <diag2=%xmm2,>in2=%r9
+movd   %xmm2,%r9
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int64#7
+# asm 2: movd   <diag3=%xmm3,>in14=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#4d
+# asm 2: xorl 40(<m=%rsi),<in10=%ecx
+xorl 40(%rsi),%ecx
+
+# qhasm: (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm: (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#6d
+# asm 2: xorl 8(<m=%rsi),<in2=%r9d
+xorl 8(%rsi),%r9d
+
+# qhasm: (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#7d
+# asm 2: xorl 56(<m=%rsi),<in14=%eax
+xorl 56(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#4d,40(<out=int64#1)
+# asm 2: movl   <in10=%ecx,40(<out=%rdi)
+movl   %ecx,40(%rdi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#6d,8(<out=int64#1)
+# asm 2: movl   <in2=%r9d,8(<out=%rdi)
+movl   %r9d,8(%rdi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#7d,56(<out=int64#1)
+# asm 2: movl   <in14=%eax,56(<out=%rdi)
+movl   %eax,56(%rdi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int64#4
+# asm 2: movd   <diag0=%xmm0,>in15=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int64#5
+# asm 2: movd   <diag1=%xmm1,>in11=%r8
+movd   %xmm1,%r8
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int64#6
+# asm 2: movd   <diag2=%xmm2,>in7=%r9
+movd   %xmm2,%r9
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int64#7
+# asm 2: movd   <diag3=%xmm3,>in3=%rax
+movd   %xmm3,%rax
+
+# qhasm: (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#4d
+# asm 2: xorl 60(<m=%rsi),<in15=%ecx
+xorl 60(%rsi),%ecx
+
+# qhasm: (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#5d
+# asm 2: xorl 44(<m=%rsi),<in11=%r8d
+xorl 44(%rsi),%r8d
+
+# qhasm: (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm: (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#7d
+# asm 2: xorl 12(<m=%rsi),<in3=%eax
+xorl 12(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#4d,60(<out=int64#1)
+# asm 2: movl   <in15=%ecx,60(<out=%rdi)
+movl   %ecx,60(%rdi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#5d,44(<out=int64#1)
+# asm 2: movl   <in11=%r8d,44(<out=%rdi)
+movl   %r8d,44(%rdi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#7d,12(<out=int64#1)
+# asm 2: movl   <in3=%eax,12(<out=%rdi)
+movl   %eax,12(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#4d
+# asm 2: movl <x2=16(%rsp),>in8=%ecx
+movl 16(%rsp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#5d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%r8d
+movl 4+32(%rsp),%r8d
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#4
+# asm 2: add  $1,<in8=%rcx
+add  $1,%rcx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#5
+# asm 2: shl  $32,<in9=%r8
+shl  $32,%r8
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#5,<in8=int64#4
+# asm 2: add  <in9=%r8,<in8=%rcx
+add  %r8,%rcx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#4,>in9=int64#5
+# asm 2: mov  <in8=%rcx,>in9=%r8
+mov  %rcx,%r8
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#5
+# asm 2: shr  $32,<in9=%r8
+shr  $32,%r8
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#5d,4+<x3=stack128#3
+# asm 2: movl <in9=%r8d,4+<x3=32(%rsp)
+movl %r8d,4+32(%rsp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm:       out = ctarget
+# asm 1: mov  <ctarget=int64#3,>out=int64#1
+# asm 2: mov  <ctarget=%rdx,>out=%rdi
+mov  %rdx,%rdi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     r11_caller = r11_stack
+# asm 1: movq <r11_stack=stack64#1,>r11_caller=int64#9
+# asm 2: movq <r11_stack=352(%rsp),>r11_caller=%r11
+movq 352(%rsp),%r11
+
+# qhasm:     r12_caller = r12_stack
+# asm 1: movq <r12_stack=stack64#2,>r12_caller=int64#10
+# asm 2: movq <r12_stack=360(%rsp),>r12_caller=%r12
+movq 360(%rsp),%r12
+
+# qhasm:     r13_caller = r13_stack
+# asm 1: movq <r13_stack=stack64#3,>r13_caller=int64#11
+# asm 2: movq <r13_stack=368(%rsp),>r13_caller=%r13
+movq 368(%rsp),%r13
+
+# qhasm:     r14_caller = r14_stack
+# asm 1: movq <r14_stack=stack64#4,>r14_caller=int64#12
+# asm 2: movq <r14_stack=376(%rsp),>r14_caller=%r14
+movq 376(%rsp),%r14
+
+# qhasm:     r15_caller = r15_stack
+# asm 1: movq <r15_stack=stack64#5,>r15_caller=int64#13
+# asm 2: movq <r15_stack=384(%rsp),>r15_caller=%r15
+movq 384(%rsp),%r15
+
+# qhasm:     rbx_caller = rbx_stack
+# asm 1: movq <rbx_stack=stack64#6,>rbx_caller=int64#14
+# asm 2: movq <rbx_stack=392(%rsp),>rbx_caller=%rbx
+movq 392(%rsp),%rbx
+
+# qhasm:     rbp_caller = rbp_stack
+# asm 1: movq <rbp_stack=stack64#7,>rbp_caller=int64#15
+# asm 2: movq <rbp_stack=400(%rsp),>rbp_caller=%rbp
+movq 400(%rsp),%rbp
+
+# qhasm:     leave
+add %r11,%rsp
+xor %rax,%rax
+xor %rdx,%rdx
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int64#6
+# asm 2: sub  $64,<bytes=%r9
+sub  $64,%r9
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int64#1
+# asm 2: add  $64,<out=%rdi
+add  $64,%rdi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int64#2
+# asm 2: add  $64,<m=%rsi
+add  $64,%rsi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/salsa20/checksum b/nacl/crypto_stream/salsa20/checksum
new file mode 100644
index 00000000..78ff05f4
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/checksum
@@ -0,0 +1 @@
+44a3966eabcd3a2b13faca2150e38f2b7e6bac187d626618f50a9f875158ae78
diff --git a/nacl/crypto_stream/salsa20/ref/api.h b/nacl/crypto_stream/salsa20/ref/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/ref/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa20/ref/implementors b/nacl/crypto_stream/salsa20/ref/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/ref/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa20/ref/stream.c b/nacl/crypto_stream/salsa20/ref/stream.c
new file mode 100644
index 00000000..2f0262eb
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/ref/stream.c
@@ -0,0 +1,49 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa20.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream(
+        unsigned char *c,unsigned long long clen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!clen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (clen >= 64) {
+    crypto_core_salsa20(c,in,k,sigma);
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    clen -= 64;
+    c += 64;
+  }
+
+  if (clen) {
+    crypto_core_salsa20(block,in,k,sigma);
+    for (i = 0;i < clen;++i) c[i] = block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa20/ref/xor.c b/nacl/crypto_stream/salsa20/ref/xor.c
new file mode 100644
index 00000000..11c7e9f0
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/ref/xor.c
@@ -0,0 +1,52 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa20.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream_xor(
+        unsigned char *c,
+  const unsigned char *m,unsigned long long mlen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!mlen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (mlen >= 64) {
+    crypto_core_salsa20(block,in,k,sigma);
+    for (i = 0;i < 64;++i) c[i] = m[i] ^ block[i];
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    mlen -= 64;
+    c += 64;
+    m += 64;
+  }
+
+  if (mlen) {
+    crypto_core_salsa20(block,in,k,sigma);
+    for (i = 0;i < mlen;++i) c[i] = m[i] ^ block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa20/used b/nacl/crypto_stream/salsa20/used
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/used
diff --git a/nacl/crypto_stream/salsa20/x86_xmm5/api.h b/nacl/crypto_stream/salsa20/x86_xmm5/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/x86_xmm5/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa20/x86_xmm5/implementors b/nacl/crypto_stream/salsa20/x86_xmm5/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/x86_xmm5/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa20/x86_xmm5/stream.s b/nacl/crypto_stream/salsa20/x86_xmm5/stream.s
new file mode 100644
index 00000000..9e32ea43
--- /dev/null
+++ b/nacl/crypto_stream/salsa20/x86_xmm5/stream.s
@@ -0,0 +1,5078 @@
+
+# qhasm: int32 a
+
+# qhasm: stack32 arg1
+
+# qhasm: stack32 arg2
+
+# qhasm: stack32 arg3
+
+# qhasm: stack32 arg4
+
+# qhasm: stack32 arg5
+
+# qhasm: stack32 arg6
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: input arg6
+
+# qhasm: int32 eax
+
+# qhasm: int32 ebx
+
+# qhasm: int32 esi
+
+# qhasm: int32 edi
+
+# qhasm: int32 ebp
+
+# qhasm: caller eax
+
+# qhasm: caller ebx
+
+# qhasm: caller esi
+
+# qhasm: caller edi
+
+# qhasm: caller ebp
+
+# qhasm: int32 k
+
+# qhasm: int32 kbits
+
+# qhasm: int32 iv
+
+# qhasm: int32 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int32 m
+
+# qhasm: stack32 out_stack
+
+# qhasm: int32 out
+
+# qhasm: stack32 bytes_stack
+
+# qhasm: int32 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int6464 p
+
+# qhasm: int6464 q
+
+# qhasm: int6464 r
+
+# qhasm: int6464 s
+
+# qhasm: int6464 t
+
+# qhasm: int6464 u
+
+# qhasm: int6464 v
+
+# qhasm: int6464 w
+
+# qhasm: int6464 mp
+
+# qhasm: int6464 mq
+
+# qhasm: int6464 mr
+
+# qhasm: int6464 ms
+
+# qhasm: int6464 mt
+
+# qhasm: int6464 mu
+
+# qhasm: int6464 mv
+
+# qhasm: int6464 mw
+
+# qhasm: int32 in0
+
+# qhasm: int32 in1
+
+# qhasm: int32 in2
+
+# qhasm: int32 in3
+
+# qhasm: int32 in4
+
+# qhasm: int32 in5
+
+# qhasm: int32 in6
+
+# qhasm: int32 in7
+
+# qhasm: int32 in8
+
+# qhasm: int32 in9
+
+# qhasm: int32 in10
+
+# qhasm: int32 in11
+
+# qhasm: int32 in12
+
+# qhasm: int32 in13
+
+# qhasm: int32 in14
+
+# qhasm: int32 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: stack32 ctarget
+
+# qhasm: enter crypto_stream_salsa20_x86_xmm5
+.text
+.p2align 5
+.globl _crypto_stream_salsa20_x86_xmm5
+.globl crypto_stream_salsa20_x86_xmm5
+_crypto_stream_salsa20_x86_xmm5:
+crypto_stream_salsa20_x86_xmm5:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: bytes = arg2
+# asm 1: movl <arg2=stack32#-2,>bytes=int32#3
+# asm 2: movl <arg2=8(%esp,%eax),>bytes=%edx
+movl 8(%esp,%eax),%edx
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm: iv = arg4
+# asm 1: movl <arg4=stack32#-4,>iv=int32#4
+# asm 2: movl <arg4=16(%esp,%eax),>iv=%ebx
+movl 16(%esp,%eax),%ebx
+
+# qhasm: k = arg5
+# asm 1: movl <arg5=stack32#-5,>k=int32#7
+# asm 2: movl <arg5=20(%esp,%eax),>k=%ebp
+movl 20(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int32#1
+# asm 2: mov  $0,>a=%eax
+mov  $0,%eax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int32#3,>i=int32#2
+# asm 2: mov  <bytes=%edx,>i=%ecx
+mov  %edx,%ecx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: subl <bytes=int32#3,<out=int32#6
+# asm 2: subl <bytes=%edx,<out=%edi
+subl %edx,%edi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa20_x86_xmm5_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa20_x86_xmm5_xor
+.globl crypto_stream_salsa20_x86_xmm5_xor
+_crypto_stream_salsa20_x86_xmm5_xor:
+crypto_stream_salsa20_x86_xmm5_xor:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = arg2
+# asm 1: movl <arg2=stack32#-2,>m=int32#5
+# asm 2: movl <arg2=8(%esp,%eax),>m=%esi
+movl 8(%esp,%eax),%esi
+
+# qhasm: bytes = arg3
+# asm 1: movl <arg3=stack32#-3,>bytes=int32#3
+# asm 2: movl <arg3=12(%esp,%eax),>bytes=%edx
+movl 12(%esp,%eax),%edx
+
+# qhasm: iv = arg5
+# asm 1: movl <arg5=stack32#-5,>iv=int32#4
+# asm 2: movl <arg5=20(%esp,%eax),>iv=%ebx
+movl 20(%esp,%eax),%ebx
+
+# qhasm: k = arg6
+# asm 1: movl <arg6=stack32#-6,>k=int32#7
+# asm 2: movl <arg6=24(%esp,%eax),>k=%ebp
+movl 24(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#3,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%edx,>bytes_stack=24(%esp)
+movl %edx,24(%esp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl 12(<k=int32#7),>in4=int32#1
+# asm 2: movl 12(<k=%ebp),>in4=%eax
+movl 12(%ebp),%eax
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl 20(<k=int32#7),>in12=int32#2
+# asm 2: movl 20(<k=%ebp),>in12=%ecx
+movl 20(%ebp),%ecx
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int32#1,>x3=stack128#1
+# asm 2: movl <in4=%eax,>x3=32(%esp)
+movl %eax,32(%esp)
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int32#2,>x1=stack128#2
+# asm 2: movl <in12=%ecx,>x1=48(%esp)
+movl %ecx,48(%esp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int32#1
+# asm 2: mov  $1634760805,>in0=%eax
+mov  $1634760805,%eax
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int32#2
+# asm 2: mov  $0,>in8=%ecx
+mov  $0,%ecx
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int32#1,>x0=stack128#3
+# asm 2: movl <in0=%eax,>x0=64(%esp)
+movl %eax,64(%esp)
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl 0(<iv=int32#4),>in6=int32#1
+# asm 2: movl 0(<iv=%ebx),>in6=%eax
+movl 0(%ebx),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl 4(<iv=int32#4),>in7=int32#2
+# asm 2: movl 4(<iv=%ebx),>in7=%ecx
+movl 4(%ebx),%ecx
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int32#1,8+<x1=stack128#2
+# asm 2: movl <in6=%eax,8+<x1=48(%esp)
+movl %eax,8+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int32#2,12+<x2=stack128#4
+# asm 2: movl <in7=%ecx,12+<x2=80(%esp)
+movl %ecx,12+80(%esp)
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int32#1
+# asm 2: mov  $0,>in9=%eax
+mov  $0,%eax
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int32#2
+# asm 2: mov  $2036477234,>in10=%ecx
+mov  $2036477234,%ecx
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#1,4+<x3=stack128#1
+# asm 2: movl <in9=%eax,4+<x3=32(%esp)
+movl %eax,4+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int32#2,8+<x0=stack128#3
+# asm 2: movl <in10=%ecx,8+<x0=64(%esp)
+movl %ecx,8+64(%esp)
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl 0(<k=int32#7),>in1=int32#1
+# asm 2: movl 0(<k=%ebp),>in1=%eax
+movl 0(%ebp),%eax
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl 4(<k=int32#7),>in2=int32#2
+# asm 2: movl 4(<k=%ebp),>in2=%ecx
+movl 4(%ebp),%ecx
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl 8(<k=int32#7),>in3=int32#3
+# asm 2: movl 8(<k=%ebp),>in3=%edx
+movl 8(%ebp),%edx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int32#4
+# asm 2: mov  $857760878,>in5=%ebx
+mov  $857760878,%ebx
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int32#1,4+<x1=stack128#2
+# asm 2: movl <in1=%eax,4+<x1=48(%esp)
+movl %eax,4+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int32#2,8+<x2=stack128#4
+# asm 2: movl <in2=%ecx,8+<x2=80(%esp)
+movl %ecx,8+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int32#3,12+<x3=stack128#1
+# asm 2: movl <in3=%edx,12+<x3=32(%esp)
+movl %edx,12+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int32#4,4+<x0=stack128#3
+# asm 2: movl <in5=%ebx,4+<x0=64(%esp)
+movl %ebx,4+64(%esp)
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl 16(<k=int32#7),>in11=int32#1
+# asm 2: movl 16(<k=%ebp),>in11=%eax
+movl 16(%ebp),%eax
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl 24(<k=int32#7),>in13=int32#2
+# asm 2: movl 24(<k=%ebp),>in13=%ecx
+movl 24(%ebp),%ecx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl 28(<k=int32#7),>in14=int32#3
+# asm 2: movl 28(<k=%ebp),>in14=%edx
+movl 28(%ebp),%edx
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int32#4
+# asm 2: mov  $1797285236,>in15=%ebx
+mov  $1797285236,%ebx
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int32#1,12+<x1=stack128#2
+# asm 2: movl <in11=%eax,12+<x1=48(%esp)
+movl %eax,12+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int32#2,4+<x2=stack128#4
+# asm 2: movl <in13=%ecx,4+<x2=80(%esp)
+movl %ecx,4+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int32#3,8+<x3=stack128#1
+# asm 2: movl <in14=%edx,8+<x3=32(%esp)
+movl %edx,8+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int32#4,12+<x0=stack128#3
+# asm 2: movl <in15=%ebx,12+<x0=64(%esp)
+movl %ebx,12+64(%esp)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#3,>z0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>z0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=96(%esp)
+movdqa %xmm1,96(%esp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=112(%esp)
+movdqa %xmm2,112(%esp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=128(%esp)
+movdqa %xmm3,128(%esp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=144(%esp)
+movdqa %xmm0,144(%esp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#2,>z1=int6464#1
+# asm 2: movdqa <x1=48(%esp),>z1=%xmm0
+movdqa 48(%esp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=160(%esp)
+movdqa %xmm1,160(%esp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=176(%esp)
+movdqa %xmm2,176(%esp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=192(%esp)
+movdqa %xmm3,192(%esp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=208(%esp)
+movdqa %xmm0,208(%esp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#4,>z2=int6464#1
+# asm 2: movdqa <x2=80(%esp),>z2=%xmm0
+movdqa 80(%esp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=224(%esp)
+movdqa %xmm1,224(%esp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=240(%esp)
+movdqa %xmm2,240(%esp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=256(%esp)
+movdqa %xmm0,256(%esp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#1,>z3=int6464#1
+# asm 2: movdqa <x3=32(%esp),>z3=%xmm0
+movdqa 32(%esp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=272(%esp)
+movdqa %xmm1,272(%esp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=288(%esp)
+movdqa %xmm2,288(%esp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=304(%esp)
+movdqa %xmm0,304(%esp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int32#2,>orig8=stack128#19
+# asm 2: movl <in8=%ecx,>orig8=320(%esp)
+movl %ecx,320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int32#3,>orig9=stack128#20
+# asm 2: movl <in9=%edx,>orig9=336(%esp)
+movl %edx,336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int32#2,4+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,4+<orig8=320(%esp)
+movl %ecx,4+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int32#3,4+<orig9=stack128#20
+# asm 2: movl <in9=%edx,4+<orig9=336(%esp)
+movl %edx,4+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int32#2,8+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,8+<orig8=320(%esp)
+movl %ecx,8+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int32#3,8+<orig9=stack128#20
+# asm 2: movl <in9=%edx,8+<orig9=336(%esp)
+movl %edx,8+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int32#2,12+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,12+<orig8=320(%esp)
+movl %ecx,12+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int32#3,12+<orig9=stack128#20
+# asm 2: movl <in9=%edx,12+<orig9=336(%esp)
+movl %edx,12+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: i = 20
+# asm 1: mov  $20,>i=int32#1
+# asm 2: mov  $20,>i=%eax
+mov  $20,%eax
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=96(%esp),>z5=%xmm0
+movdqa 96(%esp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=112(%esp),>z10=%xmm1
+movdqa 112(%esp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=128(%esp),>z15=%xmm2
+movdqa 128(%esp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=288(%esp),>z14=%xmm3
+movdqa 288(%esp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=304(%esp),>z3=%xmm4
+movdqa 304(%esp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=160(%esp),>z6=%xmm5
+movdqa 160(%esp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=176(%esp),>z11=%xmm6
+movdqa 176(%esp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=208(%esp),>z1=%xmm7
+movdqa 208(%esp),%xmm7
+
+# qhasm:   z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#21
+# asm 2: movdqa <z5=%xmm0,>z5_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:   z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#22
+# asm 2: movdqa <z10=%xmm1,>z10_stack=368(%esp)
+movdqa %xmm1,368(%esp)
+
+# qhasm:   z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#23
+# asm 2: movdqa <z15=%xmm2,>z15_stack=384(%esp)
+movdqa %xmm2,384(%esp)
+
+# qhasm:   z14_stack = z14
+# asm 1: movdqa <z14=int6464#4,>z14_stack=stack128#24
+# asm 2: movdqa <z14=%xmm3,>z14_stack=400(%esp)
+movdqa %xmm3,400(%esp)
+
+# qhasm:   z3_stack = z3
+# asm 1: movdqa <z3=int6464#5,>z3_stack=stack128#25
+# asm 2: movdqa <z3=%xmm4,>z3_stack=416(%esp)
+movdqa %xmm4,416(%esp)
+
+# qhasm:   z6_stack = z6
+# asm 1: movdqa <z6=int6464#6,>z6_stack=stack128#26
+# asm 2: movdqa <z6=%xmm5,>z6_stack=432(%esp)
+movdqa %xmm5,432(%esp)
+
+# qhasm:   z11_stack = z11
+# asm 1: movdqa <z11=int6464#7,>z11_stack=stack128#27
+# asm 2: movdqa <z11=%xmm6,>z11_stack=448(%esp)
+movdqa %xmm6,448(%esp)
+
+# qhasm:   z1_stack = z1
+# asm 1: movdqa <z1=int6464#8,>z1_stack=stack128#28
+# asm 2: movdqa <z1=%xmm7,>z1_stack=464(%esp)
+movdqa %xmm7,464(%esp)
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#5
+# asm 2: movdqa <orig7=224(%esp),>z7=%xmm4
+movdqa 224(%esp),%xmm4
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#6
+# asm 2: movdqa <orig13=240(%esp),>z13=%xmm5
+movdqa 240(%esp),%xmm5
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#7
+# asm 2: movdqa <orig2=256(%esp),>z2=%xmm6
+movdqa 256(%esp),%xmm6
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#8
+# asm 2: movdqa <orig9=336(%esp),>z9=%xmm7
+movdqa 336(%esp),%xmm7
+
+# qhasm:                   p = orig0
+# asm 1: movdqa <orig0=stack128#8,>p=int6464#1
+# asm 2: movdqa <orig0=144(%esp),>p=%xmm0
+movdqa 144(%esp),%xmm0
+
+# qhasm:                   t = orig12
+# asm 1: movdqa <orig12=stack128#11,>t=int6464#3
+# asm 2: movdqa <orig12=192(%esp),>t=%xmm2
+movdqa 192(%esp),%xmm2
+
+# qhasm:                   q = orig4
+# asm 1: movdqa <orig4=stack128#16,>q=int6464#4
+# asm 2: movdqa <orig4=272(%esp),>q=%xmm3
+movdqa 272(%esp),%xmm3
+
+# qhasm:                   r = orig8
+# asm 1: movdqa <orig8=stack128#19,>r=int6464#2
+# asm 2: movdqa <orig8=320(%esp),>r=%xmm1
+movdqa 320(%esp),%xmm1
+
+# qhasm:   z7_stack = z7
+# asm 1: movdqa <z7=int6464#5,>z7_stack=stack128#29
+# asm 2: movdqa <z7=%xmm4,>z7_stack=480(%esp)
+movdqa %xmm4,480(%esp)
+
+# qhasm:   z13_stack = z13
+# asm 1: movdqa <z13=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <z13=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm:   z2_stack = z2
+# asm 1: movdqa <z2=int6464#7,>z2_stack=stack128#31
+# asm 2: movdqa <z2=%xmm6,>z2_stack=512(%esp)
+movdqa %xmm6,512(%esp)
+
+# qhasm:   z9_stack = z9
+# asm 1: movdqa <z9=int6464#8,>z9_stack=stack128#32
+# asm 2: movdqa <z9=%xmm7,>z9_stack=528(%esp)
+movdqa %xmm7,528(%esp)
+
+# qhasm:   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:   z12_stack = t
+# asm 1: movdqa <t=int6464#3,>z12_stack=stack128#34
+# asm 2: movdqa <t=%xmm2,>z12_stack=560(%esp)
+movdqa %xmm2,560(%esp)
+
+# qhasm:   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#35
+# asm 2: movdqa <q=%xmm3,>z4_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#36
+# asm 2: movdqa <r=%xmm1,>z8_stack=592(%esp)
+movdqa %xmm1,592(%esp)
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#33
+# asm 2: movdqa <q=%xmm3,>z4_stack=544(%esp)
+movdqa %xmm3,544(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#34
+# asm 2: movdqa <r=%xmm1,>z8_stack=560(%esp)
+movdqa %xmm1,560(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>mt=int6464#3
+# asm 2: movdqa <z1_stack=464(%esp),>mt=%xmm2
+movdqa 464(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>mp=int6464#5
+# asm 2: movdqa <z5_stack=352(%esp),>mp=%xmm4
+movdqa 352(%esp),%xmm4
+
+# qhasm:                                   mq = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>mq=int6464#4
+# asm 2: movdqa <z9_stack=528(%esp),>mq=%xmm3
+movdqa 528(%esp),%xmm3
+
+# qhasm:                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>mr=int6464#6
+# asm 2: movdqa <z13_stack=496(%esp),>mr=%xmm5
+movdqa 496(%esp),%xmm5
+
+# qhasm:                   z12_stack = s
+# asm 1: movdqa <s=int6464#7,>z12_stack=stack128#30
+# asm 2: movdqa <s=%xmm6,>z12_stack=496(%esp)
+movdqa %xmm6,496(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <p=%xmm0,>z0_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z9_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z9_stack=stack128#32
+# asm 2: movdqa <mq=%xmm3,>z9_stack=528(%esp)
+movdqa %xmm3,528(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#35
+# asm 2: movdqa <mr=%xmm5,>z13_stack=576(%esp)
+movdqa %xmm5,576(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>t=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>t=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>p=int6464#1
+# asm 2: movdqa <z10_stack=368(%esp),>p=%xmm0
+movdqa 368(%esp),%xmm0
+
+# qhasm:                                                   q = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>q=int6464#4
+# asm 2: movdqa <z14_stack=400(%esp),>q=%xmm3
+movdqa 400(%esp),%xmm3
+
+# qhasm:                                                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>r=int6464#2
+# asm 2: movdqa <z2_stack=512(%esp),>r=%xmm1
+movdqa 512(%esp),%xmm1
+
+# qhasm:                                   z1_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z1_stack=stack128#22
+# asm 2: movdqa <ms=%xmm6,>z1_stack=368(%esp)
+movdqa %xmm6,368(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#24
+# asm 2: movdqa <mp=%xmm4,>z5_stack=400(%esp)
+movdqa %xmm4,400(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z14_stack = q
+# asm 1: movdqa <q=int6464#4,>z14_stack=stack128#36
+# asm 2: movdqa <q=%xmm3,>z14_stack=592(%esp)
+movdqa %xmm3,592(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#26
+# asm 2: movdqa <r=%xmm1,>z2_stack=432(%esp)
+movdqa %xmm1,432(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>mt=int6464#3
+# asm 2: movdqa <z11_stack=448(%esp),>mt=%xmm2
+movdqa 448(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>mp=int6464#5
+# asm 2: movdqa <z15_stack=384(%esp),>mp=%xmm4
+movdqa 384(%esp),%xmm4
+
+# qhasm:                                                                   mq = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>mq=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>mq=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:                                                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                                                   z6_stack = s
+# asm 1: movdqa <s=int6464#7,>z6_stack=stack128#23
+# asm 2: movdqa <s=%xmm6,>z6_stack=384(%esp)
+movdqa %xmm6,384(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#27
+# asm 2: movdqa <p=%xmm0,>z10_stack=448(%esp)
+movdqa %xmm0,448(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z3_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z3_stack=stack128#25
+# asm 2: movdqa <mq=%xmm3,>z3_stack=416(%esp)
+movdqa %xmm3,416(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>t=int6464#3
+# asm 2: movdqa <z3_stack=416(%esp),>t=%xmm2
+movdqa 416(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>p=int6464#1
+# asm 2: movdqa <z0_stack=352(%esp),>p=%xmm0
+movdqa 352(%esp),%xmm0
+
+# qhasm:                   q = z1_stack
+# asm 1: movdqa <z1_stack=stack128#22,>q=int6464#4
+# asm 2: movdqa <z1_stack=368(%esp),>q=%xmm3
+movdqa 368(%esp),%xmm3
+
+# qhasm:                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#26,>r=int6464#2
+# asm 2: movdqa <z2_stack=432(%esp),>r=%xmm1
+movdqa 432(%esp),%xmm1
+
+# qhasm:                                                                   z11_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z11_stack=stack128#21
+# asm 2: movdqa <ms=%xmm6,>z11_stack=352(%esp)
+movdqa %xmm6,352(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#22
+# asm 2: movdqa <mp=%xmm4,>z15_stack=368(%esp)
+movdqa %xmm4,368(%esp)
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z1_stack = q
+# asm 1: movdqa <q=int6464#4,>z1_stack=stack128#28
+# asm 2: movdqa <q=%xmm3,>z1_stack=464(%esp)
+movdqa %xmm3,464(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#31
+# asm 2: movdqa <r=%xmm1,>z2_stack=512(%esp)
+movdqa %xmm1,512(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z4_stack
+# asm 1: movdqa <z4_stack=stack128#33,>mt=int6464#3
+# asm 2: movdqa <z4_stack=544(%esp),>mt=%xmm2
+movdqa 544(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#24,>mp=int6464#5
+# asm 2: movdqa <z5_stack=400(%esp),>mp=%xmm4
+movdqa 400(%esp),%xmm4
+
+# qhasm:                                   mq = z6_stack
+# asm 1: movdqa <z6_stack=stack128#23,>mq=int6464#4
+# asm 2: movdqa <z6_stack=384(%esp),>mq=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                   z3_stack = s
+# asm 1: movdqa <s=int6464#7,>z3_stack=stack128#25
+# asm 2: movdqa <s=%xmm6,>z3_stack=416(%esp)
+movdqa %xmm6,416(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z6_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z6_stack=stack128#26
+# asm 2: movdqa <mq=%xmm3,>z6_stack=432(%esp)
+movdqa %xmm3,432(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>t=int6464#3
+# asm 2: movdqa <z9_stack=528(%esp),>t=%xmm2
+movdqa 528(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#27,>p=int6464#1
+# asm 2: movdqa <z10_stack=448(%esp),>p=%xmm0
+movdqa 448(%esp),%xmm0
+
+# qhasm:                                                   q = z11_stack
+# asm 1: movdqa <z11_stack=stack128#21,>q=int6464#4
+# asm 2: movdqa <z11_stack=352(%esp),>q=%xmm3
+movdqa 352(%esp),%xmm3
+
+# qhasm:                                                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#34,>r=int6464#2
+# asm 2: movdqa <z8_stack=560(%esp),>r=%xmm1
+movdqa 560(%esp),%xmm1
+
+# qhasm:                                   z4_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z4_stack=stack128#34
+# asm 2: movdqa <ms=%xmm6,>z4_stack=560(%esp)
+movdqa %xmm6,560(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#21
+# asm 2: movdqa <mp=%xmm4,>z5_stack=352(%esp)
+movdqa %xmm4,352(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z11_stack = q
+# asm 1: movdqa <q=int6464#4,>z11_stack=stack128#27
+# asm 2: movdqa <q=%xmm3,>z11_stack=448(%esp)
+movdqa %xmm3,448(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#37
+# asm 2: movdqa <r=%xmm1,>z8_stack=608(%esp)
+movdqa %xmm1,608(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z14_stack
+# asm 1: movdqa <z14_stack=stack128#36,>mt=int6464#3
+# asm 2: movdqa <z14_stack=592(%esp),>mt=%xmm2
+movdqa 592(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>mp=int6464#5
+# asm 2: movdqa <z15_stack=368(%esp),>mp=%xmm4
+movdqa 368(%esp),%xmm4
+
+# qhasm:                                                                   mq = z12_stack
+# asm 1: movdqa <z12_stack=stack128#30,>mq=int6464#4
+# asm 2: movdqa <z12_stack=496(%esp),>mq=%xmm3
+movdqa 496(%esp),%xmm3
+
+# qhasm:                                                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#35,>mr=int6464#6
+# asm 2: movdqa <z13_stack=576(%esp),>mr=%xmm5
+movdqa 576(%esp),%xmm5
+
+# qhasm:                                                   z9_stack = s
+# asm 1: movdqa <s=int6464#7,>z9_stack=stack128#32
+# asm 2: movdqa <s=%xmm6,>z9_stack=528(%esp)
+movdqa %xmm6,528(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#22
+# asm 2: movdqa <p=%xmm0,>z10_stack=368(%esp)
+movdqa %xmm0,368(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z12_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z12_stack=stack128#35
+# asm 2: movdqa <mq=%xmm3,>z12_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <mr=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>t=int6464#3
+# asm 2: movdqa <z12_stack=576(%esp),>t=%xmm2
+movdqa 576(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>p=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>p=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:                   q = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>q=int6464#4
+# asm 2: movdqa <z4_stack=560(%esp),>q=%xmm3
+movdqa 560(%esp),%xmm3
+
+# qhasm:                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>r=int6464#2
+# asm 2: movdqa <z8_stack=608(%esp),>r=%xmm1
+movdqa 608(%esp),%xmm1
+
+# qhasm:                                                                   z14_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z14_stack=stack128#24
+# asm 2: movdqa <ms=%xmm6,>z14_stack=400(%esp)
+movdqa %xmm6,400(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#23
+# asm 2: movdqa <mp=%xmm4,>z15_stack=384(%esp)
+movdqa %xmm4,384(%esp)
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int32#1
+# asm 2: sub  $2,<i=%eax
+sub  $2,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   out = out_stack
+# asm 1: movl <out_stack=stack32#6,>out=int32#6
+# asm 2: movl <out_stack=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:   z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>z0=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>z0=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:   z1 = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>z1=int6464#2
+# asm 2: movdqa <z1_stack=464(%esp),>z1=%xmm1
+movdqa 464(%esp),%xmm1
+
+# qhasm:   z2 = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>z2=int6464#3
+# asm 2: movdqa <z2_stack=512(%esp),>z2=%xmm2
+movdqa 512(%esp),%xmm2
+
+# qhasm:   z3 = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>z3=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>z3=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#1
+# asm 2: paddd <orig0=144(%esp),<z0=%xmm0
+paddd 144(%esp),%xmm0
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#2
+# asm 2: paddd <orig1=208(%esp),<z1=%xmm1
+paddd 208(%esp),%xmm1
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#3
+# asm 2: paddd <orig2=256(%esp),<z2=%xmm2
+paddd 256(%esp),%xmm2
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#4
+# asm 2: paddd <orig3=304(%esp),<z3=%xmm3
+paddd 304(%esp),%xmm3
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int32#5),<in0=int32#1
+# asm 2: xorl 64(<m=%esi),<in0=%eax
+xorl 64(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int32#5),<in1=int32#2
+# asm 2: xorl 68(<m=%esi),<in1=%ecx
+xorl 68(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int32#5),<in2=int32#3
+# asm 2: xorl 72(<m=%esi),<in2=%edx
+xorl 72(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int32#5),<in3=int32#4
+# asm 2: xorl 76(<m=%esi),<in3=%ebx
+xorl 76(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl <in0=int32#1,64(<out=int32#6)
+# asm 2: movl <in0=%eax,64(<out=%edi)
+movl %eax,64(%edi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl <in1=int32#2,68(<out=int32#6)
+# asm 2: movl <in1=%ecx,68(<out=%edi)
+movl %ecx,68(%edi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl <in2=int32#3,72(<out=int32#6)
+# asm 2: movl <in2=%edx,72(<out=%edi)
+movl %edx,72(%edi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl <in3=int32#4,76(<out=int32#6)
+# asm 2: movl <in3=%ebx,76(<out=%edi)
+movl %ebx,76(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int32#5),<in0=int32#1
+# asm 2: xorl 128(<m=%esi),<in0=%eax
+xorl 128(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int32#5),<in1=int32#2
+# asm 2: xorl 132(<m=%esi),<in1=%ecx
+xorl 132(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int32#5),<in2=int32#3
+# asm 2: xorl 136(<m=%esi),<in2=%edx
+xorl 136(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int32#5),<in3=int32#4
+# asm 2: xorl 140(<m=%esi),<in3=%ebx
+xorl 140(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl <in0=int32#1,128(<out=int32#6)
+# asm 2: movl <in0=%eax,128(<out=%edi)
+movl %eax,128(%edi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl <in1=int32#2,132(<out=int32#6)
+# asm 2: movl <in1=%ecx,132(<out=%edi)
+movl %ecx,132(%edi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl <in2=int32#3,136(<out=int32#6)
+# asm 2: movl <in2=%edx,136(<out=%edi)
+movl %edx,136(%edi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl <in3=int32#4,140(<out=int32#6)
+# asm 2: movl <in3=%ebx,140(<out=%edi)
+movl %ebx,140(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int32#5),<in0=int32#1
+# asm 2: xorl 192(<m=%esi),<in0=%eax
+xorl 192(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int32#5),<in1=int32#2
+# asm 2: xorl 196(<m=%esi),<in1=%ecx
+xorl 196(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int32#5),<in2=int32#3
+# asm 2: xorl 200(<m=%esi),<in2=%edx
+xorl 200(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int32#5),<in3=int32#4
+# asm 2: xorl 204(<m=%esi),<in3=%ebx
+xorl 204(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl <in0=int32#1,192(<out=int32#6)
+# asm 2: movl <in0=%eax,192(<out=%edi)
+movl %eax,192(%edi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl <in1=int32#2,196(<out=int32#6)
+# asm 2: movl <in1=%ecx,196(<out=%edi)
+movl %ecx,196(%edi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl <in2=int32#3,200(<out=int32#6)
+# asm 2: movl <in2=%edx,200(<out=%edi)
+movl %edx,200(%edi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl <in3=int32#4,204(<out=int32#6)
+# asm 2: movl <in3=%ebx,204(<out=%edi)
+movl %ebx,204(%edi)
+
+# qhasm:   z4 = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>z4=int6464#1
+# asm 2: movdqa <z4_stack=560(%esp),>z4=%xmm0
+movdqa 560(%esp),%xmm0
+
+# qhasm:   z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>z5=int6464#2
+# asm 2: movdqa <z5_stack=352(%esp),>z5=%xmm1
+movdqa 352(%esp),%xmm1
+
+# qhasm:   z6 = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>z6=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>z6=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:   z7 = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>z7=int6464#4
+# asm 2: movdqa <z7_stack=480(%esp),>z7=%xmm3
+movdqa 480(%esp),%xmm3
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#1
+# asm 2: paddd <orig4=272(%esp),<z4=%xmm0
+paddd 272(%esp),%xmm0
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#2
+# asm 2: paddd <orig5=96(%esp),<z5=%xmm1
+paddd 96(%esp),%xmm1
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#3
+# asm 2: paddd <orig6=160(%esp),<z6=%xmm2
+paddd 160(%esp),%xmm2
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#4
+# asm 2: paddd <orig7=224(%esp),<z7=%xmm3
+paddd 224(%esp),%xmm3
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#1
+# asm 2: xorl 16(<m=%esi),<in4=%eax
+xorl 16(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#2
+# asm 2: xorl 20(<m=%esi),<in5=%ecx
+xorl 20(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#3
+# asm 2: xorl 24(<m=%esi),<in6=%edx
+xorl 24(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#4
+# asm 2: xorl 28(<m=%esi),<in7=%ebx
+xorl 28(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#1,16(<out=int32#6)
+# asm 2: movl <in4=%eax,16(<out=%edi)
+movl %eax,16(%edi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#2,20(<out=int32#6)
+# asm 2: movl <in5=%ecx,20(<out=%edi)
+movl %ecx,20(%edi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#3,24(<out=int32#6)
+# asm 2: movl <in6=%edx,24(<out=%edi)
+movl %edx,24(%edi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#4,28(<out=int32#6)
+# asm 2: movl <in7=%ebx,28(<out=%edi)
+movl %ebx,28(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int32#5),<in4=int32#1
+# asm 2: xorl 80(<m=%esi),<in4=%eax
+xorl 80(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int32#5),<in5=int32#2
+# asm 2: xorl 84(<m=%esi),<in5=%ecx
+xorl 84(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int32#5),<in6=int32#3
+# asm 2: xorl 88(<m=%esi),<in6=%edx
+xorl 88(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int32#5),<in7=int32#4
+# asm 2: xorl 92(<m=%esi),<in7=%ebx
+xorl 92(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl <in4=int32#1,80(<out=int32#6)
+# asm 2: movl <in4=%eax,80(<out=%edi)
+movl %eax,80(%edi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl <in5=int32#2,84(<out=int32#6)
+# asm 2: movl <in5=%ecx,84(<out=%edi)
+movl %ecx,84(%edi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl <in6=int32#3,88(<out=int32#6)
+# asm 2: movl <in6=%edx,88(<out=%edi)
+movl %edx,88(%edi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl <in7=int32#4,92(<out=int32#6)
+# asm 2: movl <in7=%ebx,92(<out=%edi)
+movl %ebx,92(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int32#5),<in4=int32#1
+# asm 2: xorl 144(<m=%esi),<in4=%eax
+xorl 144(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int32#5),<in5=int32#2
+# asm 2: xorl 148(<m=%esi),<in5=%ecx
+xorl 148(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int32#5),<in6=int32#3
+# asm 2: xorl 152(<m=%esi),<in6=%edx
+xorl 152(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int32#5),<in7=int32#4
+# asm 2: xorl 156(<m=%esi),<in7=%ebx
+xorl 156(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl <in4=int32#1,144(<out=int32#6)
+# asm 2: movl <in4=%eax,144(<out=%edi)
+movl %eax,144(%edi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl <in5=int32#2,148(<out=int32#6)
+# asm 2: movl <in5=%ecx,148(<out=%edi)
+movl %ecx,148(%edi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl <in6=int32#3,152(<out=int32#6)
+# asm 2: movl <in6=%edx,152(<out=%edi)
+movl %edx,152(%edi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl <in7=int32#4,156(<out=int32#6)
+# asm 2: movl <in7=%ebx,156(<out=%edi)
+movl %ebx,156(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int32#5),<in4=int32#1
+# asm 2: xorl 208(<m=%esi),<in4=%eax
+xorl 208(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int32#5),<in5=int32#2
+# asm 2: xorl 212(<m=%esi),<in5=%ecx
+xorl 212(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int32#5),<in6=int32#3
+# asm 2: xorl 216(<m=%esi),<in6=%edx
+xorl 216(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int32#5),<in7=int32#4
+# asm 2: xorl 220(<m=%esi),<in7=%ebx
+xorl 220(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl <in4=int32#1,208(<out=int32#6)
+# asm 2: movl <in4=%eax,208(<out=%edi)
+movl %eax,208(%edi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl <in5=int32#2,212(<out=int32#6)
+# asm 2: movl <in5=%ecx,212(<out=%edi)
+movl %ecx,212(%edi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl <in6=int32#3,216(<out=int32#6)
+# asm 2: movl <in6=%edx,216(<out=%edi)
+movl %edx,216(%edi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl <in7=int32#4,220(<out=int32#6)
+# asm 2: movl <in7=%ebx,220(<out=%edi)
+movl %ebx,220(%edi)
+
+# qhasm:   z8 = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>z8=int6464#1
+# asm 2: movdqa <z8_stack=608(%esp),>z8=%xmm0
+movdqa 608(%esp),%xmm0
+
+# qhasm:   z9 = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>z9=int6464#2
+# asm 2: movdqa <z9_stack=528(%esp),>z9=%xmm1
+movdqa 528(%esp),%xmm1
+
+# qhasm:   z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>z10=int6464#3
+# asm 2: movdqa <z10_stack=368(%esp),>z10=%xmm2
+movdqa 368(%esp),%xmm2
+
+# qhasm:   z11 = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>z11=int6464#4
+# asm 2: movdqa <z11_stack=448(%esp),>z11=%xmm3
+movdqa 448(%esp),%xmm3
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#1
+# asm 2: paddd <orig8=320(%esp),<z8=%xmm0
+paddd 320(%esp),%xmm0
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#2
+# asm 2: paddd <orig9=336(%esp),<z9=%xmm1
+paddd 336(%esp),%xmm1
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#3
+# asm 2: paddd <orig10=112(%esp),<z10=%xmm2
+paddd 112(%esp),%xmm2
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#4
+# asm 2: paddd <orig11=176(%esp),<z11=%xmm3
+paddd 176(%esp),%xmm3
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#1
+# asm 2: xorl 32(<m=%esi),<in8=%eax
+xorl 32(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#2
+# asm 2: xorl 36(<m=%esi),<in9=%ecx
+xorl 36(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#3
+# asm 2: xorl 40(<m=%esi),<in10=%edx
+xorl 40(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#4
+# asm 2: xorl 44(<m=%esi),<in11=%ebx
+xorl 44(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#1,32(<out=int32#6)
+# asm 2: movl <in8=%eax,32(<out=%edi)
+movl %eax,32(%edi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#2,36(<out=int32#6)
+# asm 2: movl <in9=%ecx,36(<out=%edi)
+movl %ecx,36(%edi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#3,40(<out=int32#6)
+# asm 2: movl <in10=%edx,40(<out=%edi)
+movl %edx,40(%edi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#4,44(<out=int32#6)
+# asm 2: movl <in11=%ebx,44(<out=%edi)
+movl %ebx,44(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int32#5),<in8=int32#1
+# asm 2: xorl 96(<m=%esi),<in8=%eax
+xorl 96(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int32#5),<in9=int32#2
+# asm 2: xorl 100(<m=%esi),<in9=%ecx
+xorl 100(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int32#5),<in10=int32#3
+# asm 2: xorl 104(<m=%esi),<in10=%edx
+xorl 104(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int32#5),<in11=int32#4
+# asm 2: xorl 108(<m=%esi),<in11=%ebx
+xorl 108(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl <in8=int32#1,96(<out=int32#6)
+# asm 2: movl <in8=%eax,96(<out=%edi)
+movl %eax,96(%edi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl <in9=int32#2,100(<out=int32#6)
+# asm 2: movl <in9=%ecx,100(<out=%edi)
+movl %ecx,100(%edi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl <in10=int32#3,104(<out=int32#6)
+# asm 2: movl <in10=%edx,104(<out=%edi)
+movl %edx,104(%edi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl <in11=int32#4,108(<out=int32#6)
+# asm 2: movl <in11=%ebx,108(<out=%edi)
+movl %ebx,108(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int32#5),<in8=int32#1
+# asm 2: xorl 160(<m=%esi),<in8=%eax
+xorl 160(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int32#5),<in9=int32#2
+# asm 2: xorl 164(<m=%esi),<in9=%ecx
+xorl 164(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int32#5),<in10=int32#3
+# asm 2: xorl 168(<m=%esi),<in10=%edx
+xorl 168(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int32#5),<in11=int32#4
+# asm 2: xorl 172(<m=%esi),<in11=%ebx
+xorl 172(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl <in8=int32#1,160(<out=int32#6)
+# asm 2: movl <in8=%eax,160(<out=%edi)
+movl %eax,160(%edi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl <in9=int32#2,164(<out=int32#6)
+# asm 2: movl <in9=%ecx,164(<out=%edi)
+movl %ecx,164(%edi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl <in10=int32#3,168(<out=int32#6)
+# asm 2: movl <in10=%edx,168(<out=%edi)
+movl %edx,168(%edi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl <in11=int32#4,172(<out=int32#6)
+# asm 2: movl <in11=%ebx,172(<out=%edi)
+movl %ebx,172(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int32#5),<in8=int32#1
+# asm 2: xorl 224(<m=%esi),<in8=%eax
+xorl 224(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int32#5),<in9=int32#2
+# asm 2: xorl 228(<m=%esi),<in9=%ecx
+xorl 228(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int32#5),<in10=int32#3
+# asm 2: xorl 232(<m=%esi),<in10=%edx
+xorl 232(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int32#5),<in11=int32#4
+# asm 2: xorl 236(<m=%esi),<in11=%ebx
+xorl 236(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl <in8=int32#1,224(<out=int32#6)
+# asm 2: movl <in8=%eax,224(<out=%edi)
+movl %eax,224(%edi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl <in9=int32#2,228(<out=int32#6)
+# asm 2: movl <in9=%ecx,228(<out=%edi)
+movl %ecx,228(%edi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl <in10=int32#3,232(<out=int32#6)
+# asm 2: movl <in10=%edx,232(<out=%edi)
+movl %edx,232(%edi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl <in11=int32#4,236(<out=int32#6)
+# asm 2: movl <in11=%ebx,236(<out=%edi)
+movl %ebx,236(%edi)
+
+# qhasm:   z12 = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>z12=int6464#1
+# asm 2: movdqa <z12_stack=576(%esp),>z12=%xmm0
+movdqa 576(%esp),%xmm0
+
+# qhasm:   z13 = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>z13=int6464#2
+# asm 2: movdqa <z13_stack=496(%esp),>z13=%xmm1
+movdqa 496(%esp),%xmm1
+
+# qhasm:   z14 = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>z14=int6464#3
+# asm 2: movdqa <z14_stack=400(%esp),>z14=%xmm2
+movdqa 400(%esp),%xmm2
+
+# qhasm:   z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>z15=int6464#4
+# asm 2: movdqa <z15_stack=384(%esp),>z15=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#1
+# asm 2: paddd <orig12=192(%esp),<z12=%xmm0
+paddd 192(%esp),%xmm0
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#2
+# asm 2: paddd <orig13=240(%esp),<z13=%xmm1
+paddd 240(%esp),%xmm1
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#3
+# asm 2: paddd <orig14=288(%esp),<z14=%xmm2
+paddd 288(%esp),%xmm2
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#4
+# asm 2: paddd <orig15=128(%esp),<z15=%xmm3
+paddd 128(%esp),%xmm3
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#1
+# asm 2: xorl 48(<m=%esi),<in12=%eax
+xorl 48(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#2
+# asm 2: xorl 52(<m=%esi),<in13=%ecx
+xorl 52(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#3
+# asm 2: xorl 56(<m=%esi),<in14=%edx
+xorl 56(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#4
+# asm 2: xorl 60(<m=%esi),<in15=%ebx
+xorl 60(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#1,48(<out=int32#6)
+# asm 2: movl <in12=%eax,48(<out=%edi)
+movl %eax,48(%edi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#2,52(<out=int32#6)
+# asm 2: movl <in13=%ecx,52(<out=%edi)
+movl %ecx,52(%edi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#3,56(<out=int32#6)
+# asm 2: movl <in14=%edx,56(<out=%edi)
+movl %edx,56(%edi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#4,60(<out=int32#6)
+# asm 2: movl <in15=%ebx,60(<out=%edi)
+movl %ebx,60(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int32#5),<in12=int32#1
+# asm 2: xorl 112(<m=%esi),<in12=%eax
+xorl 112(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int32#5),<in13=int32#2
+# asm 2: xorl 116(<m=%esi),<in13=%ecx
+xorl 116(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int32#5),<in14=int32#3
+# asm 2: xorl 120(<m=%esi),<in14=%edx
+xorl 120(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int32#5),<in15=int32#4
+# asm 2: xorl 124(<m=%esi),<in15=%ebx
+xorl 124(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl <in12=int32#1,112(<out=int32#6)
+# asm 2: movl <in12=%eax,112(<out=%edi)
+movl %eax,112(%edi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl <in13=int32#2,116(<out=int32#6)
+# asm 2: movl <in13=%ecx,116(<out=%edi)
+movl %ecx,116(%edi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl <in14=int32#3,120(<out=int32#6)
+# asm 2: movl <in14=%edx,120(<out=%edi)
+movl %edx,120(%edi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl <in15=int32#4,124(<out=int32#6)
+# asm 2: movl <in15=%ebx,124(<out=%edi)
+movl %ebx,124(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int32#5),<in12=int32#1
+# asm 2: xorl 176(<m=%esi),<in12=%eax
+xorl 176(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int32#5),<in13=int32#2
+# asm 2: xorl 180(<m=%esi),<in13=%ecx
+xorl 180(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int32#5),<in14=int32#3
+# asm 2: xorl 184(<m=%esi),<in14=%edx
+xorl 184(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int32#5),<in15=int32#4
+# asm 2: xorl 188(<m=%esi),<in15=%ebx
+xorl 188(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl <in12=int32#1,176(<out=int32#6)
+# asm 2: movl <in12=%eax,176(<out=%edi)
+movl %eax,176(%edi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl <in13=int32#2,180(<out=int32#6)
+# asm 2: movl <in13=%ecx,180(<out=%edi)
+movl %ecx,180(%edi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl <in14=int32#3,184(<out=int32#6)
+# asm 2: movl <in14=%edx,184(<out=%edi)
+movl %edx,184(%edi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl <in15=int32#4,188(<out=int32#6)
+# asm 2: movl <in15=%ebx,188(<out=%edi)
+movl %ebx,188(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int32#5),<in12=int32#1
+# asm 2: xorl 240(<m=%esi),<in12=%eax
+xorl 240(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int32#5),<in13=int32#2
+# asm 2: xorl 244(<m=%esi),<in13=%ecx
+xorl 244(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int32#5),<in14=int32#3
+# asm 2: xorl 248(<m=%esi),<in14=%edx
+xorl 248(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int32#5),<in15=int32#4
+# asm 2: xorl 252(<m=%esi),<in15=%ebx
+xorl 252(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl <in12=int32#1,240(<out=int32#6)
+# asm 2: movl <in12=%eax,240(<out=%edi)
+movl %eax,240(%edi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl <in13=int32#2,244(<out=int32#6)
+# asm 2: movl <in13=%ecx,244(<out=%edi)
+movl %ecx,244(%edi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl <in14=int32#3,248(<out=int32#6)
+# asm 2: movl <in14=%edx,248(<out=%edi)
+movl %edx,248(%edi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl <in15=int32#4,252(<out=int32#6)
+# asm 2: movl <in15=%ebx,252(<out=%edi)
+movl %ebx,252(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int32#1
+# asm 2: sub  $256,<bytes=%eax
+sub  $256,%eax
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int32#5
+# asm 2: add  $256,<m=%esi
+add  $256,%esi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int32#6
+# asm 2: add  $256,<out=%edi
+add  $256,%edi
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#1
+# asm 2: cmp  $0,<bytes=%eax
+cmp  $0,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: movl <out=int32#6,>ctarget=stack32#6
+# asm 2: movl <out=%edi,>ctarget=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     m = &tmp
+# asm 1: leal <tmp=stack512#1,>m=int32#5
+# asm 2: leal <tmp=640(%esp),>m=%esi
+leal 640(%esp),%esi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#3,>diag0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>diag0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#2,>diag1=int6464#2
+# asm 2: movdqa <x1=48(%esp),>diag1=%xmm1
+movdqa 48(%esp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#4,>diag2=int6464#3
+# asm 2: movdqa <x2=80(%esp),>diag2=%xmm2
+movdqa 80(%esp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#1,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%esp),>diag3=%xmm3
+movdqa 32(%esp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 20
+# asm 1: mov  $20,>i=int32#1
+# asm 2: mov  $20,>i=%eax
+mov  $20,%eax
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int32#1
+# asm 2: sub  $4,<i=%eax
+sub  $4,%eax
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#3,<diag0=int6464#1
+# asm 2: paddd <x0=64(%esp),<diag0=%xmm0
+paddd 64(%esp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#2,<diag1=int6464#2
+# asm 2: paddd <x1=48(%esp),<diag1=%xmm1
+paddd 48(%esp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#4,<diag2=int6464#3
+# asm 2: paddd <x2=80(%esp),<diag2=%xmm2
+paddd 80(%esp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#1,<diag3=int6464#4
+# asm 2: paddd <x3=32(%esp),<diag3=%xmm3
+paddd 32(%esp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int32#1
+# asm 2: movd   <diag0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int32#2
+# asm 2: movd   <diag1=%xmm1,>in12=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int32#3
+# asm 2: movd   <diag2=%xmm2,>in8=%edx
+movd   %xmm2,%edx
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int32#4
+# asm 2: movd   <diag3=%xmm3,>in4=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm: in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#2
+# asm 2: xorl 48(<m=%esi),<in12=%ecx
+xorl 48(%esi),%ecx
+
+# qhasm: in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#3
+# asm 2: xorl 32(<m=%esi),<in8=%edx
+xorl 32(%esi),%edx
+
+# qhasm: in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#4
+# asm 2: xorl 16(<m=%esi),<in4=%ebx
+xorl 16(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#2,48(<out=int32#6)
+# asm 2: movl <in12=%ecx,48(<out=%edi)
+movl %ecx,48(%edi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#3,32(<out=int32#6)
+# asm 2: movl <in8=%edx,32(<out=%edi)
+movl %edx,32(%edi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#4,16(<out=int32#6)
+# asm 2: movl <in4=%ebx,16(<out=%edi)
+movl %ebx,16(%edi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int32#1
+# asm 2: movd   <diag0=%xmm0,>in5=%eax
+movd   %xmm0,%eax
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int32#2
+# asm 2: movd   <diag1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int32#3
+# asm 2: movd   <diag2=%xmm2,>in13=%edx
+movd   %xmm2,%edx
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int32#4
+# asm 2: movd   <diag3=%xmm3,>in9=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#1
+# asm 2: xorl 20(<m=%esi),<in5=%eax
+xorl 20(%esi),%eax
+
+# qhasm: in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm: in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#3
+# asm 2: xorl 52(<m=%esi),<in13=%edx
+xorl 52(%esi),%edx
+
+# qhasm: in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#4
+# asm 2: xorl 36(<m=%esi),<in9=%ebx
+xorl 36(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#1,20(<out=int32#6)
+# asm 2: movl <in5=%eax,20(<out=%edi)
+movl %eax,20(%edi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#3,52(<out=int32#6)
+# asm 2: movl <in13=%edx,52(<out=%edi)
+movl %edx,52(%edi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#4,36(<out=int32#6)
+# asm 2: movl <in9=%ebx,36(<out=%edi)
+movl %ebx,36(%edi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int32#1
+# asm 2: movd   <diag0=%xmm0,>in10=%eax
+movd   %xmm0,%eax
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int32#2
+# asm 2: movd   <diag1=%xmm1,>in6=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int32#3
+# asm 2: movd   <diag2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int32#4
+# asm 2: movd   <diag3=%xmm3,>in14=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#1
+# asm 2: xorl 40(<m=%esi),<in10=%eax
+xorl 40(%esi),%eax
+
+# qhasm: in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#2
+# asm 2: xorl 24(<m=%esi),<in6=%ecx
+xorl 24(%esi),%ecx
+
+# qhasm: in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm: in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#4
+# asm 2: xorl 56(<m=%esi),<in14=%ebx
+xorl 56(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#1,40(<out=int32#6)
+# asm 2: movl <in10=%eax,40(<out=%edi)
+movl %eax,40(%edi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#2,24(<out=int32#6)
+# asm 2: movl <in6=%ecx,24(<out=%edi)
+movl %ecx,24(%edi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#4,56(<out=int32#6)
+# asm 2: movl <in14=%ebx,56(<out=%edi)
+movl %ebx,56(%edi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int32#1
+# asm 2: movd   <diag0=%xmm0,>in15=%eax
+movd   %xmm0,%eax
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int32#2
+# asm 2: movd   <diag1=%xmm1,>in11=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int32#3
+# asm 2: movd   <diag2=%xmm2,>in7=%edx
+movd   %xmm2,%edx
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int32#4
+# asm 2: movd   <diag3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#1
+# asm 2: xorl 60(<m=%esi),<in15=%eax
+xorl 60(%esi),%eax
+
+# qhasm: in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#2
+# asm 2: xorl 44(<m=%esi),<in11=%ecx
+xorl 44(%esi),%ecx
+
+# qhasm: in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#3
+# asm 2: xorl 28(<m=%esi),<in7=%edx
+xorl 28(%esi),%edx
+
+# qhasm: in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#1,60(<out=int32#6)
+# asm 2: movl <in15=%eax,60(<out=%edi)
+movl %eax,60(%edi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#2,44(<out=int32#6)
+# asm 2: movl <in11=%ecx,44(<out=%edi)
+movl %ecx,44(%edi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#3,28(<out=int32#6)
+# asm 2: movl <in7=%edx,28(<out=%edi)
+movl %edx,28(%edi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm:       out = ctarget
+# asm 1: movl <ctarget=stack32#6,>out=int32#6
+# asm 2: movl <ctarget=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     eax = eax_stack
+# asm 1: movl <eax_stack=stack32#1,>eax=int32#1
+# asm 2: movl <eax_stack=0(%esp),>eax=%eax
+movl 0(%esp),%eax
+
+# qhasm:     ebx = ebx_stack
+# asm 1: movl <ebx_stack=stack32#2,>ebx=int32#4
+# asm 2: movl <ebx_stack=4(%esp),>ebx=%ebx
+movl 4(%esp),%ebx
+
+# qhasm:     esi = esi_stack
+# asm 1: movl <esi_stack=stack32#3,>esi=int32#5
+# asm 2: movl <esi_stack=8(%esp),>esi=%esi
+movl 8(%esp),%esi
+
+# qhasm:     edi = edi_stack
+# asm 1: movl <edi_stack=stack32#4,>edi=int32#6
+# asm 2: movl <edi_stack=12(%esp),>edi=%edi
+movl 12(%esp),%edi
+
+# qhasm:     ebp = ebp_stack
+# asm 1: movl <ebp_stack=stack32#5,>ebp=int32#7
+# asm 2: movl <ebp_stack=16(%esp),>ebp=%ebp
+movl 16(%esp),%ebp
+
+# qhasm:     leave
+add %eax,%esp
+xor %eax,%eax
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int32#1
+# asm 2: sub  $64,<bytes=%eax
+sub  $64,%eax
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int32#6
+# asm 2: add  $64,<out=%edi
+add  $64,%edi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int32#5
+# asm 2: add  $64,<m=%esi
+add  $64,%esi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/salsa2012/amd64_xmm6/api.h b/nacl/crypto_stream/salsa2012/amd64_xmm6/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/amd64_xmm6/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa2012/amd64_xmm6/implementors b/nacl/crypto_stream/salsa2012/amd64_xmm6/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/amd64_xmm6/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa2012/amd64_xmm6/stream.s b/nacl/crypto_stream/salsa2012/amd64_xmm6/stream.s
new file mode 100644
index 00000000..0e26dc9f
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/amd64_xmm6/stream.s
@@ -0,0 +1,4823 @@
+
+# qhasm: int64 r11_caller
+
+# qhasm: int64 r12_caller
+
+# qhasm: int64 r13_caller
+
+# qhasm: int64 r14_caller
+
+# qhasm: int64 r15_caller
+
+# qhasm: int64 rbx_caller
+
+# qhasm: int64 rbp_caller
+
+# qhasm: caller r11_caller
+
+# qhasm: caller r12_caller
+
+# qhasm: caller r13_caller
+
+# qhasm: caller r14_caller
+
+# qhasm: caller r15_caller
+
+# qhasm: caller rbx_caller
+
+# qhasm: caller rbp_caller
+
+# qhasm: stack64 r11_stack
+
+# qhasm: stack64 r12_stack
+
+# qhasm: stack64 r13_stack
+
+# qhasm: stack64 r14_stack
+
+# qhasm: stack64 r15_stack
+
+# qhasm: stack64 rbx_stack
+
+# qhasm: stack64 rbp_stack
+
+# qhasm: int64 a
+
+# qhasm: int64 arg1
+
+# qhasm: int64 arg2
+
+# qhasm: int64 arg3
+
+# qhasm: int64 arg4
+
+# qhasm: int64 arg5
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: int64 k
+
+# qhasm: int64 kbits
+
+# qhasm: int64 iv
+
+# qhasm: int64 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int64 m
+
+# qhasm: int64 out
+
+# qhasm: int64 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: int6464 y0
+
+# qhasm: int6464 y1
+
+# qhasm: int6464 y2
+
+# qhasm: int6464 y3
+
+# qhasm: int6464 y4
+
+# qhasm: int6464 y5
+
+# qhasm: int6464 y6
+
+# qhasm: int6464 y7
+
+# qhasm: int6464 y8
+
+# qhasm: int6464 y9
+
+# qhasm: int6464 y10
+
+# qhasm: int6464 y11
+
+# qhasm: int6464 y12
+
+# qhasm: int6464 y13
+
+# qhasm: int6464 y14
+
+# qhasm: int6464 y15
+
+# qhasm: int6464 r0
+
+# qhasm: int6464 r1
+
+# qhasm: int6464 r2
+
+# qhasm: int6464 r3
+
+# qhasm: int6464 r4
+
+# qhasm: int6464 r5
+
+# qhasm: int6464 r6
+
+# qhasm: int6464 r7
+
+# qhasm: int6464 r8
+
+# qhasm: int6464 r9
+
+# qhasm: int6464 r10
+
+# qhasm: int6464 r11
+
+# qhasm: int6464 r12
+
+# qhasm: int6464 r13
+
+# qhasm: int6464 r14
+
+# qhasm: int6464 r15
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int64 in0
+
+# qhasm: int64 in1
+
+# qhasm: int64 in2
+
+# qhasm: int64 in3
+
+# qhasm: int64 in4
+
+# qhasm: int64 in5
+
+# qhasm: int64 in6
+
+# qhasm: int64 in7
+
+# qhasm: int64 in8
+
+# qhasm: int64 in9
+
+# qhasm: int64 in10
+
+# qhasm: int64 in11
+
+# qhasm: int64 in12
+
+# qhasm: int64 in13
+
+# qhasm: int64 in14
+
+# qhasm: int64 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: int64 ctarget
+
+# qhasm: stack64 bytes_backup
+
+# qhasm: enter crypto_stream_salsa2012_amd64_xmm6
+.text
+.p2align 5
+.globl _crypto_stream_salsa2012_amd64_xmm6
+.globl crypto_stream_salsa2012_amd64_xmm6
+_crypto_stream_salsa2012_amd64_xmm6:
+crypto_stream_salsa2012_amd64_xmm6:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: bytes = arg2
+# asm 1: mov  <arg2=int64#2,>bytes=int64#6
+# asm 2: mov  <arg2=%rsi,>bytes=%r9
+mov  %rsi,%r9
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm: iv = arg3
+# asm 1: mov  <arg3=int64#3,>iv=int64#3
+# asm 2: mov  <arg3=%rdx,>iv=%rdx
+mov  %rdx,%rdx
+
+# qhasm: k = arg4
+# asm 1: mov  <arg4=int64#4,>k=int64#8
+# asm 2: mov  <arg4=%rcx,>k=%r10
+mov  %rcx,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int64#7
+# asm 2: mov  $0,>a=%rax
+mov  $0,%rax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: sub  <bytes=int64#6,<out=int64#1
+# asm 2: sub  <bytes=%r9,<out=%rdi
+sub  %r9,%rdi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa2012_amd64_xmm6_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa2012_amd64_xmm6_xor
+.globl crypto_stream_salsa2012_amd64_xmm6_xor
+_crypto_stream_salsa2012_amd64_xmm6_xor:
+crypto_stream_salsa2012_amd64_xmm6_xor:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = arg2
+# asm 1: mov  <arg2=int64#2,>m=int64#2
+# asm 2: mov  <arg2=%rsi,>m=%rsi
+mov  %rsi,%rsi
+
+# qhasm: bytes = arg3
+# asm 1: mov  <arg3=int64#3,>bytes=int64#6
+# asm 2: mov  <arg3=%rdx,>bytes=%r9
+mov  %rdx,%r9
+
+# qhasm: iv = arg4
+# asm 1: mov  <arg4=int64#4,>iv=int64#3
+# asm 2: mov  <arg4=%rcx,>iv=%rdx
+mov  %rcx,%rdx
+
+# qhasm: k = arg5
+# asm 1: mov  <arg5=int64#5,>k=int64#8
+# asm 2: mov  <arg5=%r8,>k=%r10
+mov  %r8,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl   20(<k=int64#8),>in12=int64#4d
+# asm 2: movl   20(<k=%r10),>in12=%ecx
+movl   20(%r10),%ecx
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl   0(<k=int64#8),>in1=int64#5d
+# asm 2: movl   0(<k=%r10),>in1=%r8d
+movl   0(%r10),%r8d
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl   0(<iv=int64#3),>in6=int64#7d
+# asm 2: movl   0(<iv=%rdx),>in6=%eax
+movl   0(%rdx),%eax
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl   16(<k=int64#8),>in11=int64#9d
+# asm 2: movl   16(<k=%r10),>in11=%r11d
+movl   16(%r10),%r11d
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int64#4d,>x1=stack128#1
+# asm 2: movl <in12=%ecx,>x1=0(%rsp)
+movl %ecx,0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int64#5d,4+<x1=stack128#1
+# asm 2: movl <in1=%r8d,4+<x1=0(%rsp)
+movl %r8d,4+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int64#7d,8+<x1=stack128#1
+# asm 2: movl <in6=%eax,8+<x1=0(%rsp)
+movl %eax,8+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int64#9d,12+<x1=stack128#1
+# asm 2: movl <in11=%r11d,12+<x1=0(%rsp)
+movl %r11d,12+0(%rsp)
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int64#4
+# asm 2: mov  $0,>in8=%rcx
+mov  $0,%rcx
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl   24(<k=int64#8),>in13=int64#5d
+# asm 2: movl   24(<k=%r10),>in13=%r8d
+movl   24(%r10),%r8d
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl   4(<k=int64#8),>in2=int64#7d
+# asm 2: movl   4(<k=%r10),>in2=%eax
+movl   4(%r10),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl   4(<iv=int64#3),>in7=int64#3d
+# asm 2: movl   4(<iv=%rdx),>in7=%edx
+movl   4(%rdx),%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int64#5d,4+<x2=stack128#2
+# asm 2: movl <in13=%r8d,4+<x2=16(%rsp)
+movl %r8d,4+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int64#7d,8+<x2=stack128#2
+# asm 2: movl <in2=%eax,8+<x2=16(%rsp)
+movl %eax,8+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int64#3d,12+<x2=stack128#2
+# asm 2: movl <in7=%edx,12+<x2=16(%rsp)
+movl %edx,12+16(%rsp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl   12(<k=int64#8),>in4=int64#3d
+# asm 2: movl   12(<k=%r10),>in4=%edx
+movl   12(%r10),%edx
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int64#4
+# asm 2: mov  $0,>in9=%rcx
+mov  $0,%rcx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl   28(<k=int64#8),>in14=int64#5d
+# asm 2: movl   28(<k=%r10),>in14=%r8d
+movl   28(%r10),%r8d
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl   8(<k=int64#8),>in3=int64#7d
+# asm 2: movl   8(<k=%r10),>in3=%eax
+movl   8(%r10),%eax
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int64#3d,>x3=stack128#3
+# asm 2: movl <in4=%edx,>x3=32(%rsp)
+movl %edx,32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int64#5d,8+<x3=stack128#3
+# asm 2: movl <in14=%r8d,8+<x3=32(%rsp)
+movl %r8d,8+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int64#7d,12+<x3=stack128#3
+# asm 2: movl <in3=%eax,12+<x3=32(%rsp)
+movl %eax,12+32(%rsp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int64#3
+# asm 2: mov  $1634760805,>in0=%rdx
+mov  $1634760805,%rdx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int64#4
+# asm 2: mov  $857760878,>in5=%rcx
+mov  $857760878,%rcx
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int64#5
+# asm 2: mov  $2036477234,>in10=%r8
+mov  $2036477234,%r8
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int64#7
+# asm 2: mov  $1797285236,>in15=%rax
+mov  $1797285236,%rax
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int64#3d,>x0=stack128#4
+# asm 2: movl <in0=%edx,>x0=48(%rsp)
+movl %edx,48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int64#4d,4+<x0=stack128#4
+# asm 2: movl <in5=%ecx,4+<x0=48(%rsp)
+movl %ecx,4+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int64#5d,8+<x0=stack128#4
+# asm 2: movl <in10=%r8d,8+<x0=48(%rsp)
+movl %r8d,8+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int64#7d,12+<x0=stack128#4
+# asm 2: movl <in15=%eax,12+<x0=48(%rsp)
+movl %eax,12+48(%rsp)
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#4,>z0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>z0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=64(%rsp)
+movdqa %xmm1,64(%rsp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=80(%rsp)
+movdqa %xmm2,80(%rsp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=96(%rsp)
+movdqa %xmm3,96(%rsp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=112(%rsp)
+movdqa %xmm0,112(%rsp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#1,>z1=int6464#1
+# asm 2: movdqa <x1=0(%rsp),>z1=%xmm0
+movdqa 0(%rsp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=128(%rsp)
+movdqa %xmm1,128(%rsp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=144(%rsp)
+movdqa %xmm2,144(%rsp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=160(%rsp)
+movdqa %xmm3,160(%rsp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=176(%rsp)
+movdqa %xmm0,176(%rsp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#2,>z2=int6464#1
+# asm 2: movdqa <x2=16(%rsp),>z2=%xmm0
+movdqa 16(%rsp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=192(%rsp)
+movdqa %xmm1,192(%rsp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=208(%rsp)
+movdqa %xmm2,208(%rsp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=224(%rsp)
+movdqa %xmm0,224(%rsp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#3,>z3=int6464#1
+# asm 2: movdqa <x3=32(%rsp),>z3=%xmm0
+movdqa 32(%rsp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=240(%rsp)
+movdqa %xmm1,240(%rsp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=256(%rsp)
+movdqa %xmm2,256(%rsp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=272(%rsp)
+movdqa %xmm0,272(%rsp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#3d
+# asm 2: movl <x2=16(%rsp),>in8=%edx
+movl 16(%rsp),%edx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#4d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%ecx
+movl 4+32(%rsp),%ecx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int64#3d,>orig8=stack128#19
+# asm 2: movl <in8=%edx,>orig8=288(%rsp)
+movl %edx,288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int64#4d,>orig9=stack128#20
+# asm 2: movl <in9=%ecx,>orig9=304(%rsp)
+movl %ecx,304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int64#3d,4+<orig8=stack128#19
+# asm 2: movl <in8=%edx,4+<orig8=288(%rsp)
+movl %edx,4+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,4+<orig9=304(%rsp)
+movl %ecx,4+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int64#3d,8+<orig8=stack128#19
+# asm 2: movl <in8=%edx,8+<orig8=288(%rsp)
+movl %edx,8+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int64#4d,8+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,8+<orig9=304(%rsp)
+movl %ecx,8+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int64#3d,12+<orig8=stack128#19
+# asm 2: movl <in8=%edx,12+<orig8=288(%rsp)
+movl %edx,12+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int64#4d,12+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,12+<orig9=304(%rsp)
+movl %ecx,12+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#3d,>x2=stack128#2
+# asm 2: movl <in8=%edx,>x2=16(%rsp)
+movl %edx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: i = 12
+# asm 1: mov  $12,>i=int64#3
+# asm 2: mov  $12,>i=%rdx
+mov  $12,%rdx
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=64(%rsp),>z5=%xmm0
+movdqa 64(%rsp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=80(%rsp),>z10=%xmm1
+movdqa 80(%rsp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=96(%rsp),>z15=%xmm2
+movdqa 96(%rsp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=256(%rsp),>z14=%xmm3
+movdqa 256(%rsp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=272(%rsp),>z3=%xmm4
+movdqa 272(%rsp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=128(%rsp),>z6=%xmm5
+movdqa 128(%rsp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=144(%rsp),>z11=%xmm6
+movdqa 144(%rsp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=176(%rsp),>z1=%xmm7
+movdqa 176(%rsp),%xmm7
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#9
+# asm 2: movdqa <orig7=192(%rsp),>z7=%xmm8
+movdqa 192(%rsp),%xmm8
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#10
+# asm 2: movdqa <orig13=208(%rsp),>z13=%xmm9
+movdqa 208(%rsp),%xmm9
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#11
+# asm 2: movdqa <orig2=224(%rsp),>z2=%xmm10
+movdqa 224(%rsp),%xmm10
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#12
+# asm 2: movdqa <orig9=304(%rsp),>z9=%xmm11
+movdqa 304(%rsp),%xmm11
+
+# qhasm:   z0 = orig0
+# asm 1: movdqa <orig0=stack128#8,>z0=int6464#13
+# asm 2: movdqa <orig0=112(%rsp),>z0=%xmm12
+movdqa 112(%rsp),%xmm12
+
+# qhasm:   z12 = orig12
+# asm 1: movdqa <orig12=stack128#11,>z12=int6464#14
+# asm 2: movdqa <orig12=160(%rsp),>z12=%xmm13
+movdqa 160(%rsp),%xmm13
+
+# qhasm:   z4 = orig4
+# asm 1: movdqa <orig4=stack128#16,>z4=int6464#15
+# asm 2: movdqa <orig4=240(%rsp),>z4=%xmm14
+movdqa 240(%rsp),%xmm14
+
+# qhasm:   z8 = orig8
+# asm 1: movdqa <orig8=stack128#19,>z8=int6464#16
+# asm 2: movdqa <orig8=288(%rsp),>z8=%xmm15
+movdqa 288(%rsp),%xmm15
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                y4 = z12
+# asm 1: movdqa <z12=int6464#14,>y4=int6464#2
+# asm 2: movdqa <z12=%xmm13,>y4=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm: uint32323232   y4 += z0
+# asm 1: paddd <z0=int6464#13,<y4=int6464#2
+# asm 2: paddd <z0=%xmm12,<y4=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y4 <<= 7
+# asm 1: pslld $7,<y4=int6464#2
+# asm 2: pslld $7,<y4=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232   r4 >>= 25
+# asm 1: psrld $25,<r4=int6464#3
+# asm 2: psrld $25,<r4=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                                y9 = z1
+# asm 1: movdqa <z1=int6464#8,>y9=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y9=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232                   y9 += z5
+# asm 1: paddd <z5=int6464#1,<y9=int6464#2
+# asm 2: paddd <z5=%xmm0,<y9=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                                r9 = y9
+# asm 1: movdqa <y9=int6464#2,>r9=int6464#3
+# asm 2: movdqa <y9=%xmm1,>r9=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y9 <<= 7
+# asm 1: pslld $7,<y9=int6464#2
+# asm 2: pslld $7,<y9=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#2,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm1,<z9=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm: uint32323232                   r9 >>= 25
+# asm 1: psrld $25,<r9=int6464#3
+# asm 2: psrld $25,<r9=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#3,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm2,<z9=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:                y8 = z0
+# asm 1: movdqa <z0=int6464#13,>y8=int6464#2
+# asm 2: movdqa <z0=%xmm12,>y8=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232   y8 += z4
+# asm 1: paddd <z4=int6464#15,<y8=int6464#2
+# asm 2: paddd <z4=%xmm14,<y8=%xmm1
+paddd %xmm14,%xmm1
+
+# qhasm:                r8 = y8
+# asm 1: movdqa <y8=int6464#2,>r8=int6464#3
+# asm 2: movdqa <y8=%xmm1,>r8=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#2
+# asm 2: pslld $9,<y8=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z8 ^= y8
+# asm 1: pxor  <y8=int6464#2,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm1,<z8=%xmm15
+pxor  %xmm1,%xmm15
+
+# qhasm: uint32323232   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#3
+# asm 2: psrld $23,<r8=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z8 ^= r8
+# asm 1: pxor  <r8=int6464#3,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm2,<z8=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:                                y13 = z5
+# asm 1: movdqa <z5=int6464#1,>y13=int6464#2
+# asm 2: movdqa <z5=%xmm0,>y13=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                   y13 += z9
+# asm 1: paddd <z9=int6464#12,<y13=int6464#2
+# asm 2: paddd <z9=%xmm11,<y13=%xmm1
+paddd %xmm11,%xmm1
+
+# qhasm:                                r13 = y13
+# asm 1: movdqa <y13=int6464#2,>r13=int6464#3
+# asm 2: movdqa <y13=%xmm1,>r13=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#2
+# asm 2: pslld $9,<y13=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#2,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm1,<z13=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm: uint32323232                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#3
+# asm 2: psrld $23,<r13=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#3,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm2,<z13=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:                y12 = z4
+# asm 1: movdqa <z4=int6464#15,>y12=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y12=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232   y12 += z8
+# asm 1: paddd <z8=int6464#16,<y12=int6464#2
+# asm 2: paddd <z8=%xmm15,<y12=%xmm1
+paddd %xmm15,%xmm1
+
+# qhasm:                r12 = y12
+# asm 1: movdqa <y12=int6464#2,>r12=int6464#3
+# asm 2: movdqa <y12=%xmm1,>r12=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y12 <<= 13
+# asm 1: pslld $13,<y12=int6464#2
+# asm 2: pslld $13,<y12=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z12 ^= y12
+# asm 1: pxor  <y12=int6464#2,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm1,<z12=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm: uint32323232   r12 >>= 19
+# asm 1: psrld $19,<r12=int6464#3
+# asm 2: psrld $19,<r12=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z12 ^= r12
+# asm 1: pxor  <r12=int6464#3,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm2,<z12=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:                                y1 = z9
+# asm 1: movdqa <z9=int6464#12,>y1=int6464#2
+# asm 2: movdqa <z9=%xmm11,>y1=%xmm1
+movdqa %xmm11,%xmm1
+
+# qhasm: uint32323232                   y1 += z13
+# asm 1: paddd <z13=int6464#10,<y1=int6464#2
+# asm 2: paddd <z13=%xmm9,<y1=%xmm1
+paddd %xmm9,%xmm1
+
+# qhasm:                                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#3
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y1 <<= 13
+# asm 1: pslld $13,<y1=int6464#2
+# asm 2: pslld $13,<y1=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232                   r1 >>= 19
+# asm 1: psrld $19,<r1=int6464#3
+# asm 2: psrld $19,<r1=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z1 ^= r1
+# asm 1: pxor  <r1=int6464#3,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm2,<z1=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:                y0 = z8
+# asm 1: movdqa <z8=int6464#16,>y0=int6464#2
+# asm 2: movdqa <z8=%xmm15,>y0=%xmm1
+movdqa %xmm15,%xmm1
+
+# qhasm: uint32323232   y0 += z12
+# asm 1: paddd <z12=int6464#14,<y0=int6464#2
+# asm 2: paddd <z12=%xmm13,<y0=%xmm1
+paddd %xmm13,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#13
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm12
+pxor  %xmm1,%xmm12
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#13
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#13,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm12,>z0_stack=320(%rsp)
+movdqa %xmm12,320(%rsp)
+
+# qhasm:                                y5 = z13
+# asm 1: movdqa <z13=int6464#10,>y5=int6464#3
+# asm 2: movdqa <z13=%xmm9,>y5=%xmm2
+movdqa %xmm9,%xmm2
+
+# qhasm: uint32323232                   y5 += z1
+# asm 1: paddd <z1=int6464#8,<y5=int6464#3
+# asm 2: paddd <z1=%xmm7,<y5=%xmm2
+paddd %xmm7,%xmm2
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#3,>r5=int6464#13
+# asm 2: movdqa <y5=%xmm2,>r5=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#3
+# asm 2: pslld $18,<y5=%xmm2
+pslld $18,%xmm2
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#3,<z5=int6464#1
+# asm 2: pxor  <y5=%xmm2,<z5=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#13
+# asm 2: psrld $14,<r5=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#13,<z5=int6464#1
+# asm 2: pxor  <r5=%xmm12,<z5=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:                                                y14 = z6
+# asm 1: movdqa <z6=int6464#6,>y14=int6464#3
+# asm 2: movdqa <z6=%xmm5,>y14=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm: uint32323232                                   y14 += z10
+# asm 1: paddd <z10=int6464#2,<y14=int6464#3
+# asm 2: paddd <z10=%xmm1,<y14=%xmm2
+paddd %xmm1,%xmm2
+
+# qhasm:                                                r14 = y14
+# asm 1: movdqa <y14=int6464#3,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm2,>r14=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                                   y14 <<= 7
+# asm 1: pslld $7,<y14=int6464#3
+# asm 2: pslld $7,<y14=%xmm2
+pslld $7,%xmm2
+
+# qhasm:                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#3,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm2,<z14=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                   r14 >>= 25
+# asm 1: psrld $25,<r14=int6464#13
+# asm 2: psrld $25,<r14=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm0,>z5_stack=336(%rsp)
+movdqa %xmm0,336(%rsp)
+
+# qhasm:                                                                y3 = z11
+# asm 1: movdqa <z11=int6464#7,>y3=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y3=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                                   y3 += z15
+# asm 1: paddd <z15=int6464#3,<y3=int6464#1
+# asm 2: paddd <z15=%xmm2,<y3=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r3 = y3
+# asm 1: movdqa <y3=int6464#1,>r3=int6464#13
+# asm 2: movdqa <y3=%xmm0,>r3=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y3 <<= 7
+# asm 1: pslld $7,<y3=int6464#1
+# asm 2: pslld $7,<y3=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z3 ^= y3
+# asm 1: pxor  <y3=int6464#1,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm0,<z3=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm: uint32323232                                                   r3 >>= 25
+# asm 1: psrld $25,<r3=int6464#13
+# asm 2: psrld $25,<r3=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z3 ^= r3
+# asm 1: pxor  <r3=int6464#13,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm12,<z3=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:                                                y2 = z10
+# asm 1: movdqa <z10=int6464#2,>y2=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y2=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y2 += z14
+# asm 1: paddd <z14=int6464#4,<y2=int6464#1
+# asm 2: paddd <z14=%xmm3,<y2=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                r2 = y2
+# asm 1: movdqa <y2=int6464#1,>r2=int6464#13
+# asm 2: movdqa <y2=%xmm0,>r2=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#1
+# asm 2: pslld $9,<y2=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z2 ^= y2
+# asm 1: pxor  <y2=int6464#1,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm0,<z2=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm: uint32323232                                   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#13
+# asm 2: psrld $23,<r2=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z2 ^= r2
+# asm 1: pxor  <r2=int6464#13,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm12,<z2=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:                                                                y7 = z15
+# asm 1: movdqa <z15=int6464#3,>y7=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y7=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y7 += z3
+# asm 1: paddd <z3=int6464#5,<y7=int6464#1
+# asm 2: paddd <z3=%xmm4,<y7=%xmm0
+paddd %xmm4,%xmm0
+
+# qhasm:                                                                r7 = y7
+# asm 1: movdqa <y7=int6464#1,>r7=int6464#13
+# asm 2: movdqa <y7=%xmm0,>r7=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#1
+# asm 2: pslld $9,<y7=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#1,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm0,<z7=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm: uint32323232                                                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#13
+# asm 2: psrld $23,<r7=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#13,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm12,<z7=%xmm8
+pxor  %xmm12,%xmm8
+
+# qhasm:                                                y6 = z14
+# asm 1: movdqa <z14=int6464#4,>y6=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y6=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                   y6 += z2
+# asm 1: paddd <z2=int6464#11,<y6=int6464#1
+# asm 2: paddd <z2=%xmm10,<y6=%xmm0
+paddd %xmm10,%xmm0
+
+# qhasm:                                                r6 = y6
+# asm 1: movdqa <y6=int6464#1,>r6=int6464#13
+# asm 2: movdqa <y6=%xmm0,>r6=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y6 <<= 13
+# asm 1: pslld $13,<y6=int6464#1
+# asm 2: pslld $13,<y6=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#1,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm0,<z6=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                   r6 >>= 19
+# asm 1: psrld $19,<r6=int6464#13
+# asm 2: psrld $19,<r6=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#13,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm12,<z6=%xmm5
+pxor  %xmm12,%xmm5
+
+# qhasm:                                                                y11 = z3
+# asm 1: movdqa <z3=int6464#5,>y11=int6464#1
+# asm 2: movdqa <z3=%xmm4,>y11=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                   y11 += z7
+# asm 1: paddd <z7=int6464#9,<y11=int6464#1
+# asm 2: paddd <z7=%xmm8,<y11=%xmm0
+paddd %xmm8,%xmm0
+
+# qhasm:                                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#13
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y11 <<= 13
+# asm 1: pslld $13,<y11=int6464#1
+# asm 2: pslld $13,<y11=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                                   r11 >>= 19
+# asm 1: psrld $19,<r11=int6464#13
+# asm 2: psrld $19,<r11=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#13,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm12,<z11=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:                                                y10 = z2
+# asm 1: movdqa <z2=int6464#11,>y10=int6464#1
+# asm 2: movdqa <z2=%xmm10,>y10=%xmm0
+movdqa %xmm10,%xmm0
+
+# qhasm: uint32323232                                   y10 += z6
+# asm 1: paddd <z6=int6464#6,<y10=int6464#1
+# asm 2: paddd <z6=%xmm5,<y10=%xmm0
+paddd %xmm5,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#1
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm0
+movdqa 320(%rsp),%xmm0
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                y1 = z3
+# asm 1: movdqa <z3=int6464#5,>y1=int6464#2
+# asm 2: movdqa <z3=%xmm4,>y1=%xmm1
+movdqa %xmm4,%xmm1
+
+# qhasm: uint32323232   y1 += z0
+# asm 1: paddd <z0=int6464#1,<y1=int6464#2
+# asm 2: paddd <z0=%xmm0,<y1=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#13
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232   y1 <<= 7
+# asm 1: pslld $7,<y1=int6464#2
+# asm 2: pslld $7,<y1=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232   r1 >>= 25
+# asm 1: psrld $25,<r1=int6464#13
+# asm 2: psrld $25,<r1=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                z1 ^= r1
+# asm 1: pxor  <r1=int6464#13,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm12,<z1=%xmm7
+pxor  %xmm12,%xmm7
+
+# qhasm:                                                                y15 = z7
+# asm 1: movdqa <z7=int6464#9,>y15=int6464#2
+# asm 2: movdqa <z7=%xmm8,>y15=%xmm1
+movdqa %xmm8,%xmm1
+
+# qhasm: uint32323232                                                   y15 += z11
+# asm 1: paddd <z11=int6464#7,<y15=int6464#2
+# asm 2: paddd <z11=%xmm6,<y15=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#2,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm1,>r15=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#2
+# asm 2: pslld $18,<y15=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#2,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm1,<z15=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#13
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm12
+movdqa 336(%rsp),%xmm12
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                                y6 = z4
+# asm 1: movdqa <z4=int6464#15,>y6=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y6=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232                   y6 += z5
+# asm 1: paddd <z5=int6464#13,<y6=int6464#2
+# asm 2: paddd <z5=%xmm12,<y6=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                                r6 = y6
+# asm 1: movdqa <y6=int6464#2,>r6=int6464#3
+# asm 2: movdqa <y6=%xmm1,>r6=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y6 <<= 7
+# asm 1: pslld $7,<y6=int6464#2
+# asm 2: pslld $7,<y6=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#2,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm1,<z6=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm: uint32323232                   r6 >>= 25
+# asm 1: psrld $25,<r6=int6464#3
+# asm 2: psrld $25,<r6=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#3,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm2,<z6=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:                y2 = z0
+# asm 1: movdqa <z0=int6464#1,>y2=int6464#2
+# asm 2: movdqa <z0=%xmm0,>y2=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232   y2 += z1
+# asm 1: paddd <z1=int6464#8,<y2=int6464#2
+# asm 2: paddd <z1=%xmm7,<y2=%xmm1
+paddd %xmm7,%xmm1
+
+# qhasm:                r2 = y2
+# asm 1: movdqa <y2=int6464#2,>r2=int6464#3
+# asm 2: movdqa <y2=%xmm1,>r2=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#2
+# asm 2: pslld $9,<y2=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z2 ^= y2
+# asm 1: pxor  <y2=int6464#2,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm1,<z2=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm: uint32323232   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#3
+# asm 2: psrld $23,<r2=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z2 ^= r2
+# asm 1: pxor  <r2=int6464#3,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm2,<z2=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:                                y7 = z5
+# asm 1: movdqa <z5=int6464#13,>y7=int6464#2
+# asm 2: movdqa <z5=%xmm12,>y7=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232                   y7 += z6
+# asm 1: paddd <z6=int6464#6,<y7=int6464#2
+# asm 2: paddd <z6=%xmm5,<y7=%xmm1
+paddd %xmm5,%xmm1
+
+# qhasm:                                r7 = y7
+# asm 1: movdqa <y7=int6464#2,>r7=int6464#3
+# asm 2: movdqa <y7=%xmm1,>r7=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#2
+# asm 2: pslld $9,<y7=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#2,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm1,<z7=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm: uint32323232                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#3
+# asm 2: psrld $23,<r7=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#3,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm2,<z7=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:                y3 = z1
+# asm 1: movdqa <z1=int6464#8,>y3=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y3=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232   y3 += z2
+# asm 1: paddd <z2=int6464#11,<y3=int6464#2
+# asm 2: paddd <z2=%xmm10,<y3=%xmm1
+paddd %xmm10,%xmm1
+
+# qhasm:                r3 = y3
+# asm 1: movdqa <y3=int6464#2,>r3=int6464#3
+# asm 2: movdqa <y3=%xmm1,>r3=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y3 <<= 13
+# asm 1: pslld $13,<y3=int6464#2
+# asm 2: pslld $13,<y3=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z3 ^= y3
+# asm 1: pxor  <y3=int6464#2,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm1,<z3=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm: uint32323232   r3 >>= 19
+# asm 1: psrld $19,<r3=int6464#3
+# asm 2: psrld $19,<r3=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z3 ^= r3
+# asm 1: pxor  <r3=int6464#3,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm2,<z3=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:                                y4 = z6
+# asm 1: movdqa <z6=int6464#6,>y4=int6464#2
+# asm 2: movdqa <z6=%xmm5,>y4=%xmm1
+movdqa %xmm5,%xmm1
+
+# qhasm: uint32323232                   y4 += z7
+# asm 1: paddd <z7=int6464#9,<y4=int6464#2
+# asm 2: paddd <z7=%xmm8,<y4=%xmm1
+paddd %xmm8,%xmm1
+
+# qhasm:                                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y4 <<= 13
+# asm 1: pslld $13,<y4=int6464#2
+# asm 2: pslld $13,<y4=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232                   r4 >>= 19
+# asm 1: psrld $19,<r4=int6464#3
+# asm 2: psrld $19,<r4=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                y0 = z2
+# asm 1: movdqa <z2=int6464#11,>y0=int6464#2
+# asm 2: movdqa <z2=%xmm10,>y0=%xmm1
+movdqa %xmm10,%xmm1
+
+# qhasm: uint32323232   y0 += z3
+# asm 1: paddd <z3=int6464#5,<y0=int6464#2
+# asm 2: paddd <z3=%xmm4,<y0=%xmm1
+paddd %xmm4,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#1
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#1
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm0,>z0_stack=320(%rsp)
+movdqa %xmm0,320(%rsp)
+
+# qhasm:                                y5 = z7
+# asm 1: movdqa <z7=int6464#9,>y5=int6464#1
+# asm 2: movdqa <z7=%xmm8,>y5=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm: uint32323232                   y5 += z4
+# asm 1: paddd <z4=int6464#15,<y5=int6464#1
+# asm 2: paddd <z4=%xmm14,<y5=%xmm0
+paddd %xmm14,%xmm0
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#1,>r5=int6464#3
+# asm 2: movdqa <y5=%xmm0,>r5=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#1
+# asm 2: pslld $18,<y5=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#1,<z5=int6464#13
+# asm 2: pxor  <y5=%xmm0,<z5=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#3
+# asm 2: psrld $14,<r5=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#3,<z5=int6464#13
+# asm 2: pxor  <r5=%xmm2,<z5=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                y11 = z9
+# asm 1: movdqa <z9=int6464#12,>y11=int6464#1
+# asm 2: movdqa <z9=%xmm11,>y11=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm: uint32323232                                   y11 += z10
+# asm 1: paddd <z10=int6464#2,<y11=int6464#1
+# asm 2: paddd <z10=%xmm1,<y11=%xmm0
+paddd %xmm1,%xmm0
+
+# qhasm:                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#3
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                   y11 <<= 7
+# asm 1: pslld $7,<y11=int6464#1
+# asm 2: pslld $7,<y11=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                   r11 >>= 25
+# asm 1: psrld $25,<r11=int6464#3
+# asm 2: psrld $25,<r11=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#3,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm2,<z11=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#13,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm12,>z5_stack=336(%rsp)
+movdqa %xmm12,336(%rsp)
+
+# qhasm:                                                                y12 = z14
+# asm 1: movdqa <z14=int6464#4,>y12=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y12=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                   y12 += z15
+# asm 1: paddd <z15=int6464#3,<y12=int6464#1
+# asm 2: paddd <z15=%xmm2,<y12=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r12 = y12
+# asm 1: movdqa <y12=int6464#1,>r12=int6464#13
+# asm 2: movdqa <y12=%xmm0,>r12=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y12 <<= 7
+# asm 1: pslld $7,<y12=int6464#1
+# asm 2: pslld $7,<y12=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z12 ^= y12
+# asm 1: pxor  <y12=int6464#1,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm0,<z12=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm: uint32323232                                                   r12 >>= 25
+# asm 1: psrld $25,<r12=int6464#13
+# asm 2: psrld $25,<r12=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z12 ^= r12
+# asm 1: pxor  <r12=int6464#13,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm12,<z12=%xmm13
+pxor  %xmm12,%xmm13
+
+# qhasm:                                                y8 = z10
+# asm 1: movdqa <z10=int6464#2,>y8=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y8=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y8 += z11
+# asm 1: paddd <z11=int6464#7,<y8=int6464#1
+# asm 2: paddd <z11=%xmm6,<y8=%xmm0
+paddd %xmm6,%xmm0
+
+# qhasm:                                                r8 = y8
+# asm 1: movdqa <y8=int6464#1,>r8=int6464#13
+# asm 2: movdqa <y8=%xmm0,>r8=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#1
+# asm 2: pslld $9,<y8=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z8 ^= y8
+# asm 1: pxor  <y8=int6464#1,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm0,<z8=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm: uint32323232                                   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#13
+# asm 2: psrld $23,<r8=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z8 ^= r8
+# asm 1: pxor  <r8=int6464#13,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm12,<z8=%xmm15
+pxor  %xmm12,%xmm15
+
+# qhasm:                                                                y13 = z15
+# asm 1: movdqa <z15=int6464#3,>y13=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y13=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y13 += z12
+# asm 1: paddd <z12=int6464#14,<y13=int6464#1
+# asm 2: paddd <z12=%xmm13,<y13=%xmm0
+paddd %xmm13,%xmm0
+
+# qhasm:                                                                r13 = y13
+# asm 1: movdqa <y13=int6464#1,>r13=int6464#13
+# asm 2: movdqa <y13=%xmm0,>r13=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#1
+# asm 2: pslld $9,<y13=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#1,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm0,<z13=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm: uint32323232                                                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#13
+# asm 2: psrld $23,<r13=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#13,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm12,<z13=%xmm9
+pxor  %xmm12,%xmm9
+
+# qhasm:                                                y9 = z11
+# asm 1: movdqa <z11=int6464#7,>y9=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y9=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                   y9 += z8
+# asm 1: paddd <z8=int6464#16,<y9=int6464#1
+# asm 2: paddd <z8=%xmm15,<y9=%xmm0
+paddd %xmm15,%xmm0
+
+# qhasm:                                                r9 = y9
+# asm 1: movdqa <y9=int6464#1,>r9=int6464#13
+# asm 2: movdqa <y9=%xmm0,>r9=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y9 <<= 13
+# asm 1: pslld $13,<y9=int6464#1
+# asm 2: pslld $13,<y9=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#1,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm0,<z9=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm: uint32323232                                   r9 >>= 19
+# asm 1: psrld $19,<r9=int6464#13
+# asm 2: psrld $19,<r9=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#13,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm12,<z9=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:                                                                y14 = z12
+# asm 1: movdqa <z12=int6464#14,>y14=int6464#1
+# asm 2: movdqa <z12=%xmm13,>y14=%xmm0
+movdqa %xmm13,%xmm0
+
+# qhasm: uint32323232                                                   y14 += z13
+# asm 1: paddd <z13=int6464#10,<y14=int6464#1
+# asm 2: paddd <z13=%xmm9,<y14=%xmm0
+paddd %xmm9,%xmm0
+
+# qhasm:                                                                r14 = y14
+# asm 1: movdqa <y14=int6464#1,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm0,>r14=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y14 <<= 13
+# asm 1: pslld $13,<y14=int6464#1
+# asm 2: pslld $13,<y14=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#1,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm0,<z14=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm: uint32323232                                                   r14 >>= 19
+# asm 1: psrld $19,<r14=int6464#13
+# asm 2: psrld $19,<r14=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                y10 = z8
+# asm 1: movdqa <z8=int6464#16,>y10=int6464#1
+# asm 2: movdqa <z8=%xmm15,>y10=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm: uint32323232                                   y10 += z9
+# asm 1: paddd <z9=int6464#12,<y10=int6464#1
+# asm 2: paddd <z9=%xmm11,<y10=%xmm0
+paddd %xmm11,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                                                                y15 = z13
+# asm 1: movdqa <z13=int6464#10,>y15=int6464#1
+# asm 2: movdqa <z13=%xmm9,>y15=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm: uint32323232                                                   y15 += z14
+# asm 1: paddd <z14=int6464#4,<y15=int6464#1
+# asm 2: paddd <z14=%xmm3,<y15=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#1,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm0,>r15=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#1
+# asm 2: pslld $18,<y15=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#1,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm0,<z15=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#13
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm12
+movdqa 320(%rsp),%xmm12
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#1
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm0
+movdqa 336(%rsp),%xmm0
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int64#3
+# asm 2: sub  $2,<i=%rdx
+sub  $2,%rdx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#13
+# asm 2: paddd <orig0=112(%rsp),<z0=%xmm12
+paddd 112(%rsp),%xmm12
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#8
+# asm 2: paddd <orig1=176(%rsp),<z1=%xmm7
+paddd 176(%rsp),%xmm7
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#11
+# asm 2: paddd <orig2=224(%rsp),<z2=%xmm10
+paddd 224(%rsp),%xmm10
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#5
+# asm 2: paddd <orig3=272(%rsp),<z3=%xmm4
+paddd 272(%rsp),%xmm4
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 0(<m=%rsi),<in0=%edx
+xorl 0(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 4(<m=%rsi),<in1=%ecx
+xorl 4(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 8(<m=%rsi),<in2=%r8d
+xorl 8(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 12(<m=%rsi),<in3=%r9d
+xorl 12(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#3d,0(<out=int64#1)
+# asm 2: movl   <in0=%edx,0(<out=%rdi)
+movl   %edx,0(%rdi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#4d,4(<out=int64#1)
+# asm 2: movl   <in1=%ecx,4(<out=%rdi)
+movl   %ecx,4(%rdi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#5d,8(<out=int64#1)
+# asm 2: movl   <in2=%r8d,8(<out=%rdi)
+movl   %r8d,8(%rdi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#6d,12(<out=int64#1)
+# asm 2: movl   <in3=%r9d,12(<out=%rdi)
+movl   %r9d,12(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 64(<m=%rsi),<in0=%edx
+xorl 64(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 68(<m=%rsi),<in1=%ecx
+xorl 68(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 72(<m=%rsi),<in2=%r8d
+xorl 72(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 76(<m=%rsi),<in3=%r9d
+xorl 76(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl   <in0=int64#3d,64(<out=int64#1)
+# asm 2: movl   <in0=%edx,64(<out=%rdi)
+movl   %edx,64(%rdi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl   <in1=int64#4d,68(<out=int64#1)
+# asm 2: movl   <in1=%ecx,68(<out=%rdi)
+movl   %ecx,68(%rdi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl   <in2=int64#5d,72(<out=int64#1)
+# asm 2: movl   <in2=%r8d,72(<out=%rdi)
+movl   %r8d,72(%rdi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl   <in3=int64#6d,76(<out=int64#1)
+# asm 2: movl   <in3=%r9d,76(<out=%rdi)
+movl   %r9d,76(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 128(<m=%rsi),<in0=%edx
+xorl 128(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 132(<m=%rsi),<in1=%ecx
+xorl 132(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 136(<m=%rsi),<in2=%r8d
+xorl 136(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 140(<m=%rsi),<in3=%r9d
+xorl 140(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl   <in0=int64#3d,128(<out=int64#1)
+# asm 2: movl   <in0=%edx,128(<out=%rdi)
+movl   %edx,128(%rdi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl   <in1=int64#4d,132(<out=int64#1)
+# asm 2: movl   <in1=%ecx,132(<out=%rdi)
+movl   %ecx,132(%rdi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl   <in2=int64#5d,136(<out=int64#1)
+# asm 2: movl   <in2=%r8d,136(<out=%rdi)
+movl   %r8d,136(%rdi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl   <in3=int64#6d,140(<out=int64#1)
+# asm 2: movl   <in3=%r9d,140(<out=%rdi)
+movl   %r9d,140(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 192(<m=%rsi),<in0=%edx
+xorl 192(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 196(<m=%rsi),<in1=%ecx
+xorl 196(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 200(<m=%rsi),<in2=%r8d
+xorl 200(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 204(<m=%rsi),<in3=%r9d
+xorl 204(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl   <in0=int64#3d,192(<out=int64#1)
+# asm 2: movl   <in0=%edx,192(<out=%rdi)
+movl   %edx,192(%rdi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl   <in1=int64#4d,196(<out=int64#1)
+# asm 2: movl   <in1=%ecx,196(<out=%rdi)
+movl   %ecx,196(%rdi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl   <in2=int64#5d,200(<out=int64#1)
+# asm 2: movl   <in2=%r8d,200(<out=%rdi)
+movl   %r8d,200(%rdi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl   <in3=int64#6d,204(<out=int64#1)
+# asm 2: movl   <in3=%r9d,204(<out=%rdi)
+movl   %r9d,204(%rdi)
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#15
+# asm 2: paddd <orig4=240(%rsp),<z4=%xmm14
+paddd 240(%rsp),%xmm14
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#1
+# asm 2: paddd <orig5=64(%rsp),<z5=%xmm0
+paddd 64(%rsp),%xmm0
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#6
+# asm 2: paddd <orig6=128(%rsp),<z6=%xmm5
+paddd 128(%rsp),%xmm5
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#9
+# asm 2: paddd <orig7=192(%rsp),<z7=%xmm8
+paddd 192(%rsp),%xmm8
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 16(<m=%rsi),<in4=%edx
+xorl 16(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#3d,16(<out=int64#1)
+# asm 2: movl   <in4=%edx,16(<out=%rdi)
+movl   %edx,16(%rdi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 80(<m=%rsi),<in4=%edx
+xorl 80(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 84(<m=%rsi),<in5=%ecx
+xorl 84(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 88(<m=%rsi),<in6=%r8d
+xorl 88(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 92(<m=%rsi),<in7=%r9d
+xorl 92(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl   <in4=int64#3d,80(<out=int64#1)
+# asm 2: movl   <in4=%edx,80(<out=%rdi)
+movl   %edx,80(%rdi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl   <in5=int64#4d,84(<out=int64#1)
+# asm 2: movl   <in5=%ecx,84(<out=%rdi)
+movl   %ecx,84(%rdi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl   <in6=int64#5d,88(<out=int64#1)
+# asm 2: movl   <in6=%r8d,88(<out=%rdi)
+movl   %r8d,88(%rdi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl   <in7=int64#6d,92(<out=int64#1)
+# asm 2: movl   <in7=%r9d,92(<out=%rdi)
+movl   %r9d,92(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 144(<m=%rsi),<in4=%edx
+xorl 144(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 148(<m=%rsi),<in5=%ecx
+xorl 148(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 152(<m=%rsi),<in6=%r8d
+xorl 152(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 156(<m=%rsi),<in7=%r9d
+xorl 156(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl   <in4=int64#3d,144(<out=int64#1)
+# asm 2: movl   <in4=%edx,144(<out=%rdi)
+movl   %edx,144(%rdi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl   <in5=int64#4d,148(<out=int64#1)
+# asm 2: movl   <in5=%ecx,148(<out=%rdi)
+movl   %ecx,148(%rdi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl   <in6=int64#5d,152(<out=int64#1)
+# asm 2: movl   <in6=%r8d,152(<out=%rdi)
+movl   %r8d,152(%rdi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl   <in7=int64#6d,156(<out=int64#1)
+# asm 2: movl   <in7=%r9d,156(<out=%rdi)
+movl   %r9d,156(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 208(<m=%rsi),<in4=%edx
+xorl 208(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 212(<m=%rsi),<in5=%ecx
+xorl 212(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 216(<m=%rsi),<in6=%r8d
+xorl 216(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 220(<m=%rsi),<in7=%r9d
+xorl 220(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl   <in4=int64#3d,208(<out=int64#1)
+# asm 2: movl   <in4=%edx,208(<out=%rdi)
+movl   %edx,208(%rdi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl   <in5=int64#4d,212(<out=int64#1)
+# asm 2: movl   <in5=%ecx,212(<out=%rdi)
+movl   %ecx,212(%rdi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl   <in6=int64#5d,216(<out=int64#1)
+# asm 2: movl   <in6=%r8d,216(<out=%rdi)
+movl   %r8d,216(%rdi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl   <in7=int64#6d,220(<out=int64#1)
+# asm 2: movl   <in7=%r9d,220(<out=%rdi)
+movl   %r9d,220(%rdi)
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#16
+# asm 2: paddd <orig8=288(%rsp),<z8=%xmm15
+paddd 288(%rsp),%xmm15
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#12
+# asm 2: paddd <orig9=304(%rsp),<z9=%xmm11
+paddd 304(%rsp),%xmm11
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#2
+# asm 2: paddd <orig10=80(%rsp),<z10=%xmm1
+paddd 80(%rsp),%xmm1
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#7
+# asm 2: paddd <orig11=144(%rsp),<z11=%xmm6
+paddd 144(%rsp),%xmm6
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 32(<m=%rsi),<in8=%edx
+xorl 32(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 36(<m=%rsi),<in9=%ecx
+xorl 36(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 40(<m=%rsi),<in10=%r8d
+xorl 40(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 44(<m=%rsi),<in11=%r9d
+xorl 44(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#3d,32(<out=int64#1)
+# asm 2: movl   <in8=%edx,32(<out=%rdi)
+movl   %edx,32(%rdi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#4d,36(<out=int64#1)
+# asm 2: movl   <in9=%ecx,36(<out=%rdi)
+movl   %ecx,36(%rdi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#5d,40(<out=int64#1)
+# asm 2: movl   <in10=%r8d,40(<out=%rdi)
+movl   %r8d,40(%rdi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#6d,44(<out=int64#1)
+# asm 2: movl   <in11=%r9d,44(<out=%rdi)
+movl   %r9d,44(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 96(<m=%rsi),<in8=%edx
+xorl 96(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 100(<m=%rsi),<in9=%ecx
+xorl 100(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 104(<m=%rsi),<in10=%r8d
+xorl 104(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 108(<m=%rsi),<in11=%r9d
+xorl 108(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl   <in8=int64#3d,96(<out=int64#1)
+# asm 2: movl   <in8=%edx,96(<out=%rdi)
+movl   %edx,96(%rdi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl   <in9=int64#4d,100(<out=int64#1)
+# asm 2: movl   <in9=%ecx,100(<out=%rdi)
+movl   %ecx,100(%rdi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl   <in10=int64#5d,104(<out=int64#1)
+# asm 2: movl   <in10=%r8d,104(<out=%rdi)
+movl   %r8d,104(%rdi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl   <in11=int64#6d,108(<out=int64#1)
+# asm 2: movl   <in11=%r9d,108(<out=%rdi)
+movl   %r9d,108(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 160(<m=%rsi),<in8=%edx
+xorl 160(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 164(<m=%rsi),<in9=%ecx
+xorl 164(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 168(<m=%rsi),<in10=%r8d
+xorl 168(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 172(<m=%rsi),<in11=%r9d
+xorl 172(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl   <in8=int64#3d,160(<out=int64#1)
+# asm 2: movl   <in8=%edx,160(<out=%rdi)
+movl   %edx,160(%rdi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl   <in9=int64#4d,164(<out=int64#1)
+# asm 2: movl   <in9=%ecx,164(<out=%rdi)
+movl   %ecx,164(%rdi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl   <in10=int64#5d,168(<out=int64#1)
+# asm 2: movl   <in10=%r8d,168(<out=%rdi)
+movl   %r8d,168(%rdi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl   <in11=int64#6d,172(<out=int64#1)
+# asm 2: movl   <in11=%r9d,172(<out=%rdi)
+movl   %r9d,172(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 224(<m=%rsi),<in8=%edx
+xorl 224(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 228(<m=%rsi),<in9=%ecx
+xorl 228(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 232(<m=%rsi),<in10=%r8d
+xorl 232(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 236(<m=%rsi),<in11=%r9d
+xorl 236(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl   <in8=int64#3d,224(<out=int64#1)
+# asm 2: movl   <in8=%edx,224(<out=%rdi)
+movl   %edx,224(%rdi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl   <in9=int64#4d,228(<out=int64#1)
+# asm 2: movl   <in9=%ecx,228(<out=%rdi)
+movl   %ecx,228(%rdi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl   <in10=int64#5d,232(<out=int64#1)
+# asm 2: movl   <in10=%r8d,232(<out=%rdi)
+movl   %r8d,232(%rdi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl   <in11=int64#6d,236(<out=int64#1)
+# asm 2: movl   <in11=%r9d,236(<out=%rdi)
+movl   %r9d,236(%rdi)
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#14
+# asm 2: paddd <orig12=160(%rsp),<z12=%xmm13
+paddd 160(%rsp),%xmm13
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#10
+# asm 2: paddd <orig13=208(%rsp),<z13=%xmm9
+paddd 208(%rsp),%xmm9
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#4
+# asm 2: paddd <orig14=256(%rsp),<z14=%xmm3
+paddd 256(%rsp),%xmm3
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#3
+# asm 2: paddd <orig15=96(%rsp),<z15=%xmm2
+paddd 96(%rsp),%xmm2
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 48(<m=%rsi),<in12=%edx
+xorl 48(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 52(<m=%rsi),<in13=%ecx
+xorl 52(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 56(<m=%rsi),<in14=%r8d
+xorl 56(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 60(<m=%rsi),<in15=%r9d
+xorl 60(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#3d,48(<out=int64#1)
+# asm 2: movl   <in12=%edx,48(<out=%rdi)
+movl   %edx,48(%rdi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#4d,52(<out=int64#1)
+# asm 2: movl   <in13=%ecx,52(<out=%rdi)
+movl   %ecx,52(%rdi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#5d,56(<out=int64#1)
+# asm 2: movl   <in14=%r8d,56(<out=%rdi)
+movl   %r8d,56(%rdi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#6d,60(<out=int64#1)
+# asm 2: movl   <in15=%r9d,60(<out=%rdi)
+movl   %r9d,60(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 112(<m=%rsi),<in12=%edx
+xorl 112(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 116(<m=%rsi),<in13=%ecx
+xorl 116(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 120(<m=%rsi),<in14=%r8d
+xorl 120(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 124(<m=%rsi),<in15=%r9d
+xorl 124(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl   <in12=int64#3d,112(<out=int64#1)
+# asm 2: movl   <in12=%edx,112(<out=%rdi)
+movl   %edx,112(%rdi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl   <in13=int64#4d,116(<out=int64#1)
+# asm 2: movl   <in13=%ecx,116(<out=%rdi)
+movl   %ecx,116(%rdi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl   <in14=int64#5d,120(<out=int64#1)
+# asm 2: movl   <in14=%r8d,120(<out=%rdi)
+movl   %r8d,120(%rdi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl   <in15=int64#6d,124(<out=int64#1)
+# asm 2: movl   <in15=%r9d,124(<out=%rdi)
+movl   %r9d,124(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 176(<m=%rsi),<in12=%edx
+xorl 176(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 180(<m=%rsi),<in13=%ecx
+xorl 180(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 184(<m=%rsi),<in14=%r8d
+xorl 184(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 188(<m=%rsi),<in15=%r9d
+xorl 188(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl   <in12=int64#3d,176(<out=int64#1)
+# asm 2: movl   <in12=%edx,176(<out=%rdi)
+movl   %edx,176(%rdi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl   <in13=int64#4d,180(<out=int64#1)
+# asm 2: movl   <in13=%ecx,180(<out=%rdi)
+movl   %ecx,180(%rdi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl   <in14=int64#5d,184(<out=int64#1)
+# asm 2: movl   <in14=%r8d,184(<out=%rdi)
+movl   %r8d,184(%rdi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl   <in15=int64#6d,188(<out=int64#1)
+# asm 2: movl   <in15=%r9d,188(<out=%rdi)
+movl   %r9d,188(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 240(<m=%rsi),<in12=%edx
+xorl 240(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 244(<m=%rsi),<in13=%ecx
+xorl 244(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 248(<m=%rsi),<in14=%r8d
+xorl 248(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 252(<m=%rsi),<in15=%r9d
+xorl 252(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl   <in12=int64#3d,240(<out=int64#1)
+# asm 2: movl   <in12=%edx,240(<out=%rdi)
+movl   %edx,240(%rdi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl   <in13=int64#4d,244(<out=int64#1)
+# asm 2: movl   <in13=%ecx,244(<out=%rdi)
+movl   %ecx,244(%rdi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl   <in14=int64#5d,248(<out=int64#1)
+# asm 2: movl   <in14=%r8d,248(<out=%rdi)
+movl   %r8d,248(%rdi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl   <in15=int64#6d,252(<out=int64#1)
+# asm 2: movl   <in15=%r9d,252(<out=%rdi)
+movl   %r9d,252(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int64#6
+# asm 2: sub  $256,<bytes=%r9
+sub  $256,%r9
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int64#2
+# asm 2: add  $256,<m=%rsi
+add  $256,%rsi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int64#1
+# asm 2: add  $256,<out=%rdi
+add  $256,%rdi
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: mov  <out=int64#1,>ctarget=int64#3
+# asm 2: mov  <out=%rdi,>ctarget=%rdx
+mov  %rdi,%rdx
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     m = &tmp
+# asm 1: leaq <tmp=stack512#1,>m=int64#2
+# asm 2: leaq <tmp=416(%rsp),>m=%rsi
+leaq 416(%rsp),%rsi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#4,>diag0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>diag0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#1,>diag1=int6464#2
+# asm 2: movdqa <x1=0(%rsp),>diag1=%xmm1
+movdqa 0(%rsp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#2,>diag2=int6464#3
+# asm 2: movdqa <x2=16(%rsp),>diag2=%xmm2
+movdqa 16(%rsp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#3,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%rsp),>diag3=%xmm3
+movdqa 32(%rsp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 12
+# asm 1: mov  $12,>i=int64#4
+# asm 2: mov  $12,>i=%rcx
+mov  $12,%rcx
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int64#4
+# asm 2: sub  $4,<i=%rcx
+sub  $4,%rcx
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#4,<diag0=int6464#1
+# asm 2: paddd <x0=48(%rsp),<diag0=%xmm0
+paddd 48(%rsp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#1,<diag1=int6464#2
+# asm 2: paddd <x1=0(%rsp),<diag1=%xmm1
+paddd 0(%rsp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#2,<diag2=int6464#3
+# asm 2: paddd <x2=16(%rsp),<diag2=%xmm2
+paddd 16(%rsp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#3,<diag3=int6464#4
+# asm 2: paddd <x3=32(%rsp),<diag3=%xmm3
+paddd 32(%rsp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int64#4
+# asm 2: movd   <diag0=%xmm0,>in0=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int64#5
+# asm 2: movd   <diag1=%xmm1,>in12=%r8
+movd   %xmm1,%r8
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int64#6
+# asm 2: movd   <diag2=%xmm2,>in8=%r9
+movd   %xmm2,%r9
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int64#7
+# asm 2: movd   <diag3=%xmm3,>in4=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#4d
+# asm 2: xorl 0(<m=%rsi),<in0=%ecx
+xorl 0(%rsi),%ecx
+
+# qhasm: (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#5d
+# asm 2: xorl 48(<m=%rsi),<in12=%r8d
+xorl 48(%rsi),%r8d
+
+# qhasm: (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#6d
+# asm 2: xorl 32(<m=%rsi),<in8=%r9d
+xorl 32(%rsi),%r9d
+
+# qhasm: (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#7d
+# asm 2: xorl 16(<m=%rsi),<in4=%eax
+xorl 16(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#4d,0(<out=int64#1)
+# asm 2: movl   <in0=%ecx,0(<out=%rdi)
+movl   %ecx,0(%rdi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#5d,48(<out=int64#1)
+# asm 2: movl   <in12=%r8d,48(<out=%rdi)
+movl   %r8d,48(%rdi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#6d,32(<out=int64#1)
+# asm 2: movl   <in8=%r9d,32(<out=%rdi)
+movl   %r9d,32(%rdi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#7d,16(<out=int64#1)
+# asm 2: movl   <in4=%eax,16(<out=%rdi)
+movl   %eax,16(%rdi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int64#4
+# asm 2: movd   <diag0=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int64#5
+# asm 2: movd   <diag1=%xmm1,>in1=%r8
+movd   %xmm1,%r8
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int64#6
+# asm 2: movd   <diag2=%xmm2,>in13=%r9
+movd   %xmm2,%r9
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int64#7
+# asm 2: movd   <diag3=%xmm3,>in9=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm: (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#5d
+# asm 2: xorl 4(<m=%rsi),<in1=%r8d
+xorl 4(%rsi),%r8d
+
+# qhasm: (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#6d
+# asm 2: xorl 52(<m=%rsi),<in13=%r9d
+xorl 52(%rsi),%r9d
+
+# qhasm: (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#7d
+# asm 2: xorl 36(<m=%rsi),<in9=%eax
+xorl 36(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#5d,4(<out=int64#1)
+# asm 2: movl   <in1=%r8d,4(<out=%rdi)
+movl   %r8d,4(%rdi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#6d,52(<out=int64#1)
+# asm 2: movl   <in13=%r9d,52(<out=%rdi)
+movl   %r9d,52(%rdi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#7d,36(<out=int64#1)
+# asm 2: movl   <in9=%eax,36(<out=%rdi)
+movl   %eax,36(%rdi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int64#4
+# asm 2: movd   <diag0=%xmm0,>in10=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int64#5
+# asm 2: movd   <diag1=%xmm1,>in6=%r8
+movd   %xmm1,%r8
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int64#6
+# asm 2: movd   <diag2=%xmm2,>in2=%r9
+movd   %xmm2,%r9
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int64#7
+# asm 2: movd   <diag3=%xmm3,>in14=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#4d
+# asm 2: xorl 40(<m=%rsi),<in10=%ecx
+xorl 40(%rsi),%ecx
+
+# qhasm: (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm: (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#6d
+# asm 2: xorl 8(<m=%rsi),<in2=%r9d
+xorl 8(%rsi),%r9d
+
+# qhasm: (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#7d
+# asm 2: xorl 56(<m=%rsi),<in14=%eax
+xorl 56(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#4d,40(<out=int64#1)
+# asm 2: movl   <in10=%ecx,40(<out=%rdi)
+movl   %ecx,40(%rdi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#6d,8(<out=int64#1)
+# asm 2: movl   <in2=%r9d,8(<out=%rdi)
+movl   %r9d,8(%rdi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#7d,56(<out=int64#1)
+# asm 2: movl   <in14=%eax,56(<out=%rdi)
+movl   %eax,56(%rdi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int64#4
+# asm 2: movd   <diag0=%xmm0,>in15=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int64#5
+# asm 2: movd   <diag1=%xmm1,>in11=%r8
+movd   %xmm1,%r8
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int64#6
+# asm 2: movd   <diag2=%xmm2,>in7=%r9
+movd   %xmm2,%r9
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int64#7
+# asm 2: movd   <diag3=%xmm3,>in3=%rax
+movd   %xmm3,%rax
+
+# qhasm: (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#4d
+# asm 2: xorl 60(<m=%rsi),<in15=%ecx
+xorl 60(%rsi),%ecx
+
+# qhasm: (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#5d
+# asm 2: xorl 44(<m=%rsi),<in11=%r8d
+xorl 44(%rsi),%r8d
+
+# qhasm: (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm: (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#7d
+# asm 2: xorl 12(<m=%rsi),<in3=%eax
+xorl 12(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#4d,60(<out=int64#1)
+# asm 2: movl   <in15=%ecx,60(<out=%rdi)
+movl   %ecx,60(%rdi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#5d,44(<out=int64#1)
+# asm 2: movl   <in11=%r8d,44(<out=%rdi)
+movl   %r8d,44(%rdi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#7d,12(<out=int64#1)
+# asm 2: movl   <in3=%eax,12(<out=%rdi)
+movl   %eax,12(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#4d
+# asm 2: movl <x2=16(%rsp),>in8=%ecx
+movl 16(%rsp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#5d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%r8d
+movl 4+32(%rsp),%r8d
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#4
+# asm 2: add  $1,<in8=%rcx
+add  $1,%rcx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#5
+# asm 2: shl  $32,<in9=%r8
+shl  $32,%r8
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#5,<in8=int64#4
+# asm 2: add  <in9=%r8,<in8=%rcx
+add  %r8,%rcx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#4,>in9=int64#5
+# asm 2: mov  <in8=%rcx,>in9=%r8
+mov  %rcx,%r8
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#5
+# asm 2: shr  $32,<in9=%r8
+shr  $32,%r8
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#5d,4+<x3=stack128#3
+# asm 2: movl <in9=%r8d,4+<x3=32(%rsp)
+movl %r8d,4+32(%rsp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm:       out = ctarget
+# asm 1: mov  <ctarget=int64#3,>out=int64#1
+# asm 2: mov  <ctarget=%rdx,>out=%rdi
+mov  %rdx,%rdi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     r11_caller = r11_stack
+# asm 1: movq <r11_stack=stack64#1,>r11_caller=int64#9
+# asm 2: movq <r11_stack=352(%rsp),>r11_caller=%r11
+movq 352(%rsp),%r11
+
+# qhasm:     r12_caller = r12_stack
+# asm 1: movq <r12_stack=stack64#2,>r12_caller=int64#10
+# asm 2: movq <r12_stack=360(%rsp),>r12_caller=%r12
+movq 360(%rsp),%r12
+
+# qhasm:     r13_caller = r13_stack
+# asm 1: movq <r13_stack=stack64#3,>r13_caller=int64#11
+# asm 2: movq <r13_stack=368(%rsp),>r13_caller=%r13
+movq 368(%rsp),%r13
+
+# qhasm:     r14_caller = r14_stack
+# asm 1: movq <r14_stack=stack64#4,>r14_caller=int64#12
+# asm 2: movq <r14_stack=376(%rsp),>r14_caller=%r14
+movq 376(%rsp),%r14
+
+# qhasm:     r15_caller = r15_stack
+# asm 1: movq <r15_stack=stack64#5,>r15_caller=int64#13
+# asm 2: movq <r15_stack=384(%rsp),>r15_caller=%r15
+movq 384(%rsp),%r15
+
+# qhasm:     rbx_caller = rbx_stack
+# asm 1: movq <rbx_stack=stack64#6,>rbx_caller=int64#14
+# asm 2: movq <rbx_stack=392(%rsp),>rbx_caller=%rbx
+movq 392(%rsp),%rbx
+
+# qhasm:     rbp_caller = rbp_stack
+# asm 1: movq <rbp_stack=stack64#7,>rbp_caller=int64#15
+# asm 2: movq <rbp_stack=400(%rsp),>rbp_caller=%rbp
+movq 400(%rsp),%rbp
+
+# qhasm:     leave
+add %r11,%rsp
+xor %rax,%rax
+xor %rdx,%rdx
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int64#6
+# asm 2: sub  $64,<bytes=%r9
+sub  $64,%r9
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int64#1
+# asm 2: add  $64,<out=%rdi
+add  $64,%rdi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int64#2
+# asm 2: add  $64,<m=%rsi
+add  $64,%rsi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/salsa2012/checksum b/nacl/crypto_stream/salsa2012/checksum
new file mode 100644
index 00000000..f801d9e3
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/checksum
@@ -0,0 +1 @@
+ecc758f200061c3cc770b25797da73583548d4f90f69a967fbbe1a6d94d1705c
diff --git a/nacl/crypto_stream/salsa2012/ref/api.h b/nacl/crypto_stream/salsa2012/ref/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/ref/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa2012/ref/implementors b/nacl/crypto_stream/salsa2012/ref/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/ref/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa2012/ref/stream.c b/nacl/crypto_stream/salsa2012/ref/stream.c
new file mode 100644
index 00000000..86053337
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/ref/stream.c
@@ -0,0 +1,49 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa2012.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream(
+        unsigned char *c,unsigned long long clen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!clen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (clen >= 64) {
+    crypto_core_salsa2012(c,in,k,sigma);
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    clen -= 64;
+    c += 64;
+  }
+
+  if (clen) {
+    crypto_core_salsa2012(block,in,k,sigma);
+    for (i = 0;i < clen;++i) c[i] = block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa2012/ref/xor.c b/nacl/crypto_stream/salsa2012/ref/xor.c
new file mode 100644
index 00000000..90206426
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/ref/xor.c
@@ -0,0 +1,52 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa2012.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream_xor(
+        unsigned char *c,
+  const unsigned char *m,unsigned long long mlen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!mlen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (mlen >= 64) {
+    crypto_core_salsa2012(block,in,k,sigma);
+    for (i = 0;i < 64;++i) c[i] = m[i] ^ block[i];
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    mlen -= 64;
+    c += 64;
+    m += 64;
+  }
+
+  if (mlen) {
+    crypto_core_salsa2012(block,in,k,sigma);
+    for (i = 0;i < mlen;++i) c[i] = m[i] ^ block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa2012/used b/nacl/crypto_stream/salsa2012/used
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/used
diff --git a/nacl/crypto_stream/salsa2012/x86_xmm5/api.h b/nacl/crypto_stream/salsa2012/x86_xmm5/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/x86_xmm5/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa2012/x86_xmm5/implementors b/nacl/crypto_stream/salsa2012/x86_xmm5/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/x86_xmm5/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa2012/x86_xmm5/stream.s b/nacl/crypto_stream/salsa2012/x86_xmm5/stream.s
new file mode 100644
index 00000000..c511b0d3
--- /dev/null
+++ b/nacl/crypto_stream/salsa2012/x86_xmm5/stream.s
@@ -0,0 +1,5078 @@
+
+# qhasm: int32 a
+
+# qhasm: stack32 arg1
+
+# qhasm: stack32 arg2
+
+# qhasm: stack32 arg3
+
+# qhasm: stack32 arg4
+
+# qhasm: stack32 arg5
+
+# qhasm: stack32 arg6
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: input arg6
+
+# qhasm: int32 eax
+
+# qhasm: int32 ebx
+
+# qhasm: int32 esi
+
+# qhasm: int32 edi
+
+# qhasm: int32 ebp
+
+# qhasm: caller eax
+
+# qhasm: caller ebx
+
+# qhasm: caller esi
+
+# qhasm: caller edi
+
+# qhasm: caller ebp
+
+# qhasm: int32 k
+
+# qhasm: int32 kbits
+
+# qhasm: int32 iv
+
+# qhasm: int32 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int32 m
+
+# qhasm: stack32 out_stack
+
+# qhasm: int32 out
+
+# qhasm: stack32 bytes_stack
+
+# qhasm: int32 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int6464 p
+
+# qhasm: int6464 q
+
+# qhasm: int6464 r
+
+# qhasm: int6464 s
+
+# qhasm: int6464 t
+
+# qhasm: int6464 u
+
+# qhasm: int6464 v
+
+# qhasm: int6464 w
+
+# qhasm: int6464 mp
+
+# qhasm: int6464 mq
+
+# qhasm: int6464 mr
+
+# qhasm: int6464 ms
+
+# qhasm: int6464 mt
+
+# qhasm: int6464 mu
+
+# qhasm: int6464 mv
+
+# qhasm: int6464 mw
+
+# qhasm: int32 in0
+
+# qhasm: int32 in1
+
+# qhasm: int32 in2
+
+# qhasm: int32 in3
+
+# qhasm: int32 in4
+
+# qhasm: int32 in5
+
+# qhasm: int32 in6
+
+# qhasm: int32 in7
+
+# qhasm: int32 in8
+
+# qhasm: int32 in9
+
+# qhasm: int32 in10
+
+# qhasm: int32 in11
+
+# qhasm: int32 in12
+
+# qhasm: int32 in13
+
+# qhasm: int32 in14
+
+# qhasm: int32 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: stack32 ctarget
+
+# qhasm: enter crypto_stream_salsa2012_x86_xmm5
+.text
+.p2align 5
+.globl _crypto_stream_salsa2012_x86_xmm5
+.globl crypto_stream_salsa2012_x86_xmm5
+_crypto_stream_salsa2012_x86_xmm5:
+crypto_stream_salsa2012_x86_xmm5:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: bytes = arg2
+# asm 1: movl <arg2=stack32#-2,>bytes=int32#3
+# asm 2: movl <arg2=8(%esp,%eax),>bytes=%edx
+movl 8(%esp,%eax),%edx
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm: iv = arg4
+# asm 1: movl <arg4=stack32#-4,>iv=int32#4
+# asm 2: movl <arg4=16(%esp,%eax),>iv=%ebx
+movl 16(%esp,%eax),%ebx
+
+# qhasm: k = arg5
+# asm 1: movl <arg5=stack32#-5,>k=int32#7
+# asm 2: movl <arg5=20(%esp,%eax),>k=%ebp
+movl 20(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int32#1
+# asm 2: mov  $0,>a=%eax
+mov  $0,%eax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int32#3,>i=int32#2
+# asm 2: mov  <bytes=%edx,>i=%ecx
+mov  %edx,%ecx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: subl <bytes=int32#3,<out=int32#6
+# asm 2: subl <bytes=%edx,<out=%edi
+subl %edx,%edi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa2012_x86_xmm5_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa2012_x86_xmm5_xor
+.globl crypto_stream_salsa2012_x86_xmm5_xor
+_crypto_stream_salsa2012_x86_xmm5_xor:
+crypto_stream_salsa2012_x86_xmm5_xor:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = arg2
+# asm 1: movl <arg2=stack32#-2,>m=int32#5
+# asm 2: movl <arg2=8(%esp,%eax),>m=%esi
+movl 8(%esp,%eax),%esi
+
+# qhasm: bytes = arg3
+# asm 1: movl <arg3=stack32#-3,>bytes=int32#3
+# asm 2: movl <arg3=12(%esp,%eax),>bytes=%edx
+movl 12(%esp,%eax),%edx
+
+# qhasm: iv = arg5
+# asm 1: movl <arg5=stack32#-5,>iv=int32#4
+# asm 2: movl <arg5=20(%esp,%eax),>iv=%ebx
+movl 20(%esp,%eax),%ebx
+
+# qhasm: k = arg6
+# asm 1: movl <arg6=stack32#-6,>k=int32#7
+# asm 2: movl <arg6=24(%esp,%eax),>k=%ebp
+movl 24(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#3,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%edx,>bytes_stack=24(%esp)
+movl %edx,24(%esp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl 12(<k=int32#7),>in4=int32#1
+# asm 2: movl 12(<k=%ebp),>in4=%eax
+movl 12(%ebp),%eax
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl 20(<k=int32#7),>in12=int32#2
+# asm 2: movl 20(<k=%ebp),>in12=%ecx
+movl 20(%ebp),%ecx
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int32#1,>x3=stack128#1
+# asm 2: movl <in4=%eax,>x3=32(%esp)
+movl %eax,32(%esp)
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int32#2,>x1=stack128#2
+# asm 2: movl <in12=%ecx,>x1=48(%esp)
+movl %ecx,48(%esp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int32#1
+# asm 2: mov  $1634760805,>in0=%eax
+mov  $1634760805,%eax
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int32#2
+# asm 2: mov  $0,>in8=%ecx
+mov  $0,%ecx
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int32#1,>x0=stack128#3
+# asm 2: movl <in0=%eax,>x0=64(%esp)
+movl %eax,64(%esp)
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl 0(<iv=int32#4),>in6=int32#1
+# asm 2: movl 0(<iv=%ebx),>in6=%eax
+movl 0(%ebx),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl 4(<iv=int32#4),>in7=int32#2
+# asm 2: movl 4(<iv=%ebx),>in7=%ecx
+movl 4(%ebx),%ecx
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int32#1,8+<x1=stack128#2
+# asm 2: movl <in6=%eax,8+<x1=48(%esp)
+movl %eax,8+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int32#2,12+<x2=stack128#4
+# asm 2: movl <in7=%ecx,12+<x2=80(%esp)
+movl %ecx,12+80(%esp)
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int32#1
+# asm 2: mov  $0,>in9=%eax
+mov  $0,%eax
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int32#2
+# asm 2: mov  $2036477234,>in10=%ecx
+mov  $2036477234,%ecx
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#1,4+<x3=stack128#1
+# asm 2: movl <in9=%eax,4+<x3=32(%esp)
+movl %eax,4+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int32#2,8+<x0=stack128#3
+# asm 2: movl <in10=%ecx,8+<x0=64(%esp)
+movl %ecx,8+64(%esp)
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl 0(<k=int32#7),>in1=int32#1
+# asm 2: movl 0(<k=%ebp),>in1=%eax
+movl 0(%ebp),%eax
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl 4(<k=int32#7),>in2=int32#2
+# asm 2: movl 4(<k=%ebp),>in2=%ecx
+movl 4(%ebp),%ecx
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl 8(<k=int32#7),>in3=int32#3
+# asm 2: movl 8(<k=%ebp),>in3=%edx
+movl 8(%ebp),%edx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int32#4
+# asm 2: mov  $857760878,>in5=%ebx
+mov  $857760878,%ebx
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int32#1,4+<x1=stack128#2
+# asm 2: movl <in1=%eax,4+<x1=48(%esp)
+movl %eax,4+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int32#2,8+<x2=stack128#4
+# asm 2: movl <in2=%ecx,8+<x2=80(%esp)
+movl %ecx,8+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int32#3,12+<x3=stack128#1
+# asm 2: movl <in3=%edx,12+<x3=32(%esp)
+movl %edx,12+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int32#4,4+<x0=stack128#3
+# asm 2: movl <in5=%ebx,4+<x0=64(%esp)
+movl %ebx,4+64(%esp)
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl 16(<k=int32#7),>in11=int32#1
+# asm 2: movl 16(<k=%ebp),>in11=%eax
+movl 16(%ebp),%eax
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl 24(<k=int32#7),>in13=int32#2
+# asm 2: movl 24(<k=%ebp),>in13=%ecx
+movl 24(%ebp),%ecx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl 28(<k=int32#7),>in14=int32#3
+# asm 2: movl 28(<k=%ebp),>in14=%edx
+movl 28(%ebp),%edx
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int32#4
+# asm 2: mov  $1797285236,>in15=%ebx
+mov  $1797285236,%ebx
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int32#1,12+<x1=stack128#2
+# asm 2: movl <in11=%eax,12+<x1=48(%esp)
+movl %eax,12+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int32#2,4+<x2=stack128#4
+# asm 2: movl <in13=%ecx,4+<x2=80(%esp)
+movl %ecx,4+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int32#3,8+<x3=stack128#1
+# asm 2: movl <in14=%edx,8+<x3=32(%esp)
+movl %edx,8+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int32#4,12+<x0=stack128#3
+# asm 2: movl <in15=%ebx,12+<x0=64(%esp)
+movl %ebx,12+64(%esp)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#3,>z0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>z0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=96(%esp)
+movdqa %xmm1,96(%esp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=112(%esp)
+movdqa %xmm2,112(%esp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=128(%esp)
+movdqa %xmm3,128(%esp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=144(%esp)
+movdqa %xmm0,144(%esp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#2,>z1=int6464#1
+# asm 2: movdqa <x1=48(%esp),>z1=%xmm0
+movdqa 48(%esp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=160(%esp)
+movdqa %xmm1,160(%esp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=176(%esp)
+movdqa %xmm2,176(%esp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=192(%esp)
+movdqa %xmm3,192(%esp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=208(%esp)
+movdqa %xmm0,208(%esp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#4,>z2=int6464#1
+# asm 2: movdqa <x2=80(%esp),>z2=%xmm0
+movdqa 80(%esp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=224(%esp)
+movdqa %xmm1,224(%esp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=240(%esp)
+movdqa %xmm2,240(%esp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=256(%esp)
+movdqa %xmm0,256(%esp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#1,>z3=int6464#1
+# asm 2: movdqa <x3=32(%esp),>z3=%xmm0
+movdqa 32(%esp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=272(%esp)
+movdqa %xmm1,272(%esp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=288(%esp)
+movdqa %xmm2,288(%esp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=304(%esp)
+movdqa %xmm0,304(%esp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int32#2,>orig8=stack128#19
+# asm 2: movl <in8=%ecx,>orig8=320(%esp)
+movl %ecx,320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int32#3,>orig9=stack128#20
+# asm 2: movl <in9=%edx,>orig9=336(%esp)
+movl %edx,336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int32#2,4+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,4+<orig8=320(%esp)
+movl %ecx,4+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int32#3,4+<orig9=stack128#20
+# asm 2: movl <in9=%edx,4+<orig9=336(%esp)
+movl %edx,4+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int32#2,8+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,8+<orig8=320(%esp)
+movl %ecx,8+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int32#3,8+<orig9=stack128#20
+# asm 2: movl <in9=%edx,8+<orig9=336(%esp)
+movl %edx,8+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int32#2,12+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,12+<orig8=320(%esp)
+movl %ecx,12+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int32#3,12+<orig9=stack128#20
+# asm 2: movl <in9=%edx,12+<orig9=336(%esp)
+movl %edx,12+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: i = 12
+# asm 1: mov  $12,>i=int32#1
+# asm 2: mov  $12,>i=%eax
+mov  $12,%eax
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=96(%esp),>z5=%xmm0
+movdqa 96(%esp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=112(%esp),>z10=%xmm1
+movdqa 112(%esp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=128(%esp),>z15=%xmm2
+movdqa 128(%esp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=288(%esp),>z14=%xmm3
+movdqa 288(%esp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=304(%esp),>z3=%xmm4
+movdqa 304(%esp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=160(%esp),>z6=%xmm5
+movdqa 160(%esp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=176(%esp),>z11=%xmm6
+movdqa 176(%esp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=208(%esp),>z1=%xmm7
+movdqa 208(%esp),%xmm7
+
+# qhasm:   z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#21
+# asm 2: movdqa <z5=%xmm0,>z5_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:   z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#22
+# asm 2: movdqa <z10=%xmm1,>z10_stack=368(%esp)
+movdqa %xmm1,368(%esp)
+
+# qhasm:   z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#23
+# asm 2: movdqa <z15=%xmm2,>z15_stack=384(%esp)
+movdqa %xmm2,384(%esp)
+
+# qhasm:   z14_stack = z14
+# asm 1: movdqa <z14=int6464#4,>z14_stack=stack128#24
+# asm 2: movdqa <z14=%xmm3,>z14_stack=400(%esp)
+movdqa %xmm3,400(%esp)
+
+# qhasm:   z3_stack = z3
+# asm 1: movdqa <z3=int6464#5,>z3_stack=stack128#25
+# asm 2: movdqa <z3=%xmm4,>z3_stack=416(%esp)
+movdqa %xmm4,416(%esp)
+
+# qhasm:   z6_stack = z6
+# asm 1: movdqa <z6=int6464#6,>z6_stack=stack128#26
+# asm 2: movdqa <z6=%xmm5,>z6_stack=432(%esp)
+movdqa %xmm5,432(%esp)
+
+# qhasm:   z11_stack = z11
+# asm 1: movdqa <z11=int6464#7,>z11_stack=stack128#27
+# asm 2: movdqa <z11=%xmm6,>z11_stack=448(%esp)
+movdqa %xmm6,448(%esp)
+
+# qhasm:   z1_stack = z1
+# asm 1: movdqa <z1=int6464#8,>z1_stack=stack128#28
+# asm 2: movdqa <z1=%xmm7,>z1_stack=464(%esp)
+movdqa %xmm7,464(%esp)
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#5
+# asm 2: movdqa <orig7=224(%esp),>z7=%xmm4
+movdqa 224(%esp),%xmm4
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#6
+# asm 2: movdqa <orig13=240(%esp),>z13=%xmm5
+movdqa 240(%esp),%xmm5
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#7
+# asm 2: movdqa <orig2=256(%esp),>z2=%xmm6
+movdqa 256(%esp),%xmm6
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#8
+# asm 2: movdqa <orig9=336(%esp),>z9=%xmm7
+movdqa 336(%esp),%xmm7
+
+# qhasm:                   p = orig0
+# asm 1: movdqa <orig0=stack128#8,>p=int6464#1
+# asm 2: movdqa <orig0=144(%esp),>p=%xmm0
+movdqa 144(%esp),%xmm0
+
+# qhasm:                   t = orig12
+# asm 1: movdqa <orig12=stack128#11,>t=int6464#3
+# asm 2: movdqa <orig12=192(%esp),>t=%xmm2
+movdqa 192(%esp),%xmm2
+
+# qhasm:                   q = orig4
+# asm 1: movdqa <orig4=stack128#16,>q=int6464#4
+# asm 2: movdqa <orig4=272(%esp),>q=%xmm3
+movdqa 272(%esp),%xmm3
+
+# qhasm:                   r = orig8
+# asm 1: movdqa <orig8=stack128#19,>r=int6464#2
+# asm 2: movdqa <orig8=320(%esp),>r=%xmm1
+movdqa 320(%esp),%xmm1
+
+# qhasm:   z7_stack = z7
+# asm 1: movdqa <z7=int6464#5,>z7_stack=stack128#29
+# asm 2: movdqa <z7=%xmm4,>z7_stack=480(%esp)
+movdqa %xmm4,480(%esp)
+
+# qhasm:   z13_stack = z13
+# asm 1: movdqa <z13=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <z13=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm:   z2_stack = z2
+# asm 1: movdqa <z2=int6464#7,>z2_stack=stack128#31
+# asm 2: movdqa <z2=%xmm6,>z2_stack=512(%esp)
+movdqa %xmm6,512(%esp)
+
+# qhasm:   z9_stack = z9
+# asm 1: movdqa <z9=int6464#8,>z9_stack=stack128#32
+# asm 2: movdqa <z9=%xmm7,>z9_stack=528(%esp)
+movdqa %xmm7,528(%esp)
+
+# qhasm:   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:   z12_stack = t
+# asm 1: movdqa <t=int6464#3,>z12_stack=stack128#34
+# asm 2: movdqa <t=%xmm2,>z12_stack=560(%esp)
+movdqa %xmm2,560(%esp)
+
+# qhasm:   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#35
+# asm 2: movdqa <q=%xmm3,>z4_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#36
+# asm 2: movdqa <r=%xmm1,>z8_stack=592(%esp)
+movdqa %xmm1,592(%esp)
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#33
+# asm 2: movdqa <q=%xmm3,>z4_stack=544(%esp)
+movdqa %xmm3,544(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#34
+# asm 2: movdqa <r=%xmm1,>z8_stack=560(%esp)
+movdqa %xmm1,560(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>mt=int6464#3
+# asm 2: movdqa <z1_stack=464(%esp),>mt=%xmm2
+movdqa 464(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>mp=int6464#5
+# asm 2: movdqa <z5_stack=352(%esp),>mp=%xmm4
+movdqa 352(%esp),%xmm4
+
+# qhasm:                                   mq = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>mq=int6464#4
+# asm 2: movdqa <z9_stack=528(%esp),>mq=%xmm3
+movdqa 528(%esp),%xmm3
+
+# qhasm:                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>mr=int6464#6
+# asm 2: movdqa <z13_stack=496(%esp),>mr=%xmm5
+movdqa 496(%esp),%xmm5
+
+# qhasm:                   z12_stack = s
+# asm 1: movdqa <s=int6464#7,>z12_stack=stack128#30
+# asm 2: movdqa <s=%xmm6,>z12_stack=496(%esp)
+movdqa %xmm6,496(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <p=%xmm0,>z0_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z9_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z9_stack=stack128#32
+# asm 2: movdqa <mq=%xmm3,>z9_stack=528(%esp)
+movdqa %xmm3,528(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#35
+# asm 2: movdqa <mr=%xmm5,>z13_stack=576(%esp)
+movdqa %xmm5,576(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>t=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>t=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>p=int6464#1
+# asm 2: movdqa <z10_stack=368(%esp),>p=%xmm0
+movdqa 368(%esp),%xmm0
+
+# qhasm:                                                   q = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>q=int6464#4
+# asm 2: movdqa <z14_stack=400(%esp),>q=%xmm3
+movdqa 400(%esp),%xmm3
+
+# qhasm:                                                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>r=int6464#2
+# asm 2: movdqa <z2_stack=512(%esp),>r=%xmm1
+movdqa 512(%esp),%xmm1
+
+# qhasm:                                   z1_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z1_stack=stack128#22
+# asm 2: movdqa <ms=%xmm6,>z1_stack=368(%esp)
+movdqa %xmm6,368(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#24
+# asm 2: movdqa <mp=%xmm4,>z5_stack=400(%esp)
+movdqa %xmm4,400(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z14_stack = q
+# asm 1: movdqa <q=int6464#4,>z14_stack=stack128#36
+# asm 2: movdqa <q=%xmm3,>z14_stack=592(%esp)
+movdqa %xmm3,592(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#26
+# asm 2: movdqa <r=%xmm1,>z2_stack=432(%esp)
+movdqa %xmm1,432(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>mt=int6464#3
+# asm 2: movdqa <z11_stack=448(%esp),>mt=%xmm2
+movdqa 448(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>mp=int6464#5
+# asm 2: movdqa <z15_stack=384(%esp),>mp=%xmm4
+movdqa 384(%esp),%xmm4
+
+# qhasm:                                                                   mq = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>mq=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>mq=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:                                                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                                                   z6_stack = s
+# asm 1: movdqa <s=int6464#7,>z6_stack=stack128#23
+# asm 2: movdqa <s=%xmm6,>z6_stack=384(%esp)
+movdqa %xmm6,384(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#27
+# asm 2: movdqa <p=%xmm0,>z10_stack=448(%esp)
+movdqa %xmm0,448(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z3_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z3_stack=stack128#25
+# asm 2: movdqa <mq=%xmm3,>z3_stack=416(%esp)
+movdqa %xmm3,416(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>t=int6464#3
+# asm 2: movdqa <z3_stack=416(%esp),>t=%xmm2
+movdqa 416(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>p=int6464#1
+# asm 2: movdqa <z0_stack=352(%esp),>p=%xmm0
+movdqa 352(%esp),%xmm0
+
+# qhasm:                   q = z1_stack
+# asm 1: movdqa <z1_stack=stack128#22,>q=int6464#4
+# asm 2: movdqa <z1_stack=368(%esp),>q=%xmm3
+movdqa 368(%esp),%xmm3
+
+# qhasm:                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#26,>r=int6464#2
+# asm 2: movdqa <z2_stack=432(%esp),>r=%xmm1
+movdqa 432(%esp),%xmm1
+
+# qhasm:                                                                   z11_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z11_stack=stack128#21
+# asm 2: movdqa <ms=%xmm6,>z11_stack=352(%esp)
+movdqa %xmm6,352(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#22
+# asm 2: movdqa <mp=%xmm4,>z15_stack=368(%esp)
+movdqa %xmm4,368(%esp)
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z1_stack = q
+# asm 1: movdqa <q=int6464#4,>z1_stack=stack128#28
+# asm 2: movdqa <q=%xmm3,>z1_stack=464(%esp)
+movdqa %xmm3,464(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#31
+# asm 2: movdqa <r=%xmm1,>z2_stack=512(%esp)
+movdqa %xmm1,512(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z4_stack
+# asm 1: movdqa <z4_stack=stack128#33,>mt=int6464#3
+# asm 2: movdqa <z4_stack=544(%esp),>mt=%xmm2
+movdqa 544(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#24,>mp=int6464#5
+# asm 2: movdqa <z5_stack=400(%esp),>mp=%xmm4
+movdqa 400(%esp),%xmm4
+
+# qhasm:                                   mq = z6_stack
+# asm 1: movdqa <z6_stack=stack128#23,>mq=int6464#4
+# asm 2: movdqa <z6_stack=384(%esp),>mq=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                   z3_stack = s
+# asm 1: movdqa <s=int6464#7,>z3_stack=stack128#25
+# asm 2: movdqa <s=%xmm6,>z3_stack=416(%esp)
+movdqa %xmm6,416(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z6_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z6_stack=stack128#26
+# asm 2: movdqa <mq=%xmm3,>z6_stack=432(%esp)
+movdqa %xmm3,432(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>t=int6464#3
+# asm 2: movdqa <z9_stack=528(%esp),>t=%xmm2
+movdqa 528(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#27,>p=int6464#1
+# asm 2: movdqa <z10_stack=448(%esp),>p=%xmm0
+movdqa 448(%esp),%xmm0
+
+# qhasm:                                                   q = z11_stack
+# asm 1: movdqa <z11_stack=stack128#21,>q=int6464#4
+# asm 2: movdqa <z11_stack=352(%esp),>q=%xmm3
+movdqa 352(%esp),%xmm3
+
+# qhasm:                                                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#34,>r=int6464#2
+# asm 2: movdqa <z8_stack=560(%esp),>r=%xmm1
+movdqa 560(%esp),%xmm1
+
+# qhasm:                                   z4_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z4_stack=stack128#34
+# asm 2: movdqa <ms=%xmm6,>z4_stack=560(%esp)
+movdqa %xmm6,560(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#21
+# asm 2: movdqa <mp=%xmm4,>z5_stack=352(%esp)
+movdqa %xmm4,352(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z11_stack = q
+# asm 1: movdqa <q=int6464#4,>z11_stack=stack128#27
+# asm 2: movdqa <q=%xmm3,>z11_stack=448(%esp)
+movdqa %xmm3,448(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#37
+# asm 2: movdqa <r=%xmm1,>z8_stack=608(%esp)
+movdqa %xmm1,608(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z14_stack
+# asm 1: movdqa <z14_stack=stack128#36,>mt=int6464#3
+# asm 2: movdqa <z14_stack=592(%esp),>mt=%xmm2
+movdqa 592(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>mp=int6464#5
+# asm 2: movdqa <z15_stack=368(%esp),>mp=%xmm4
+movdqa 368(%esp),%xmm4
+
+# qhasm:                                                                   mq = z12_stack
+# asm 1: movdqa <z12_stack=stack128#30,>mq=int6464#4
+# asm 2: movdqa <z12_stack=496(%esp),>mq=%xmm3
+movdqa 496(%esp),%xmm3
+
+# qhasm:                                                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#35,>mr=int6464#6
+# asm 2: movdqa <z13_stack=576(%esp),>mr=%xmm5
+movdqa 576(%esp),%xmm5
+
+# qhasm:                                                   z9_stack = s
+# asm 1: movdqa <s=int6464#7,>z9_stack=stack128#32
+# asm 2: movdqa <s=%xmm6,>z9_stack=528(%esp)
+movdqa %xmm6,528(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#22
+# asm 2: movdqa <p=%xmm0,>z10_stack=368(%esp)
+movdqa %xmm0,368(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z12_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z12_stack=stack128#35
+# asm 2: movdqa <mq=%xmm3,>z12_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <mr=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>t=int6464#3
+# asm 2: movdqa <z12_stack=576(%esp),>t=%xmm2
+movdqa 576(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>p=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>p=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:                   q = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>q=int6464#4
+# asm 2: movdqa <z4_stack=560(%esp),>q=%xmm3
+movdqa 560(%esp),%xmm3
+
+# qhasm:                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>r=int6464#2
+# asm 2: movdqa <z8_stack=608(%esp),>r=%xmm1
+movdqa 608(%esp),%xmm1
+
+# qhasm:                                                                   z14_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z14_stack=stack128#24
+# asm 2: movdqa <ms=%xmm6,>z14_stack=400(%esp)
+movdqa %xmm6,400(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#23
+# asm 2: movdqa <mp=%xmm4,>z15_stack=384(%esp)
+movdqa %xmm4,384(%esp)
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int32#1
+# asm 2: sub  $2,<i=%eax
+sub  $2,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   out = out_stack
+# asm 1: movl <out_stack=stack32#6,>out=int32#6
+# asm 2: movl <out_stack=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:   z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>z0=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>z0=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:   z1 = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>z1=int6464#2
+# asm 2: movdqa <z1_stack=464(%esp),>z1=%xmm1
+movdqa 464(%esp),%xmm1
+
+# qhasm:   z2 = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>z2=int6464#3
+# asm 2: movdqa <z2_stack=512(%esp),>z2=%xmm2
+movdqa 512(%esp),%xmm2
+
+# qhasm:   z3 = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>z3=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>z3=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#1
+# asm 2: paddd <orig0=144(%esp),<z0=%xmm0
+paddd 144(%esp),%xmm0
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#2
+# asm 2: paddd <orig1=208(%esp),<z1=%xmm1
+paddd 208(%esp),%xmm1
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#3
+# asm 2: paddd <orig2=256(%esp),<z2=%xmm2
+paddd 256(%esp),%xmm2
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#4
+# asm 2: paddd <orig3=304(%esp),<z3=%xmm3
+paddd 304(%esp),%xmm3
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int32#5),<in0=int32#1
+# asm 2: xorl 64(<m=%esi),<in0=%eax
+xorl 64(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int32#5),<in1=int32#2
+# asm 2: xorl 68(<m=%esi),<in1=%ecx
+xorl 68(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int32#5),<in2=int32#3
+# asm 2: xorl 72(<m=%esi),<in2=%edx
+xorl 72(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int32#5),<in3=int32#4
+# asm 2: xorl 76(<m=%esi),<in3=%ebx
+xorl 76(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl <in0=int32#1,64(<out=int32#6)
+# asm 2: movl <in0=%eax,64(<out=%edi)
+movl %eax,64(%edi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl <in1=int32#2,68(<out=int32#6)
+# asm 2: movl <in1=%ecx,68(<out=%edi)
+movl %ecx,68(%edi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl <in2=int32#3,72(<out=int32#6)
+# asm 2: movl <in2=%edx,72(<out=%edi)
+movl %edx,72(%edi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl <in3=int32#4,76(<out=int32#6)
+# asm 2: movl <in3=%ebx,76(<out=%edi)
+movl %ebx,76(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int32#5),<in0=int32#1
+# asm 2: xorl 128(<m=%esi),<in0=%eax
+xorl 128(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int32#5),<in1=int32#2
+# asm 2: xorl 132(<m=%esi),<in1=%ecx
+xorl 132(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int32#5),<in2=int32#3
+# asm 2: xorl 136(<m=%esi),<in2=%edx
+xorl 136(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int32#5),<in3=int32#4
+# asm 2: xorl 140(<m=%esi),<in3=%ebx
+xorl 140(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl <in0=int32#1,128(<out=int32#6)
+# asm 2: movl <in0=%eax,128(<out=%edi)
+movl %eax,128(%edi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl <in1=int32#2,132(<out=int32#6)
+# asm 2: movl <in1=%ecx,132(<out=%edi)
+movl %ecx,132(%edi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl <in2=int32#3,136(<out=int32#6)
+# asm 2: movl <in2=%edx,136(<out=%edi)
+movl %edx,136(%edi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl <in3=int32#4,140(<out=int32#6)
+# asm 2: movl <in3=%ebx,140(<out=%edi)
+movl %ebx,140(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int32#5),<in0=int32#1
+# asm 2: xorl 192(<m=%esi),<in0=%eax
+xorl 192(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int32#5),<in1=int32#2
+# asm 2: xorl 196(<m=%esi),<in1=%ecx
+xorl 196(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int32#5),<in2=int32#3
+# asm 2: xorl 200(<m=%esi),<in2=%edx
+xorl 200(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int32#5),<in3=int32#4
+# asm 2: xorl 204(<m=%esi),<in3=%ebx
+xorl 204(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl <in0=int32#1,192(<out=int32#6)
+# asm 2: movl <in0=%eax,192(<out=%edi)
+movl %eax,192(%edi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl <in1=int32#2,196(<out=int32#6)
+# asm 2: movl <in1=%ecx,196(<out=%edi)
+movl %ecx,196(%edi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl <in2=int32#3,200(<out=int32#6)
+# asm 2: movl <in2=%edx,200(<out=%edi)
+movl %edx,200(%edi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl <in3=int32#4,204(<out=int32#6)
+# asm 2: movl <in3=%ebx,204(<out=%edi)
+movl %ebx,204(%edi)
+
+# qhasm:   z4 = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>z4=int6464#1
+# asm 2: movdqa <z4_stack=560(%esp),>z4=%xmm0
+movdqa 560(%esp),%xmm0
+
+# qhasm:   z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>z5=int6464#2
+# asm 2: movdqa <z5_stack=352(%esp),>z5=%xmm1
+movdqa 352(%esp),%xmm1
+
+# qhasm:   z6 = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>z6=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>z6=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:   z7 = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>z7=int6464#4
+# asm 2: movdqa <z7_stack=480(%esp),>z7=%xmm3
+movdqa 480(%esp),%xmm3
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#1
+# asm 2: paddd <orig4=272(%esp),<z4=%xmm0
+paddd 272(%esp),%xmm0
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#2
+# asm 2: paddd <orig5=96(%esp),<z5=%xmm1
+paddd 96(%esp),%xmm1
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#3
+# asm 2: paddd <orig6=160(%esp),<z6=%xmm2
+paddd 160(%esp),%xmm2
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#4
+# asm 2: paddd <orig7=224(%esp),<z7=%xmm3
+paddd 224(%esp),%xmm3
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#1
+# asm 2: xorl 16(<m=%esi),<in4=%eax
+xorl 16(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#2
+# asm 2: xorl 20(<m=%esi),<in5=%ecx
+xorl 20(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#3
+# asm 2: xorl 24(<m=%esi),<in6=%edx
+xorl 24(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#4
+# asm 2: xorl 28(<m=%esi),<in7=%ebx
+xorl 28(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#1,16(<out=int32#6)
+# asm 2: movl <in4=%eax,16(<out=%edi)
+movl %eax,16(%edi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#2,20(<out=int32#6)
+# asm 2: movl <in5=%ecx,20(<out=%edi)
+movl %ecx,20(%edi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#3,24(<out=int32#6)
+# asm 2: movl <in6=%edx,24(<out=%edi)
+movl %edx,24(%edi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#4,28(<out=int32#6)
+# asm 2: movl <in7=%ebx,28(<out=%edi)
+movl %ebx,28(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int32#5),<in4=int32#1
+# asm 2: xorl 80(<m=%esi),<in4=%eax
+xorl 80(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int32#5),<in5=int32#2
+# asm 2: xorl 84(<m=%esi),<in5=%ecx
+xorl 84(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int32#5),<in6=int32#3
+# asm 2: xorl 88(<m=%esi),<in6=%edx
+xorl 88(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int32#5),<in7=int32#4
+# asm 2: xorl 92(<m=%esi),<in7=%ebx
+xorl 92(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl <in4=int32#1,80(<out=int32#6)
+# asm 2: movl <in4=%eax,80(<out=%edi)
+movl %eax,80(%edi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl <in5=int32#2,84(<out=int32#6)
+# asm 2: movl <in5=%ecx,84(<out=%edi)
+movl %ecx,84(%edi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl <in6=int32#3,88(<out=int32#6)
+# asm 2: movl <in6=%edx,88(<out=%edi)
+movl %edx,88(%edi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl <in7=int32#4,92(<out=int32#6)
+# asm 2: movl <in7=%ebx,92(<out=%edi)
+movl %ebx,92(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int32#5),<in4=int32#1
+# asm 2: xorl 144(<m=%esi),<in4=%eax
+xorl 144(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int32#5),<in5=int32#2
+# asm 2: xorl 148(<m=%esi),<in5=%ecx
+xorl 148(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int32#5),<in6=int32#3
+# asm 2: xorl 152(<m=%esi),<in6=%edx
+xorl 152(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int32#5),<in7=int32#4
+# asm 2: xorl 156(<m=%esi),<in7=%ebx
+xorl 156(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl <in4=int32#1,144(<out=int32#6)
+# asm 2: movl <in4=%eax,144(<out=%edi)
+movl %eax,144(%edi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl <in5=int32#2,148(<out=int32#6)
+# asm 2: movl <in5=%ecx,148(<out=%edi)
+movl %ecx,148(%edi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl <in6=int32#3,152(<out=int32#6)
+# asm 2: movl <in6=%edx,152(<out=%edi)
+movl %edx,152(%edi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl <in7=int32#4,156(<out=int32#6)
+# asm 2: movl <in7=%ebx,156(<out=%edi)
+movl %ebx,156(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int32#5),<in4=int32#1
+# asm 2: xorl 208(<m=%esi),<in4=%eax
+xorl 208(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int32#5),<in5=int32#2
+# asm 2: xorl 212(<m=%esi),<in5=%ecx
+xorl 212(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int32#5),<in6=int32#3
+# asm 2: xorl 216(<m=%esi),<in6=%edx
+xorl 216(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int32#5),<in7=int32#4
+# asm 2: xorl 220(<m=%esi),<in7=%ebx
+xorl 220(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl <in4=int32#1,208(<out=int32#6)
+# asm 2: movl <in4=%eax,208(<out=%edi)
+movl %eax,208(%edi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl <in5=int32#2,212(<out=int32#6)
+# asm 2: movl <in5=%ecx,212(<out=%edi)
+movl %ecx,212(%edi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl <in6=int32#3,216(<out=int32#6)
+# asm 2: movl <in6=%edx,216(<out=%edi)
+movl %edx,216(%edi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl <in7=int32#4,220(<out=int32#6)
+# asm 2: movl <in7=%ebx,220(<out=%edi)
+movl %ebx,220(%edi)
+
+# qhasm:   z8 = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>z8=int6464#1
+# asm 2: movdqa <z8_stack=608(%esp),>z8=%xmm0
+movdqa 608(%esp),%xmm0
+
+# qhasm:   z9 = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>z9=int6464#2
+# asm 2: movdqa <z9_stack=528(%esp),>z9=%xmm1
+movdqa 528(%esp),%xmm1
+
+# qhasm:   z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>z10=int6464#3
+# asm 2: movdqa <z10_stack=368(%esp),>z10=%xmm2
+movdqa 368(%esp),%xmm2
+
+# qhasm:   z11 = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>z11=int6464#4
+# asm 2: movdqa <z11_stack=448(%esp),>z11=%xmm3
+movdqa 448(%esp),%xmm3
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#1
+# asm 2: paddd <orig8=320(%esp),<z8=%xmm0
+paddd 320(%esp),%xmm0
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#2
+# asm 2: paddd <orig9=336(%esp),<z9=%xmm1
+paddd 336(%esp),%xmm1
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#3
+# asm 2: paddd <orig10=112(%esp),<z10=%xmm2
+paddd 112(%esp),%xmm2
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#4
+# asm 2: paddd <orig11=176(%esp),<z11=%xmm3
+paddd 176(%esp),%xmm3
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#1
+# asm 2: xorl 32(<m=%esi),<in8=%eax
+xorl 32(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#2
+# asm 2: xorl 36(<m=%esi),<in9=%ecx
+xorl 36(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#3
+# asm 2: xorl 40(<m=%esi),<in10=%edx
+xorl 40(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#4
+# asm 2: xorl 44(<m=%esi),<in11=%ebx
+xorl 44(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#1,32(<out=int32#6)
+# asm 2: movl <in8=%eax,32(<out=%edi)
+movl %eax,32(%edi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#2,36(<out=int32#6)
+# asm 2: movl <in9=%ecx,36(<out=%edi)
+movl %ecx,36(%edi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#3,40(<out=int32#6)
+# asm 2: movl <in10=%edx,40(<out=%edi)
+movl %edx,40(%edi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#4,44(<out=int32#6)
+# asm 2: movl <in11=%ebx,44(<out=%edi)
+movl %ebx,44(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int32#5),<in8=int32#1
+# asm 2: xorl 96(<m=%esi),<in8=%eax
+xorl 96(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int32#5),<in9=int32#2
+# asm 2: xorl 100(<m=%esi),<in9=%ecx
+xorl 100(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int32#5),<in10=int32#3
+# asm 2: xorl 104(<m=%esi),<in10=%edx
+xorl 104(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int32#5),<in11=int32#4
+# asm 2: xorl 108(<m=%esi),<in11=%ebx
+xorl 108(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl <in8=int32#1,96(<out=int32#6)
+# asm 2: movl <in8=%eax,96(<out=%edi)
+movl %eax,96(%edi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl <in9=int32#2,100(<out=int32#6)
+# asm 2: movl <in9=%ecx,100(<out=%edi)
+movl %ecx,100(%edi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl <in10=int32#3,104(<out=int32#6)
+# asm 2: movl <in10=%edx,104(<out=%edi)
+movl %edx,104(%edi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl <in11=int32#4,108(<out=int32#6)
+# asm 2: movl <in11=%ebx,108(<out=%edi)
+movl %ebx,108(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int32#5),<in8=int32#1
+# asm 2: xorl 160(<m=%esi),<in8=%eax
+xorl 160(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int32#5),<in9=int32#2
+# asm 2: xorl 164(<m=%esi),<in9=%ecx
+xorl 164(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int32#5),<in10=int32#3
+# asm 2: xorl 168(<m=%esi),<in10=%edx
+xorl 168(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int32#5),<in11=int32#4
+# asm 2: xorl 172(<m=%esi),<in11=%ebx
+xorl 172(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl <in8=int32#1,160(<out=int32#6)
+# asm 2: movl <in8=%eax,160(<out=%edi)
+movl %eax,160(%edi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl <in9=int32#2,164(<out=int32#6)
+# asm 2: movl <in9=%ecx,164(<out=%edi)
+movl %ecx,164(%edi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl <in10=int32#3,168(<out=int32#6)
+# asm 2: movl <in10=%edx,168(<out=%edi)
+movl %edx,168(%edi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl <in11=int32#4,172(<out=int32#6)
+# asm 2: movl <in11=%ebx,172(<out=%edi)
+movl %ebx,172(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int32#5),<in8=int32#1
+# asm 2: xorl 224(<m=%esi),<in8=%eax
+xorl 224(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int32#5),<in9=int32#2
+# asm 2: xorl 228(<m=%esi),<in9=%ecx
+xorl 228(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int32#5),<in10=int32#3
+# asm 2: xorl 232(<m=%esi),<in10=%edx
+xorl 232(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int32#5),<in11=int32#4
+# asm 2: xorl 236(<m=%esi),<in11=%ebx
+xorl 236(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl <in8=int32#1,224(<out=int32#6)
+# asm 2: movl <in8=%eax,224(<out=%edi)
+movl %eax,224(%edi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl <in9=int32#2,228(<out=int32#6)
+# asm 2: movl <in9=%ecx,228(<out=%edi)
+movl %ecx,228(%edi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl <in10=int32#3,232(<out=int32#6)
+# asm 2: movl <in10=%edx,232(<out=%edi)
+movl %edx,232(%edi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl <in11=int32#4,236(<out=int32#6)
+# asm 2: movl <in11=%ebx,236(<out=%edi)
+movl %ebx,236(%edi)
+
+# qhasm:   z12 = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>z12=int6464#1
+# asm 2: movdqa <z12_stack=576(%esp),>z12=%xmm0
+movdqa 576(%esp),%xmm0
+
+# qhasm:   z13 = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>z13=int6464#2
+# asm 2: movdqa <z13_stack=496(%esp),>z13=%xmm1
+movdqa 496(%esp),%xmm1
+
+# qhasm:   z14 = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>z14=int6464#3
+# asm 2: movdqa <z14_stack=400(%esp),>z14=%xmm2
+movdqa 400(%esp),%xmm2
+
+# qhasm:   z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>z15=int6464#4
+# asm 2: movdqa <z15_stack=384(%esp),>z15=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#1
+# asm 2: paddd <orig12=192(%esp),<z12=%xmm0
+paddd 192(%esp),%xmm0
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#2
+# asm 2: paddd <orig13=240(%esp),<z13=%xmm1
+paddd 240(%esp),%xmm1
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#3
+# asm 2: paddd <orig14=288(%esp),<z14=%xmm2
+paddd 288(%esp),%xmm2
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#4
+# asm 2: paddd <orig15=128(%esp),<z15=%xmm3
+paddd 128(%esp),%xmm3
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#1
+# asm 2: xorl 48(<m=%esi),<in12=%eax
+xorl 48(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#2
+# asm 2: xorl 52(<m=%esi),<in13=%ecx
+xorl 52(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#3
+# asm 2: xorl 56(<m=%esi),<in14=%edx
+xorl 56(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#4
+# asm 2: xorl 60(<m=%esi),<in15=%ebx
+xorl 60(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#1,48(<out=int32#6)
+# asm 2: movl <in12=%eax,48(<out=%edi)
+movl %eax,48(%edi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#2,52(<out=int32#6)
+# asm 2: movl <in13=%ecx,52(<out=%edi)
+movl %ecx,52(%edi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#3,56(<out=int32#6)
+# asm 2: movl <in14=%edx,56(<out=%edi)
+movl %edx,56(%edi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#4,60(<out=int32#6)
+# asm 2: movl <in15=%ebx,60(<out=%edi)
+movl %ebx,60(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int32#5),<in12=int32#1
+# asm 2: xorl 112(<m=%esi),<in12=%eax
+xorl 112(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int32#5),<in13=int32#2
+# asm 2: xorl 116(<m=%esi),<in13=%ecx
+xorl 116(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int32#5),<in14=int32#3
+# asm 2: xorl 120(<m=%esi),<in14=%edx
+xorl 120(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int32#5),<in15=int32#4
+# asm 2: xorl 124(<m=%esi),<in15=%ebx
+xorl 124(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl <in12=int32#1,112(<out=int32#6)
+# asm 2: movl <in12=%eax,112(<out=%edi)
+movl %eax,112(%edi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl <in13=int32#2,116(<out=int32#6)
+# asm 2: movl <in13=%ecx,116(<out=%edi)
+movl %ecx,116(%edi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl <in14=int32#3,120(<out=int32#6)
+# asm 2: movl <in14=%edx,120(<out=%edi)
+movl %edx,120(%edi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl <in15=int32#4,124(<out=int32#6)
+# asm 2: movl <in15=%ebx,124(<out=%edi)
+movl %ebx,124(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int32#5),<in12=int32#1
+# asm 2: xorl 176(<m=%esi),<in12=%eax
+xorl 176(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int32#5),<in13=int32#2
+# asm 2: xorl 180(<m=%esi),<in13=%ecx
+xorl 180(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int32#5),<in14=int32#3
+# asm 2: xorl 184(<m=%esi),<in14=%edx
+xorl 184(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int32#5),<in15=int32#4
+# asm 2: xorl 188(<m=%esi),<in15=%ebx
+xorl 188(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl <in12=int32#1,176(<out=int32#6)
+# asm 2: movl <in12=%eax,176(<out=%edi)
+movl %eax,176(%edi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl <in13=int32#2,180(<out=int32#6)
+# asm 2: movl <in13=%ecx,180(<out=%edi)
+movl %ecx,180(%edi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl <in14=int32#3,184(<out=int32#6)
+# asm 2: movl <in14=%edx,184(<out=%edi)
+movl %edx,184(%edi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl <in15=int32#4,188(<out=int32#6)
+# asm 2: movl <in15=%ebx,188(<out=%edi)
+movl %ebx,188(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int32#5),<in12=int32#1
+# asm 2: xorl 240(<m=%esi),<in12=%eax
+xorl 240(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int32#5),<in13=int32#2
+# asm 2: xorl 244(<m=%esi),<in13=%ecx
+xorl 244(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int32#5),<in14=int32#3
+# asm 2: xorl 248(<m=%esi),<in14=%edx
+xorl 248(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int32#5),<in15=int32#4
+# asm 2: xorl 252(<m=%esi),<in15=%ebx
+xorl 252(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl <in12=int32#1,240(<out=int32#6)
+# asm 2: movl <in12=%eax,240(<out=%edi)
+movl %eax,240(%edi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl <in13=int32#2,244(<out=int32#6)
+# asm 2: movl <in13=%ecx,244(<out=%edi)
+movl %ecx,244(%edi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl <in14=int32#3,248(<out=int32#6)
+# asm 2: movl <in14=%edx,248(<out=%edi)
+movl %edx,248(%edi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl <in15=int32#4,252(<out=int32#6)
+# asm 2: movl <in15=%ebx,252(<out=%edi)
+movl %ebx,252(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int32#1
+# asm 2: sub  $256,<bytes=%eax
+sub  $256,%eax
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int32#5
+# asm 2: add  $256,<m=%esi
+add  $256,%esi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int32#6
+# asm 2: add  $256,<out=%edi
+add  $256,%edi
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#1
+# asm 2: cmp  $0,<bytes=%eax
+cmp  $0,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: movl <out=int32#6,>ctarget=stack32#6
+# asm 2: movl <out=%edi,>ctarget=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     m = &tmp
+# asm 1: leal <tmp=stack512#1,>m=int32#5
+# asm 2: leal <tmp=640(%esp),>m=%esi
+leal 640(%esp),%esi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#3,>diag0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>diag0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#2,>diag1=int6464#2
+# asm 2: movdqa <x1=48(%esp),>diag1=%xmm1
+movdqa 48(%esp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#4,>diag2=int6464#3
+# asm 2: movdqa <x2=80(%esp),>diag2=%xmm2
+movdqa 80(%esp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#1,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%esp),>diag3=%xmm3
+movdqa 32(%esp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 12
+# asm 1: mov  $12,>i=int32#1
+# asm 2: mov  $12,>i=%eax
+mov  $12,%eax
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int32#1
+# asm 2: sub  $4,<i=%eax
+sub  $4,%eax
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#3,<diag0=int6464#1
+# asm 2: paddd <x0=64(%esp),<diag0=%xmm0
+paddd 64(%esp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#2,<diag1=int6464#2
+# asm 2: paddd <x1=48(%esp),<diag1=%xmm1
+paddd 48(%esp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#4,<diag2=int6464#3
+# asm 2: paddd <x2=80(%esp),<diag2=%xmm2
+paddd 80(%esp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#1,<diag3=int6464#4
+# asm 2: paddd <x3=32(%esp),<diag3=%xmm3
+paddd 32(%esp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int32#1
+# asm 2: movd   <diag0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int32#2
+# asm 2: movd   <diag1=%xmm1,>in12=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int32#3
+# asm 2: movd   <diag2=%xmm2,>in8=%edx
+movd   %xmm2,%edx
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int32#4
+# asm 2: movd   <diag3=%xmm3,>in4=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm: in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#2
+# asm 2: xorl 48(<m=%esi),<in12=%ecx
+xorl 48(%esi),%ecx
+
+# qhasm: in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#3
+# asm 2: xorl 32(<m=%esi),<in8=%edx
+xorl 32(%esi),%edx
+
+# qhasm: in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#4
+# asm 2: xorl 16(<m=%esi),<in4=%ebx
+xorl 16(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#2,48(<out=int32#6)
+# asm 2: movl <in12=%ecx,48(<out=%edi)
+movl %ecx,48(%edi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#3,32(<out=int32#6)
+# asm 2: movl <in8=%edx,32(<out=%edi)
+movl %edx,32(%edi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#4,16(<out=int32#6)
+# asm 2: movl <in4=%ebx,16(<out=%edi)
+movl %ebx,16(%edi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int32#1
+# asm 2: movd   <diag0=%xmm0,>in5=%eax
+movd   %xmm0,%eax
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int32#2
+# asm 2: movd   <diag1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int32#3
+# asm 2: movd   <diag2=%xmm2,>in13=%edx
+movd   %xmm2,%edx
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int32#4
+# asm 2: movd   <diag3=%xmm3,>in9=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#1
+# asm 2: xorl 20(<m=%esi),<in5=%eax
+xorl 20(%esi),%eax
+
+# qhasm: in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm: in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#3
+# asm 2: xorl 52(<m=%esi),<in13=%edx
+xorl 52(%esi),%edx
+
+# qhasm: in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#4
+# asm 2: xorl 36(<m=%esi),<in9=%ebx
+xorl 36(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#1,20(<out=int32#6)
+# asm 2: movl <in5=%eax,20(<out=%edi)
+movl %eax,20(%edi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#3,52(<out=int32#6)
+# asm 2: movl <in13=%edx,52(<out=%edi)
+movl %edx,52(%edi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#4,36(<out=int32#6)
+# asm 2: movl <in9=%ebx,36(<out=%edi)
+movl %ebx,36(%edi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int32#1
+# asm 2: movd   <diag0=%xmm0,>in10=%eax
+movd   %xmm0,%eax
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int32#2
+# asm 2: movd   <diag1=%xmm1,>in6=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int32#3
+# asm 2: movd   <diag2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int32#4
+# asm 2: movd   <diag3=%xmm3,>in14=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#1
+# asm 2: xorl 40(<m=%esi),<in10=%eax
+xorl 40(%esi),%eax
+
+# qhasm: in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#2
+# asm 2: xorl 24(<m=%esi),<in6=%ecx
+xorl 24(%esi),%ecx
+
+# qhasm: in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm: in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#4
+# asm 2: xorl 56(<m=%esi),<in14=%ebx
+xorl 56(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#1,40(<out=int32#6)
+# asm 2: movl <in10=%eax,40(<out=%edi)
+movl %eax,40(%edi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#2,24(<out=int32#6)
+# asm 2: movl <in6=%ecx,24(<out=%edi)
+movl %ecx,24(%edi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#4,56(<out=int32#6)
+# asm 2: movl <in14=%ebx,56(<out=%edi)
+movl %ebx,56(%edi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int32#1
+# asm 2: movd   <diag0=%xmm0,>in15=%eax
+movd   %xmm0,%eax
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int32#2
+# asm 2: movd   <diag1=%xmm1,>in11=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int32#3
+# asm 2: movd   <diag2=%xmm2,>in7=%edx
+movd   %xmm2,%edx
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int32#4
+# asm 2: movd   <diag3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#1
+# asm 2: xorl 60(<m=%esi),<in15=%eax
+xorl 60(%esi),%eax
+
+# qhasm: in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#2
+# asm 2: xorl 44(<m=%esi),<in11=%ecx
+xorl 44(%esi),%ecx
+
+# qhasm: in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#3
+# asm 2: xorl 28(<m=%esi),<in7=%edx
+xorl 28(%esi),%edx
+
+# qhasm: in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#1,60(<out=int32#6)
+# asm 2: movl <in15=%eax,60(<out=%edi)
+movl %eax,60(%edi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#2,44(<out=int32#6)
+# asm 2: movl <in11=%ecx,44(<out=%edi)
+movl %ecx,44(%edi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#3,28(<out=int32#6)
+# asm 2: movl <in7=%edx,28(<out=%edi)
+movl %edx,28(%edi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm:       out = ctarget
+# asm 1: movl <ctarget=stack32#6,>out=int32#6
+# asm 2: movl <ctarget=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     eax = eax_stack
+# asm 1: movl <eax_stack=stack32#1,>eax=int32#1
+# asm 2: movl <eax_stack=0(%esp),>eax=%eax
+movl 0(%esp),%eax
+
+# qhasm:     ebx = ebx_stack
+# asm 1: movl <ebx_stack=stack32#2,>ebx=int32#4
+# asm 2: movl <ebx_stack=4(%esp),>ebx=%ebx
+movl 4(%esp),%ebx
+
+# qhasm:     esi = esi_stack
+# asm 1: movl <esi_stack=stack32#3,>esi=int32#5
+# asm 2: movl <esi_stack=8(%esp),>esi=%esi
+movl 8(%esp),%esi
+
+# qhasm:     edi = edi_stack
+# asm 1: movl <edi_stack=stack32#4,>edi=int32#6
+# asm 2: movl <edi_stack=12(%esp),>edi=%edi
+movl 12(%esp),%edi
+
+# qhasm:     ebp = ebp_stack
+# asm 1: movl <ebp_stack=stack32#5,>ebp=int32#7
+# asm 2: movl <ebp_stack=16(%esp),>ebp=%ebp
+movl 16(%esp),%ebp
+
+# qhasm:     leave
+add %eax,%esp
+xor %eax,%eax
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int32#1
+# asm 2: sub  $64,<bytes=%eax
+sub  $64,%eax
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int32#6
+# asm 2: add  $64,<out=%edi
+add  $64,%edi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int32#5
+# asm 2: add  $64,<m=%esi
+add  $64,%esi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/salsa208/amd64_xmm6/api.h b/nacl/crypto_stream/salsa208/amd64_xmm6/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/amd64_xmm6/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa208/amd64_xmm6/implementors b/nacl/crypto_stream/salsa208/amd64_xmm6/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/amd64_xmm6/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa208/amd64_xmm6/stream.s b/nacl/crypto_stream/salsa208/amd64_xmm6/stream.s
new file mode 100644
index 00000000..f27411fe
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/amd64_xmm6/stream.s
@@ -0,0 +1,4823 @@
+
+# qhasm: int64 r11_caller
+
+# qhasm: int64 r12_caller
+
+# qhasm: int64 r13_caller
+
+# qhasm: int64 r14_caller
+
+# qhasm: int64 r15_caller
+
+# qhasm: int64 rbx_caller
+
+# qhasm: int64 rbp_caller
+
+# qhasm: caller r11_caller
+
+# qhasm: caller r12_caller
+
+# qhasm: caller r13_caller
+
+# qhasm: caller r14_caller
+
+# qhasm: caller r15_caller
+
+# qhasm: caller rbx_caller
+
+# qhasm: caller rbp_caller
+
+# qhasm: stack64 r11_stack
+
+# qhasm: stack64 r12_stack
+
+# qhasm: stack64 r13_stack
+
+# qhasm: stack64 r14_stack
+
+# qhasm: stack64 r15_stack
+
+# qhasm: stack64 rbx_stack
+
+# qhasm: stack64 rbp_stack
+
+# qhasm: int64 a
+
+# qhasm: int64 arg1
+
+# qhasm: int64 arg2
+
+# qhasm: int64 arg3
+
+# qhasm: int64 arg4
+
+# qhasm: int64 arg5
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: int64 k
+
+# qhasm: int64 kbits
+
+# qhasm: int64 iv
+
+# qhasm: int64 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int64 m
+
+# qhasm: int64 out
+
+# qhasm: int64 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: int6464 y0
+
+# qhasm: int6464 y1
+
+# qhasm: int6464 y2
+
+# qhasm: int6464 y3
+
+# qhasm: int6464 y4
+
+# qhasm: int6464 y5
+
+# qhasm: int6464 y6
+
+# qhasm: int6464 y7
+
+# qhasm: int6464 y8
+
+# qhasm: int6464 y9
+
+# qhasm: int6464 y10
+
+# qhasm: int6464 y11
+
+# qhasm: int6464 y12
+
+# qhasm: int6464 y13
+
+# qhasm: int6464 y14
+
+# qhasm: int6464 y15
+
+# qhasm: int6464 r0
+
+# qhasm: int6464 r1
+
+# qhasm: int6464 r2
+
+# qhasm: int6464 r3
+
+# qhasm: int6464 r4
+
+# qhasm: int6464 r5
+
+# qhasm: int6464 r6
+
+# qhasm: int6464 r7
+
+# qhasm: int6464 r8
+
+# qhasm: int6464 r9
+
+# qhasm: int6464 r10
+
+# qhasm: int6464 r11
+
+# qhasm: int6464 r12
+
+# qhasm: int6464 r13
+
+# qhasm: int6464 r14
+
+# qhasm: int6464 r15
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int64 in0
+
+# qhasm: int64 in1
+
+# qhasm: int64 in2
+
+# qhasm: int64 in3
+
+# qhasm: int64 in4
+
+# qhasm: int64 in5
+
+# qhasm: int64 in6
+
+# qhasm: int64 in7
+
+# qhasm: int64 in8
+
+# qhasm: int64 in9
+
+# qhasm: int64 in10
+
+# qhasm: int64 in11
+
+# qhasm: int64 in12
+
+# qhasm: int64 in13
+
+# qhasm: int64 in14
+
+# qhasm: int64 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: int64 ctarget
+
+# qhasm: stack64 bytes_backup
+
+# qhasm: enter crypto_stream_salsa208_amd64_xmm6
+.text
+.p2align 5
+.globl _crypto_stream_salsa208_amd64_xmm6
+.globl crypto_stream_salsa208_amd64_xmm6
+_crypto_stream_salsa208_amd64_xmm6:
+crypto_stream_salsa208_amd64_xmm6:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: bytes = arg2
+# asm 1: mov  <arg2=int64#2,>bytes=int64#6
+# asm 2: mov  <arg2=%rsi,>bytes=%r9
+mov  %rsi,%r9
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm: iv = arg3
+# asm 1: mov  <arg3=int64#3,>iv=int64#3
+# asm 2: mov  <arg3=%rdx,>iv=%rdx
+mov  %rdx,%rdx
+
+# qhasm: k = arg4
+# asm 1: mov  <arg4=int64#4,>k=int64#8
+# asm 2: mov  <arg4=%rcx,>k=%r10
+mov  %rcx,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int64#7
+# asm 2: mov  $0,>a=%rax
+mov  $0,%rax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: sub  <bytes=int64#6,<out=int64#1
+# asm 2: sub  <bytes=%r9,<out=%rdi
+sub  %r9,%rdi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa208_amd64_xmm6_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa208_amd64_xmm6_xor
+.globl crypto_stream_salsa208_amd64_xmm6_xor
+_crypto_stream_salsa208_amd64_xmm6_xor:
+crypto_stream_salsa208_amd64_xmm6_xor:
+mov %rsp,%r11
+and $31,%r11
+add $480,%r11
+sub %r11,%rsp
+
+# qhasm: r11_stack = r11_caller
+# asm 1: movq <r11_caller=int64#9,>r11_stack=stack64#1
+# asm 2: movq <r11_caller=%r11,>r11_stack=352(%rsp)
+movq %r11,352(%rsp)
+
+# qhasm: r12_stack = r12_caller
+# asm 1: movq <r12_caller=int64#10,>r12_stack=stack64#2
+# asm 2: movq <r12_caller=%r12,>r12_stack=360(%rsp)
+movq %r12,360(%rsp)
+
+# qhasm: r13_stack = r13_caller
+# asm 1: movq <r13_caller=int64#11,>r13_stack=stack64#3
+# asm 2: movq <r13_caller=%r13,>r13_stack=368(%rsp)
+movq %r13,368(%rsp)
+
+# qhasm: r14_stack = r14_caller
+# asm 1: movq <r14_caller=int64#12,>r14_stack=stack64#4
+# asm 2: movq <r14_caller=%r14,>r14_stack=376(%rsp)
+movq %r14,376(%rsp)
+
+# qhasm: r15_stack = r15_caller
+# asm 1: movq <r15_caller=int64#13,>r15_stack=stack64#5
+# asm 2: movq <r15_caller=%r15,>r15_stack=384(%rsp)
+movq %r15,384(%rsp)
+
+# qhasm: rbx_stack = rbx_caller
+# asm 1: movq <rbx_caller=int64#14,>rbx_stack=stack64#6
+# asm 2: movq <rbx_caller=%rbx,>rbx_stack=392(%rsp)
+movq %rbx,392(%rsp)
+
+# qhasm: rbp_stack = rbp_caller
+# asm 1: movq <rbp_caller=int64#15,>rbp_stack=stack64#7
+# asm 2: movq <rbp_caller=%rbp,>rbp_stack=400(%rsp)
+movq %rbp,400(%rsp)
+
+# qhasm: out = arg1
+# asm 1: mov  <arg1=int64#1,>out=int64#1
+# asm 2: mov  <arg1=%rdi,>out=%rdi
+mov  %rdi,%rdi
+
+# qhasm: m = arg2
+# asm 1: mov  <arg2=int64#2,>m=int64#2
+# asm 2: mov  <arg2=%rsi,>m=%rsi
+mov  %rsi,%rsi
+
+# qhasm: bytes = arg3
+# asm 1: mov  <arg3=int64#3,>bytes=int64#6
+# asm 2: mov  <arg3=%rdx,>bytes=%r9
+mov  %rdx,%r9
+
+# qhasm: iv = arg4
+# asm 1: mov  <arg4=int64#4,>iv=int64#3
+# asm 2: mov  <arg4=%rcx,>iv=%rdx
+mov  %rcx,%rdx
+
+# qhasm: k = arg5
+# asm 1: mov  <arg5=int64#5,>k=int64#8
+# asm 2: mov  <arg5=%r8,>k=%r10
+mov  %r8,%r10
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl   20(<k=int64#8),>in12=int64#4d
+# asm 2: movl   20(<k=%r10),>in12=%ecx
+movl   20(%r10),%ecx
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl   0(<k=int64#8),>in1=int64#5d
+# asm 2: movl   0(<k=%r10),>in1=%r8d
+movl   0(%r10),%r8d
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl   0(<iv=int64#3),>in6=int64#7d
+# asm 2: movl   0(<iv=%rdx),>in6=%eax
+movl   0(%rdx),%eax
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl   16(<k=int64#8),>in11=int64#9d
+# asm 2: movl   16(<k=%r10),>in11=%r11d
+movl   16(%r10),%r11d
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int64#4d,>x1=stack128#1
+# asm 2: movl <in12=%ecx,>x1=0(%rsp)
+movl %ecx,0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int64#5d,4+<x1=stack128#1
+# asm 2: movl <in1=%r8d,4+<x1=0(%rsp)
+movl %r8d,4+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int64#7d,8+<x1=stack128#1
+# asm 2: movl <in6=%eax,8+<x1=0(%rsp)
+movl %eax,8+0(%rsp)
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int64#9d,12+<x1=stack128#1
+# asm 2: movl <in11=%r11d,12+<x1=0(%rsp)
+movl %r11d,12+0(%rsp)
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int64#4
+# asm 2: mov  $0,>in8=%rcx
+mov  $0,%rcx
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl   24(<k=int64#8),>in13=int64#5d
+# asm 2: movl   24(<k=%r10),>in13=%r8d
+movl   24(%r10),%r8d
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl   4(<k=int64#8),>in2=int64#7d
+# asm 2: movl   4(<k=%r10),>in2=%eax
+movl   4(%r10),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl   4(<iv=int64#3),>in7=int64#3d
+# asm 2: movl   4(<iv=%rdx),>in7=%edx
+movl   4(%rdx),%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int64#5d,4+<x2=stack128#2
+# asm 2: movl <in13=%r8d,4+<x2=16(%rsp)
+movl %r8d,4+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int64#7d,8+<x2=stack128#2
+# asm 2: movl <in2=%eax,8+<x2=16(%rsp)
+movl %eax,8+16(%rsp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int64#3d,12+<x2=stack128#2
+# asm 2: movl <in7=%edx,12+<x2=16(%rsp)
+movl %edx,12+16(%rsp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl   12(<k=int64#8),>in4=int64#3d
+# asm 2: movl   12(<k=%r10),>in4=%edx
+movl   12(%r10),%edx
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int64#4
+# asm 2: mov  $0,>in9=%rcx
+mov  $0,%rcx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl   28(<k=int64#8),>in14=int64#5d
+# asm 2: movl   28(<k=%r10),>in14=%r8d
+movl   28(%r10),%r8d
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl   8(<k=int64#8),>in3=int64#7d
+# asm 2: movl   8(<k=%r10),>in3=%eax
+movl   8(%r10),%eax
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int64#3d,>x3=stack128#3
+# asm 2: movl <in4=%edx,>x3=32(%rsp)
+movl %edx,32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int64#5d,8+<x3=stack128#3
+# asm 2: movl <in14=%r8d,8+<x3=32(%rsp)
+movl %r8d,8+32(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int64#7d,12+<x3=stack128#3
+# asm 2: movl <in3=%eax,12+<x3=32(%rsp)
+movl %eax,12+32(%rsp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int64#3
+# asm 2: mov  $1634760805,>in0=%rdx
+mov  $1634760805,%rdx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int64#4
+# asm 2: mov  $857760878,>in5=%rcx
+mov  $857760878,%rcx
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int64#5
+# asm 2: mov  $2036477234,>in10=%r8
+mov  $2036477234,%r8
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int64#7
+# asm 2: mov  $1797285236,>in15=%rax
+mov  $1797285236,%rax
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int64#3d,>x0=stack128#4
+# asm 2: movl <in0=%edx,>x0=48(%rsp)
+movl %edx,48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int64#4d,4+<x0=stack128#4
+# asm 2: movl <in5=%ecx,4+<x0=48(%rsp)
+movl %ecx,4+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int64#5d,8+<x0=stack128#4
+# asm 2: movl <in10=%r8d,8+<x0=48(%rsp)
+movl %r8d,8+48(%rsp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int64#7d,12+<x0=stack128#4
+# asm 2: movl <in15=%eax,12+<x0=48(%rsp)
+movl %eax,12+48(%rsp)
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#4,>z0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>z0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=64(%rsp)
+movdqa %xmm1,64(%rsp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=80(%rsp)
+movdqa %xmm2,80(%rsp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=96(%rsp)
+movdqa %xmm3,96(%rsp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=112(%rsp)
+movdqa %xmm0,112(%rsp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#1,>z1=int6464#1
+# asm 2: movdqa <x1=0(%rsp),>z1=%xmm0
+movdqa 0(%rsp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=128(%rsp)
+movdqa %xmm1,128(%rsp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=144(%rsp)
+movdqa %xmm2,144(%rsp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=160(%rsp)
+movdqa %xmm3,160(%rsp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=176(%rsp)
+movdqa %xmm0,176(%rsp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#2,>z2=int6464#1
+# asm 2: movdqa <x2=16(%rsp),>z2=%xmm0
+movdqa 16(%rsp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=192(%rsp)
+movdqa %xmm1,192(%rsp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=208(%rsp)
+movdqa %xmm2,208(%rsp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=224(%rsp)
+movdqa %xmm0,224(%rsp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#3,>z3=int6464#1
+# asm 2: movdqa <x3=32(%rsp),>z3=%xmm0
+movdqa 32(%rsp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=240(%rsp)
+movdqa %xmm1,240(%rsp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=256(%rsp)
+movdqa %xmm2,256(%rsp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=272(%rsp)
+movdqa %xmm0,272(%rsp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#3d
+# asm 2: movl <x2=16(%rsp),>in8=%edx
+movl 16(%rsp),%edx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#4d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%ecx
+movl 4+32(%rsp),%ecx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int64#3d,>orig8=stack128#19
+# asm 2: movl <in8=%edx,>orig8=288(%rsp)
+movl %edx,288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int64#4d,>orig9=stack128#20
+# asm 2: movl <in9=%ecx,>orig9=304(%rsp)
+movl %ecx,304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int64#3d,4+<orig8=stack128#19
+# asm 2: movl <in8=%edx,4+<orig8=288(%rsp)
+movl %edx,4+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,4+<orig9=304(%rsp)
+movl %ecx,4+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int64#3d,8+<orig8=stack128#19
+# asm 2: movl <in8=%edx,8+<orig8=288(%rsp)
+movl %edx,8+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int64#4d,8+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,8+<orig9=304(%rsp)
+movl %ecx,8+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int64#3d,12+<orig8=stack128#19
+# asm 2: movl <in8=%edx,12+<orig8=288(%rsp)
+movl %edx,12+288(%rsp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int64#4d,12+<orig9=stack128#20
+# asm 2: movl <in9=%ecx,12+<orig9=304(%rsp)
+movl %ecx,12+304(%rsp)
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#3
+# asm 2: add  $1,<in8=%rdx
+add  $1,%rdx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#4
+# asm 2: shl  $32,<in9=%rcx
+shl  $32,%rcx
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#4,<in8=int64#3
+# asm 2: add  <in9=%rcx,<in8=%rdx
+add  %rcx,%rdx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#3,>in9=int64#4
+# asm 2: mov  <in8=%rdx,>in9=%rcx
+mov  %rdx,%rcx
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#4
+# asm 2: shr  $32,<in9=%rcx
+shr  $32,%rcx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#3d,>x2=stack128#2
+# asm 2: movl <in8=%edx,>x2=16(%rsp)
+movl %edx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#4d,4+<x3=stack128#3
+# asm 2: movl <in9=%ecx,4+<x3=32(%rsp)
+movl %ecx,4+32(%rsp)
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: i = 8
+# asm 1: mov  $8,>i=int64#3
+# asm 2: mov  $8,>i=%rdx
+mov  $8,%rdx
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=64(%rsp),>z5=%xmm0
+movdqa 64(%rsp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=80(%rsp),>z10=%xmm1
+movdqa 80(%rsp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=96(%rsp),>z15=%xmm2
+movdqa 96(%rsp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=256(%rsp),>z14=%xmm3
+movdqa 256(%rsp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=272(%rsp),>z3=%xmm4
+movdqa 272(%rsp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=128(%rsp),>z6=%xmm5
+movdqa 128(%rsp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=144(%rsp),>z11=%xmm6
+movdqa 144(%rsp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=176(%rsp),>z1=%xmm7
+movdqa 176(%rsp),%xmm7
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#9
+# asm 2: movdqa <orig7=192(%rsp),>z7=%xmm8
+movdqa 192(%rsp),%xmm8
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#10
+# asm 2: movdqa <orig13=208(%rsp),>z13=%xmm9
+movdqa 208(%rsp),%xmm9
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#11
+# asm 2: movdqa <orig2=224(%rsp),>z2=%xmm10
+movdqa 224(%rsp),%xmm10
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#12
+# asm 2: movdqa <orig9=304(%rsp),>z9=%xmm11
+movdqa 304(%rsp),%xmm11
+
+# qhasm:   z0 = orig0
+# asm 1: movdqa <orig0=stack128#8,>z0=int6464#13
+# asm 2: movdqa <orig0=112(%rsp),>z0=%xmm12
+movdqa 112(%rsp),%xmm12
+
+# qhasm:   z12 = orig12
+# asm 1: movdqa <orig12=stack128#11,>z12=int6464#14
+# asm 2: movdqa <orig12=160(%rsp),>z12=%xmm13
+movdqa 160(%rsp),%xmm13
+
+# qhasm:   z4 = orig4
+# asm 1: movdqa <orig4=stack128#16,>z4=int6464#15
+# asm 2: movdqa <orig4=240(%rsp),>z4=%xmm14
+movdqa 240(%rsp),%xmm14
+
+# qhasm:   z8 = orig8
+# asm 1: movdqa <orig8=stack128#19,>z8=int6464#16
+# asm 2: movdqa <orig8=288(%rsp),>z8=%xmm15
+movdqa 288(%rsp),%xmm15
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                y4 = z12
+# asm 1: movdqa <z12=int6464#14,>y4=int6464#2
+# asm 2: movdqa <z12=%xmm13,>y4=%xmm1
+movdqa %xmm13,%xmm1
+
+# qhasm: uint32323232   y4 += z0
+# asm 1: paddd <z0=int6464#13,<y4=int6464#2
+# asm 2: paddd <z0=%xmm12,<y4=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y4 <<= 7
+# asm 1: pslld $7,<y4=int6464#2
+# asm 2: pslld $7,<y4=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232   r4 >>= 25
+# asm 1: psrld $25,<r4=int6464#3
+# asm 2: psrld $25,<r4=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                                y9 = z1
+# asm 1: movdqa <z1=int6464#8,>y9=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y9=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232                   y9 += z5
+# asm 1: paddd <z5=int6464#1,<y9=int6464#2
+# asm 2: paddd <z5=%xmm0,<y9=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                                r9 = y9
+# asm 1: movdqa <y9=int6464#2,>r9=int6464#3
+# asm 2: movdqa <y9=%xmm1,>r9=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y9 <<= 7
+# asm 1: pslld $7,<y9=int6464#2
+# asm 2: pslld $7,<y9=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#2,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm1,<z9=%xmm11
+pxor  %xmm1,%xmm11
+
+# qhasm: uint32323232                   r9 >>= 25
+# asm 1: psrld $25,<r9=int6464#3
+# asm 2: psrld $25,<r9=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#3,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm2,<z9=%xmm11
+pxor  %xmm2,%xmm11
+
+# qhasm:                y8 = z0
+# asm 1: movdqa <z0=int6464#13,>y8=int6464#2
+# asm 2: movdqa <z0=%xmm12,>y8=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232   y8 += z4
+# asm 1: paddd <z4=int6464#15,<y8=int6464#2
+# asm 2: paddd <z4=%xmm14,<y8=%xmm1
+paddd %xmm14,%xmm1
+
+# qhasm:                r8 = y8
+# asm 1: movdqa <y8=int6464#2,>r8=int6464#3
+# asm 2: movdqa <y8=%xmm1,>r8=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#2
+# asm 2: pslld $9,<y8=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z8 ^= y8
+# asm 1: pxor  <y8=int6464#2,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm1,<z8=%xmm15
+pxor  %xmm1,%xmm15
+
+# qhasm: uint32323232   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#3
+# asm 2: psrld $23,<r8=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z8 ^= r8
+# asm 1: pxor  <r8=int6464#3,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm2,<z8=%xmm15
+pxor  %xmm2,%xmm15
+
+# qhasm:                                y13 = z5
+# asm 1: movdqa <z5=int6464#1,>y13=int6464#2
+# asm 2: movdqa <z5=%xmm0,>y13=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                   y13 += z9
+# asm 1: paddd <z9=int6464#12,<y13=int6464#2
+# asm 2: paddd <z9=%xmm11,<y13=%xmm1
+paddd %xmm11,%xmm1
+
+# qhasm:                                r13 = y13
+# asm 1: movdqa <y13=int6464#2,>r13=int6464#3
+# asm 2: movdqa <y13=%xmm1,>r13=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#2
+# asm 2: pslld $9,<y13=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#2,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm1,<z13=%xmm9
+pxor  %xmm1,%xmm9
+
+# qhasm: uint32323232                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#3
+# asm 2: psrld $23,<r13=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#3,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm2,<z13=%xmm9
+pxor  %xmm2,%xmm9
+
+# qhasm:                y12 = z4
+# asm 1: movdqa <z4=int6464#15,>y12=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y12=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232   y12 += z8
+# asm 1: paddd <z8=int6464#16,<y12=int6464#2
+# asm 2: paddd <z8=%xmm15,<y12=%xmm1
+paddd %xmm15,%xmm1
+
+# qhasm:                r12 = y12
+# asm 1: movdqa <y12=int6464#2,>r12=int6464#3
+# asm 2: movdqa <y12=%xmm1,>r12=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y12 <<= 13
+# asm 1: pslld $13,<y12=int6464#2
+# asm 2: pslld $13,<y12=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z12 ^= y12
+# asm 1: pxor  <y12=int6464#2,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm1,<z12=%xmm13
+pxor  %xmm1,%xmm13
+
+# qhasm: uint32323232   r12 >>= 19
+# asm 1: psrld $19,<r12=int6464#3
+# asm 2: psrld $19,<r12=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z12 ^= r12
+# asm 1: pxor  <r12=int6464#3,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm2,<z12=%xmm13
+pxor  %xmm2,%xmm13
+
+# qhasm:                                y1 = z9
+# asm 1: movdqa <z9=int6464#12,>y1=int6464#2
+# asm 2: movdqa <z9=%xmm11,>y1=%xmm1
+movdqa %xmm11,%xmm1
+
+# qhasm: uint32323232                   y1 += z13
+# asm 1: paddd <z13=int6464#10,<y1=int6464#2
+# asm 2: paddd <z13=%xmm9,<y1=%xmm1
+paddd %xmm9,%xmm1
+
+# qhasm:                                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#3
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y1 <<= 13
+# asm 1: pslld $13,<y1=int6464#2
+# asm 2: pslld $13,<y1=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232                   r1 >>= 19
+# asm 1: psrld $19,<r1=int6464#3
+# asm 2: psrld $19,<r1=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z1 ^= r1
+# asm 1: pxor  <r1=int6464#3,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm2,<z1=%xmm7
+pxor  %xmm2,%xmm7
+
+# qhasm:                y0 = z8
+# asm 1: movdqa <z8=int6464#16,>y0=int6464#2
+# asm 2: movdqa <z8=%xmm15,>y0=%xmm1
+movdqa %xmm15,%xmm1
+
+# qhasm: uint32323232   y0 += z12
+# asm 1: paddd <z12=int6464#14,<y0=int6464#2
+# asm 2: paddd <z12=%xmm13,<y0=%xmm1
+paddd %xmm13,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#13
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm12
+pxor  %xmm1,%xmm12
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#13
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#13,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm12,>z0_stack=320(%rsp)
+movdqa %xmm12,320(%rsp)
+
+# qhasm:                                y5 = z13
+# asm 1: movdqa <z13=int6464#10,>y5=int6464#3
+# asm 2: movdqa <z13=%xmm9,>y5=%xmm2
+movdqa %xmm9,%xmm2
+
+# qhasm: uint32323232                   y5 += z1
+# asm 1: paddd <z1=int6464#8,<y5=int6464#3
+# asm 2: paddd <z1=%xmm7,<y5=%xmm2
+paddd %xmm7,%xmm2
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#3,>r5=int6464#13
+# asm 2: movdqa <y5=%xmm2,>r5=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#3
+# asm 2: pslld $18,<y5=%xmm2
+pslld $18,%xmm2
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#3,<z5=int6464#1
+# asm 2: pxor  <y5=%xmm2,<z5=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#13
+# asm 2: psrld $14,<r5=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#13,<z5=int6464#1
+# asm 2: pxor  <r5=%xmm12,<z5=%xmm0
+pxor  %xmm12,%xmm0
+
+# qhasm:                                                y14 = z6
+# asm 1: movdqa <z6=int6464#6,>y14=int6464#3
+# asm 2: movdqa <z6=%xmm5,>y14=%xmm2
+movdqa %xmm5,%xmm2
+
+# qhasm: uint32323232                                   y14 += z10
+# asm 1: paddd <z10=int6464#2,<y14=int6464#3
+# asm 2: paddd <z10=%xmm1,<y14=%xmm2
+paddd %xmm1,%xmm2
+
+# qhasm:                                                r14 = y14
+# asm 1: movdqa <y14=int6464#3,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm2,>r14=%xmm12
+movdqa %xmm2,%xmm12
+
+# qhasm: uint32323232                                   y14 <<= 7
+# asm 1: pslld $7,<y14=int6464#3
+# asm 2: pslld $7,<y14=%xmm2
+pslld $7,%xmm2
+
+# qhasm:                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#3,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm2,<z14=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                   r14 >>= 25
+# asm 1: psrld $25,<r14=int6464#13
+# asm 2: psrld $25,<r14=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm0,>z5_stack=336(%rsp)
+movdqa %xmm0,336(%rsp)
+
+# qhasm:                                                                y3 = z11
+# asm 1: movdqa <z11=int6464#7,>y3=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y3=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                                   y3 += z15
+# asm 1: paddd <z15=int6464#3,<y3=int6464#1
+# asm 2: paddd <z15=%xmm2,<y3=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r3 = y3
+# asm 1: movdqa <y3=int6464#1,>r3=int6464#13
+# asm 2: movdqa <y3=%xmm0,>r3=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y3 <<= 7
+# asm 1: pslld $7,<y3=int6464#1
+# asm 2: pslld $7,<y3=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z3 ^= y3
+# asm 1: pxor  <y3=int6464#1,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm0,<z3=%xmm4
+pxor  %xmm0,%xmm4
+
+# qhasm: uint32323232                                                   r3 >>= 25
+# asm 1: psrld $25,<r3=int6464#13
+# asm 2: psrld $25,<r3=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z3 ^= r3
+# asm 1: pxor  <r3=int6464#13,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm12,<z3=%xmm4
+pxor  %xmm12,%xmm4
+
+# qhasm:                                                y2 = z10
+# asm 1: movdqa <z10=int6464#2,>y2=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y2=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y2 += z14
+# asm 1: paddd <z14=int6464#4,<y2=int6464#1
+# asm 2: paddd <z14=%xmm3,<y2=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                r2 = y2
+# asm 1: movdqa <y2=int6464#1,>r2=int6464#13
+# asm 2: movdqa <y2=%xmm0,>r2=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#1
+# asm 2: pslld $9,<y2=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z2 ^= y2
+# asm 1: pxor  <y2=int6464#1,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm0,<z2=%xmm10
+pxor  %xmm0,%xmm10
+
+# qhasm: uint32323232                                   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#13
+# asm 2: psrld $23,<r2=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z2 ^= r2
+# asm 1: pxor  <r2=int6464#13,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm12,<z2=%xmm10
+pxor  %xmm12,%xmm10
+
+# qhasm:                                                                y7 = z15
+# asm 1: movdqa <z15=int6464#3,>y7=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y7=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y7 += z3
+# asm 1: paddd <z3=int6464#5,<y7=int6464#1
+# asm 2: paddd <z3=%xmm4,<y7=%xmm0
+paddd %xmm4,%xmm0
+
+# qhasm:                                                                r7 = y7
+# asm 1: movdqa <y7=int6464#1,>r7=int6464#13
+# asm 2: movdqa <y7=%xmm0,>r7=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#1
+# asm 2: pslld $9,<y7=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#1,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm0,<z7=%xmm8
+pxor  %xmm0,%xmm8
+
+# qhasm: uint32323232                                                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#13
+# asm 2: psrld $23,<r7=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#13,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm12,<z7=%xmm8
+pxor  %xmm12,%xmm8
+
+# qhasm:                                                y6 = z14
+# asm 1: movdqa <z14=int6464#4,>y6=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y6=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                   y6 += z2
+# asm 1: paddd <z2=int6464#11,<y6=int6464#1
+# asm 2: paddd <z2=%xmm10,<y6=%xmm0
+paddd %xmm10,%xmm0
+
+# qhasm:                                                r6 = y6
+# asm 1: movdqa <y6=int6464#1,>r6=int6464#13
+# asm 2: movdqa <y6=%xmm0,>r6=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y6 <<= 13
+# asm 1: pslld $13,<y6=int6464#1
+# asm 2: pslld $13,<y6=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#1,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm0,<z6=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                   r6 >>= 19
+# asm 1: psrld $19,<r6=int6464#13
+# asm 2: psrld $19,<r6=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#13,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm12,<z6=%xmm5
+pxor  %xmm12,%xmm5
+
+# qhasm:                                                                y11 = z3
+# asm 1: movdqa <z3=int6464#5,>y11=int6464#1
+# asm 2: movdqa <z3=%xmm4,>y11=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                   y11 += z7
+# asm 1: paddd <z7=int6464#9,<y11=int6464#1
+# asm 2: paddd <z7=%xmm8,<y11=%xmm0
+paddd %xmm8,%xmm0
+
+# qhasm:                                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#13
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y11 <<= 13
+# asm 1: pslld $13,<y11=int6464#1
+# asm 2: pslld $13,<y11=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                                   r11 >>= 19
+# asm 1: psrld $19,<r11=int6464#13
+# asm 2: psrld $19,<r11=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#13,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm12,<z11=%xmm6
+pxor  %xmm12,%xmm6
+
+# qhasm:                                                y10 = z2
+# asm 1: movdqa <z2=int6464#11,>y10=int6464#1
+# asm 2: movdqa <z2=%xmm10,>y10=%xmm0
+movdqa %xmm10,%xmm0
+
+# qhasm: uint32323232                                   y10 += z6
+# asm 1: paddd <z6=int6464#6,<y10=int6464#1
+# asm 2: paddd <z6=%xmm5,<y10=%xmm0
+paddd %xmm5,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#1
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm0
+movdqa 320(%rsp),%xmm0
+
+# qhasm:                                                z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#21
+# asm 2: movdqa <z10=%xmm1,>z10_stack=320(%rsp)
+movdqa %xmm1,320(%rsp)
+
+# qhasm:                y1 = z3
+# asm 1: movdqa <z3=int6464#5,>y1=int6464#2
+# asm 2: movdqa <z3=%xmm4,>y1=%xmm1
+movdqa %xmm4,%xmm1
+
+# qhasm: uint32323232   y1 += z0
+# asm 1: paddd <z0=int6464#1,<y1=int6464#2
+# asm 2: paddd <z0=%xmm0,<y1=%xmm1
+paddd %xmm0,%xmm1
+
+# qhasm:                r1 = y1
+# asm 1: movdqa <y1=int6464#2,>r1=int6464#13
+# asm 2: movdqa <y1=%xmm1,>r1=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232   y1 <<= 7
+# asm 1: pslld $7,<y1=int6464#2
+# asm 2: pslld $7,<y1=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                z1 ^= y1
+# asm 1: pxor  <y1=int6464#2,<z1=int6464#8
+# asm 2: pxor  <y1=%xmm1,<z1=%xmm7
+pxor  %xmm1,%xmm7
+
+# qhasm: uint32323232   r1 >>= 25
+# asm 1: psrld $25,<r1=int6464#13
+# asm 2: psrld $25,<r1=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                z1 ^= r1
+# asm 1: pxor  <r1=int6464#13,<z1=int6464#8
+# asm 2: pxor  <r1=%xmm12,<z1=%xmm7
+pxor  %xmm12,%xmm7
+
+# qhasm:                                                                y15 = z7
+# asm 1: movdqa <z7=int6464#9,>y15=int6464#2
+# asm 2: movdqa <z7=%xmm8,>y15=%xmm1
+movdqa %xmm8,%xmm1
+
+# qhasm: uint32323232                                                   y15 += z11
+# asm 1: paddd <z11=int6464#7,<y15=int6464#2
+# asm 2: paddd <z11=%xmm6,<y15=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#2,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm1,>r15=%xmm12
+movdqa %xmm1,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#2
+# asm 2: pslld $18,<y15=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#2,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm1,<z15=%xmm2
+pxor  %xmm1,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#13
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm12
+movdqa 336(%rsp),%xmm12
+
+# qhasm:                                                                z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#22
+# asm 2: movdqa <z15=%xmm2,>z15_stack=336(%rsp)
+movdqa %xmm2,336(%rsp)
+
+# qhasm:                                y6 = z4
+# asm 1: movdqa <z4=int6464#15,>y6=int6464#2
+# asm 2: movdqa <z4=%xmm14,>y6=%xmm1
+movdqa %xmm14,%xmm1
+
+# qhasm: uint32323232                   y6 += z5
+# asm 1: paddd <z5=int6464#13,<y6=int6464#2
+# asm 2: paddd <z5=%xmm12,<y6=%xmm1
+paddd %xmm12,%xmm1
+
+# qhasm:                                r6 = y6
+# asm 1: movdqa <y6=int6464#2,>r6=int6464#3
+# asm 2: movdqa <y6=%xmm1,>r6=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y6 <<= 7
+# asm 1: pslld $7,<y6=int6464#2
+# asm 2: pslld $7,<y6=%xmm1
+pslld $7,%xmm1
+
+# qhasm:                                z6 ^= y6
+# asm 1: pxor  <y6=int6464#2,<z6=int6464#6
+# asm 2: pxor  <y6=%xmm1,<z6=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm: uint32323232                   r6 >>= 25
+# asm 1: psrld $25,<r6=int6464#3
+# asm 2: psrld $25,<r6=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                z6 ^= r6
+# asm 1: pxor  <r6=int6464#3,<z6=int6464#6
+# asm 2: pxor  <r6=%xmm2,<z6=%xmm5
+pxor  %xmm2,%xmm5
+
+# qhasm:                y2 = z0
+# asm 1: movdqa <z0=int6464#1,>y2=int6464#2
+# asm 2: movdqa <z0=%xmm0,>y2=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232   y2 += z1
+# asm 1: paddd <z1=int6464#8,<y2=int6464#2
+# asm 2: paddd <z1=%xmm7,<y2=%xmm1
+paddd %xmm7,%xmm1
+
+# qhasm:                r2 = y2
+# asm 1: movdqa <y2=int6464#2,>r2=int6464#3
+# asm 2: movdqa <y2=%xmm1,>r2=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y2 <<= 9
+# asm 1: pslld $9,<y2=int6464#2
+# asm 2: pslld $9,<y2=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                z2 ^= y2
+# asm 1: pxor  <y2=int6464#2,<z2=int6464#11
+# asm 2: pxor  <y2=%xmm1,<z2=%xmm10
+pxor  %xmm1,%xmm10
+
+# qhasm: uint32323232   r2 >>= 23
+# asm 1: psrld $23,<r2=int6464#3
+# asm 2: psrld $23,<r2=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                z2 ^= r2
+# asm 1: pxor  <r2=int6464#3,<z2=int6464#11
+# asm 2: pxor  <r2=%xmm2,<z2=%xmm10
+pxor  %xmm2,%xmm10
+
+# qhasm:                                y7 = z5
+# asm 1: movdqa <z5=int6464#13,>y7=int6464#2
+# asm 2: movdqa <z5=%xmm12,>y7=%xmm1
+movdqa %xmm12,%xmm1
+
+# qhasm: uint32323232                   y7 += z6
+# asm 1: paddd <z6=int6464#6,<y7=int6464#2
+# asm 2: paddd <z6=%xmm5,<y7=%xmm1
+paddd %xmm5,%xmm1
+
+# qhasm:                                r7 = y7
+# asm 1: movdqa <y7=int6464#2,>r7=int6464#3
+# asm 2: movdqa <y7=%xmm1,>r7=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y7 <<= 9
+# asm 1: pslld $9,<y7=int6464#2
+# asm 2: pslld $9,<y7=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                z7 ^= y7
+# asm 1: pxor  <y7=int6464#2,<z7=int6464#9
+# asm 2: pxor  <y7=%xmm1,<z7=%xmm8
+pxor  %xmm1,%xmm8
+
+# qhasm: uint32323232                   r7 >>= 23
+# asm 1: psrld $23,<r7=int6464#3
+# asm 2: psrld $23,<r7=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                z7 ^= r7
+# asm 1: pxor  <r7=int6464#3,<z7=int6464#9
+# asm 2: pxor  <r7=%xmm2,<z7=%xmm8
+pxor  %xmm2,%xmm8
+
+# qhasm:                y3 = z1
+# asm 1: movdqa <z1=int6464#8,>y3=int6464#2
+# asm 2: movdqa <z1=%xmm7,>y3=%xmm1
+movdqa %xmm7,%xmm1
+
+# qhasm: uint32323232   y3 += z2
+# asm 1: paddd <z2=int6464#11,<y3=int6464#2
+# asm 2: paddd <z2=%xmm10,<y3=%xmm1
+paddd %xmm10,%xmm1
+
+# qhasm:                r3 = y3
+# asm 1: movdqa <y3=int6464#2,>r3=int6464#3
+# asm 2: movdqa <y3=%xmm1,>r3=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y3 <<= 13
+# asm 1: pslld $13,<y3=int6464#2
+# asm 2: pslld $13,<y3=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                z3 ^= y3
+# asm 1: pxor  <y3=int6464#2,<z3=int6464#5
+# asm 2: pxor  <y3=%xmm1,<z3=%xmm4
+pxor  %xmm1,%xmm4
+
+# qhasm: uint32323232   r3 >>= 19
+# asm 1: psrld $19,<r3=int6464#3
+# asm 2: psrld $19,<r3=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                z3 ^= r3
+# asm 1: pxor  <r3=int6464#3,<z3=int6464#5
+# asm 2: pxor  <r3=%xmm2,<z3=%xmm4
+pxor  %xmm2,%xmm4
+
+# qhasm:                                y4 = z6
+# asm 1: movdqa <z6=int6464#6,>y4=int6464#2
+# asm 2: movdqa <z6=%xmm5,>y4=%xmm1
+movdqa %xmm5,%xmm1
+
+# qhasm: uint32323232                   y4 += z7
+# asm 1: paddd <z7=int6464#9,<y4=int6464#2
+# asm 2: paddd <z7=%xmm8,<y4=%xmm1
+paddd %xmm8,%xmm1
+
+# qhasm:                                r4 = y4
+# asm 1: movdqa <y4=int6464#2,>r4=int6464#3
+# asm 2: movdqa <y4=%xmm1,>r4=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232                   y4 <<= 13
+# asm 1: pslld $13,<y4=int6464#2
+# asm 2: pslld $13,<y4=%xmm1
+pslld $13,%xmm1
+
+# qhasm:                                z4 ^= y4
+# asm 1: pxor  <y4=int6464#2,<z4=int6464#15
+# asm 2: pxor  <y4=%xmm1,<z4=%xmm14
+pxor  %xmm1,%xmm14
+
+# qhasm: uint32323232                   r4 >>= 19
+# asm 1: psrld $19,<r4=int6464#3
+# asm 2: psrld $19,<r4=%xmm2
+psrld $19,%xmm2
+
+# qhasm:                                z4 ^= r4
+# asm 1: pxor  <r4=int6464#3,<z4=int6464#15
+# asm 2: pxor  <r4=%xmm2,<z4=%xmm14
+pxor  %xmm2,%xmm14
+
+# qhasm:                y0 = z2
+# asm 1: movdqa <z2=int6464#11,>y0=int6464#2
+# asm 2: movdqa <z2=%xmm10,>y0=%xmm1
+movdqa %xmm10,%xmm1
+
+# qhasm: uint32323232   y0 += z3
+# asm 1: paddd <z3=int6464#5,<y0=int6464#2
+# asm 2: paddd <z3=%xmm4,<y0=%xmm1
+paddd %xmm4,%xmm1
+
+# qhasm:                r0 = y0
+# asm 1: movdqa <y0=int6464#2,>r0=int6464#3
+# asm 2: movdqa <y0=%xmm1,>r0=%xmm2
+movdqa %xmm1,%xmm2
+
+# qhasm: uint32323232   y0 <<= 18
+# asm 1: pslld $18,<y0=int6464#2
+# asm 2: pslld $18,<y0=%xmm1
+pslld $18,%xmm1
+
+# qhasm:                z0 ^= y0
+# asm 1: pxor  <y0=int6464#2,<z0=int6464#1
+# asm 2: pxor  <y0=%xmm1,<z0=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232   r0 >>= 14
+# asm 1: psrld $14,<r0=int6464#3
+# asm 2: psrld $14,<r0=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                z0 ^= r0
+# asm 1: pxor  <r0=int6464#3,<z0=int6464#1
+# asm 2: pxor  <r0=%xmm2,<z0=%xmm0
+pxor  %xmm2,%xmm0
+
+# qhasm:                                                z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#21,>z10=int6464#2
+# asm 2: movdqa <z10_stack=320(%rsp),>z10=%xmm1
+movdqa 320(%rsp),%xmm1
+
+# qhasm:                z0_stack = z0
+# asm 1: movdqa <z0=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <z0=%xmm0,>z0_stack=320(%rsp)
+movdqa %xmm0,320(%rsp)
+
+# qhasm:                                y5 = z7
+# asm 1: movdqa <z7=int6464#9,>y5=int6464#1
+# asm 2: movdqa <z7=%xmm8,>y5=%xmm0
+movdqa %xmm8,%xmm0
+
+# qhasm: uint32323232                   y5 += z4
+# asm 1: paddd <z4=int6464#15,<y5=int6464#1
+# asm 2: paddd <z4=%xmm14,<y5=%xmm0
+paddd %xmm14,%xmm0
+
+# qhasm:                                r5 = y5
+# asm 1: movdqa <y5=int6464#1,>r5=int6464#3
+# asm 2: movdqa <y5=%xmm0,>r5=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                   y5 <<= 18
+# asm 1: pslld $18,<y5=int6464#1
+# asm 2: pslld $18,<y5=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                z5 ^= y5
+# asm 1: pxor  <y5=int6464#1,<z5=int6464#13
+# asm 2: pxor  <y5=%xmm0,<z5=%xmm12
+pxor  %xmm0,%xmm12
+
+# qhasm: uint32323232                   r5 >>= 14
+# asm 1: psrld $14,<r5=int6464#3
+# asm 2: psrld $14,<r5=%xmm2
+psrld $14,%xmm2
+
+# qhasm:                                z5 ^= r5
+# asm 1: pxor  <r5=int6464#3,<z5=int6464#13
+# asm 2: pxor  <r5=%xmm2,<z5=%xmm12
+pxor  %xmm2,%xmm12
+
+# qhasm:                                                y11 = z9
+# asm 1: movdqa <z9=int6464#12,>y11=int6464#1
+# asm 2: movdqa <z9=%xmm11,>y11=%xmm0
+movdqa %xmm11,%xmm0
+
+# qhasm: uint32323232                                   y11 += z10
+# asm 1: paddd <z10=int6464#2,<y11=int6464#1
+# asm 2: paddd <z10=%xmm1,<y11=%xmm0
+paddd %xmm1,%xmm0
+
+# qhasm:                                                r11 = y11
+# asm 1: movdqa <y11=int6464#1,>r11=int6464#3
+# asm 2: movdqa <y11=%xmm0,>r11=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                   y11 <<= 7
+# asm 1: pslld $7,<y11=int6464#1
+# asm 2: pslld $7,<y11=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                z11 ^= y11
+# asm 1: pxor  <y11=int6464#1,<z11=int6464#7
+# asm 2: pxor  <y11=%xmm0,<z11=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm: uint32323232                                   r11 >>= 25
+# asm 1: psrld $25,<r11=int6464#3
+# asm 2: psrld $25,<r11=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                z11 ^= r11
+# asm 1: pxor  <r11=int6464#3,<z11=int6464#7
+# asm 2: pxor  <r11=%xmm2,<z11=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>z15=int6464#3
+# asm 2: movdqa <z15_stack=336(%rsp),>z15=%xmm2
+movdqa 336(%rsp),%xmm2
+
+# qhasm:                                z5_stack = z5
+# asm 1: movdqa <z5=int6464#13,>z5_stack=stack128#22
+# asm 2: movdqa <z5=%xmm12,>z5_stack=336(%rsp)
+movdqa %xmm12,336(%rsp)
+
+# qhasm:                                                                y12 = z14
+# asm 1: movdqa <z14=int6464#4,>y12=int6464#1
+# asm 2: movdqa <z14=%xmm3,>y12=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                   y12 += z15
+# asm 1: paddd <z15=int6464#3,<y12=int6464#1
+# asm 2: paddd <z15=%xmm2,<y12=%xmm0
+paddd %xmm2,%xmm0
+
+# qhasm:                                                                r12 = y12
+# asm 1: movdqa <y12=int6464#1,>r12=int6464#13
+# asm 2: movdqa <y12=%xmm0,>r12=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y12 <<= 7
+# asm 1: pslld $7,<y12=int6464#1
+# asm 2: pslld $7,<y12=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                z12 ^= y12
+# asm 1: pxor  <y12=int6464#1,<z12=int6464#14
+# asm 2: pxor  <y12=%xmm0,<z12=%xmm13
+pxor  %xmm0,%xmm13
+
+# qhasm: uint32323232                                                   r12 >>= 25
+# asm 1: psrld $25,<r12=int6464#13
+# asm 2: psrld $25,<r12=%xmm12
+psrld $25,%xmm12
+
+# qhasm:                                                                z12 ^= r12
+# asm 1: pxor  <r12=int6464#13,<z12=int6464#14
+# asm 2: pxor  <r12=%xmm12,<z12=%xmm13
+pxor  %xmm12,%xmm13
+
+# qhasm:                                                y8 = z10
+# asm 1: movdqa <z10=int6464#2,>y8=int6464#1
+# asm 2: movdqa <z10=%xmm1,>y8=%xmm0
+movdqa %xmm1,%xmm0
+
+# qhasm: uint32323232                                   y8 += z11
+# asm 1: paddd <z11=int6464#7,<y8=int6464#1
+# asm 2: paddd <z11=%xmm6,<y8=%xmm0
+paddd %xmm6,%xmm0
+
+# qhasm:                                                r8 = y8
+# asm 1: movdqa <y8=int6464#1,>r8=int6464#13
+# asm 2: movdqa <y8=%xmm0,>r8=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y8 <<= 9
+# asm 1: pslld $9,<y8=int6464#1
+# asm 2: pslld $9,<y8=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                z8 ^= y8
+# asm 1: pxor  <y8=int6464#1,<z8=int6464#16
+# asm 2: pxor  <y8=%xmm0,<z8=%xmm15
+pxor  %xmm0,%xmm15
+
+# qhasm: uint32323232                                   r8 >>= 23
+# asm 1: psrld $23,<r8=int6464#13
+# asm 2: psrld $23,<r8=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                z8 ^= r8
+# asm 1: pxor  <r8=int6464#13,<z8=int6464#16
+# asm 2: pxor  <r8=%xmm12,<z8=%xmm15
+pxor  %xmm12,%xmm15
+
+# qhasm:                                                                y13 = z15
+# asm 1: movdqa <z15=int6464#3,>y13=int6464#1
+# asm 2: movdqa <z15=%xmm2,>y13=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                   y13 += z12
+# asm 1: paddd <z12=int6464#14,<y13=int6464#1
+# asm 2: paddd <z12=%xmm13,<y13=%xmm0
+paddd %xmm13,%xmm0
+
+# qhasm:                                                                r13 = y13
+# asm 1: movdqa <y13=int6464#1,>r13=int6464#13
+# asm 2: movdqa <y13=%xmm0,>r13=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y13 <<= 9
+# asm 1: pslld $9,<y13=int6464#1
+# asm 2: pslld $9,<y13=%xmm0
+pslld $9,%xmm0
+
+# qhasm:                                                                z13 ^= y13
+# asm 1: pxor  <y13=int6464#1,<z13=int6464#10
+# asm 2: pxor  <y13=%xmm0,<z13=%xmm9
+pxor  %xmm0,%xmm9
+
+# qhasm: uint32323232                                                   r13 >>= 23
+# asm 1: psrld $23,<r13=int6464#13
+# asm 2: psrld $23,<r13=%xmm12
+psrld $23,%xmm12
+
+# qhasm:                                                                z13 ^= r13
+# asm 1: pxor  <r13=int6464#13,<z13=int6464#10
+# asm 2: pxor  <r13=%xmm12,<z13=%xmm9
+pxor  %xmm12,%xmm9
+
+# qhasm:                                                y9 = z11
+# asm 1: movdqa <z11=int6464#7,>y9=int6464#1
+# asm 2: movdqa <z11=%xmm6,>y9=%xmm0
+movdqa %xmm6,%xmm0
+
+# qhasm: uint32323232                                   y9 += z8
+# asm 1: paddd <z8=int6464#16,<y9=int6464#1
+# asm 2: paddd <z8=%xmm15,<y9=%xmm0
+paddd %xmm15,%xmm0
+
+# qhasm:                                                r9 = y9
+# asm 1: movdqa <y9=int6464#1,>r9=int6464#13
+# asm 2: movdqa <y9=%xmm0,>r9=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y9 <<= 13
+# asm 1: pslld $13,<y9=int6464#1
+# asm 2: pslld $13,<y9=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                z9 ^= y9
+# asm 1: pxor  <y9=int6464#1,<z9=int6464#12
+# asm 2: pxor  <y9=%xmm0,<z9=%xmm11
+pxor  %xmm0,%xmm11
+
+# qhasm: uint32323232                                   r9 >>= 19
+# asm 1: psrld $19,<r9=int6464#13
+# asm 2: psrld $19,<r9=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                z9 ^= r9
+# asm 1: pxor  <r9=int6464#13,<z9=int6464#12
+# asm 2: pxor  <r9=%xmm12,<z9=%xmm11
+pxor  %xmm12,%xmm11
+
+# qhasm:                                                                y14 = z12
+# asm 1: movdqa <z12=int6464#14,>y14=int6464#1
+# asm 2: movdqa <z12=%xmm13,>y14=%xmm0
+movdqa %xmm13,%xmm0
+
+# qhasm: uint32323232                                                   y14 += z13
+# asm 1: paddd <z13=int6464#10,<y14=int6464#1
+# asm 2: paddd <z13=%xmm9,<y14=%xmm0
+paddd %xmm9,%xmm0
+
+# qhasm:                                                                r14 = y14
+# asm 1: movdqa <y14=int6464#1,>r14=int6464#13
+# asm 2: movdqa <y14=%xmm0,>r14=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y14 <<= 13
+# asm 1: pslld $13,<y14=int6464#1
+# asm 2: pslld $13,<y14=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                z14 ^= y14
+# asm 1: pxor  <y14=int6464#1,<z14=int6464#4
+# asm 2: pxor  <y14=%xmm0,<z14=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm: uint32323232                                                   r14 >>= 19
+# asm 1: psrld $19,<r14=int6464#13
+# asm 2: psrld $19,<r14=%xmm12
+psrld $19,%xmm12
+
+# qhasm:                                                                z14 ^= r14
+# asm 1: pxor  <r14=int6464#13,<z14=int6464#4
+# asm 2: pxor  <r14=%xmm12,<z14=%xmm3
+pxor  %xmm12,%xmm3
+
+# qhasm:                                                y10 = z8
+# asm 1: movdqa <z8=int6464#16,>y10=int6464#1
+# asm 2: movdqa <z8=%xmm15,>y10=%xmm0
+movdqa %xmm15,%xmm0
+
+# qhasm: uint32323232                                   y10 += z9
+# asm 1: paddd <z9=int6464#12,<y10=int6464#1
+# asm 2: paddd <z9=%xmm11,<y10=%xmm0
+paddd %xmm11,%xmm0
+
+# qhasm:                                                r10 = y10
+# asm 1: movdqa <y10=int6464#1,>r10=int6464#13
+# asm 2: movdqa <y10=%xmm0,>r10=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                   y10 <<= 18
+# asm 1: pslld $18,<y10=int6464#1
+# asm 2: pslld $18,<y10=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                z10 ^= y10
+# asm 1: pxor  <y10=int6464#1,<z10=int6464#2
+# asm 2: pxor  <y10=%xmm0,<z10=%xmm1
+pxor  %xmm0,%xmm1
+
+# qhasm: uint32323232                                   r10 >>= 14
+# asm 1: psrld $14,<r10=int6464#13
+# asm 2: psrld $14,<r10=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                z10 ^= r10
+# asm 1: pxor  <r10=int6464#13,<z10=int6464#2
+# asm 2: pxor  <r10=%xmm12,<z10=%xmm1
+pxor  %xmm12,%xmm1
+
+# qhasm:                                                                y15 = z13
+# asm 1: movdqa <z13=int6464#10,>y15=int6464#1
+# asm 2: movdqa <z13=%xmm9,>y15=%xmm0
+movdqa %xmm9,%xmm0
+
+# qhasm: uint32323232                                                   y15 += z14
+# asm 1: paddd <z14=int6464#4,<y15=int6464#1
+# asm 2: paddd <z14=%xmm3,<y15=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                r15 = y15
+# asm 1: movdqa <y15=int6464#1,>r15=int6464#13
+# asm 2: movdqa <y15=%xmm0,>r15=%xmm12
+movdqa %xmm0,%xmm12
+
+# qhasm: uint32323232                                                   y15 <<= 18
+# asm 1: pslld $18,<y15=int6464#1
+# asm 2: pslld $18,<y15=%xmm0
+pslld $18,%xmm0
+
+# qhasm:                                                                z15 ^= y15
+# asm 1: pxor  <y15=int6464#1,<z15=int6464#3
+# asm 2: pxor  <y15=%xmm0,<z15=%xmm2
+pxor  %xmm0,%xmm2
+
+# qhasm: uint32323232                                                   r15 >>= 14
+# asm 1: psrld $14,<r15=int6464#13
+# asm 2: psrld $14,<r15=%xmm12
+psrld $14,%xmm12
+
+# qhasm:                                                                z15 ^= r15
+# asm 1: pxor  <r15=int6464#13,<z15=int6464#3
+# asm 2: pxor  <r15=%xmm12,<z15=%xmm2
+pxor  %xmm12,%xmm2
+
+# qhasm:                z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>z0=int6464#13
+# asm 2: movdqa <z0_stack=320(%rsp),>z0=%xmm12
+movdqa 320(%rsp),%xmm12
+
+# qhasm:                                z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#22,>z5=int6464#1
+# asm 2: movdqa <z5_stack=336(%rsp),>z5=%xmm0
+movdqa 336(%rsp),%xmm0
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int64#3
+# asm 2: sub  $2,<i=%rdx
+sub  $2,%rdx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#13
+# asm 2: paddd <orig0=112(%rsp),<z0=%xmm12
+paddd 112(%rsp),%xmm12
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#8
+# asm 2: paddd <orig1=176(%rsp),<z1=%xmm7
+paddd 176(%rsp),%xmm7
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#11
+# asm 2: paddd <orig2=224(%rsp),<z2=%xmm10
+paddd 224(%rsp),%xmm10
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#5
+# asm 2: paddd <orig3=272(%rsp),<z3=%xmm4
+paddd 272(%rsp),%xmm4
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 0(<m=%rsi),<in0=%edx
+xorl 0(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 4(<m=%rsi),<in1=%ecx
+xorl 4(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 8(<m=%rsi),<in2=%r8d
+xorl 8(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 12(<m=%rsi),<in3=%r9d
+xorl 12(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#3d,0(<out=int64#1)
+# asm 2: movl   <in0=%edx,0(<out=%rdi)
+movl   %edx,0(%rdi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#4d,4(<out=int64#1)
+# asm 2: movl   <in1=%ecx,4(<out=%rdi)
+movl   %ecx,4(%rdi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#5d,8(<out=int64#1)
+# asm 2: movl   <in2=%r8d,8(<out=%rdi)
+movl   %r8d,8(%rdi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#6d,12(<out=int64#1)
+# asm 2: movl   <in3=%r9d,12(<out=%rdi)
+movl   %r9d,12(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 64(<m=%rsi),<in0=%edx
+xorl 64(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 68(<m=%rsi),<in1=%ecx
+xorl 68(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 72(<m=%rsi),<in2=%r8d
+xorl 72(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 76(<m=%rsi),<in3=%r9d
+xorl 76(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl   <in0=int64#3d,64(<out=int64#1)
+# asm 2: movl   <in0=%edx,64(<out=%rdi)
+movl   %edx,64(%rdi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl   <in1=int64#4d,68(<out=int64#1)
+# asm 2: movl   <in1=%ecx,68(<out=%rdi)
+movl   %ecx,68(%rdi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl   <in2=int64#5d,72(<out=int64#1)
+# asm 2: movl   <in2=%r8d,72(<out=%rdi)
+movl   %r8d,72(%rdi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl   <in3=int64#6d,76(<out=int64#1)
+# asm 2: movl   <in3=%r9d,76(<out=%rdi)
+movl   %r9d,76(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#13,<z0=int6464#13
+# asm 2: pshufd $0x39,<z0=%xmm12,<z0=%xmm12
+pshufd $0x39,%xmm12,%xmm12
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#8,<z1=int6464#8
+# asm 2: pshufd $0x39,<z1=%xmm7,<z1=%xmm7
+pshufd $0x39,%xmm7,%xmm7
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#11,<z2=int6464#11
+# asm 2: pshufd $0x39,<z2=%xmm10,<z2=%xmm10
+pshufd $0x39,%xmm10,%xmm10
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#5,<z3=int6464#5
+# asm 2: pshufd $0x39,<z3=%xmm4,<z3=%xmm4
+pshufd $0x39,%xmm4,%xmm4
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 128(<m=%rsi),<in0=%edx
+xorl 128(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 132(<m=%rsi),<in1=%ecx
+xorl 132(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 136(<m=%rsi),<in2=%r8d
+xorl 136(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 140(<m=%rsi),<in3=%r9d
+xorl 140(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl   <in0=int64#3d,128(<out=int64#1)
+# asm 2: movl   <in0=%edx,128(<out=%rdi)
+movl   %edx,128(%rdi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl   <in1=int64#4d,132(<out=int64#1)
+# asm 2: movl   <in1=%ecx,132(<out=%rdi)
+movl   %ecx,132(%rdi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl   <in2=int64#5d,136(<out=int64#1)
+# asm 2: movl   <in2=%r8d,136(<out=%rdi)
+movl   %r8d,136(%rdi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl   <in3=int64#6d,140(<out=int64#1)
+# asm 2: movl   <in3=%r9d,140(<out=%rdi)
+movl   %r9d,140(%rdi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#13,>in0=int64#3
+# asm 2: movd   <z0=%xmm12,>in0=%rdx
+movd   %xmm12,%rdx
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#8,>in1=int64#4
+# asm 2: movd   <z1=%xmm7,>in1=%rcx
+movd   %xmm7,%rcx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#11,>in2=int64#5
+# asm 2: movd   <z2=%xmm10,>in2=%r8
+movd   %xmm10,%r8
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#5,>in3=int64#6
+# asm 2: movd   <z3=%xmm4,>in3=%r9
+movd   %xmm4,%r9
+
+# qhasm:   (uint32) in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int64#2),<in0=int64#3d
+# asm 2: xorl 192(<m=%rsi),<in0=%edx
+xorl 192(%rsi),%edx
+
+# qhasm:   (uint32) in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int64#2),<in1=int64#4d
+# asm 2: xorl 196(<m=%rsi),<in1=%ecx
+xorl 196(%rsi),%ecx
+
+# qhasm:   (uint32) in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int64#2),<in2=int64#5d
+# asm 2: xorl 200(<m=%rsi),<in2=%r8d
+xorl 200(%rsi),%r8d
+
+# qhasm:   (uint32) in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int64#2),<in3=int64#6d
+# asm 2: xorl 204(<m=%rsi),<in3=%r9d
+xorl 204(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl   <in0=int64#3d,192(<out=int64#1)
+# asm 2: movl   <in0=%edx,192(<out=%rdi)
+movl   %edx,192(%rdi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl   <in1=int64#4d,196(<out=int64#1)
+# asm 2: movl   <in1=%ecx,196(<out=%rdi)
+movl   %ecx,196(%rdi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl   <in2=int64#5d,200(<out=int64#1)
+# asm 2: movl   <in2=%r8d,200(<out=%rdi)
+movl   %r8d,200(%rdi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl   <in3=int64#6d,204(<out=int64#1)
+# asm 2: movl   <in3=%r9d,204(<out=%rdi)
+movl   %r9d,204(%rdi)
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#15
+# asm 2: paddd <orig4=240(%rsp),<z4=%xmm14
+paddd 240(%rsp),%xmm14
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#1
+# asm 2: paddd <orig5=64(%rsp),<z5=%xmm0
+paddd 64(%rsp),%xmm0
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#6
+# asm 2: paddd <orig6=128(%rsp),<z6=%xmm5
+paddd 128(%rsp),%xmm5
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#9
+# asm 2: paddd <orig7=192(%rsp),<z7=%xmm8
+paddd 192(%rsp),%xmm8
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 16(<m=%rsi),<in4=%edx
+xorl 16(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#3d,16(<out=int64#1)
+# asm 2: movl   <in4=%edx,16(<out=%rdi)
+movl   %edx,16(%rdi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 80(<m=%rsi),<in4=%edx
+xorl 80(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 84(<m=%rsi),<in5=%ecx
+xorl 84(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 88(<m=%rsi),<in6=%r8d
+xorl 88(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 92(<m=%rsi),<in7=%r9d
+xorl 92(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl   <in4=int64#3d,80(<out=int64#1)
+# asm 2: movl   <in4=%edx,80(<out=%rdi)
+movl   %edx,80(%rdi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl   <in5=int64#4d,84(<out=int64#1)
+# asm 2: movl   <in5=%ecx,84(<out=%rdi)
+movl   %ecx,84(%rdi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl   <in6=int64#5d,88(<out=int64#1)
+# asm 2: movl   <in6=%r8d,88(<out=%rdi)
+movl   %r8d,88(%rdi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl   <in7=int64#6d,92(<out=int64#1)
+# asm 2: movl   <in7=%r9d,92(<out=%rdi)
+movl   %r9d,92(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#15,<z4=int6464#15
+# asm 2: pshufd $0x39,<z4=%xmm14,<z4=%xmm14
+pshufd $0x39,%xmm14,%xmm14
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#1,<z5=int6464#1
+# asm 2: pshufd $0x39,<z5=%xmm0,<z5=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#6,<z6=int6464#6
+# asm 2: pshufd $0x39,<z6=%xmm5,<z6=%xmm5
+pshufd $0x39,%xmm5,%xmm5
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#9,<z7=int6464#9
+# asm 2: pshufd $0x39,<z7=%xmm8,<z7=%xmm8
+pshufd $0x39,%xmm8,%xmm8
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 144(<m=%rsi),<in4=%edx
+xorl 144(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 148(<m=%rsi),<in5=%ecx
+xorl 148(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 152(<m=%rsi),<in6=%r8d
+xorl 152(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 156(<m=%rsi),<in7=%r9d
+xorl 156(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl   <in4=int64#3d,144(<out=int64#1)
+# asm 2: movl   <in4=%edx,144(<out=%rdi)
+movl   %edx,144(%rdi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl   <in5=int64#4d,148(<out=int64#1)
+# asm 2: movl   <in5=%ecx,148(<out=%rdi)
+movl   %ecx,148(%rdi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl   <in6=int64#5d,152(<out=int64#1)
+# asm 2: movl   <in6=%r8d,152(<out=%rdi)
+movl   %r8d,152(%rdi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl   <in7=int64#6d,156(<out=int64#1)
+# asm 2: movl   <in7=%r9d,156(<out=%rdi)
+movl   %r9d,156(%rdi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#15,>in4=int64#3
+# asm 2: movd   <z4=%xmm14,>in4=%rdx
+movd   %xmm14,%rdx
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#1,>in5=int64#4
+# asm 2: movd   <z5=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#6,>in6=int64#5
+# asm 2: movd   <z6=%xmm5,>in6=%r8
+movd   %xmm5,%r8
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#9,>in7=int64#6
+# asm 2: movd   <z7=%xmm8,>in7=%r9
+movd   %xmm8,%r9
+
+# qhasm:   (uint32) in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int64#2),<in4=int64#3d
+# asm 2: xorl 208(<m=%rsi),<in4=%edx
+xorl 208(%rsi),%edx
+
+# qhasm:   (uint32) in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 212(<m=%rsi),<in5=%ecx
+xorl 212(%rsi),%ecx
+
+# qhasm:   (uint32) in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 216(<m=%rsi),<in6=%r8d
+xorl 216(%rsi),%r8d
+
+# qhasm:   (uint32) in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 220(<m=%rsi),<in7=%r9d
+xorl 220(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl   <in4=int64#3d,208(<out=int64#1)
+# asm 2: movl   <in4=%edx,208(<out=%rdi)
+movl   %edx,208(%rdi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl   <in5=int64#4d,212(<out=int64#1)
+# asm 2: movl   <in5=%ecx,212(<out=%rdi)
+movl   %ecx,212(%rdi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl   <in6=int64#5d,216(<out=int64#1)
+# asm 2: movl   <in6=%r8d,216(<out=%rdi)
+movl   %r8d,216(%rdi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl   <in7=int64#6d,220(<out=int64#1)
+# asm 2: movl   <in7=%r9d,220(<out=%rdi)
+movl   %r9d,220(%rdi)
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#16
+# asm 2: paddd <orig8=288(%rsp),<z8=%xmm15
+paddd 288(%rsp),%xmm15
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#12
+# asm 2: paddd <orig9=304(%rsp),<z9=%xmm11
+paddd 304(%rsp),%xmm11
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#2
+# asm 2: paddd <orig10=80(%rsp),<z10=%xmm1
+paddd 80(%rsp),%xmm1
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#7
+# asm 2: paddd <orig11=144(%rsp),<z11=%xmm6
+paddd 144(%rsp),%xmm6
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 32(<m=%rsi),<in8=%edx
+xorl 32(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 36(<m=%rsi),<in9=%ecx
+xorl 36(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 40(<m=%rsi),<in10=%r8d
+xorl 40(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 44(<m=%rsi),<in11=%r9d
+xorl 44(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#3d,32(<out=int64#1)
+# asm 2: movl   <in8=%edx,32(<out=%rdi)
+movl   %edx,32(%rdi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#4d,36(<out=int64#1)
+# asm 2: movl   <in9=%ecx,36(<out=%rdi)
+movl   %ecx,36(%rdi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#5d,40(<out=int64#1)
+# asm 2: movl   <in10=%r8d,40(<out=%rdi)
+movl   %r8d,40(%rdi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#6d,44(<out=int64#1)
+# asm 2: movl   <in11=%r9d,44(<out=%rdi)
+movl   %r9d,44(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 96(<m=%rsi),<in8=%edx
+xorl 96(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 100(<m=%rsi),<in9=%ecx
+xorl 100(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 104(<m=%rsi),<in10=%r8d
+xorl 104(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 108(<m=%rsi),<in11=%r9d
+xorl 108(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl   <in8=int64#3d,96(<out=int64#1)
+# asm 2: movl   <in8=%edx,96(<out=%rdi)
+movl   %edx,96(%rdi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl   <in9=int64#4d,100(<out=int64#1)
+# asm 2: movl   <in9=%ecx,100(<out=%rdi)
+movl   %ecx,100(%rdi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl   <in10=int64#5d,104(<out=int64#1)
+# asm 2: movl   <in10=%r8d,104(<out=%rdi)
+movl   %r8d,104(%rdi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl   <in11=int64#6d,108(<out=int64#1)
+# asm 2: movl   <in11=%r9d,108(<out=%rdi)
+movl   %r9d,108(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#16,<z8=int6464#16
+# asm 2: pshufd $0x39,<z8=%xmm15,<z8=%xmm15
+pshufd $0x39,%xmm15,%xmm15
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#12,<z9=int6464#12
+# asm 2: pshufd $0x39,<z9=%xmm11,<z9=%xmm11
+pshufd $0x39,%xmm11,%xmm11
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#2,<z10=int6464#2
+# asm 2: pshufd $0x39,<z10=%xmm1,<z10=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#7,<z11=int6464#7
+# asm 2: pshufd $0x39,<z11=%xmm6,<z11=%xmm6
+pshufd $0x39,%xmm6,%xmm6
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 160(<m=%rsi),<in8=%edx
+xorl 160(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 164(<m=%rsi),<in9=%ecx
+xorl 164(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 168(<m=%rsi),<in10=%r8d
+xorl 168(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 172(<m=%rsi),<in11=%r9d
+xorl 172(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl   <in8=int64#3d,160(<out=int64#1)
+# asm 2: movl   <in8=%edx,160(<out=%rdi)
+movl   %edx,160(%rdi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl   <in9=int64#4d,164(<out=int64#1)
+# asm 2: movl   <in9=%ecx,164(<out=%rdi)
+movl   %ecx,164(%rdi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl   <in10=int64#5d,168(<out=int64#1)
+# asm 2: movl   <in10=%r8d,168(<out=%rdi)
+movl   %r8d,168(%rdi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl   <in11=int64#6d,172(<out=int64#1)
+# asm 2: movl   <in11=%r9d,172(<out=%rdi)
+movl   %r9d,172(%rdi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#16,>in8=int64#3
+# asm 2: movd   <z8=%xmm15,>in8=%rdx
+movd   %xmm15,%rdx
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#12,>in9=int64#4
+# asm 2: movd   <z9=%xmm11,>in9=%rcx
+movd   %xmm11,%rcx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#2,>in10=int64#5
+# asm 2: movd   <z10=%xmm1,>in10=%r8
+movd   %xmm1,%r8
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#7,>in11=int64#6
+# asm 2: movd   <z11=%xmm6,>in11=%r9
+movd   %xmm6,%r9
+
+# qhasm:   (uint32) in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int64#2),<in8=int64#3d
+# asm 2: xorl 224(<m=%rsi),<in8=%edx
+xorl 224(%rsi),%edx
+
+# qhasm:   (uint32) in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int64#2),<in9=int64#4d
+# asm 2: xorl 228(<m=%rsi),<in9=%ecx
+xorl 228(%rsi),%ecx
+
+# qhasm:   (uint32) in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int64#2),<in10=int64#5d
+# asm 2: xorl 232(<m=%rsi),<in10=%r8d
+xorl 232(%rsi),%r8d
+
+# qhasm:   (uint32) in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int64#2),<in11=int64#6d
+# asm 2: xorl 236(<m=%rsi),<in11=%r9d
+xorl 236(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl   <in8=int64#3d,224(<out=int64#1)
+# asm 2: movl   <in8=%edx,224(<out=%rdi)
+movl   %edx,224(%rdi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl   <in9=int64#4d,228(<out=int64#1)
+# asm 2: movl   <in9=%ecx,228(<out=%rdi)
+movl   %ecx,228(%rdi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl   <in10=int64#5d,232(<out=int64#1)
+# asm 2: movl   <in10=%r8d,232(<out=%rdi)
+movl   %r8d,232(%rdi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl   <in11=int64#6d,236(<out=int64#1)
+# asm 2: movl   <in11=%r9d,236(<out=%rdi)
+movl   %r9d,236(%rdi)
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#14
+# asm 2: paddd <orig12=160(%rsp),<z12=%xmm13
+paddd 160(%rsp),%xmm13
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#10
+# asm 2: paddd <orig13=208(%rsp),<z13=%xmm9
+paddd 208(%rsp),%xmm9
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#4
+# asm 2: paddd <orig14=256(%rsp),<z14=%xmm3
+paddd 256(%rsp),%xmm3
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#3
+# asm 2: paddd <orig15=96(%rsp),<z15=%xmm2
+paddd 96(%rsp),%xmm2
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 48(<m=%rsi),<in12=%edx
+xorl 48(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 52(<m=%rsi),<in13=%ecx
+xorl 52(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 56(<m=%rsi),<in14=%r8d
+xorl 56(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 60(<m=%rsi),<in15=%r9d
+xorl 60(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#3d,48(<out=int64#1)
+# asm 2: movl   <in12=%edx,48(<out=%rdi)
+movl   %edx,48(%rdi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#4d,52(<out=int64#1)
+# asm 2: movl   <in13=%ecx,52(<out=%rdi)
+movl   %ecx,52(%rdi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#5d,56(<out=int64#1)
+# asm 2: movl   <in14=%r8d,56(<out=%rdi)
+movl   %r8d,56(%rdi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#6d,60(<out=int64#1)
+# asm 2: movl   <in15=%r9d,60(<out=%rdi)
+movl   %r9d,60(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 112(<m=%rsi),<in12=%edx
+xorl 112(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 116(<m=%rsi),<in13=%ecx
+xorl 116(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 120(<m=%rsi),<in14=%r8d
+xorl 120(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 124(<m=%rsi),<in15=%r9d
+xorl 124(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl   <in12=int64#3d,112(<out=int64#1)
+# asm 2: movl   <in12=%edx,112(<out=%rdi)
+movl   %edx,112(%rdi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl   <in13=int64#4d,116(<out=int64#1)
+# asm 2: movl   <in13=%ecx,116(<out=%rdi)
+movl   %ecx,116(%rdi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl   <in14=int64#5d,120(<out=int64#1)
+# asm 2: movl   <in14=%r8d,120(<out=%rdi)
+movl   %r8d,120(%rdi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl   <in15=int64#6d,124(<out=int64#1)
+# asm 2: movl   <in15=%r9d,124(<out=%rdi)
+movl   %r9d,124(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#14,<z12=int6464#14
+# asm 2: pshufd $0x39,<z12=%xmm13,<z12=%xmm13
+pshufd $0x39,%xmm13,%xmm13
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#10,<z13=int6464#10
+# asm 2: pshufd $0x39,<z13=%xmm9,<z13=%xmm9
+pshufd $0x39,%xmm9,%xmm9
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#4,<z14=int6464#4
+# asm 2: pshufd $0x39,<z14=%xmm3,<z14=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#3,<z15=int6464#3
+# asm 2: pshufd $0x39,<z15=%xmm2,<z15=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 176(<m=%rsi),<in12=%edx
+xorl 176(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 180(<m=%rsi),<in13=%ecx
+xorl 180(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 184(<m=%rsi),<in14=%r8d
+xorl 184(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 188(<m=%rsi),<in15=%r9d
+xorl 188(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl   <in12=int64#3d,176(<out=int64#1)
+# asm 2: movl   <in12=%edx,176(<out=%rdi)
+movl   %edx,176(%rdi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl   <in13=int64#4d,180(<out=int64#1)
+# asm 2: movl   <in13=%ecx,180(<out=%rdi)
+movl   %ecx,180(%rdi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl   <in14=int64#5d,184(<out=int64#1)
+# asm 2: movl   <in14=%r8d,184(<out=%rdi)
+movl   %r8d,184(%rdi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl   <in15=int64#6d,188(<out=int64#1)
+# asm 2: movl   <in15=%r9d,188(<out=%rdi)
+movl   %r9d,188(%rdi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#14,>in12=int64#3
+# asm 2: movd   <z12=%xmm13,>in12=%rdx
+movd   %xmm13,%rdx
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#10,>in13=int64#4
+# asm 2: movd   <z13=%xmm9,>in13=%rcx
+movd   %xmm9,%rcx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#4,>in14=int64#5
+# asm 2: movd   <z14=%xmm3,>in14=%r8
+movd   %xmm3,%r8
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#3,>in15=int64#6
+# asm 2: movd   <z15=%xmm2,>in15=%r9
+movd   %xmm2,%r9
+
+# qhasm:   (uint32) in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int64#2),<in12=int64#3d
+# asm 2: xorl 240(<m=%rsi),<in12=%edx
+xorl 240(%rsi),%edx
+
+# qhasm:   (uint32) in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int64#2),<in13=int64#4d
+# asm 2: xorl 244(<m=%rsi),<in13=%ecx
+xorl 244(%rsi),%ecx
+
+# qhasm:   (uint32) in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int64#2),<in14=int64#5d
+# asm 2: xorl 248(<m=%rsi),<in14=%r8d
+xorl 248(%rsi),%r8d
+
+# qhasm:   (uint32) in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int64#2),<in15=int64#6d
+# asm 2: xorl 252(<m=%rsi),<in15=%r9d
+xorl 252(%rsi),%r9d
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl   <in12=int64#3d,240(<out=int64#1)
+# asm 2: movl   <in12=%edx,240(<out=%rdi)
+movl   %edx,240(%rdi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl   <in13=int64#4d,244(<out=int64#1)
+# asm 2: movl   <in13=%ecx,244(<out=%rdi)
+movl   %ecx,244(%rdi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl   <in14=int64#5d,248(<out=int64#1)
+# asm 2: movl   <in14=%r8d,248(<out=%rdi)
+movl   %r8d,248(%rdi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl   <in15=int64#6d,252(<out=int64#1)
+# asm 2: movl   <in15=%r9d,252(<out=%rdi)
+movl   %r9d,252(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int64#6
+# asm 2: sub  $256,<bytes=%r9
+sub  $256,%r9
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int64#2
+# asm 2: add  $256,<m=%rsi
+add  $256,%rsi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int64#1
+# asm 2: add  $256,<out=%rdi
+add  $256,%rdi
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int64#6
+# asm 2: cmp  $256,<bytes=%r9
+cmp  $256,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int64#6
+# asm 2: cmp  $0,<bytes=%r9
+cmp  $0,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: mov  <out=int64#1,>ctarget=int64#3
+# asm 2: mov  <out=%rdi,>ctarget=%rdx
+mov  %rdi,%rdx
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leaq <tmp=stack512#1,>out=int64#1
+# asm 2: leaq <tmp=416(%rsp),>out=%rdi
+leaq 416(%rsp),%rdi
+
+# qhasm:     m = &tmp
+# asm 1: leaq <tmp=stack512#1,>m=int64#2
+# asm 2: leaq <tmp=416(%rsp),>m=%rsi
+leaq 416(%rsp),%rsi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_backup = bytes
+# asm 1: movq <bytes=int64#6,>bytes_backup=stack64#8
+# asm 2: movq <bytes=%r9,>bytes_backup=408(%rsp)
+movq %r9,408(%rsp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#4,>diag0=int6464#1
+# asm 2: movdqa <x0=48(%rsp),>diag0=%xmm0
+movdqa 48(%rsp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#1,>diag1=int6464#2
+# asm 2: movdqa <x1=0(%rsp),>diag1=%xmm1
+movdqa 0(%rsp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#2,>diag2=int6464#3
+# asm 2: movdqa <x2=16(%rsp),>diag2=%xmm2
+movdqa 16(%rsp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#3,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%rsp),>diag3=%xmm3
+movdqa 32(%rsp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 8
+# asm 1: mov  $8,>i=int64#4
+# asm 2: mov  $8,>i=%rcx
+mov  $8,%rcx
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int64#4
+# asm 2: sub  $4,<i=%rcx
+sub  $4,%rcx
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#4,<diag0=int6464#1
+# asm 2: paddd <x0=48(%rsp),<diag0=%xmm0
+paddd 48(%rsp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#1,<diag1=int6464#2
+# asm 2: paddd <x1=0(%rsp),<diag1=%xmm1
+paddd 0(%rsp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#2,<diag2=int6464#3
+# asm 2: paddd <x2=16(%rsp),<diag2=%xmm2
+paddd 16(%rsp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#3,<diag3=int6464#4
+# asm 2: paddd <x3=32(%rsp),<diag3=%xmm3
+paddd 32(%rsp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int64#4
+# asm 2: movd   <diag0=%xmm0,>in0=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int64#5
+# asm 2: movd   <diag1=%xmm1,>in12=%r8
+movd   %xmm1,%r8
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int64#6
+# asm 2: movd   <diag2=%xmm2,>in8=%r9
+movd   %xmm2,%r9
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int64#7
+# asm 2: movd   <diag3=%xmm3,>in4=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int64#2),<in0=int64#4d
+# asm 2: xorl 0(<m=%rsi),<in0=%ecx
+xorl 0(%rsi),%ecx
+
+# qhasm: (uint32) in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int64#2),<in12=int64#5d
+# asm 2: xorl 48(<m=%rsi),<in12=%r8d
+xorl 48(%rsi),%r8d
+
+# qhasm: (uint32) in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int64#2),<in8=int64#6d
+# asm 2: xorl 32(<m=%rsi),<in8=%r9d
+xorl 32(%rsi),%r9d
+
+# qhasm: (uint32) in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int64#2),<in4=int64#7d
+# asm 2: xorl 16(<m=%rsi),<in4=%eax
+xorl 16(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl   <in0=int64#4d,0(<out=int64#1)
+# asm 2: movl   <in0=%ecx,0(<out=%rdi)
+movl   %ecx,0(%rdi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl   <in12=int64#5d,48(<out=int64#1)
+# asm 2: movl   <in12=%r8d,48(<out=%rdi)
+movl   %r8d,48(%rdi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl   <in8=int64#6d,32(<out=int64#1)
+# asm 2: movl   <in8=%r9d,32(<out=%rdi)
+movl   %r9d,32(%rdi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl   <in4=int64#7d,16(<out=int64#1)
+# asm 2: movl   <in4=%eax,16(<out=%rdi)
+movl   %eax,16(%rdi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int64#4
+# asm 2: movd   <diag0=%xmm0,>in5=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int64#5
+# asm 2: movd   <diag1=%xmm1,>in1=%r8
+movd   %xmm1,%r8
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int64#6
+# asm 2: movd   <diag2=%xmm2,>in13=%r9
+movd   %xmm2,%r9
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int64#7
+# asm 2: movd   <diag3=%xmm3,>in9=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int64#2),<in5=int64#4d
+# asm 2: xorl 20(<m=%rsi),<in5=%ecx
+xorl 20(%rsi),%ecx
+
+# qhasm: (uint32) in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int64#2),<in1=int64#5d
+# asm 2: xorl 4(<m=%rsi),<in1=%r8d
+xorl 4(%rsi),%r8d
+
+# qhasm: (uint32) in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int64#2),<in13=int64#6d
+# asm 2: xorl 52(<m=%rsi),<in13=%r9d
+xorl 52(%rsi),%r9d
+
+# qhasm: (uint32) in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int64#2),<in9=int64#7d
+# asm 2: xorl 36(<m=%rsi),<in9=%eax
+xorl 36(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl   <in5=int64#4d,20(<out=int64#1)
+# asm 2: movl   <in5=%ecx,20(<out=%rdi)
+movl   %ecx,20(%rdi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl   <in1=int64#5d,4(<out=int64#1)
+# asm 2: movl   <in1=%r8d,4(<out=%rdi)
+movl   %r8d,4(%rdi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl   <in13=int64#6d,52(<out=int64#1)
+# asm 2: movl   <in13=%r9d,52(<out=%rdi)
+movl   %r9d,52(%rdi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl   <in9=int64#7d,36(<out=int64#1)
+# asm 2: movl   <in9=%eax,36(<out=%rdi)
+movl   %eax,36(%rdi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int64#4
+# asm 2: movd   <diag0=%xmm0,>in10=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int64#5
+# asm 2: movd   <diag1=%xmm1,>in6=%r8
+movd   %xmm1,%r8
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int64#6
+# asm 2: movd   <diag2=%xmm2,>in2=%r9
+movd   %xmm2,%r9
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int64#7
+# asm 2: movd   <diag3=%xmm3,>in14=%rax
+movd   %xmm3,%rax
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: (uint32) in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int64#2),<in10=int64#4d
+# asm 2: xorl 40(<m=%rsi),<in10=%ecx
+xorl 40(%rsi),%ecx
+
+# qhasm: (uint32) in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int64#2),<in6=int64#5d
+# asm 2: xorl 24(<m=%rsi),<in6=%r8d
+xorl 24(%rsi),%r8d
+
+# qhasm: (uint32) in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int64#2),<in2=int64#6d
+# asm 2: xorl 8(<m=%rsi),<in2=%r9d
+xorl 8(%rsi),%r9d
+
+# qhasm: (uint32) in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int64#2),<in14=int64#7d
+# asm 2: xorl 56(<m=%rsi),<in14=%eax
+xorl 56(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl   <in10=int64#4d,40(<out=int64#1)
+# asm 2: movl   <in10=%ecx,40(<out=%rdi)
+movl   %ecx,40(%rdi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl   <in6=int64#5d,24(<out=int64#1)
+# asm 2: movl   <in6=%r8d,24(<out=%rdi)
+movl   %r8d,24(%rdi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl   <in2=int64#6d,8(<out=int64#1)
+# asm 2: movl   <in2=%r9d,8(<out=%rdi)
+movl   %r9d,8(%rdi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl   <in14=int64#7d,56(<out=int64#1)
+# asm 2: movl   <in14=%eax,56(<out=%rdi)
+movl   %eax,56(%rdi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int64#4
+# asm 2: movd   <diag0=%xmm0,>in15=%rcx
+movd   %xmm0,%rcx
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int64#5
+# asm 2: movd   <diag1=%xmm1,>in11=%r8
+movd   %xmm1,%r8
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int64#6
+# asm 2: movd   <diag2=%xmm2,>in7=%r9
+movd   %xmm2,%r9
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int64#7
+# asm 2: movd   <diag3=%xmm3,>in3=%rax
+movd   %xmm3,%rax
+
+# qhasm: (uint32) in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int64#2),<in15=int64#4d
+# asm 2: xorl 60(<m=%rsi),<in15=%ecx
+xorl 60(%rsi),%ecx
+
+# qhasm: (uint32) in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int64#2),<in11=int64#5d
+# asm 2: xorl 44(<m=%rsi),<in11=%r8d
+xorl 44(%rsi),%r8d
+
+# qhasm: (uint32) in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int64#2),<in7=int64#6d
+# asm 2: xorl 28(<m=%rsi),<in7=%r9d
+xorl 28(%rsi),%r9d
+
+# qhasm: (uint32) in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int64#2),<in3=int64#7d
+# asm 2: xorl 12(<m=%rsi),<in3=%eax
+xorl 12(%rsi),%eax
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl   <in15=int64#4d,60(<out=int64#1)
+# asm 2: movl   <in15=%ecx,60(<out=%rdi)
+movl   %ecx,60(%rdi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl   <in11=int64#5d,44(<out=int64#1)
+# asm 2: movl   <in11=%r8d,44(<out=%rdi)
+movl   %r8d,44(%rdi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl   <in7=int64#6d,28(<out=int64#1)
+# asm 2: movl   <in7=%r9d,28(<out=%rdi)
+movl   %r9d,28(%rdi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl   <in3=int64#7d,12(<out=int64#1)
+# asm 2: movl   <in3=%eax,12(<out=%rdi)
+movl   %eax,12(%rdi)
+
+# qhasm:   bytes = bytes_backup
+# asm 1: movq <bytes_backup=stack64#8,>bytes=int64#6
+# asm 2: movq <bytes_backup=408(%rsp),>bytes=%r9
+movq 408(%rsp),%r9
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#2,>in8=int64#4d
+# asm 2: movl <x2=16(%rsp),>in8=%ecx
+movl 16(%rsp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#3,>in9=int64#5d
+# asm 2: movl 4+<x3=32(%rsp),>in9=%r8d
+movl 4+32(%rsp),%r8d
+
+# qhasm:   in8 += 1
+# asm 1: add  $1,<in8=int64#4
+# asm 2: add  $1,<in8=%rcx
+add  $1,%rcx
+
+# qhasm:   in9 <<= 32
+# asm 1: shl  $32,<in9=int64#5
+# asm 2: shl  $32,<in9=%r8
+shl  $32,%r8
+
+# qhasm:   in8 += in9
+# asm 1: add  <in9=int64#5,<in8=int64#4
+# asm 2: add  <in9=%r8,<in8=%rcx
+add  %r8,%rcx
+
+# qhasm:   in9 = in8
+# asm 1: mov  <in8=int64#4,>in9=int64#5
+# asm 2: mov  <in8=%rcx,>in9=%r8
+mov  %rcx,%r8
+
+# qhasm:   (uint64) in9 >>= 32
+# asm 1: shr  $32,<in9=int64#5
+# asm 2: shr  $32,<in9=%r8
+shr  $32,%r8
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int64#4d,>x2=stack128#2
+# asm 2: movl <in8=%ecx,>x2=16(%rsp)
+movl %ecx,16(%rsp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int64#5d,4+<x3=stack128#3
+# asm 2: movl <in9=%r8d,4+<x3=32(%rsp)
+movl %r8d,4+32(%rsp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int64#6
+# asm 2: cmp  $64,<bytes=%r9
+cmp  $64,%r9
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int64#1,>m=int64#2
+# asm 2: mov  <out=%rdi,>m=%rsi
+mov  %rdi,%rsi
+
+# qhasm:       out = ctarget
+# asm 1: mov  <ctarget=int64#3,>out=int64#1
+# asm 2: mov  <ctarget=%rdx,>out=%rdi
+mov  %rdx,%rdi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int64#6,>i=int64#4
+# asm 2: mov  <bytes=%r9,>i=%rcx
+mov  %r9,%rcx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     r11_caller = r11_stack
+# asm 1: movq <r11_stack=stack64#1,>r11_caller=int64#9
+# asm 2: movq <r11_stack=352(%rsp),>r11_caller=%r11
+movq 352(%rsp),%r11
+
+# qhasm:     r12_caller = r12_stack
+# asm 1: movq <r12_stack=stack64#2,>r12_caller=int64#10
+# asm 2: movq <r12_stack=360(%rsp),>r12_caller=%r12
+movq 360(%rsp),%r12
+
+# qhasm:     r13_caller = r13_stack
+# asm 1: movq <r13_stack=stack64#3,>r13_caller=int64#11
+# asm 2: movq <r13_stack=368(%rsp),>r13_caller=%r13
+movq 368(%rsp),%r13
+
+# qhasm:     r14_caller = r14_stack
+# asm 1: movq <r14_stack=stack64#4,>r14_caller=int64#12
+# asm 2: movq <r14_stack=376(%rsp),>r14_caller=%r14
+movq 376(%rsp),%r14
+
+# qhasm:     r15_caller = r15_stack
+# asm 1: movq <r15_stack=stack64#5,>r15_caller=int64#13
+# asm 2: movq <r15_stack=384(%rsp),>r15_caller=%r15
+movq 384(%rsp),%r15
+
+# qhasm:     rbx_caller = rbx_stack
+# asm 1: movq <rbx_stack=stack64#6,>rbx_caller=int64#14
+# asm 2: movq <rbx_stack=392(%rsp),>rbx_caller=%rbx
+movq 392(%rsp),%rbx
+
+# qhasm:     rbp_caller = rbp_stack
+# asm 1: movq <rbp_stack=stack64#7,>rbp_caller=int64#15
+# asm 2: movq <rbp_stack=400(%rsp),>rbp_caller=%rbp
+movq 400(%rsp),%rbp
+
+# qhasm:     leave
+add %r11,%rsp
+xor %rax,%rax
+xor %rdx,%rdx
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int64#6
+# asm 2: sub  $64,<bytes=%r9
+sub  $64,%r9
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int64#1
+# asm 2: add  $64,<out=%rdi
+add  $64,%rdi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int64#2
+# asm 2: add  $64,<m=%rsi
+add  $64,%rsi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/salsa208/checksum b/nacl/crypto_stream/salsa208/checksum
new file mode 100644
index 00000000..c87364e6
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/checksum
@@ -0,0 +1 @@
+05f32b0647417aaa446b0b3127318133cf9af32b771869eab267000bf02710cd
diff --git a/nacl/crypto_stream/salsa208/ref/api.h b/nacl/crypto_stream/salsa208/ref/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/ref/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa208/ref/implementors b/nacl/crypto_stream/salsa208/ref/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/ref/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa208/ref/stream.c b/nacl/crypto_stream/salsa208/ref/stream.c
new file mode 100644
index 00000000..cdcfbc0e
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/ref/stream.c
@@ -0,0 +1,49 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa208.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream(
+        unsigned char *c,unsigned long long clen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!clen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (clen >= 64) {
+    crypto_core_salsa208(c,in,k,sigma);
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    clen -= 64;
+    c += 64;
+  }
+
+  if (clen) {
+    crypto_core_salsa208(block,in,k,sigma);
+    for (i = 0;i < clen;++i) c[i] = block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa208/ref/xor.c b/nacl/crypto_stream/salsa208/ref/xor.c
new file mode 100644
index 00000000..c017ac42
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/ref/xor.c
@@ -0,0 +1,52 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_salsa208.h"
+#include "crypto_stream.h"
+
+typedef unsigned int uint32;
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream_xor(
+        unsigned char *c,
+  const unsigned char *m,unsigned long long mlen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char in[16];
+  unsigned char block[64];
+  int i;
+  unsigned int u;
+
+  if (!mlen) return 0;
+
+  for (i = 0;i < 8;++i) in[i] = n[i];
+  for (i = 8;i < 16;++i) in[i] = 0;
+
+  while (mlen >= 64) {
+    crypto_core_salsa208(block,in,k,sigma);
+    for (i = 0;i < 64;++i) c[i] = m[i] ^ block[i];
+
+    u = 1;
+    for (i = 8;i < 16;++i) {
+      u += (unsigned int) in[i];
+      in[i] = u;
+      u >>= 8;
+    }
+
+    mlen -= 64;
+    c += 64;
+    m += 64;
+  }
+
+  if (mlen) {
+    crypto_core_salsa208(block,in,k,sigma);
+    for (i = 0;i < mlen;++i) c[i] = m[i] ^ block[i];
+  }
+  return 0;
+}
diff --git a/nacl/crypto_stream/salsa208/used b/nacl/crypto_stream/salsa208/used
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/used
diff --git a/nacl/crypto_stream/salsa208/x86_xmm5/api.h b/nacl/crypto_stream/salsa208/x86_xmm5/api.h
new file mode 100644
index 00000000..c2b18461
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/x86_xmm5/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 8
diff --git a/nacl/crypto_stream/salsa208/x86_xmm5/implementors b/nacl/crypto_stream/salsa208/x86_xmm5/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/x86_xmm5/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/salsa208/x86_xmm5/stream.s b/nacl/crypto_stream/salsa208/x86_xmm5/stream.s
new file mode 100644
index 00000000..065253a8
--- /dev/null
+++ b/nacl/crypto_stream/salsa208/x86_xmm5/stream.s
@@ -0,0 +1,5078 @@
+
+# qhasm: int32 a
+
+# qhasm: stack32 arg1
+
+# qhasm: stack32 arg2
+
+# qhasm: stack32 arg3
+
+# qhasm: stack32 arg4
+
+# qhasm: stack32 arg5
+
+# qhasm: stack32 arg6
+
+# qhasm: input arg1
+
+# qhasm: input arg2
+
+# qhasm: input arg3
+
+# qhasm: input arg4
+
+# qhasm: input arg5
+
+# qhasm: input arg6
+
+# qhasm: int32 eax
+
+# qhasm: int32 ebx
+
+# qhasm: int32 esi
+
+# qhasm: int32 edi
+
+# qhasm: int32 ebp
+
+# qhasm: caller eax
+
+# qhasm: caller ebx
+
+# qhasm: caller esi
+
+# qhasm: caller edi
+
+# qhasm: caller ebp
+
+# qhasm: int32 k
+
+# qhasm: int32 kbits
+
+# qhasm: int32 iv
+
+# qhasm: int32 i
+
+# qhasm: stack128 x0
+
+# qhasm: stack128 x1
+
+# qhasm: stack128 x2
+
+# qhasm: stack128 x3
+
+# qhasm: int32 m
+
+# qhasm: stack32 out_stack
+
+# qhasm: int32 out
+
+# qhasm: stack32 bytes_stack
+
+# qhasm: int32 bytes
+
+# qhasm: stack32 eax_stack
+
+# qhasm: stack32 ebx_stack
+
+# qhasm: stack32 esi_stack
+
+# qhasm: stack32 edi_stack
+
+# qhasm: stack32 ebp_stack
+
+# qhasm: int6464 diag0
+
+# qhasm: int6464 diag1
+
+# qhasm: int6464 diag2
+
+# qhasm: int6464 diag3
+
+# qhasm: int6464 a0
+
+# qhasm: int6464 a1
+
+# qhasm: int6464 a2
+
+# qhasm: int6464 a3
+
+# qhasm: int6464 a4
+
+# qhasm: int6464 a5
+
+# qhasm: int6464 a6
+
+# qhasm: int6464 a7
+
+# qhasm: int6464 b0
+
+# qhasm: int6464 b1
+
+# qhasm: int6464 b2
+
+# qhasm: int6464 b3
+
+# qhasm: int6464 b4
+
+# qhasm: int6464 b5
+
+# qhasm: int6464 b6
+
+# qhasm: int6464 b7
+
+# qhasm: int6464 z0
+
+# qhasm: int6464 z1
+
+# qhasm: int6464 z2
+
+# qhasm: int6464 z3
+
+# qhasm: int6464 z4
+
+# qhasm: int6464 z5
+
+# qhasm: int6464 z6
+
+# qhasm: int6464 z7
+
+# qhasm: int6464 z8
+
+# qhasm: int6464 z9
+
+# qhasm: int6464 z10
+
+# qhasm: int6464 z11
+
+# qhasm: int6464 z12
+
+# qhasm: int6464 z13
+
+# qhasm: int6464 z14
+
+# qhasm: int6464 z15
+
+# qhasm: stack128 z0_stack
+
+# qhasm: stack128 z1_stack
+
+# qhasm: stack128 z2_stack
+
+# qhasm: stack128 z3_stack
+
+# qhasm: stack128 z4_stack
+
+# qhasm: stack128 z5_stack
+
+# qhasm: stack128 z6_stack
+
+# qhasm: stack128 z7_stack
+
+# qhasm: stack128 z8_stack
+
+# qhasm: stack128 z9_stack
+
+# qhasm: stack128 z10_stack
+
+# qhasm: stack128 z11_stack
+
+# qhasm: stack128 z12_stack
+
+# qhasm: stack128 z13_stack
+
+# qhasm: stack128 z14_stack
+
+# qhasm: stack128 z15_stack
+
+# qhasm: stack128 orig0
+
+# qhasm: stack128 orig1
+
+# qhasm: stack128 orig2
+
+# qhasm: stack128 orig3
+
+# qhasm: stack128 orig4
+
+# qhasm: stack128 orig5
+
+# qhasm: stack128 orig6
+
+# qhasm: stack128 orig7
+
+# qhasm: stack128 orig8
+
+# qhasm: stack128 orig9
+
+# qhasm: stack128 orig10
+
+# qhasm: stack128 orig11
+
+# qhasm: stack128 orig12
+
+# qhasm: stack128 orig13
+
+# qhasm: stack128 orig14
+
+# qhasm: stack128 orig15
+
+# qhasm: int6464 p
+
+# qhasm: int6464 q
+
+# qhasm: int6464 r
+
+# qhasm: int6464 s
+
+# qhasm: int6464 t
+
+# qhasm: int6464 u
+
+# qhasm: int6464 v
+
+# qhasm: int6464 w
+
+# qhasm: int6464 mp
+
+# qhasm: int6464 mq
+
+# qhasm: int6464 mr
+
+# qhasm: int6464 ms
+
+# qhasm: int6464 mt
+
+# qhasm: int6464 mu
+
+# qhasm: int6464 mv
+
+# qhasm: int6464 mw
+
+# qhasm: int32 in0
+
+# qhasm: int32 in1
+
+# qhasm: int32 in2
+
+# qhasm: int32 in3
+
+# qhasm: int32 in4
+
+# qhasm: int32 in5
+
+# qhasm: int32 in6
+
+# qhasm: int32 in7
+
+# qhasm: int32 in8
+
+# qhasm: int32 in9
+
+# qhasm: int32 in10
+
+# qhasm: int32 in11
+
+# qhasm: int32 in12
+
+# qhasm: int32 in13
+
+# qhasm: int32 in14
+
+# qhasm: int32 in15
+
+# qhasm: stack512 tmp
+
+# qhasm: stack32 ctarget
+
+# qhasm: enter crypto_stream_salsa208_x86_xmm5
+.text
+.p2align 5
+.globl _crypto_stream_salsa208_x86_xmm5
+.globl crypto_stream_salsa208_x86_xmm5
+_crypto_stream_salsa208_x86_xmm5:
+crypto_stream_salsa208_x86_xmm5:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: bytes = arg2
+# asm 1: movl <arg2=stack32#-2,>bytes=int32#3
+# asm 2: movl <arg2=8(%esp,%eax),>bytes=%edx
+movl 8(%esp,%eax),%edx
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm: iv = arg4
+# asm 1: movl <arg4=stack32#-4,>iv=int32#4
+# asm 2: movl <arg4=16(%esp,%eax),>iv=%ebx
+movl 16(%esp,%eax),%ebx
+
+# qhasm: k = arg5
+# asm 1: movl <arg5=stack32#-5,>k=int32#7
+# asm 2: movl <arg5=20(%esp,%eax),>k=%ebp
+movl 20(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+
+# qhasm: a = 0
+# asm 1: mov  $0,>a=int32#1
+# asm 2: mov  $0,>a=%eax
+mov  $0,%eax
+
+# qhasm: i = bytes
+# asm 1: mov  <bytes=int32#3,>i=int32#2
+# asm 2: mov  <bytes=%edx,>i=%ecx
+mov  %edx,%ecx
+
+# qhasm: while (i) { *out++ = a; --i }
+rep stosb
+
+# qhasm: out -= bytes
+# asm 1: subl <bytes=int32#3,<out=int32#6
+# asm 2: subl <bytes=%edx,<out=%edi
+subl %edx,%edi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto start
+jmp ._start
+
+# qhasm: enter crypto_stream_salsa208_x86_xmm5_xor
+.text
+.p2align 5
+.globl _crypto_stream_salsa208_x86_xmm5_xor
+.globl crypto_stream_salsa208_x86_xmm5_xor
+_crypto_stream_salsa208_x86_xmm5_xor:
+crypto_stream_salsa208_x86_xmm5_xor:
+mov %esp,%eax
+and $31,%eax
+add $704,%eax
+sub %eax,%esp
+
+# qhasm: eax_stack = eax
+# asm 1: movl <eax=int32#1,>eax_stack=stack32#1
+# asm 2: movl <eax=%eax,>eax_stack=0(%esp)
+movl %eax,0(%esp)
+
+# qhasm: ebx_stack = ebx
+# asm 1: movl <ebx=int32#4,>ebx_stack=stack32#2
+# asm 2: movl <ebx=%ebx,>ebx_stack=4(%esp)
+movl %ebx,4(%esp)
+
+# qhasm: esi_stack = esi
+# asm 1: movl <esi=int32#5,>esi_stack=stack32#3
+# asm 2: movl <esi=%esi,>esi_stack=8(%esp)
+movl %esi,8(%esp)
+
+# qhasm: edi_stack = edi
+# asm 1: movl <edi=int32#6,>edi_stack=stack32#4
+# asm 2: movl <edi=%edi,>edi_stack=12(%esp)
+movl %edi,12(%esp)
+
+# qhasm: ebp_stack = ebp
+# asm 1: movl <ebp=int32#7,>ebp_stack=stack32#5
+# asm 2: movl <ebp=%ebp,>ebp_stack=16(%esp)
+movl %ebp,16(%esp)
+
+# qhasm: out = arg1
+# asm 1: movl <arg1=stack32#-1,>out=int32#6
+# asm 2: movl <arg1=4(%esp,%eax),>out=%edi
+movl 4(%esp,%eax),%edi
+
+# qhasm: m = arg2
+# asm 1: movl <arg2=stack32#-2,>m=int32#5
+# asm 2: movl <arg2=8(%esp,%eax),>m=%esi
+movl 8(%esp,%eax),%esi
+
+# qhasm: bytes = arg3
+# asm 1: movl <arg3=stack32#-3,>bytes=int32#3
+# asm 2: movl <arg3=12(%esp,%eax),>bytes=%edx
+movl 12(%esp,%eax),%edx
+
+# qhasm: iv = arg5
+# asm 1: movl <arg5=stack32#-5,>iv=int32#4
+# asm 2: movl <arg5=20(%esp,%eax),>iv=%ebx
+movl 20(%esp,%eax),%ebx
+
+# qhasm: k = arg6
+# asm 1: movl <arg6=stack32#-6,>k=int32#7
+# asm 2: movl <arg6=24(%esp,%eax),>k=%ebp
+movl 24(%esp,%eax),%ebp
+
+# qhasm:               unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#3
+# asm 2: cmp  $0,<bytes=%edx
+cmp  $0,%edx
+# comment:fp stack unchanged by jump
+
+# qhasm: goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: start:
+._start:
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#3,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%edx,>bytes_stack=24(%esp)
+movl %edx,24(%esp)
+
+# qhasm:   in4 = *(uint32 *) (k + 12)
+# asm 1: movl 12(<k=int32#7),>in4=int32#1
+# asm 2: movl 12(<k=%ebp),>in4=%eax
+movl 12(%ebp),%eax
+
+# qhasm:   in12 = *(uint32 *) (k + 20)
+# asm 1: movl 20(<k=int32#7),>in12=int32#2
+# asm 2: movl 20(<k=%ebp),>in12=%ecx
+movl 20(%ebp),%ecx
+
+# qhasm:   ((uint32 *)&x3)[0] = in4
+# asm 1: movl <in4=int32#1,>x3=stack128#1
+# asm 2: movl <in4=%eax,>x3=32(%esp)
+movl %eax,32(%esp)
+
+# qhasm:   ((uint32 *)&x1)[0] = in12
+# asm 1: movl <in12=int32#2,>x1=stack128#2
+# asm 2: movl <in12=%ecx,>x1=48(%esp)
+movl %ecx,48(%esp)
+
+# qhasm:   in0 = 1634760805
+# asm 1: mov  $1634760805,>in0=int32#1
+# asm 2: mov  $1634760805,>in0=%eax
+mov  $1634760805,%eax
+
+# qhasm:   in8 = 0
+# asm 1: mov  $0,>in8=int32#2
+# asm 2: mov  $0,>in8=%ecx
+mov  $0,%ecx
+
+# qhasm:   ((uint32 *)&x0)[0] = in0
+# asm 1: movl <in0=int32#1,>x0=stack128#3
+# asm 2: movl <in0=%eax,>x0=64(%esp)
+movl %eax,64(%esp)
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   in6 = *(uint32 *) (iv + 0)
+# asm 1: movl 0(<iv=int32#4),>in6=int32#1
+# asm 2: movl 0(<iv=%ebx),>in6=%eax
+movl 0(%ebx),%eax
+
+# qhasm:   in7 = *(uint32 *) (iv + 4)
+# asm 1: movl 4(<iv=int32#4),>in7=int32#2
+# asm 2: movl 4(<iv=%ebx),>in7=%ecx
+movl 4(%ebx),%ecx
+
+# qhasm:   ((uint32 *)&x1)[2] = in6
+# asm 1: movl <in6=int32#1,8+<x1=stack128#2
+# asm 2: movl <in6=%eax,8+<x1=48(%esp)
+movl %eax,8+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[3] = in7
+# asm 1: movl <in7=int32#2,12+<x2=stack128#4
+# asm 2: movl <in7=%ecx,12+<x2=80(%esp)
+movl %ecx,12+80(%esp)
+
+# qhasm:   in9 = 0
+# asm 1: mov  $0,>in9=int32#1
+# asm 2: mov  $0,>in9=%eax
+mov  $0,%eax
+
+# qhasm:   in10 = 2036477234
+# asm 1: mov  $2036477234,>in10=int32#2
+# asm 2: mov  $2036477234,>in10=%ecx
+mov  $2036477234,%ecx
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#1,4+<x3=stack128#1
+# asm 2: movl <in9=%eax,4+<x3=32(%esp)
+movl %eax,4+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[2] = in10
+# asm 1: movl <in10=int32#2,8+<x0=stack128#3
+# asm 2: movl <in10=%ecx,8+<x0=64(%esp)
+movl %ecx,8+64(%esp)
+
+# qhasm:   in1 = *(uint32 *) (k + 0)
+# asm 1: movl 0(<k=int32#7),>in1=int32#1
+# asm 2: movl 0(<k=%ebp),>in1=%eax
+movl 0(%ebp),%eax
+
+# qhasm:   in2 = *(uint32 *) (k + 4)
+# asm 1: movl 4(<k=int32#7),>in2=int32#2
+# asm 2: movl 4(<k=%ebp),>in2=%ecx
+movl 4(%ebp),%ecx
+
+# qhasm:   in3 = *(uint32 *) (k + 8)
+# asm 1: movl 8(<k=int32#7),>in3=int32#3
+# asm 2: movl 8(<k=%ebp),>in3=%edx
+movl 8(%ebp),%edx
+
+# qhasm:   in5 = 857760878
+# asm 1: mov  $857760878,>in5=int32#4
+# asm 2: mov  $857760878,>in5=%ebx
+mov  $857760878,%ebx
+
+# qhasm:   ((uint32 *)&x1)[1] = in1
+# asm 1: movl <in1=int32#1,4+<x1=stack128#2
+# asm 2: movl <in1=%eax,4+<x1=48(%esp)
+movl %eax,4+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[2] = in2
+# asm 1: movl <in2=int32#2,8+<x2=stack128#4
+# asm 2: movl <in2=%ecx,8+<x2=80(%esp)
+movl %ecx,8+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[3] = in3
+# asm 1: movl <in3=int32#3,12+<x3=stack128#1
+# asm 2: movl <in3=%edx,12+<x3=32(%esp)
+movl %edx,12+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[1] = in5
+# asm 1: movl <in5=int32#4,4+<x0=stack128#3
+# asm 2: movl <in5=%ebx,4+<x0=64(%esp)
+movl %ebx,4+64(%esp)
+
+# qhasm:   in11 = *(uint32 *) (k + 16)
+# asm 1: movl 16(<k=int32#7),>in11=int32#1
+# asm 2: movl 16(<k=%ebp),>in11=%eax
+movl 16(%ebp),%eax
+
+# qhasm:   in13 = *(uint32 *) (k + 24)
+# asm 1: movl 24(<k=int32#7),>in13=int32#2
+# asm 2: movl 24(<k=%ebp),>in13=%ecx
+movl 24(%ebp),%ecx
+
+# qhasm:   in14 = *(uint32 *) (k + 28)
+# asm 1: movl 28(<k=int32#7),>in14=int32#3
+# asm 2: movl 28(<k=%ebp),>in14=%edx
+movl 28(%ebp),%edx
+
+# qhasm:   in15 = 1797285236
+# asm 1: mov  $1797285236,>in15=int32#4
+# asm 2: mov  $1797285236,>in15=%ebx
+mov  $1797285236,%ebx
+
+# qhasm:   ((uint32 *)&x1)[3] = in11
+# asm 1: movl <in11=int32#1,12+<x1=stack128#2
+# asm 2: movl <in11=%eax,12+<x1=48(%esp)
+movl %eax,12+48(%esp)
+
+# qhasm:   ((uint32 *)&x2)[1] = in13
+# asm 1: movl <in13=int32#2,4+<x2=stack128#4
+# asm 2: movl <in13=%ecx,4+<x2=80(%esp)
+movl %ecx,4+80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[2] = in14
+# asm 1: movl <in14=int32#3,8+<x3=stack128#1
+# asm 2: movl <in14=%edx,8+<x3=32(%esp)
+movl %edx,8+32(%esp)
+
+# qhasm:   ((uint32 *)&x0)[3] = in15
+# asm 1: movl <in15=int32#4,12+<x0=stack128#3
+# asm 2: movl <in15=%ebx,12+<x0=64(%esp)
+movl %ebx,12+64(%esp)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:                               unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesbetween1and255 if unsigned<
+jb ._bytesbetween1and255
+
+# qhasm:   z0 = x0
+# asm 1: movdqa <x0=stack128#3,>z0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>z0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm:   z5 = z0[1,1,1,1]
+# asm 1: pshufd $0x55,<z0=int6464#1,>z5=int6464#2
+# asm 2: pshufd $0x55,<z0=%xmm0,>z5=%xmm1
+pshufd $0x55,%xmm0,%xmm1
+
+# qhasm:   z10 = z0[2,2,2,2]
+# asm 1: pshufd $0xaa,<z0=int6464#1,>z10=int6464#3
+# asm 2: pshufd $0xaa,<z0=%xmm0,>z10=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z15 = z0[3,3,3,3]
+# asm 1: pshufd $0xff,<z0=int6464#1,>z15=int6464#4
+# asm 2: pshufd $0xff,<z0=%xmm0,>z15=%xmm3
+pshufd $0xff,%xmm0,%xmm3
+
+# qhasm:   z0 = z0[0,0,0,0]
+# asm 1: pshufd $0x00,<z0=int6464#1,>z0=int6464#1
+# asm 2: pshufd $0x00,<z0=%xmm0,>z0=%xmm0
+pshufd $0x00,%xmm0,%xmm0
+
+# qhasm:   orig5 = z5
+# asm 1: movdqa <z5=int6464#2,>orig5=stack128#5
+# asm 2: movdqa <z5=%xmm1,>orig5=96(%esp)
+movdqa %xmm1,96(%esp)
+
+# qhasm:   orig10 = z10
+# asm 1: movdqa <z10=int6464#3,>orig10=stack128#6
+# asm 2: movdqa <z10=%xmm2,>orig10=112(%esp)
+movdqa %xmm2,112(%esp)
+
+# qhasm:   orig15 = z15
+# asm 1: movdqa <z15=int6464#4,>orig15=stack128#7
+# asm 2: movdqa <z15=%xmm3,>orig15=128(%esp)
+movdqa %xmm3,128(%esp)
+
+# qhasm:   orig0 = z0
+# asm 1: movdqa <z0=int6464#1,>orig0=stack128#8
+# asm 2: movdqa <z0=%xmm0,>orig0=144(%esp)
+movdqa %xmm0,144(%esp)
+
+# qhasm:   z1 = x1
+# asm 1: movdqa <x1=stack128#2,>z1=int6464#1
+# asm 2: movdqa <x1=48(%esp),>z1=%xmm0
+movdqa 48(%esp),%xmm0
+
+# qhasm:   z6 = z1[2,2,2,2]
+# asm 1: pshufd $0xaa,<z1=int6464#1,>z6=int6464#2
+# asm 2: pshufd $0xaa,<z1=%xmm0,>z6=%xmm1
+pshufd $0xaa,%xmm0,%xmm1
+
+# qhasm:   z11 = z1[3,3,3,3]
+# asm 1: pshufd $0xff,<z1=int6464#1,>z11=int6464#3
+# asm 2: pshufd $0xff,<z1=%xmm0,>z11=%xmm2
+pshufd $0xff,%xmm0,%xmm2
+
+# qhasm:   z12 = z1[0,0,0,0]
+# asm 1: pshufd $0x00,<z1=int6464#1,>z12=int6464#4
+# asm 2: pshufd $0x00,<z1=%xmm0,>z12=%xmm3
+pshufd $0x00,%xmm0,%xmm3
+
+# qhasm:   z1 = z1[1,1,1,1]
+# asm 1: pshufd $0x55,<z1=int6464#1,>z1=int6464#1
+# asm 2: pshufd $0x55,<z1=%xmm0,>z1=%xmm0
+pshufd $0x55,%xmm0,%xmm0
+
+# qhasm:   orig6 = z6
+# asm 1: movdqa <z6=int6464#2,>orig6=stack128#9
+# asm 2: movdqa <z6=%xmm1,>orig6=160(%esp)
+movdqa %xmm1,160(%esp)
+
+# qhasm:   orig11 = z11
+# asm 1: movdqa <z11=int6464#3,>orig11=stack128#10
+# asm 2: movdqa <z11=%xmm2,>orig11=176(%esp)
+movdqa %xmm2,176(%esp)
+
+# qhasm:   orig12 = z12
+# asm 1: movdqa <z12=int6464#4,>orig12=stack128#11
+# asm 2: movdqa <z12=%xmm3,>orig12=192(%esp)
+movdqa %xmm3,192(%esp)
+
+# qhasm:   orig1 = z1
+# asm 1: movdqa <z1=int6464#1,>orig1=stack128#12
+# asm 2: movdqa <z1=%xmm0,>orig1=208(%esp)
+movdqa %xmm0,208(%esp)
+
+# qhasm:   z2 = x2
+# asm 1: movdqa <x2=stack128#4,>z2=int6464#1
+# asm 2: movdqa <x2=80(%esp),>z2=%xmm0
+movdqa 80(%esp),%xmm0
+
+# qhasm:   z7 = z2[3,3,3,3]
+# asm 1: pshufd $0xff,<z2=int6464#1,>z7=int6464#2
+# asm 2: pshufd $0xff,<z2=%xmm0,>z7=%xmm1
+pshufd $0xff,%xmm0,%xmm1
+
+# qhasm:   z13 = z2[1,1,1,1]
+# asm 1: pshufd $0x55,<z2=int6464#1,>z13=int6464#3
+# asm 2: pshufd $0x55,<z2=%xmm0,>z13=%xmm2
+pshufd $0x55,%xmm0,%xmm2
+
+# qhasm:   z2 = z2[2,2,2,2]
+# asm 1: pshufd $0xaa,<z2=int6464#1,>z2=int6464#1
+# asm 2: pshufd $0xaa,<z2=%xmm0,>z2=%xmm0
+pshufd $0xaa,%xmm0,%xmm0
+
+# qhasm:   orig7 = z7
+# asm 1: movdqa <z7=int6464#2,>orig7=stack128#13
+# asm 2: movdqa <z7=%xmm1,>orig7=224(%esp)
+movdqa %xmm1,224(%esp)
+
+# qhasm:   orig13 = z13
+# asm 1: movdqa <z13=int6464#3,>orig13=stack128#14
+# asm 2: movdqa <z13=%xmm2,>orig13=240(%esp)
+movdqa %xmm2,240(%esp)
+
+# qhasm:   orig2 = z2
+# asm 1: movdqa <z2=int6464#1,>orig2=stack128#15
+# asm 2: movdqa <z2=%xmm0,>orig2=256(%esp)
+movdqa %xmm0,256(%esp)
+
+# qhasm:   z3 = x3
+# asm 1: movdqa <x3=stack128#1,>z3=int6464#1
+# asm 2: movdqa <x3=32(%esp),>z3=%xmm0
+movdqa 32(%esp),%xmm0
+
+# qhasm:   z4 = z3[0,0,0,0]
+# asm 1: pshufd $0x00,<z3=int6464#1,>z4=int6464#2
+# asm 2: pshufd $0x00,<z3=%xmm0,>z4=%xmm1
+pshufd $0x00,%xmm0,%xmm1
+
+# qhasm:   z14 = z3[2,2,2,2]
+# asm 1: pshufd $0xaa,<z3=int6464#1,>z14=int6464#3
+# asm 2: pshufd $0xaa,<z3=%xmm0,>z14=%xmm2
+pshufd $0xaa,%xmm0,%xmm2
+
+# qhasm:   z3 = z3[3,3,3,3]
+# asm 1: pshufd $0xff,<z3=int6464#1,>z3=int6464#1
+# asm 2: pshufd $0xff,<z3=%xmm0,>z3=%xmm0
+pshufd $0xff,%xmm0,%xmm0
+
+# qhasm:   orig4 = z4
+# asm 1: movdqa <z4=int6464#2,>orig4=stack128#16
+# asm 2: movdqa <z4=%xmm1,>orig4=272(%esp)
+movdqa %xmm1,272(%esp)
+
+# qhasm:   orig14 = z14
+# asm 1: movdqa <z14=int6464#3,>orig14=stack128#17
+# asm 2: movdqa <z14=%xmm2,>orig14=288(%esp)
+movdqa %xmm2,288(%esp)
+
+# qhasm:   orig3 = z3
+# asm 1: movdqa <z3=int6464#1,>orig3=stack128#18
+# asm 2: movdqa <z3=%xmm0,>orig3=304(%esp)
+movdqa %xmm0,304(%esp)
+
+# qhasm: bytesatleast256:
+._bytesatleast256:
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   ((uint32 *) &orig8)[0] = in8
+# asm 1: movl <in8=int32#2,>orig8=stack128#19
+# asm 2: movl <in8=%ecx,>orig8=320(%esp)
+movl %ecx,320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[0] = in9
+# asm 1: movl <in9=int32#3,>orig9=stack128#20
+# asm 2: movl <in9=%edx,>orig9=336(%esp)
+movl %edx,336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[1] = in8
+# asm 1: movl <in8=int32#2,4+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,4+<orig8=320(%esp)
+movl %ecx,4+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[1] = in9
+# asm 1: movl <in9=int32#3,4+<orig9=stack128#20
+# asm 2: movl <in9=%edx,4+<orig9=336(%esp)
+movl %edx,4+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[2] = in8
+# asm 1: movl <in8=int32#2,8+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,8+<orig8=320(%esp)
+movl %ecx,8+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[2] = in9
+# asm 1: movl <in9=int32#3,8+<orig9=stack128#20
+# asm 2: movl <in9=%edx,8+<orig9=336(%esp)
+movl %edx,8+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *) &orig8)[3] = in8
+# asm 1: movl <in8=int32#2,12+<orig8=stack128#19
+# asm 2: movl <in8=%ecx,12+<orig8=320(%esp)
+movl %ecx,12+320(%esp)
+
+# qhasm:   ((uint32 *) &orig9)[3] = in9
+# asm 1: movl <in9=int32#3,12+<orig9=stack128#20
+# asm 2: movl <in9=%edx,12+<orig9=336(%esp)
+movl %edx,12+336(%esp)
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: i = 8
+# asm 1: mov  $8,>i=int32#1
+# asm 2: mov  $8,>i=%eax
+mov  $8,%eax
+
+# qhasm:   z5 = orig5
+# asm 1: movdqa <orig5=stack128#5,>z5=int6464#1
+# asm 2: movdqa <orig5=96(%esp),>z5=%xmm0
+movdqa 96(%esp),%xmm0
+
+# qhasm:   z10 = orig10
+# asm 1: movdqa <orig10=stack128#6,>z10=int6464#2
+# asm 2: movdqa <orig10=112(%esp),>z10=%xmm1
+movdqa 112(%esp),%xmm1
+
+# qhasm:   z15 = orig15
+# asm 1: movdqa <orig15=stack128#7,>z15=int6464#3
+# asm 2: movdqa <orig15=128(%esp),>z15=%xmm2
+movdqa 128(%esp),%xmm2
+
+# qhasm:   z14 = orig14
+# asm 1: movdqa <orig14=stack128#17,>z14=int6464#4
+# asm 2: movdqa <orig14=288(%esp),>z14=%xmm3
+movdqa 288(%esp),%xmm3
+
+# qhasm:   z3 = orig3
+# asm 1: movdqa <orig3=stack128#18,>z3=int6464#5
+# asm 2: movdqa <orig3=304(%esp),>z3=%xmm4
+movdqa 304(%esp),%xmm4
+
+# qhasm:   z6 = orig6
+# asm 1: movdqa <orig6=stack128#9,>z6=int6464#6
+# asm 2: movdqa <orig6=160(%esp),>z6=%xmm5
+movdqa 160(%esp),%xmm5
+
+# qhasm:   z11 = orig11
+# asm 1: movdqa <orig11=stack128#10,>z11=int6464#7
+# asm 2: movdqa <orig11=176(%esp),>z11=%xmm6
+movdqa 176(%esp),%xmm6
+
+# qhasm:   z1 = orig1
+# asm 1: movdqa <orig1=stack128#12,>z1=int6464#8
+# asm 2: movdqa <orig1=208(%esp),>z1=%xmm7
+movdqa 208(%esp),%xmm7
+
+# qhasm:   z5_stack = z5
+# asm 1: movdqa <z5=int6464#1,>z5_stack=stack128#21
+# asm 2: movdqa <z5=%xmm0,>z5_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:   z10_stack = z10
+# asm 1: movdqa <z10=int6464#2,>z10_stack=stack128#22
+# asm 2: movdqa <z10=%xmm1,>z10_stack=368(%esp)
+movdqa %xmm1,368(%esp)
+
+# qhasm:   z15_stack = z15
+# asm 1: movdqa <z15=int6464#3,>z15_stack=stack128#23
+# asm 2: movdqa <z15=%xmm2,>z15_stack=384(%esp)
+movdqa %xmm2,384(%esp)
+
+# qhasm:   z14_stack = z14
+# asm 1: movdqa <z14=int6464#4,>z14_stack=stack128#24
+# asm 2: movdqa <z14=%xmm3,>z14_stack=400(%esp)
+movdqa %xmm3,400(%esp)
+
+# qhasm:   z3_stack = z3
+# asm 1: movdqa <z3=int6464#5,>z3_stack=stack128#25
+# asm 2: movdqa <z3=%xmm4,>z3_stack=416(%esp)
+movdqa %xmm4,416(%esp)
+
+# qhasm:   z6_stack = z6
+# asm 1: movdqa <z6=int6464#6,>z6_stack=stack128#26
+# asm 2: movdqa <z6=%xmm5,>z6_stack=432(%esp)
+movdqa %xmm5,432(%esp)
+
+# qhasm:   z11_stack = z11
+# asm 1: movdqa <z11=int6464#7,>z11_stack=stack128#27
+# asm 2: movdqa <z11=%xmm6,>z11_stack=448(%esp)
+movdqa %xmm6,448(%esp)
+
+# qhasm:   z1_stack = z1
+# asm 1: movdqa <z1=int6464#8,>z1_stack=stack128#28
+# asm 2: movdqa <z1=%xmm7,>z1_stack=464(%esp)
+movdqa %xmm7,464(%esp)
+
+# qhasm:   z7 = orig7
+# asm 1: movdqa <orig7=stack128#13,>z7=int6464#5
+# asm 2: movdqa <orig7=224(%esp),>z7=%xmm4
+movdqa 224(%esp),%xmm4
+
+# qhasm:   z13 = orig13
+# asm 1: movdqa <orig13=stack128#14,>z13=int6464#6
+# asm 2: movdqa <orig13=240(%esp),>z13=%xmm5
+movdqa 240(%esp),%xmm5
+
+# qhasm:   z2 = orig2
+# asm 1: movdqa <orig2=stack128#15,>z2=int6464#7
+# asm 2: movdqa <orig2=256(%esp),>z2=%xmm6
+movdqa 256(%esp),%xmm6
+
+# qhasm:   z9 = orig9
+# asm 1: movdqa <orig9=stack128#20,>z9=int6464#8
+# asm 2: movdqa <orig9=336(%esp),>z9=%xmm7
+movdqa 336(%esp),%xmm7
+
+# qhasm:                   p = orig0
+# asm 1: movdqa <orig0=stack128#8,>p=int6464#1
+# asm 2: movdqa <orig0=144(%esp),>p=%xmm0
+movdqa 144(%esp),%xmm0
+
+# qhasm:                   t = orig12
+# asm 1: movdqa <orig12=stack128#11,>t=int6464#3
+# asm 2: movdqa <orig12=192(%esp),>t=%xmm2
+movdqa 192(%esp),%xmm2
+
+# qhasm:                   q = orig4
+# asm 1: movdqa <orig4=stack128#16,>q=int6464#4
+# asm 2: movdqa <orig4=272(%esp),>q=%xmm3
+movdqa 272(%esp),%xmm3
+
+# qhasm:                   r = orig8
+# asm 1: movdqa <orig8=stack128#19,>r=int6464#2
+# asm 2: movdqa <orig8=320(%esp),>r=%xmm1
+movdqa 320(%esp),%xmm1
+
+# qhasm:   z7_stack = z7
+# asm 1: movdqa <z7=int6464#5,>z7_stack=stack128#29
+# asm 2: movdqa <z7=%xmm4,>z7_stack=480(%esp)
+movdqa %xmm4,480(%esp)
+
+# qhasm:   z13_stack = z13
+# asm 1: movdqa <z13=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <z13=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm:   z2_stack = z2
+# asm 1: movdqa <z2=int6464#7,>z2_stack=stack128#31
+# asm 2: movdqa <z2=%xmm6,>z2_stack=512(%esp)
+movdqa %xmm6,512(%esp)
+
+# qhasm:   z9_stack = z9
+# asm 1: movdqa <z9=int6464#8,>z9_stack=stack128#32
+# asm 2: movdqa <z9=%xmm7,>z9_stack=528(%esp)
+movdqa %xmm7,528(%esp)
+
+# qhasm:   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:   z12_stack = t
+# asm 1: movdqa <t=int6464#3,>z12_stack=stack128#34
+# asm 2: movdqa <t=%xmm2,>z12_stack=560(%esp)
+movdqa %xmm2,560(%esp)
+
+# qhasm:   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#35
+# asm 2: movdqa <q=%xmm3,>z4_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#36
+# asm 2: movdqa <r=%xmm1,>z8_stack=592(%esp)
+movdqa %xmm1,592(%esp)
+
+# qhasm: mainloop1:
+._mainloop1:
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z4_stack = q
+# asm 1: movdqa <q=int6464#4,>z4_stack=stack128#33
+# asm 2: movdqa <q=%xmm3,>z4_stack=544(%esp)
+movdqa %xmm3,544(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#34
+# asm 2: movdqa <r=%xmm1,>z8_stack=560(%esp)
+movdqa %xmm1,560(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>mt=int6464#3
+# asm 2: movdqa <z1_stack=464(%esp),>mt=%xmm2
+movdqa 464(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>mp=int6464#5
+# asm 2: movdqa <z5_stack=352(%esp),>mp=%xmm4
+movdqa 352(%esp),%xmm4
+
+# qhasm:                                   mq = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>mq=int6464#4
+# asm 2: movdqa <z9_stack=528(%esp),>mq=%xmm3
+movdqa 528(%esp),%xmm3
+
+# qhasm:                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>mr=int6464#6
+# asm 2: movdqa <z13_stack=496(%esp),>mr=%xmm5
+movdqa 496(%esp),%xmm5
+
+# qhasm:                   z12_stack = s
+# asm 1: movdqa <s=int6464#7,>z12_stack=stack128#30
+# asm 2: movdqa <s=%xmm6,>z12_stack=496(%esp)
+movdqa %xmm6,496(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#21
+# asm 2: movdqa <p=%xmm0,>z0_stack=352(%esp)
+movdqa %xmm0,352(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z9_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z9_stack=stack128#32
+# asm 2: movdqa <mq=%xmm3,>z9_stack=528(%esp)
+movdqa %xmm3,528(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#35
+# asm 2: movdqa <mr=%xmm5,>z13_stack=576(%esp)
+movdqa %xmm5,576(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>t=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>t=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>p=int6464#1
+# asm 2: movdqa <z10_stack=368(%esp),>p=%xmm0
+movdqa 368(%esp),%xmm0
+
+# qhasm:                                                   q = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>q=int6464#4
+# asm 2: movdqa <z14_stack=400(%esp),>q=%xmm3
+movdqa 400(%esp),%xmm3
+
+# qhasm:                                                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>r=int6464#2
+# asm 2: movdqa <z2_stack=512(%esp),>r=%xmm1
+movdqa 512(%esp),%xmm1
+
+# qhasm:                                   z1_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z1_stack=stack128#22
+# asm 2: movdqa <ms=%xmm6,>z1_stack=368(%esp)
+movdqa %xmm6,368(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#24
+# asm 2: movdqa <mp=%xmm4,>z5_stack=400(%esp)
+movdqa %xmm4,400(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z14_stack = q
+# asm 1: movdqa <q=int6464#4,>z14_stack=stack128#36
+# asm 2: movdqa <q=%xmm3,>z14_stack=592(%esp)
+movdqa %xmm3,592(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#26
+# asm 2: movdqa <r=%xmm1,>z2_stack=432(%esp)
+movdqa %xmm1,432(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>mt=int6464#3
+# asm 2: movdqa <z11_stack=448(%esp),>mt=%xmm2
+movdqa 448(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>mp=int6464#5
+# asm 2: movdqa <z15_stack=384(%esp),>mp=%xmm4
+movdqa 384(%esp),%xmm4
+
+# qhasm:                                                                   mq = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>mq=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>mq=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:                                                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                                                   z6_stack = s
+# asm 1: movdqa <s=int6464#7,>z6_stack=stack128#23
+# asm 2: movdqa <s=%xmm6,>z6_stack=384(%esp)
+movdqa %xmm6,384(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#27
+# asm 2: movdqa <p=%xmm0,>z10_stack=448(%esp)
+movdqa %xmm0,448(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z3_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z3_stack=stack128#25
+# asm 2: movdqa <mq=%xmm3,>z3_stack=416(%esp)
+movdqa %xmm3,416(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>t=int6464#3
+# asm 2: movdqa <z3_stack=416(%esp),>t=%xmm2
+movdqa 416(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#21,>p=int6464#1
+# asm 2: movdqa <z0_stack=352(%esp),>p=%xmm0
+movdqa 352(%esp),%xmm0
+
+# qhasm:                   q = z1_stack
+# asm 1: movdqa <z1_stack=stack128#22,>q=int6464#4
+# asm 2: movdqa <z1_stack=368(%esp),>q=%xmm3
+movdqa 368(%esp),%xmm3
+
+# qhasm:                   r = z2_stack
+# asm 1: movdqa <z2_stack=stack128#26,>r=int6464#2
+# asm 2: movdqa <z2_stack=432(%esp),>r=%xmm1
+movdqa 432(%esp),%xmm1
+
+# qhasm:                                                                   z11_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z11_stack=stack128#21
+# asm 2: movdqa <ms=%xmm6,>z11_stack=352(%esp)
+movdqa %xmm6,352(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#22
+# asm 2: movdqa <mp=%xmm4,>z15_stack=368(%esp)
+movdqa %xmm4,368(%esp)
+
+# qhasm:                   assign xmm0 to p
+
+# qhasm:                   assign xmm1 to r
+
+# qhasm:                   assign xmm2 to t
+
+# qhasm:                   assign xmm3 to q
+
+# qhasm:                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                   z1_stack = q
+# asm 1: movdqa <q=int6464#4,>z1_stack=stack128#28
+# asm 2: movdqa <q=%xmm3,>z1_stack=464(%esp)
+movdqa %xmm3,464(%esp)
+
+# qhasm:                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                   z2_stack = r
+# asm 1: movdqa <r=int6464#2,>z2_stack=stack128#31
+# asm 2: movdqa <r=%xmm1,>z2_stack=512(%esp)
+movdqa %xmm1,512(%esp)
+
+# qhasm: uint32323232      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                   mt = z4_stack
+# asm 1: movdqa <z4_stack=stack128#33,>mt=int6464#3
+# asm 2: movdqa <z4_stack=544(%esp),>mt=%xmm2
+movdqa 544(%esp),%xmm2
+
+# qhasm:                                   mp = z5_stack
+# asm 1: movdqa <z5_stack=stack128#24,>mp=int6464#5
+# asm 2: movdqa <z5_stack=400(%esp),>mp=%xmm4
+movdqa 400(%esp),%xmm4
+
+# qhasm:                                   mq = z6_stack
+# asm 1: movdqa <z6_stack=stack128#23,>mq=int6464#4
+# asm 2: movdqa <z6_stack=384(%esp),>mq=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:                                   mr = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>mr=int6464#6
+# asm 2: movdqa <z7_stack=480(%esp),>mr=%xmm5
+movdqa 480(%esp),%xmm5
+
+# qhasm:                   z3_stack = s
+# asm 1: movdqa <s=int6464#7,>z3_stack=stack128#25
+# asm 2: movdqa <s=%xmm6,>z3_stack=416(%esp)
+movdqa %xmm6,416(%esp)
+
+# qhasm: uint32323232      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                   z0_stack = p
+# asm 1: movdqa <p=int6464#1,>z0_stack=stack128#33
+# asm 2: movdqa <p=%xmm0,>z0_stack=544(%esp)
+movdqa %xmm0,544(%esp)
+
+# qhasm:                                   assign xmm2 to mt
+
+# qhasm:                                   assign xmm3 to mq
+
+# qhasm:                                   assign xmm4 to mp
+
+# qhasm:                                   assign xmm5 to mr
+
+# qhasm:                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                   z6_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z6_stack=stack128#26
+# asm 2: movdqa <mq=%xmm3,>z6_stack=432(%esp)
+movdqa %xmm3,432(%esp)
+
+# qhasm:                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                   z7_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z7_stack=stack128#29
+# asm 2: movdqa <mr=%xmm5,>z7_stack=480(%esp)
+movdqa %xmm5,480(%esp)
+
+# qhasm: uint32323232                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                                                   t = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>t=int6464#3
+# asm 2: movdqa <z9_stack=528(%esp),>t=%xmm2
+movdqa 528(%esp),%xmm2
+
+# qhasm:                                                   p = z10_stack
+# asm 1: movdqa <z10_stack=stack128#27,>p=int6464#1
+# asm 2: movdqa <z10_stack=448(%esp),>p=%xmm0
+movdqa 448(%esp),%xmm0
+
+# qhasm:                                                   q = z11_stack
+# asm 1: movdqa <z11_stack=stack128#21,>q=int6464#4
+# asm 2: movdqa <z11_stack=352(%esp),>q=%xmm3
+movdqa 352(%esp),%xmm3
+
+# qhasm:                                                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#34,>r=int6464#2
+# asm 2: movdqa <z8_stack=560(%esp),>r=%xmm1
+movdqa 560(%esp),%xmm1
+
+# qhasm:                                   z4_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z4_stack=stack128#34
+# asm 2: movdqa <ms=%xmm6,>z4_stack=560(%esp)
+movdqa %xmm6,560(%esp)
+
+# qhasm: uint32323232                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                   z5_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z5_stack=stack128#21
+# asm 2: movdqa <mp=%xmm4,>z5_stack=352(%esp)
+movdqa %xmm4,352(%esp)
+
+# qhasm:                                                   assign xmm0 to p
+
+# qhasm:                                                   assign xmm1 to r
+
+# qhasm:                                                   assign xmm2 to t
+
+# qhasm:                                                   assign xmm3 to q
+
+# qhasm:                                                   s = t
+# asm 1: movdqa <t=int6464#3,>s=int6464#7
+# asm 2: movdqa <t=%xmm2,>s=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                      t += p
+# asm 1: paddd <p=int6464#1,<t=int6464#3
+# asm 2: paddd <p=%xmm0,<t=%xmm2
+paddd %xmm0,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 25
+# asm 1: psrld $25,<t=int6464#3
+# asm 2: psrld $25,<t=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                   q ^= t
+# asm 1: pxor  <t=int6464#3,<q=int6464#4
+# asm 2: pxor  <t=%xmm2,<q=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                      u <<= 7
+# asm 1: pslld $7,<u=int6464#5
+# asm 2: pslld $7,<u=%xmm4
+pslld $7,%xmm4
+
+# qhasm:                                                   q ^= u
+# asm 1: pxor  <u=int6464#5,<q=int6464#4
+# asm 2: pxor  <u=%xmm4,<q=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                                   z11_stack = q
+# asm 1: movdqa <q=int6464#4,>z11_stack=stack128#27
+# asm 2: movdqa <q=%xmm3,>z11_stack=448(%esp)
+movdqa %xmm3,448(%esp)
+
+# qhasm:                                                   t = p
+# asm 1: movdqa <p=int6464#1,>t=int6464#3
+# asm 2: movdqa <p=%xmm0,>t=%xmm2
+movdqa %xmm0,%xmm2
+
+# qhasm: uint32323232                                      t += q
+# asm 1: paddd <q=int6464#4,<t=int6464#3
+# asm 2: paddd <q=%xmm3,<t=%xmm2
+paddd %xmm3,%xmm2
+
+# qhasm:                                                   u = t
+# asm 1: movdqa <t=int6464#3,>u=int6464#5
+# asm 2: movdqa <t=%xmm2,>u=%xmm4
+movdqa %xmm2,%xmm4
+
+# qhasm: uint32323232                                      t >>= 23
+# asm 1: psrld $23,<t=int6464#3
+# asm 2: psrld $23,<t=%xmm2
+psrld $23,%xmm2
+
+# qhasm:                                                   r ^= t
+# asm 1: pxor  <t=int6464#3,<r=int6464#2
+# asm 2: pxor  <t=%xmm2,<r=%xmm1
+pxor  %xmm2,%xmm1
+
+# qhasm: uint32323232                                      u <<= 9
+# asm 1: pslld $9,<u=int6464#5
+# asm 2: pslld $9,<u=%xmm4
+pslld $9,%xmm4
+
+# qhasm:                                                   r ^= u
+# asm 1: pxor  <u=int6464#5,<r=int6464#2
+# asm 2: pxor  <u=%xmm4,<r=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                                   z8_stack = r
+# asm 1: movdqa <r=int6464#2,>z8_stack=stack128#37
+# asm 2: movdqa <r=%xmm1,>z8_stack=608(%esp)
+movdqa %xmm1,608(%esp)
+
+# qhasm: uint32323232                                      q += r
+# asm 1: paddd <r=int6464#2,<q=int6464#4
+# asm 2: paddd <r=%xmm1,<q=%xmm3
+paddd %xmm1,%xmm3
+
+# qhasm:                                                   u = q
+# asm 1: movdqa <q=int6464#4,>u=int6464#3
+# asm 2: movdqa <q=%xmm3,>u=%xmm2
+movdqa %xmm3,%xmm2
+
+# qhasm: uint32323232                                      q >>= 19
+# asm 1: psrld $19,<q=int6464#4
+# asm 2: psrld $19,<q=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                   s ^= q
+# asm 1: pxor  <q=int6464#4,<s=int6464#7
+# asm 2: pxor  <q=%xmm3,<s=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                      u <<= 13
+# asm 1: pslld $13,<u=int6464#3
+# asm 2: pslld $13,<u=%xmm2
+pslld $13,%xmm2
+
+# qhasm:                                                   s ^= u
+# asm 1: pxor  <u=int6464#3,<s=int6464#7
+# asm 2: pxor  <u=%xmm2,<s=%xmm6
+pxor  %xmm2,%xmm6
+
+# qhasm:                                                                   mt = z14_stack
+# asm 1: movdqa <z14_stack=stack128#36,>mt=int6464#3
+# asm 2: movdqa <z14_stack=592(%esp),>mt=%xmm2
+movdqa 592(%esp),%xmm2
+
+# qhasm:                                                                   mp = z15_stack
+# asm 1: movdqa <z15_stack=stack128#22,>mp=int6464#5
+# asm 2: movdqa <z15_stack=368(%esp),>mp=%xmm4
+movdqa 368(%esp),%xmm4
+
+# qhasm:                                                                   mq = z12_stack
+# asm 1: movdqa <z12_stack=stack128#30,>mq=int6464#4
+# asm 2: movdqa <z12_stack=496(%esp),>mq=%xmm3
+movdqa 496(%esp),%xmm3
+
+# qhasm:                                                                   mr = z13_stack
+# asm 1: movdqa <z13_stack=stack128#35,>mr=int6464#6
+# asm 2: movdqa <z13_stack=576(%esp),>mr=%xmm5
+movdqa 576(%esp),%xmm5
+
+# qhasm:                                                   z9_stack = s
+# asm 1: movdqa <s=int6464#7,>z9_stack=stack128#32
+# asm 2: movdqa <s=%xmm6,>z9_stack=528(%esp)
+movdqa %xmm6,528(%esp)
+
+# qhasm: uint32323232                                      r += s
+# asm 1: paddd <s=int6464#7,<r=int6464#2
+# asm 2: paddd <s=%xmm6,<r=%xmm1
+paddd %xmm6,%xmm1
+
+# qhasm:                                                   u = r
+# asm 1: movdqa <r=int6464#2,>u=int6464#7
+# asm 2: movdqa <r=%xmm1,>u=%xmm6
+movdqa %xmm1,%xmm6
+
+# qhasm: uint32323232                                      r >>= 14
+# asm 1: psrld $14,<r=int6464#2
+# asm 2: psrld $14,<r=%xmm1
+psrld $14,%xmm1
+
+# qhasm:                                                   p ^= r
+# asm 1: pxor  <r=int6464#2,<p=int6464#1
+# asm 2: pxor  <r=%xmm1,<p=%xmm0
+pxor  %xmm1,%xmm0
+
+# qhasm: uint32323232                                      u <<= 18
+# asm 1: pslld $18,<u=int6464#7
+# asm 2: pslld $18,<u=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                   p ^= u
+# asm 1: pxor  <u=int6464#7,<p=int6464#1
+# asm 2: pxor  <u=%xmm6,<p=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm:                                                   z10_stack = p
+# asm 1: movdqa <p=int6464#1,>z10_stack=stack128#22
+# asm 2: movdqa <p=%xmm0,>z10_stack=368(%esp)
+movdqa %xmm0,368(%esp)
+
+# qhasm:                                                                   assign xmm2 to mt
+
+# qhasm:                                                                   assign xmm3 to mq
+
+# qhasm:                                                                   assign xmm4 to mp
+
+# qhasm:                                                                   assign xmm5 to mr
+
+# qhasm:                                                                   ms = mt
+# asm 1: movdqa <mt=int6464#3,>ms=int6464#7
+# asm 2: movdqa <mt=%xmm2,>ms=%xmm6
+movdqa %xmm2,%xmm6
+
+# qhasm: uint32323232                                                      mt += mp
+# asm 1: paddd <mp=int6464#5,<mt=int6464#3
+# asm 2: paddd <mp=%xmm4,<mt=%xmm2
+paddd %xmm4,%xmm2
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#3,>mu=int6464#1
+# asm 2: movdqa <mt=%xmm2,>mu=%xmm0
+movdqa %xmm2,%xmm0
+
+# qhasm: uint32323232                                                      mt >>= 25
+# asm 1: psrld $25,<mt=int6464#3
+# asm 2: psrld $25,<mt=%xmm2
+psrld $25,%xmm2
+
+# qhasm:                                                                   mq ^= mt
+# asm 1: pxor  <mt=int6464#3,<mq=int6464#4
+# asm 2: pxor  <mt=%xmm2,<mq=%xmm3
+pxor  %xmm2,%xmm3
+
+# qhasm: uint32323232                                                      mu <<= 7
+# asm 1: pslld $7,<mu=int6464#1
+# asm 2: pslld $7,<mu=%xmm0
+pslld $7,%xmm0
+
+# qhasm:                                                                   mq ^= mu
+# asm 1: pxor  <mu=int6464#1,<mq=int6464#4
+# asm 2: pxor  <mu=%xmm0,<mq=%xmm3
+pxor  %xmm0,%xmm3
+
+# qhasm:                                                                   z12_stack = mq
+# asm 1: movdqa <mq=int6464#4,>z12_stack=stack128#35
+# asm 2: movdqa <mq=%xmm3,>z12_stack=576(%esp)
+movdqa %xmm3,576(%esp)
+
+# qhasm:                                                                   mt = mp
+# asm 1: movdqa <mp=int6464#5,>mt=int6464#1
+# asm 2: movdqa <mp=%xmm4,>mt=%xmm0
+movdqa %xmm4,%xmm0
+
+# qhasm: uint32323232                                                      mt += mq
+# asm 1: paddd <mq=int6464#4,<mt=int6464#1
+# asm 2: paddd <mq=%xmm3,<mt=%xmm0
+paddd %xmm3,%xmm0
+
+# qhasm:                                                                   mu = mt
+# asm 1: movdqa <mt=int6464#1,>mu=int6464#2
+# asm 2: movdqa <mt=%xmm0,>mu=%xmm1
+movdqa %xmm0,%xmm1
+
+# qhasm: uint32323232                                                      mt >>= 23
+# asm 1: psrld $23,<mt=int6464#1
+# asm 2: psrld $23,<mt=%xmm0
+psrld $23,%xmm0
+
+# qhasm:                                                                   mr ^= mt
+# asm 1: pxor  <mt=int6464#1,<mr=int6464#6
+# asm 2: pxor  <mt=%xmm0,<mr=%xmm5
+pxor  %xmm0,%xmm5
+
+# qhasm: uint32323232                                                      mu <<= 9
+# asm 1: pslld $9,<mu=int6464#2
+# asm 2: pslld $9,<mu=%xmm1
+pslld $9,%xmm1
+
+# qhasm:                                                                   mr ^= mu
+# asm 1: pxor  <mu=int6464#2,<mr=int6464#6
+# asm 2: pxor  <mu=%xmm1,<mr=%xmm5
+pxor  %xmm1,%xmm5
+
+# qhasm:                                                                   z13_stack = mr
+# asm 1: movdqa <mr=int6464#6,>z13_stack=stack128#30
+# asm 2: movdqa <mr=%xmm5,>z13_stack=496(%esp)
+movdqa %xmm5,496(%esp)
+
+# qhasm: uint32323232                                                      mq += mr
+# asm 1: paddd <mr=int6464#6,<mq=int6464#4
+# asm 2: paddd <mr=%xmm5,<mq=%xmm3
+paddd %xmm5,%xmm3
+
+# qhasm:                                                                   mu = mq
+# asm 1: movdqa <mq=int6464#4,>mu=int6464#1
+# asm 2: movdqa <mq=%xmm3,>mu=%xmm0
+movdqa %xmm3,%xmm0
+
+# qhasm: uint32323232                                                      mq >>= 19
+# asm 1: psrld $19,<mq=int6464#4
+# asm 2: psrld $19,<mq=%xmm3
+psrld $19,%xmm3
+
+# qhasm:                                                                   ms ^= mq
+# asm 1: pxor  <mq=int6464#4,<ms=int6464#7
+# asm 2: pxor  <mq=%xmm3,<ms=%xmm6
+pxor  %xmm3,%xmm6
+
+# qhasm: uint32323232                                                      mu <<= 13
+# asm 1: pslld $13,<mu=int6464#1
+# asm 2: pslld $13,<mu=%xmm0
+pslld $13,%xmm0
+
+# qhasm:                                                                   ms ^= mu
+# asm 1: pxor  <mu=int6464#1,<ms=int6464#7
+# asm 2: pxor  <mu=%xmm0,<ms=%xmm6
+pxor  %xmm0,%xmm6
+
+# qhasm:                   t = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>t=int6464#3
+# asm 2: movdqa <z12_stack=576(%esp),>t=%xmm2
+movdqa 576(%esp),%xmm2
+
+# qhasm:                   p = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>p=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>p=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:                   q = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>q=int6464#4
+# asm 2: movdqa <z4_stack=560(%esp),>q=%xmm3
+movdqa 560(%esp),%xmm3
+
+# qhasm:                   r = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>r=int6464#2
+# asm 2: movdqa <z8_stack=608(%esp),>r=%xmm1
+movdqa 608(%esp),%xmm1
+
+# qhasm:                                                                   z14_stack = ms
+# asm 1: movdqa <ms=int6464#7,>z14_stack=stack128#24
+# asm 2: movdqa <ms=%xmm6,>z14_stack=400(%esp)
+movdqa %xmm6,400(%esp)
+
+# qhasm: uint32323232                                                      mr += ms
+# asm 1: paddd <ms=int6464#7,<mr=int6464#6
+# asm 2: paddd <ms=%xmm6,<mr=%xmm5
+paddd %xmm6,%xmm5
+
+# qhasm:                                                                   mu = mr
+# asm 1: movdqa <mr=int6464#6,>mu=int6464#7
+# asm 2: movdqa <mr=%xmm5,>mu=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                      mr >>= 14
+# asm 1: psrld $14,<mr=int6464#6
+# asm 2: psrld $14,<mr=%xmm5
+psrld $14,%xmm5
+
+# qhasm:                                                                   mp ^= mr
+# asm 1: pxor  <mr=int6464#6,<mp=int6464#5
+# asm 2: pxor  <mr=%xmm5,<mp=%xmm4
+pxor  %xmm5,%xmm4
+
+# qhasm: uint32323232                                                      mu <<= 18
+# asm 1: pslld $18,<mu=int6464#7
+# asm 2: pslld $18,<mu=%xmm6
+pslld $18,%xmm6
+
+# qhasm:                                                                   mp ^= mu
+# asm 1: pxor  <mu=int6464#7,<mp=int6464#5
+# asm 2: pxor  <mu=%xmm6,<mp=%xmm4
+pxor  %xmm6,%xmm4
+
+# qhasm:                                                                   z15_stack = mp
+# asm 1: movdqa <mp=int6464#5,>z15_stack=stack128#23
+# asm 2: movdqa <mp=%xmm4,>z15_stack=384(%esp)
+movdqa %xmm4,384(%esp)
+
+# qhasm:                   unsigned>? i -= 2
+# asm 1: sub  $2,<i=int32#1
+# asm 2: sub  $2,<i=%eax
+sub  $2,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop1 if unsigned>
+ja ._mainloop1
+
+# qhasm:   out = out_stack
+# asm 1: movl <out_stack=stack32#6,>out=int32#6
+# asm 2: movl <out_stack=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:   z0 = z0_stack
+# asm 1: movdqa <z0_stack=stack128#33,>z0=int6464#1
+# asm 2: movdqa <z0_stack=544(%esp),>z0=%xmm0
+movdqa 544(%esp),%xmm0
+
+# qhasm:   z1 = z1_stack
+# asm 1: movdqa <z1_stack=stack128#28,>z1=int6464#2
+# asm 2: movdqa <z1_stack=464(%esp),>z1=%xmm1
+movdqa 464(%esp),%xmm1
+
+# qhasm:   z2 = z2_stack
+# asm 1: movdqa <z2_stack=stack128#31,>z2=int6464#3
+# asm 2: movdqa <z2_stack=512(%esp),>z2=%xmm2
+movdqa 512(%esp),%xmm2
+
+# qhasm:   z3 = z3_stack
+# asm 1: movdqa <z3_stack=stack128#25,>z3=int6464#4
+# asm 2: movdqa <z3_stack=416(%esp),>z3=%xmm3
+movdqa 416(%esp),%xmm3
+
+# qhasm:   uint32323232 z0 += orig0
+# asm 1: paddd <orig0=stack128#8,<z0=int6464#1
+# asm 2: paddd <orig0=144(%esp),<z0=%xmm0
+paddd 144(%esp),%xmm0
+
+# qhasm:   uint32323232 z1 += orig1
+# asm 1: paddd <orig1=stack128#12,<z1=int6464#2
+# asm 2: paddd <orig1=208(%esp),<z1=%xmm1
+paddd 208(%esp),%xmm1
+
+# qhasm:   uint32323232 z2 += orig2
+# asm 1: paddd <orig2=stack128#15,<z2=int6464#3
+# asm 2: paddd <orig2=256(%esp),<z2=%xmm2
+paddd 256(%esp),%xmm2
+
+# qhasm:   uint32323232 z3 += orig3
+# asm 1: paddd <orig3=stack128#18,<z3=int6464#4
+# asm 2: paddd <orig3=304(%esp),<z3=%xmm3
+paddd 304(%esp),%xmm3
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm:   *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm:   *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm:   *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 64)
+# asm 1: xorl 64(<m=int32#5),<in0=int32#1
+# asm 2: xorl 64(<m=%esi),<in0=%eax
+xorl 64(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 68)
+# asm 1: xorl 68(<m=int32#5),<in1=int32#2
+# asm 2: xorl 68(<m=%esi),<in1=%ecx
+xorl 68(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 72)
+# asm 1: xorl 72(<m=int32#5),<in2=int32#3
+# asm 2: xorl 72(<m=%esi),<in2=%edx
+xorl 72(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 76)
+# asm 1: xorl 76(<m=int32#5),<in3=int32#4
+# asm 2: xorl 76(<m=%esi),<in3=%ebx
+xorl 76(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 64) = in0
+# asm 1: movl <in0=int32#1,64(<out=int32#6)
+# asm 2: movl <in0=%eax,64(<out=%edi)
+movl %eax,64(%edi)
+
+# qhasm:   *(uint32 *) (out + 68) = in1
+# asm 1: movl <in1=int32#2,68(<out=int32#6)
+# asm 2: movl <in1=%ecx,68(<out=%edi)
+movl %ecx,68(%edi)
+
+# qhasm:   *(uint32 *) (out + 72) = in2
+# asm 1: movl <in2=int32#3,72(<out=int32#6)
+# asm 2: movl <in2=%edx,72(<out=%edi)
+movl %edx,72(%edi)
+
+# qhasm:   *(uint32 *) (out + 76) = in3
+# asm 1: movl <in3=int32#4,76(<out=int32#6)
+# asm 2: movl <in3=%ebx,76(<out=%edi)
+movl %ebx,76(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z0 <<<= 96
+# asm 1: pshufd $0x39,<z0=int6464#1,<z0=int6464#1
+# asm 2: pshufd $0x39,<z0=%xmm0,<z0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z1 <<<= 96
+# asm 1: pshufd $0x39,<z1=int6464#2,<z1=int6464#2
+# asm 2: pshufd $0x39,<z1=%xmm1,<z1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z2 <<<= 96
+# asm 1: pshufd $0x39,<z2=int6464#3,<z2=int6464#3
+# asm 2: pshufd $0x39,<z2=%xmm2,<z2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z3 <<<= 96
+# asm 1: pshufd $0x39,<z3=int6464#4,<z3=int6464#4
+# asm 2: pshufd $0x39,<z3=%xmm3,<z3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in0 ^= *(uint32 *) (m + 128)
+# asm 1: xorl 128(<m=int32#5),<in0=int32#1
+# asm 2: xorl 128(<m=%esi),<in0=%eax
+xorl 128(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 132)
+# asm 1: xorl 132(<m=int32#5),<in1=int32#2
+# asm 2: xorl 132(<m=%esi),<in1=%ecx
+xorl 132(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 136)
+# asm 1: xorl 136(<m=int32#5),<in2=int32#3
+# asm 2: xorl 136(<m=%esi),<in2=%edx
+xorl 136(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 140)
+# asm 1: xorl 140(<m=int32#5),<in3=int32#4
+# asm 2: xorl 140(<m=%esi),<in3=%ebx
+xorl 140(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 128) = in0
+# asm 1: movl <in0=int32#1,128(<out=int32#6)
+# asm 2: movl <in0=%eax,128(<out=%edi)
+movl %eax,128(%edi)
+
+# qhasm:   *(uint32 *) (out + 132) = in1
+# asm 1: movl <in1=int32#2,132(<out=int32#6)
+# asm 2: movl <in1=%ecx,132(<out=%edi)
+movl %ecx,132(%edi)
+
+# qhasm:   *(uint32 *) (out + 136) = in2
+# asm 1: movl <in2=int32#3,136(<out=int32#6)
+# asm 2: movl <in2=%edx,136(<out=%edi)
+movl %edx,136(%edi)
+
+# qhasm:   *(uint32 *) (out + 140) = in3
+# asm 1: movl <in3=int32#4,140(<out=int32#6)
+# asm 2: movl <in3=%ebx,140(<out=%edi)
+movl %ebx,140(%edi)
+
+# qhasm:   in0 = z0
+# asm 1: movd   <z0=int6464#1,>in0=int32#1
+# asm 2: movd   <z0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in1 = z1
+# asm 1: movd   <z1=int6464#2,>in1=int32#2
+# asm 2: movd   <z1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in2 = z2
+# asm 1: movd   <z2=int6464#3,>in2=int32#3
+# asm 2: movd   <z2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in3 = z3
+# asm 1: movd   <z3=int6464#4,>in3=int32#4
+# asm 2: movd   <z3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in0 ^= *(uint32 *) (m + 192)
+# asm 1: xorl 192(<m=int32#5),<in0=int32#1
+# asm 2: xorl 192(<m=%esi),<in0=%eax
+xorl 192(%esi),%eax
+
+# qhasm:   in1 ^= *(uint32 *) (m + 196)
+# asm 1: xorl 196(<m=int32#5),<in1=int32#2
+# asm 2: xorl 196(<m=%esi),<in1=%ecx
+xorl 196(%esi),%ecx
+
+# qhasm:   in2 ^= *(uint32 *) (m + 200)
+# asm 1: xorl 200(<m=int32#5),<in2=int32#3
+# asm 2: xorl 200(<m=%esi),<in2=%edx
+xorl 200(%esi),%edx
+
+# qhasm:   in3 ^= *(uint32 *) (m + 204)
+# asm 1: xorl 204(<m=int32#5),<in3=int32#4
+# asm 2: xorl 204(<m=%esi),<in3=%ebx
+xorl 204(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 192) = in0
+# asm 1: movl <in0=int32#1,192(<out=int32#6)
+# asm 2: movl <in0=%eax,192(<out=%edi)
+movl %eax,192(%edi)
+
+# qhasm:   *(uint32 *) (out + 196) = in1
+# asm 1: movl <in1=int32#2,196(<out=int32#6)
+# asm 2: movl <in1=%ecx,196(<out=%edi)
+movl %ecx,196(%edi)
+
+# qhasm:   *(uint32 *) (out + 200) = in2
+# asm 1: movl <in2=int32#3,200(<out=int32#6)
+# asm 2: movl <in2=%edx,200(<out=%edi)
+movl %edx,200(%edi)
+
+# qhasm:   *(uint32 *) (out + 204) = in3
+# asm 1: movl <in3=int32#4,204(<out=int32#6)
+# asm 2: movl <in3=%ebx,204(<out=%edi)
+movl %ebx,204(%edi)
+
+# qhasm:   z4 = z4_stack
+# asm 1: movdqa <z4_stack=stack128#34,>z4=int6464#1
+# asm 2: movdqa <z4_stack=560(%esp),>z4=%xmm0
+movdqa 560(%esp),%xmm0
+
+# qhasm:   z5 = z5_stack
+# asm 1: movdqa <z5_stack=stack128#21,>z5=int6464#2
+# asm 2: movdqa <z5_stack=352(%esp),>z5=%xmm1
+movdqa 352(%esp),%xmm1
+
+# qhasm:   z6 = z6_stack
+# asm 1: movdqa <z6_stack=stack128#26,>z6=int6464#3
+# asm 2: movdqa <z6_stack=432(%esp),>z6=%xmm2
+movdqa 432(%esp),%xmm2
+
+# qhasm:   z7 = z7_stack
+# asm 1: movdqa <z7_stack=stack128#29,>z7=int6464#4
+# asm 2: movdqa <z7_stack=480(%esp),>z7=%xmm3
+movdqa 480(%esp),%xmm3
+
+# qhasm:   uint32323232 z4 += orig4
+# asm 1: paddd <orig4=stack128#16,<z4=int6464#1
+# asm 2: paddd <orig4=272(%esp),<z4=%xmm0
+paddd 272(%esp),%xmm0
+
+# qhasm:   uint32323232 z5 += orig5
+# asm 1: paddd <orig5=stack128#5,<z5=int6464#2
+# asm 2: paddd <orig5=96(%esp),<z5=%xmm1
+paddd 96(%esp),%xmm1
+
+# qhasm:   uint32323232 z6 += orig6
+# asm 1: paddd <orig6=stack128#9,<z6=int6464#3
+# asm 2: paddd <orig6=160(%esp),<z6=%xmm2
+paddd 160(%esp),%xmm2
+
+# qhasm:   uint32323232 z7 += orig7
+# asm 1: paddd <orig7=stack128#13,<z7=int6464#4
+# asm 2: paddd <orig7=224(%esp),<z7=%xmm3
+paddd 224(%esp),%xmm3
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#1
+# asm 2: xorl 16(<m=%esi),<in4=%eax
+xorl 16(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#2
+# asm 2: xorl 20(<m=%esi),<in5=%ecx
+xorl 20(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#3
+# asm 2: xorl 24(<m=%esi),<in6=%edx
+xorl 24(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#4
+# asm 2: xorl 28(<m=%esi),<in7=%ebx
+xorl 28(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#1,16(<out=int32#6)
+# asm 2: movl <in4=%eax,16(<out=%edi)
+movl %eax,16(%edi)
+
+# qhasm:   *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#2,20(<out=int32#6)
+# asm 2: movl <in5=%ecx,20(<out=%edi)
+movl %ecx,20(%edi)
+
+# qhasm:   *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#3,24(<out=int32#6)
+# asm 2: movl <in6=%edx,24(<out=%edi)
+movl %edx,24(%edi)
+
+# qhasm:   *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#4,28(<out=int32#6)
+# asm 2: movl <in7=%ebx,28(<out=%edi)
+movl %ebx,28(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 80)
+# asm 1: xorl 80(<m=int32#5),<in4=int32#1
+# asm 2: xorl 80(<m=%esi),<in4=%eax
+xorl 80(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 84)
+# asm 1: xorl 84(<m=int32#5),<in5=int32#2
+# asm 2: xorl 84(<m=%esi),<in5=%ecx
+xorl 84(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 88)
+# asm 1: xorl 88(<m=int32#5),<in6=int32#3
+# asm 2: xorl 88(<m=%esi),<in6=%edx
+xorl 88(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 92)
+# asm 1: xorl 92(<m=int32#5),<in7=int32#4
+# asm 2: xorl 92(<m=%esi),<in7=%ebx
+xorl 92(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 80) = in4
+# asm 1: movl <in4=int32#1,80(<out=int32#6)
+# asm 2: movl <in4=%eax,80(<out=%edi)
+movl %eax,80(%edi)
+
+# qhasm:   *(uint32 *) (out + 84) = in5
+# asm 1: movl <in5=int32#2,84(<out=int32#6)
+# asm 2: movl <in5=%ecx,84(<out=%edi)
+movl %ecx,84(%edi)
+
+# qhasm:   *(uint32 *) (out + 88) = in6
+# asm 1: movl <in6=int32#3,88(<out=int32#6)
+# asm 2: movl <in6=%edx,88(<out=%edi)
+movl %edx,88(%edi)
+
+# qhasm:   *(uint32 *) (out + 92) = in7
+# asm 1: movl <in7=int32#4,92(<out=int32#6)
+# asm 2: movl <in7=%ebx,92(<out=%edi)
+movl %ebx,92(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z4 <<<= 96
+# asm 1: pshufd $0x39,<z4=int6464#1,<z4=int6464#1
+# asm 2: pshufd $0x39,<z4=%xmm0,<z4=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z5 <<<= 96
+# asm 1: pshufd $0x39,<z5=int6464#2,<z5=int6464#2
+# asm 2: pshufd $0x39,<z5=%xmm1,<z5=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z6 <<<= 96
+# asm 1: pshufd $0x39,<z6=int6464#3,<z6=int6464#3
+# asm 2: pshufd $0x39,<z6=%xmm2,<z6=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z7 <<<= 96
+# asm 1: pshufd $0x39,<z7=int6464#4,<z7=int6464#4
+# asm 2: pshufd $0x39,<z7=%xmm3,<z7=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in4 ^= *(uint32 *) (m + 144)
+# asm 1: xorl 144(<m=int32#5),<in4=int32#1
+# asm 2: xorl 144(<m=%esi),<in4=%eax
+xorl 144(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 148)
+# asm 1: xorl 148(<m=int32#5),<in5=int32#2
+# asm 2: xorl 148(<m=%esi),<in5=%ecx
+xorl 148(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 152)
+# asm 1: xorl 152(<m=int32#5),<in6=int32#3
+# asm 2: xorl 152(<m=%esi),<in6=%edx
+xorl 152(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 156)
+# asm 1: xorl 156(<m=int32#5),<in7=int32#4
+# asm 2: xorl 156(<m=%esi),<in7=%ebx
+xorl 156(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 144) = in4
+# asm 1: movl <in4=int32#1,144(<out=int32#6)
+# asm 2: movl <in4=%eax,144(<out=%edi)
+movl %eax,144(%edi)
+
+# qhasm:   *(uint32 *) (out + 148) = in5
+# asm 1: movl <in5=int32#2,148(<out=int32#6)
+# asm 2: movl <in5=%ecx,148(<out=%edi)
+movl %ecx,148(%edi)
+
+# qhasm:   *(uint32 *) (out + 152) = in6
+# asm 1: movl <in6=int32#3,152(<out=int32#6)
+# asm 2: movl <in6=%edx,152(<out=%edi)
+movl %edx,152(%edi)
+
+# qhasm:   *(uint32 *) (out + 156) = in7
+# asm 1: movl <in7=int32#4,156(<out=int32#6)
+# asm 2: movl <in7=%ebx,156(<out=%edi)
+movl %ebx,156(%edi)
+
+# qhasm:   in4 = z4
+# asm 1: movd   <z4=int6464#1,>in4=int32#1
+# asm 2: movd   <z4=%xmm0,>in4=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in5 = z5
+# asm 1: movd   <z5=int6464#2,>in5=int32#2
+# asm 2: movd   <z5=%xmm1,>in5=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in6 = z6
+# asm 1: movd   <z6=int6464#3,>in6=int32#3
+# asm 2: movd   <z6=%xmm2,>in6=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in7 = z7
+# asm 1: movd   <z7=int6464#4,>in7=int32#4
+# asm 2: movd   <z7=%xmm3,>in7=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in4 ^= *(uint32 *) (m + 208)
+# asm 1: xorl 208(<m=int32#5),<in4=int32#1
+# asm 2: xorl 208(<m=%esi),<in4=%eax
+xorl 208(%esi),%eax
+
+# qhasm:   in5 ^= *(uint32 *) (m + 212)
+# asm 1: xorl 212(<m=int32#5),<in5=int32#2
+# asm 2: xorl 212(<m=%esi),<in5=%ecx
+xorl 212(%esi),%ecx
+
+# qhasm:   in6 ^= *(uint32 *) (m + 216)
+# asm 1: xorl 216(<m=int32#5),<in6=int32#3
+# asm 2: xorl 216(<m=%esi),<in6=%edx
+xorl 216(%esi),%edx
+
+# qhasm:   in7 ^= *(uint32 *) (m + 220)
+# asm 1: xorl 220(<m=int32#5),<in7=int32#4
+# asm 2: xorl 220(<m=%esi),<in7=%ebx
+xorl 220(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 208) = in4
+# asm 1: movl <in4=int32#1,208(<out=int32#6)
+# asm 2: movl <in4=%eax,208(<out=%edi)
+movl %eax,208(%edi)
+
+# qhasm:   *(uint32 *) (out + 212) = in5
+# asm 1: movl <in5=int32#2,212(<out=int32#6)
+# asm 2: movl <in5=%ecx,212(<out=%edi)
+movl %ecx,212(%edi)
+
+# qhasm:   *(uint32 *) (out + 216) = in6
+# asm 1: movl <in6=int32#3,216(<out=int32#6)
+# asm 2: movl <in6=%edx,216(<out=%edi)
+movl %edx,216(%edi)
+
+# qhasm:   *(uint32 *) (out + 220) = in7
+# asm 1: movl <in7=int32#4,220(<out=int32#6)
+# asm 2: movl <in7=%ebx,220(<out=%edi)
+movl %ebx,220(%edi)
+
+# qhasm:   z8 = z8_stack
+# asm 1: movdqa <z8_stack=stack128#37,>z8=int6464#1
+# asm 2: movdqa <z8_stack=608(%esp),>z8=%xmm0
+movdqa 608(%esp),%xmm0
+
+# qhasm:   z9 = z9_stack
+# asm 1: movdqa <z9_stack=stack128#32,>z9=int6464#2
+# asm 2: movdqa <z9_stack=528(%esp),>z9=%xmm1
+movdqa 528(%esp),%xmm1
+
+# qhasm:   z10 = z10_stack
+# asm 1: movdqa <z10_stack=stack128#22,>z10=int6464#3
+# asm 2: movdqa <z10_stack=368(%esp),>z10=%xmm2
+movdqa 368(%esp),%xmm2
+
+# qhasm:   z11 = z11_stack
+# asm 1: movdqa <z11_stack=stack128#27,>z11=int6464#4
+# asm 2: movdqa <z11_stack=448(%esp),>z11=%xmm3
+movdqa 448(%esp),%xmm3
+
+# qhasm:   uint32323232 z8 += orig8
+# asm 1: paddd <orig8=stack128#19,<z8=int6464#1
+# asm 2: paddd <orig8=320(%esp),<z8=%xmm0
+paddd 320(%esp),%xmm0
+
+# qhasm:   uint32323232 z9 += orig9
+# asm 1: paddd <orig9=stack128#20,<z9=int6464#2
+# asm 2: paddd <orig9=336(%esp),<z9=%xmm1
+paddd 336(%esp),%xmm1
+
+# qhasm:   uint32323232 z10 += orig10
+# asm 1: paddd <orig10=stack128#6,<z10=int6464#3
+# asm 2: paddd <orig10=112(%esp),<z10=%xmm2
+paddd 112(%esp),%xmm2
+
+# qhasm:   uint32323232 z11 += orig11
+# asm 1: paddd <orig11=stack128#10,<z11=int6464#4
+# asm 2: paddd <orig11=176(%esp),<z11=%xmm3
+paddd 176(%esp),%xmm3
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#1
+# asm 2: xorl 32(<m=%esi),<in8=%eax
+xorl 32(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#2
+# asm 2: xorl 36(<m=%esi),<in9=%ecx
+xorl 36(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#3
+# asm 2: xorl 40(<m=%esi),<in10=%edx
+xorl 40(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#4
+# asm 2: xorl 44(<m=%esi),<in11=%ebx
+xorl 44(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#1,32(<out=int32#6)
+# asm 2: movl <in8=%eax,32(<out=%edi)
+movl %eax,32(%edi)
+
+# qhasm:   *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#2,36(<out=int32#6)
+# asm 2: movl <in9=%ecx,36(<out=%edi)
+movl %ecx,36(%edi)
+
+# qhasm:   *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#3,40(<out=int32#6)
+# asm 2: movl <in10=%edx,40(<out=%edi)
+movl %edx,40(%edi)
+
+# qhasm:   *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#4,44(<out=int32#6)
+# asm 2: movl <in11=%ebx,44(<out=%edi)
+movl %ebx,44(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 96)
+# asm 1: xorl 96(<m=int32#5),<in8=int32#1
+# asm 2: xorl 96(<m=%esi),<in8=%eax
+xorl 96(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 100)
+# asm 1: xorl 100(<m=int32#5),<in9=int32#2
+# asm 2: xorl 100(<m=%esi),<in9=%ecx
+xorl 100(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 104)
+# asm 1: xorl 104(<m=int32#5),<in10=int32#3
+# asm 2: xorl 104(<m=%esi),<in10=%edx
+xorl 104(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 108)
+# asm 1: xorl 108(<m=int32#5),<in11=int32#4
+# asm 2: xorl 108(<m=%esi),<in11=%ebx
+xorl 108(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 96) = in8
+# asm 1: movl <in8=int32#1,96(<out=int32#6)
+# asm 2: movl <in8=%eax,96(<out=%edi)
+movl %eax,96(%edi)
+
+# qhasm:   *(uint32 *) (out + 100) = in9
+# asm 1: movl <in9=int32#2,100(<out=int32#6)
+# asm 2: movl <in9=%ecx,100(<out=%edi)
+movl %ecx,100(%edi)
+
+# qhasm:   *(uint32 *) (out + 104) = in10
+# asm 1: movl <in10=int32#3,104(<out=int32#6)
+# asm 2: movl <in10=%edx,104(<out=%edi)
+movl %edx,104(%edi)
+
+# qhasm:   *(uint32 *) (out + 108) = in11
+# asm 1: movl <in11=int32#4,108(<out=int32#6)
+# asm 2: movl <in11=%ebx,108(<out=%edi)
+movl %ebx,108(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z8 <<<= 96
+# asm 1: pshufd $0x39,<z8=int6464#1,<z8=int6464#1
+# asm 2: pshufd $0x39,<z8=%xmm0,<z8=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z9 <<<= 96
+# asm 1: pshufd $0x39,<z9=int6464#2,<z9=int6464#2
+# asm 2: pshufd $0x39,<z9=%xmm1,<z9=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z10 <<<= 96
+# asm 1: pshufd $0x39,<z10=int6464#3,<z10=int6464#3
+# asm 2: pshufd $0x39,<z10=%xmm2,<z10=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z11 <<<= 96
+# asm 1: pshufd $0x39,<z11=int6464#4,<z11=int6464#4
+# asm 2: pshufd $0x39,<z11=%xmm3,<z11=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in8 ^= *(uint32 *) (m + 160)
+# asm 1: xorl 160(<m=int32#5),<in8=int32#1
+# asm 2: xorl 160(<m=%esi),<in8=%eax
+xorl 160(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 164)
+# asm 1: xorl 164(<m=int32#5),<in9=int32#2
+# asm 2: xorl 164(<m=%esi),<in9=%ecx
+xorl 164(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 168)
+# asm 1: xorl 168(<m=int32#5),<in10=int32#3
+# asm 2: xorl 168(<m=%esi),<in10=%edx
+xorl 168(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 172)
+# asm 1: xorl 172(<m=int32#5),<in11=int32#4
+# asm 2: xorl 172(<m=%esi),<in11=%ebx
+xorl 172(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 160) = in8
+# asm 1: movl <in8=int32#1,160(<out=int32#6)
+# asm 2: movl <in8=%eax,160(<out=%edi)
+movl %eax,160(%edi)
+
+# qhasm:   *(uint32 *) (out + 164) = in9
+# asm 1: movl <in9=int32#2,164(<out=int32#6)
+# asm 2: movl <in9=%ecx,164(<out=%edi)
+movl %ecx,164(%edi)
+
+# qhasm:   *(uint32 *) (out + 168) = in10
+# asm 1: movl <in10=int32#3,168(<out=int32#6)
+# asm 2: movl <in10=%edx,168(<out=%edi)
+movl %edx,168(%edi)
+
+# qhasm:   *(uint32 *) (out + 172) = in11
+# asm 1: movl <in11=int32#4,172(<out=int32#6)
+# asm 2: movl <in11=%ebx,172(<out=%edi)
+movl %ebx,172(%edi)
+
+# qhasm:   in8 = z8
+# asm 1: movd   <z8=int6464#1,>in8=int32#1
+# asm 2: movd   <z8=%xmm0,>in8=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in9 = z9
+# asm 1: movd   <z9=int6464#2,>in9=int32#2
+# asm 2: movd   <z9=%xmm1,>in9=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in10 = z10
+# asm 1: movd   <z10=int6464#3,>in10=int32#3
+# asm 2: movd   <z10=%xmm2,>in10=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in11 = z11
+# asm 1: movd   <z11=int6464#4,>in11=int32#4
+# asm 2: movd   <z11=%xmm3,>in11=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in8 ^= *(uint32 *) (m + 224)
+# asm 1: xorl 224(<m=int32#5),<in8=int32#1
+# asm 2: xorl 224(<m=%esi),<in8=%eax
+xorl 224(%esi),%eax
+
+# qhasm:   in9 ^= *(uint32 *) (m + 228)
+# asm 1: xorl 228(<m=int32#5),<in9=int32#2
+# asm 2: xorl 228(<m=%esi),<in9=%ecx
+xorl 228(%esi),%ecx
+
+# qhasm:   in10 ^= *(uint32 *) (m + 232)
+# asm 1: xorl 232(<m=int32#5),<in10=int32#3
+# asm 2: xorl 232(<m=%esi),<in10=%edx
+xorl 232(%esi),%edx
+
+# qhasm:   in11 ^= *(uint32 *) (m + 236)
+# asm 1: xorl 236(<m=int32#5),<in11=int32#4
+# asm 2: xorl 236(<m=%esi),<in11=%ebx
+xorl 236(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 224) = in8
+# asm 1: movl <in8=int32#1,224(<out=int32#6)
+# asm 2: movl <in8=%eax,224(<out=%edi)
+movl %eax,224(%edi)
+
+# qhasm:   *(uint32 *) (out + 228) = in9
+# asm 1: movl <in9=int32#2,228(<out=int32#6)
+# asm 2: movl <in9=%ecx,228(<out=%edi)
+movl %ecx,228(%edi)
+
+# qhasm:   *(uint32 *) (out + 232) = in10
+# asm 1: movl <in10=int32#3,232(<out=int32#6)
+# asm 2: movl <in10=%edx,232(<out=%edi)
+movl %edx,232(%edi)
+
+# qhasm:   *(uint32 *) (out + 236) = in11
+# asm 1: movl <in11=int32#4,236(<out=int32#6)
+# asm 2: movl <in11=%ebx,236(<out=%edi)
+movl %ebx,236(%edi)
+
+# qhasm:   z12 = z12_stack
+# asm 1: movdqa <z12_stack=stack128#35,>z12=int6464#1
+# asm 2: movdqa <z12_stack=576(%esp),>z12=%xmm0
+movdqa 576(%esp),%xmm0
+
+# qhasm:   z13 = z13_stack
+# asm 1: movdqa <z13_stack=stack128#30,>z13=int6464#2
+# asm 2: movdqa <z13_stack=496(%esp),>z13=%xmm1
+movdqa 496(%esp),%xmm1
+
+# qhasm:   z14 = z14_stack
+# asm 1: movdqa <z14_stack=stack128#24,>z14=int6464#3
+# asm 2: movdqa <z14_stack=400(%esp),>z14=%xmm2
+movdqa 400(%esp),%xmm2
+
+# qhasm:   z15 = z15_stack
+# asm 1: movdqa <z15_stack=stack128#23,>z15=int6464#4
+# asm 2: movdqa <z15_stack=384(%esp),>z15=%xmm3
+movdqa 384(%esp),%xmm3
+
+# qhasm:   uint32323232 z12 += orig12
+# asm 1: paddd <orig12=stack128#11,<z12=int6464#1
+# asm 2: paddd <orig12=192(%esp),<z12=%xmm0
+paddd 192(%esp),%xmm0
+
+# qhasm:   uint32323232 z13 += orig13
+# asm 1: paddd <orig13=stack128#14,<z13=int6464#2
+# asm 2: paddd <orig13=240(%esp),<z13=%xmm1
+paddd 240(%esp),%xmm1
+
+# qhasm:   uint32323232 z14 += orig14
+# asm 1: paddd <orig14=stack128#17,<z14=int6464#3
+# asm 2: paddd <orig14=288(%esp),<z14=%xmm2
+paddd 288(%esp),%xmm2
+
+# qhasm:   uint32323232 z15 += orig15
+# asm 1: paddd <orig15=stack128#7,<z15=int6464#4
+# asm 2: paddd <orig15=128(%esp),<z15=%xmm3
+paddd 128(%esp),%xmm3
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#1
+# asm 2: xorl 48(<m=%esi),<in12=%eax
+xorl 48(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#2
+# asm 2: xorl 52(<m=%esi),<in13=%ecx
+xorl 52(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#3
+# asm 2: xorl 56(<m=%esi),<in14=%edx
+xorl 56(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#4
+# asm 2: xorl 60(<m=%esi),<in15=%ebx
+xorl 60(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#1,48(<out=int32#6)
+# asm 2: movl <in12=%eax,48(<out=%edi)
+movl %eax,48(%edi)
+
+# qhasm:   *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#2,52(<out=int32#6)
+# asm 2: movl <in13=%ecx,52(<out=%edi)
+movl %ecx,52(%edi)
+
+# qhasm:   *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#3,56(<out=int32#6)
+# asm 2: movl <in14=%edx,56(<out=%edi)
+movl %edx,56(%edi)
+
+# qhasm:   *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#4,60(<out=int32#6)
+# asm 2: movl <in15=%ebx,60(<out=%edi)
+movl %ebx,60(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 112)
+# asm 1: xorl 112(<m=int32#5),<in12=int32#1
+# asm 2: xorl 112(<m=%esi),<in12=%eax
+xorl 112(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 116)
+# asm 1: xorl 116(<m=int32#5),<in13=int32#2
+# asm 2: xorl 116(<m=%esi),<in13=%ecx
+xorl 116(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 120)
+# asm 1: xorl 120(<m=int32#5),<in14=int32#3
+# asm 2: xorl 120(<m=%esi),<in14=%edx
+xorl 120(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 124)
+# asm 1: xorl 124(<m=int32#5),<in15=int32#4
+# asm 2: xorl 124(<m=%esi),<in15=%ebx
+xorl 124(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 112) = in12
+# asm 1: movl <in12=int32#1,112(<out=int32#6)
+# asm 2: movl <in12=%eax,112(<out=%edi)
+movl %eax,112(%edi)
+
+# qhasm:   *(uint32 *) (out + 116) = in13
+# asm 1: movl <in13=int32#2,116(<out=int32#6)
+# asm 2: movl <in13=%ecx,116(<out=%edi)
+movl %ecx,116(%edi)
+
+# qhasm:   *(uint32 *) (out + 120) = in14
+# asm 1: movl <in14=int32#3,120(<out=int32#6)
+# asm 2: movl <in14=%edx,120(<out=%edi)
+movl %edx,120(%edi)
+
+# qhasm:   *(uint32 *) (out + 124) = in15
+# asm 1: movl <in15=int32#4,124(<out=int32#6)
+# asm 2: movl <in15=%ebx,124(<out=%edi)
+movl %ebx,124(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   z12 <<<= 96
+# asm 1: pshufd $0x39,<z12=int6464#1,<z12=int6464#1
+# asm 2: pshufd $0x39,<z12=%xmm0,<z12=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm:   z13 <<<= 96
+# asm 1: pshufd $0x39,<z13=int6464#2,<z13=int6464#2
+# asm 2: pshufd $0x39,<z13=%xmm1,<z13=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:   z14 <<<= 96
+# asm 1: pshufd $0x39,<z14=int6464#3,<z14=int6464#3
+# asm 2: pshufd $0x39,<z14=%xmm2,<z14=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm:   z15 <<<= 96
+# asm 1: pshufd $0x39,<z15=int6464#4,<z15=int6464#4
+# asm 2: pshufd $0x39,<z15=%xmm3,<z15=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:   in12 ^= *(uint32 *) (m + 176)
+# asm 1: xorl 176(<m=int32#5),<in12=int32#1
+# asm 2: xorl 176(<m=%esi),<in12=%eax
+xorl 176(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 180)
+# asm 1: xorl 180(<m=int32#5),<in13=int32#2
+# asm 2: xorl 180(<m=%esi),<in13=%ecx
+xorl 180(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 184)
+# asm 1: xorl 184(<m=int32#5),<in14=int32#3
+# asm 2: xorl 184(<m=%esi),<in14=%edx
+xorl 184(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 188)
+# asm 1: xorl 188(<m=int32#5),<in15=int32#4
+# asm 2: xorl 188(<m=%esi),<in15=%ebx
+xorl 188(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 176) = in12
+# asm 1: movl <in12=int32#1,176(<out=int32#6)
+# asm 2: movl <in12=%eax,176(<out=%edi)
+movl %eax,176(%edi)
+
+# qhasm:   *(uint32 *) (out + 180) = in13
+# asm 1: movl <in13=int32#2,180(<out=int32#6)
+# asm 2: movl <in13=%ecx,180(<out=%edi)
+movl %ecx,180(%edi)
+
+# qhasm:   *(uint32 *) (out + 184) = in14
+# asm 1: movl <in14=int32#3,184(<out=int32#6)
+# asm 2: movl <in14=%edx,184(<out=%edi)
+movl %edx,184(%edi)
+
+# qhasm:   *(uint32 *) (out + 188) = in15
+# asm 1: movl <in15=int32#4,188(<out=int32#6)
+# asm 2: movl <in15=%ebx,188(<out=%edi)
+movl %ebx,188(%edi)
+
+# qhasm:   in12 = z12
+# asm 1: movd   <z12=int6464#1,>in12=int32#1
+# asm 2: movd   <z12=%xmm0,>in12=%eax
+movd   %xmm0,%eax
+
+# qhasm:   in13 = z13
+# asm 1: movd   <z13=int6464#2,>in13=int32#2
+# asm 2: movd   <z13=%xmm1,>in13=%ecx
+movd   %xmm1,%ecx
+
+# qhasm:   in14 = z14
+# asm 1: movd   <z14=int6464#3,>in14=int32#3
+# asm 2: movd   <z14=%xmm2,>in14=%edx
+movd   %xmm2,%edx
+
+# qhasm:   in15 = z15
+# asm 1: movd   <z15=int6464#4,>in15=int32#4
+# asm 2: movd   <z15=%xmm3,>in15=%ebx
+movd   %xmm3,%ebx
+
+# qhasm:   in12 ^= *(uint32 *) (m + 240)
+# asm 1: xorl 240(<m=int32#5),<in12=int32#1
+# asm 2: xorl 240(<m=%esi),<in12=%eax
+xorl 240(%esi),%eax
+
+# qhasm:   in13 ^= *(uint32 *) (m + 244)
+# asm 1: xorl 244(<m=int32#5),<in13=int32#2
+# asm 2: xorl 244(<m=%esi),<in13=%ecx
+xorl 244(%esi),%ecx
+
+# qhasm:   in14 ^= *(uint32 *) (m + 248)
+# asm 1: xorl 248(<m=int32#5),<in14=int32#3
+# asm 2: xorl 248(<m=%esi),<in14=%edx
+xorl 248(%esi),%edx
+
+# qhasm:   in15 ^= *(uint32 *) (m + 252)
+# asm 1: xorl 252(<m=int32#5),<in15=int32#4
+# asm 2: xorl 252(<m=%esi),<in15=%ebx
+xorl 252(%esi),%ebx
+
+# qhasm:   *(uint32 *) (out + 240) = in12
+# asm 1: movl <in12=int32#1,240(<out=int32#6)
+# asm 2: movl <in12=%eax,240(<out=%edi)
+movl %eax,240(%edi)
+
+# qhasm:   *(uint32 *) (out + 244) = in13
+# asm 1: movl <in13=int32#2,244(<out=int32#6)
+# asm 2: movl <in13=%ecx,244(<out=%edi)
+movl %ecx,244(%edi)
+
+# qhasm:   *(uint32 *) (out + 248) = in14
+# asm 1: movl <in14=int32#3,248(<out=int32#6)
+# asm 2: movl <in14=%edx,248(<out=%edi)
+movl %edx,248(%edi)
+
+# qhasm:   *(uint32 *) (out + 252) = in15
+# asm 1: movl <in15=int32#4,252(<out=int32#6)
+# asm 2: movl <in15=%ebx,252(<out=%edi)
+movl %ebx,252(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   bytes -= 256
+# asm 1: sub  $256,<bytes=int32#1
+# asm 2: sub  $256,<bytes=%eax
+sub  $256,%eax
+
+# qhasm:   m += 256
+# asm 1: add  $256,<m=int32#5
+# asm 2: add  $256,<m=%esi
+add  $256,%esi
+
+# qhasm:   out += 256
+# asm 1: add  $256,<out=int32#6
+# asm 2: add  $256,<out=%edi
+add  $256,%edi
+
+# qhasm:   out_stack = out
+# asm 1: movl <out=int32#6,>out_stack=stack32#6
+# asm 2: movl <out=%edi,>out_stack=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:                            unsigned<? bytes - 256
+# asm 1: cmp  $256,<bytes=int32#1
+# asm 2: cmp  $256,<bytes=%eax
+cmp  $256,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast256 if !unsigned<
+jae ._bytesatleast256
+
+# qhasm:                 unsigned>? bytes - 0
+# asm 1: cmp  $0,<bytes=int32#1
+# asm 2: cmp  $0,<bytes=%eax
+cmp  $0,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto done if !unsigned>
+jbe ._done
+# comment:fp stack unchanged by fallthrough
+
+# qhasm: bytesbetween1and255:
+._bytesbetween1and255:
+
+# qhasm:                   unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto nocopy if !unsigned<
+jae ._nocopy
+
+# qhasm:     ctarget = out
+# asm 1: movl <out=int32#6,>ctarget=stack32#6
+# asm 2: movl <out=%edi,>ctarget=20(%esp)
+movl %edi,20(%esp)
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:     while (i) { *out++ = *m++; --i }
+rep movsb
+
+# qhasm:     out = &tmp
+# asm 1: leal <tmp=stack512#1,>out=int32#6
+# asm 2: leal <tmp=640(%esp),>out=%edi
+leal 640(%esp),%edi
+
+# qhasm:     m = &tmp
+# asm 1: leal <tmp=stack512#1,>m=int32#5
+# asm 2: leal <tmp=640(%esp),>m=%esi
+leal 640(%esp),%esi
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:   nocopy:
+._nocopy:
+
+# qhasm:   bytes_stack = bytes
+# asm 1: movl <bytes=int32#1,>bytes_stack=stack32#7
+# asm 2: movl <bytes=%eax,>bytes_stack=24(%esp)
+movl %eax,24(%esp)
+
+# qhasm: diag0 = x0
+# asm 1: movdqa <x0=stack128#3,>diag0=int6464#1
+# asm 2: movdqa <x0=64(%esp),>diag0=%xmm0
+movdqa 64(%esp),%xmm0
+
+# qhasm: diag1 = x1
+# asm 1: movdqa <x1=stack128#2,>diag1=int6464#2
+# asm 2: movdqa <x1=48(%esp),>diag1=%xmm1
+movdqa 48(%esp),%xmm1
+
+# qhasm: diag2 = x2
+# asm 1: movdqa <x2=stack128#4,>diag2=int6464#3
+# asm 2: movdqa <x2=80(%esp),>diag2=%xmm2
+movdqa 80(%esp),%xmm2
+
+# qhasm: diag3 = x3
+# asm 1: movdqa <x3=stack128#1,>diag3=int6464#4
+# asm 2: movdqa <x3=32(%esp),>diag3=%xmm3
+movdqa 32(%esp),%xmm3
+
+# qhasm:                     a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm: i = 8
+# asm 1: mov  $8,>i=int32#1
+# asm 2: mov  $8,>i=%eax
+mov  $8,%eax
+
+# qhasm: mainloop2:
+._mainloop2:
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a0 += diag0
+# asm 1: paddd <diag0=int6464#1,<a0=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a0=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a1 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a1=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a1=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b0 = a0
+# asm 1: movdqa <a0=int6464#5,>b0=int6464#7
+# asm 2: movdqa <a0=%xmm4,>b0=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a0 <<= 7
+# asm 1: pslld $7,<a0=int6464#5
+# asm 2: pslld $7,<a0=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b0 >>= 25
+# asm 1: psrld $25,<b0=int6464#7
+# asm 2: psrld $25,<b0=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag3 ^= a0
+# asm 1: pxor  <a0=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a0=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                 diag3 ^= b0
+# asm 1: pxor  <b0=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b0=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm: uint32323232                        a1 += diag3
+# asm 1: paddd <diag3=int6464#4,<a1=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a1=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                                                 a2 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a2=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a2=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                     b1 = a1
+# asm 1: movdqa <a1=int6464#6,>b1=int6464#7
+# asm 2: movdqa <a1=%xmm5,>b1=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a1 <<= 9
+# asm 1: pslld $9,<a1=int6464#6
+# asm 2: pslld $9,<a1=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b1 >>= 23
+# asm 1: psrld $23,<b1=int6464#7
+# asm 2: psrld $23,<b1=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a1
+# asm 1: pxor  <a1=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a1=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag3 <<<= 32
+# asm 1: pshufd $0x93,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x93,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x93,%xmm3,%xmm3
+
+# qhasm:                                 diag2 ^= b1
+# asm 1: pxor  <b1=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b1=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a2 += diag2
+# asm 1: paddd <diag2=int6464#3,<a2=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a2=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a3 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a3=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a3=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b2 = a2
+# asm 1: movdqa <a2=int6464#5,>b2=int6464#7
+# asm 2: movdqa <a2=%xmm4,>b2=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a2 <<= 13
+# asm 1: pslld $13,<a2=int6464#5
+# asm 2: pslld $13,<a2=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b2 >>= 19
+# asm 1: psrld $19,<b2=int6464#7
+# asm 2: psrld $19,<b2=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag1 ^= a2
+# asm 1: pxor  <a2=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a2=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag1 ^= b2
+# asm 1: pxor  <b2=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b2=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                                                        a3 += diag1
+# asm 1: paddd <diag1=int6464#2,<a3=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a3=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                 a4 = diag3
+# asm 1: movdqa <diag3=int6464#4,>a4=int6464#5
+# asm 2: movdqa <diag3=%xmm3,>a4=%xmm4
+movdqa %xmm3,%xmm4
+
+# qhasm:                                                                     b3 = a3
+# asm 1: movdqa <a3=int6464#6,>b3=int6464#7
+# asm 2: movdqa <a3=%xmm5,>b3=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a3 <<= 18
+# asm 1: pslld $18,<a3=int6464#6
+# asm 2: pslld $18,<a3=%xmm5
+pslld $18,%xmm5
+
+# qhasm: uint32323232                                                        b3 >>= 14
+# asm 1: psrld $14,<b3=int6464#7
+# asm 2: psrld $14,<b3=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a3
+# asm 1: pxor  <a3=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a3=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm:                                                                 diag0 ^= b3
+# asm 1: pxor  <b3=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b3=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+
+# qhasm: uint32323232        a4 += diag0
+# asm 1: paddd <diag0=int6464#1,<a4=int6464#5
+# asm 2: paddd <diag0=%xmm0,<a4=%xmm4
+paddd %xmm0,%xmm4
+
+# qhasm:                                 a5 = diag0
+# asm 1: movdqa <diag0=int6464#1,>a5=int6464#6
+# asm 2: movdqa <diag0=%xmm0,>a5=%xmm5
+movdqa %xmm0,%xmm5
+
+# qhasm:                     b4 = a4
+# asm 1: movdqa <a4=int6464#5,>b4=int6464#7
+# asm 2: movdqa <a4=%xmm4,>b4=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232        a4 <<= 7
+# asm 1: pslld $7,<a4=int6464#5
+# asm 2: pslld $7,<a4=%xmm4
+pslld $7,%xmm4
+
+# qhasm: uint32323232        b4 >>= 25
+# asm 1: psrld $25,<b4=int6464#7
+# asm 2: psrld $25,<b4=%xmm6
+psrld $25,%xmm6
+
+# qhasm:                 diag1 ^= a4
+# asm 1: pxor  <a4=int6464#5,<diag1=int6464#2
+# asm 2: pxor  <a4=%xmm4,<diag1=%xmm1
+pxor  %xmm4,%xmm1
+
+# qhasm:                 diag1 ^= b4
+# asm 1: pxor  <b4=int6464#7,<diag1=int6464#2
+# asm 2: pxor  <b4=%xmm6,<diag1=%xmm1
+pxor  %xmm6,%xmm1
+
+# qhasm: uint32323232                        a5 += diag1
+# asm 1: paddd <diag1=int6464#2,<a5=int6464#6
+# asm 2: paddd <diag1=%xmm1,<a5=%xmm5
+paddd %xmm1,%xmm5
+
+# qhasm:                                                 a6 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a6=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a6=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                     b5 = a5
+# asm 1: movdqa <a5=int6464#6,>b5=int6464#7
+# asm 2: movdqa <a5=%xmm5,>b5=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                        a5 <<= 9
+# asm 1: pslld $9,<a5=int6464#6
+# asm 2: pslld $9,<a5=%xmm5
+pslld $9,%xmm5
+
+# qhasm: uint32323232                        b5 >>= 23
+# asm 1: psrld $23,<b5=int6464#7
+# asm 2: psrld $23,<b5=%xmm6
+psrld $23,%xmm6
+
+# qhasm:                                 diag2 ^= a5
+# asm 1: pxor  <a5=int6464#6,<diag2=int6464#3
+# asm 2: pxor  <a5=%xmm5,<diag2=%xmm2
+pxor  %xmm5,%xmm2
+
+# qhasm:                 diag1 <<<= 32
+# asm 1: pshufd $0x93,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x93,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x93,%xmm1,%xmm1
+
+# qhasm:                                 diag2 ^= b5
+# asm 1: pxor  <b5=int6464#7,<diag2=int6464#3
+# asm 2: pxor  <b5=%xmm6,<diag2=%xmm2
+pxor  %xmm6,%xmm2
+
+# qhasm: uint32323232                                        a6 += diag2
+# asm 1: paddd <diag2=int6464#3,<a6=int6464#5
+# asm 2: paddd <diag2=%xmm2,<a6=%xmm4
+paddd %xmm2,%xmm4
+
+# qhasm:                                                                 a7 = diag2
+# asm 1: movdqa <diag2=int6464#3,>a7=int6464#6
+# asm 2: movdqa <diag2=%xmm2,>a7=%xmm5
+movdqa %xmm2,%xmm5
+
+# qhasm:                                                     b6 = a6
+# asm 1: movdqa <a6=int6464#5,>b6=int6464#7
+# asm 2: movdqa <a6=%xmm4,>b6=%xmm6
+movdqa %xmm4,%xmm6
+
+# qhasm: uint32323232                                        a6 <<= 13
+# asm 1: pslld $13,<a6=int6464#5
+# asm 2: pslld $13,<a6=%xmm4
+pslld $13,%xmm4
+
+# qhasm: uint32323232                                        b6 >>= 19
+# asm 1: psrld $19,<b6=int6464#7
+# asm 2: psrld $19,<b6=%xmm6
+psrld $19,%xmm6
+
+# qhasm:                                                 diag3 ^= a6
+# asm 1: pxor  <a6=int6464#5,<diag3=int6464#4
+# asm 2: pxor  <a6=%xmm4,<diag3=%xmm3
+pxor  %xmm4,%xmm3
+
+# qhasm:                                 diag2 <<<= 64
+# asm 1: pshufd $0x4e,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x4e,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x4e,%xmm2,%xmm2
+
+# qhasm:                                                 diag3 ^= b6
+# asm 1: pxor  <b6=int6464#7,<diag3=int6464#4
+# asm 2: pxor  <b6=%xmm6,<diag3=%xmm3
+pxor  %xmm6,%xmm3
+
+# qhasm:                  unsigned>? i -= 4
+# asm 1: sub  $4,<i=int32#1
+# asm 2: sub  $4,<i=%eax
+sub  $4,%eax
+
+# qhasm: uint32323232                                                        a7 += diag3
+# asm 1: paddd <diag3=int6464#4,<a7=int6464#6
+# asm 2: paddd <diag3=%xmm3,<a7=%xmm5
+paddd %xmm3,%xmm5
+
+# qhasm:                 a0 = diag1
+# asm 1: movdqa <diag1=int6464#2,>a0=int6464#5
+# asm 2: movdqa <diag1=%xmm1,>a0=%xmm4
+movdqa %xmm1,%xmm4
+
+# qhasm:                                                                     b7 = a7
+# asm 1: movdqa <a7=int6464#6,>b7=int6464#7
+# asm 2: movdqa <a7=%xmm5,>b7=%xmm6
+movdqa %xmm5,%xmm6
+
+# qhasm: uint32323232                                                        a7 <<= 18
+# asm 1: pslld $18,<a7=int6464#6
+# asm 2: pslld $18,<a7=%xmm5
+pslld $18,%xmm5
+
+# qhasm:                 b0 = 0
+# asm 1: pxor   >b0=int6464#8,>b0=int6464#8
+# asm 2: pxor   >b0=%xmm7,>b0=%xmm7
+pxor   %xmm7,%xmm7
+
+# qhasm: uint32323232                                                        b7 >>= 14
+# asm 1: psrld $14,<b7=int6464#7
+# asm 2: psrld $14,<b7=%xmm6
+psrld $14,%xmm6
+
+# qhasm:                                                                 diag0 ^= a7
+# asm 1: pxor  <a7=int6464#6,<diag0=int6464#1
+# asm 2: pxor  <a7=%xmm5,<diag0=%xmm0
+pxor  %xmm5,%xmm0
+
+# qhasm:                                                 diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm:                                                                 diag0 ^= b7
+# asm 1: pxor  <b7=int6464#7,<diag0=int6464#1
+# asm 2: pxor  <b7=%xmm6,<diag0=%xmm0
+pxor  %xmm6,%xmm0
+# comment:fp stack unchanged by jump
+
+# qhasm: goto mainloop2 if unsigned>
+ja ._mainloop2
+
+# qhasm: uint32323232 diag0 += x0
+# asm 1: paddd <x0=stack128#3,<diag0=int6464#1
+# asm 2: paddd <x0=64(%esp),<diag0=%xmm0
+paddd 64(%esp),%xmm0
+
+# qhasm: uint32323232 diag1 += x1
+# asm 1: paddd <x1=stack128#2,<diag1=int6464#2
+# asm 2: paddd <x1=48(%esp),<diag1=%xmm1
+paddd 48(%esp),%xmm1
+
+# qhasm: uint32323232 diag2 += x2
+# asm 1: paddd <x2=stack128#4,<diag2=int6464#3
+# asm 2: paddd <x2=80(%esp),<diag2=%xmm2
+paddd 80(%esp),%xmm2
+
+# qhasm: uint32323232 diag3 += x3
+# asm 1: paddd <x3=stack128#1,<diag3=int6464#4
+# asm 2: paddd <x3=32(%esp),<diag3=%xmm3
+paddd 32(%esp),%xmm3
+
+# qhasm: in0 = diag0
+# asm 1: movd   <diag0=int6464#1,>in0=int32#1
+# asm 2: movd   <diag0=%xmm0,>in0=%eax
+movd   %xmm0,%eax
+
+# qhasm: in12 = diag1
+# asm 1: movd   <diag1=int6464#2,>in12=int32#2
+# asm 2: movd   <diag1=%xmm1,>in12=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in8 = diag2
+# asm 1: movd   <diag2=int6464#3,>in8=int32#3
+# asm 2: movd   <diag2=%xmm2,>in8=%edx
+movd   %xmm2,%edx
+
+# qhasm: in4 = diag3
+# asm 1: movd   <diag3=int6464#4,>in4=int32#4
+# asm 2: movd   <diag3=%xmm3,>in4=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in0 ^= *(uint32 *) (m + 0)
+# asm 1: xorl 0(<m=int32#5),<in0=int32#1
+# asm 2: xorl 0(<m=%esi),<in0=%eax
+xorl 0(%esi),%eax
+
+# qhasm: in12 ^= *(uint32 *) (m + 48)
+# asm 1: xorl 48(<m=int32#5),<in12=int32#2
+# asm 2: xorl 48(<m=%esi),<in12=%ecx
+xorl 48(%esi),%ecx
+
+# qhasm: in8 ^= *(uint32 *) (m + 32)
+# asm 1: xorl 32(<m=int32#5),<in8=int32#3
+# asm 2: xorl 32(<m=%esi),<in8=%edx
+xorl 32(%esi),%edx
+
+# qhasm: in4 ^= *(uint32 *) (m + 16)
+# asm 1: xorl 16(<m=int32#5),<in4=int32#4
+# asm 2: xorl 16(<m=%esi),<in4=%ebx
+xorl 16(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 0) = in0
+# asm 1: movl <in0=int32#1,0(<out=int32#6)
+# asm 2: movl <in0=%eax,0(<out=%edi)
+movl %eax,0(%edi)
+
+# qhasm: *(uint32 *) (out + 48) = in12
+# asm 1: movl <in12=int32#2,48(<out=int32#6)
+# asm 2: movl <in12=%ecx,48(<out=%edi)
+movl %ecx,48(%edi)
+
+# qhasm: *(uint32 *) (out + 32) = in8
+# asm 1: movl <in8=int32#3,32(<out=int32#6)
+# asm 2: movl <in8=%edx,32(<out=%edi)
+movl %edx,32(%edi)
+
+# qhasm: *(uint32 *) (out + 16) = in4
+# asm 1: movl <in4=int32#4,16(<out=int32#6)
+# asm 2: movl <in4=%ebx,16(<out=%edi)
+movl %ebx,16(%edi)
+
+# qhasm: in5 = diag0
+# asm 1: movd   <diag0=int6464#1,>in5=int32#1
+# asm 2: movd   <diag0=%xmm0,>in5=%eax
+movd   %xmm0,%eax
+
+# qhasm: in1 = diag1
+# asm 1: movd   <diag1=int6464#2,>in1=int32#2
+# asm 2: movd   <diag1=%xmm1,>in1=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in13 = diag2
+# asm 1: movd   <diag2=int6464#3,>in13=int32#3
+# asm 2: movd   <diag2=%xmm2,>in13=%edx
+movd   %xmm2,%edx
+
+# qhasm: in9 = diag3
+# asm 1: movd   <diag3=int6464#4,>in9=int32#4
+# asm 2: movd   <diag3=%xmm3,>in9=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in5 ^= *(uint32 *) (m + 20)
+# asm 1: xorl 20(<m=int32#5),<in5=int32#1
+# asm 2: xorl 20(<m=%esi),<in5=%eax
+xorl 20(%esi),%eax
+
+# qhasm: in1 ^= *(uint32 *) (m + 4)
+# asm 1: xorl 4(<m=int32#5),<in1=int32#2
+# asm 2: xorl 4(<m=%esi),<in1=%ecx
+xorl 4(%esi),%ecx
+
+# qhasm: in13 ^= *(uint32 *) (m + 52)
+# asm 1: xorl 52(<m=int32#5),<in13=int32#3
+# asm 2: xorl 52(<m=%esi),<in13=%edx
+xorl 52(%esi),%edx
+
+# qhasm: in9 ^= *(uint32 *) (m + 36)
+# asm 1: xorl 36(<m=int32#5),<in9=int32#4
+# asm 2: xorl 36(<m=%esi),<in9=%ebx
+xorl 36(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 20) = in5
+# asm 1: movl <in5=int32#1,20(<out=int32#6)
+# asm 2: movl <in5=%eax,20(<out=%edi)
+movl %eax,20(%edi)
+
+# qhasm: *(uint32 *) (out + 4) = in1
+# asm 1: movl <in1=int32#2,4(<out=int32#6)
+# asm 2: movl <in1=%ecx,4(<out=%edi)
+movl %ecx,4(%edi)
+
+# qhasm: *(uint32 *) (out + 52) = in13
+# asm 1: movl <in13=int32#3,52(<out=int32#6)
+# asm 2: movl <in13=%edx,52(<out=%edi)
+movl %edx,52(%edi)
+
+# qhasm: *(uint32 *) (out + 36) = in9
+# asm 1: movl <in9=int32#4,36(<out=int32#6)
+# asm 2: movl <in9=%ebx,36(<out=%edi)
+movl %ebx,36(%edi)
+
+# qhasm: in10 = diag0
+# asm 1: movd   <diag0=int6464#1,>in10=int32#1
+# asm 2: movd   <diag0=%xmm0,>in10=%eax
+movd   %xmm0,%eax
+
+# qhasm: in6 = diag1
+# asm 1: movd   <diag1=int6464#2,>in6=int32#2
+# asm 2: movd   <diag1=%xmm1,>in6=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in2 = diag2
+# asm 1: movd   <diag2=int6464#3,>in2=int32#3
+# asm 2: movd   <diag2=%xmm2,>in2=%edx
+movd   %xmm2,%edx
+
+# qhasm: in14 = diag3
+# asm 1: movd   <diag3=int6464#4,>in14=int32#4
+# asm 2: movd   <diag3=%xmm3,>in14=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: diag0 <<<= 96
+# asm 1: pshufd $0x39,<diag0=int6464#1,<diag0=int6464#1
+# asm 2: pshufd $0x39,<diag0=%xmm0,<diag0=%xmm0
+pshufd $0x39,%xmm0,%xmm0
+
+# qhasm: diag1 <<<= 96
+# asm 1: pshufd $0x39,<diag1=int6464#2,<diag1=int6464#2
+# asm 2: pshufd $0x39,<diag1=%xmm1,<diag1=%xmm1
+pshufd $0x39,%xmm1,%xmm1
+
+# qhasm: diag2 <<<= 96
+# asm 1: pshufd $0x39,<diag2=int6464#3,<diag2=int6464#3
+# asm 2: pshufd $0x39,<diag2=%xmm2,<diag2=%xmm2
+pshufd $0x39,%xmm2,%xmm2
+
+# qhasm: diag3 <<<= 96
+# asm 1: pshufd $0x39,<diag3=int6464#4,<diag3=int6464#4
+# asm 2: pshufd $0x39,<diag3=%xmm3,<diag3=%xmm3
+pshufd $0x39,%xmm3,%xmm3
+
+# qhasm: in10 ^= *(uint32 *) (m + 40)
+# asm 1: xorl 40(<m=int32#5),<in10=int32#1
+# asm 2: xorl 40(<m=%esi),<in10=%eax
+xorl 40(%esi),%eax
+
+# qhasm: in6 ^= *(uint32 *) (m + 24)
+# asm 1: xorl 24(<m=int32#5),<in6=int32#2
+# asm 2: xorl 24(<m=%esi),<in6=%ecx
+xorl 24(%esi),%ecx
+
+# qhasm: in2 ^= *(uint32 *) (m + 8)
+# asm 1: xorl 8(<m=int32#5),<in2=int32#3
+# asm 2: xorl 8(<m=%esi),<in2=%edx
+xorl 8(%esi),%edx
+
+# qhasm: in14 ^= *(uint32 *) (m + 56)
+# asm 1: xorl 56(<m=int32#5),<in14=int32#4
+# asm 2: xorl 56(<m=%esi),<in14=%ebx
+xorl 56(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 40) = in10
+# asm 1: movl <in10=int32#1,40(<out=int32#6)
+# asm 2: movl <in10=%eax,40(<out=%edi)
+movl %eax,40(%edi)
+
+# qhasm: *(uint32 *) (out + 24) = in6
+# asm 1: movl <in6=int32#2,24(<out=int32#6)
+# asm 2: movl <in6=%ecx,24(<out=%edi)
+movl %ecx,24(%edi)
+
+# qhasm: *(uint32 *) (out + 8) = in2
+# asm 1: movl <in2=int32#3,8(<out=int32#6)
+# asm 2: movl <in2=%edx,8(<out=%edi)
+movl %edx,8(%edi)
+
+# qhasm: *(uint32 *) (out + 56) = in14
+# asm 1: movl <in14=int32#4,56(<out=int32#6)
+# asm 2: movl <in14=%ebx,56(<out=%edi)
+movl %ebx,56(%edi)
+
+# qhasm: in15 = diag0
+# asm 1: movd   <diag0=int6464#1,>in15=int32#1
+# asm 2: movd   <diag0=%xmm0,>in15=%eax
+movd   %xmm0,%eax
+
+# qhasm: in11 = diag1
+# asm 1: movd   <diag1=int6464#2,>in11=int32#2
+# asm 2: movd   <diag1=%xmm1,>in11=%ecx
+movd   %xmm1,%ecx
+
+# qhasm: in7 = diag2
+# asm 1: movd   <diag2=int6464#3,>in7=int32#3
+# asm 2: movd   <diag2=%xmm2,>in7=%edx
+movd   %xmm2,%edx
+
+# qhasm: in3 = diag3
+# asm 1: movd   <diag3=int6464#4,>in3=int32#4
+# asm 2: movd   <diag3=%xmm3,>in3=%ebx
+movd   %xmm3,%ebx
+
+# qhasm: in15 ^= *(uint32 *) (m + 60)
+# asm 1: xorl 60(<m=int32#5),<in15=int32#1
+# asm 2: xorl 60(<m=%esi),<in15=%eax
+xorl 60(%esi),%eax
+
+# qhasm: in11 ^= *(uint32 *) (m + 44)
+# asm 1: xorl 44(<m=int32#5),<in11=int32#2
+# asm 2: xorl 44(<m=%esi),<in11=%ecx
+xorl 44(%esi),%ecx
+
+# qhasm: in7 ^= *(uint32 *) (m + 28)
+# asm 1: xorl 28(<m=int32#5),<in7=int32#3
+# asm 2: xorl 28(<m=%esi),<in7=%edx
+xorl 28(%esi),%edx
+
+# qhasm: in3 ^= *(uint32 *) (m + 12)
+# asm 1: xorl 12(<m=int32#5),<in3=int32#4
+# asm 2: xorl 12(<m=%esi),<in3=%ebx
+xorl 12(%esi),%ebx
+
+# qhasm: *(uint32 *) (out + 60) = in15
+# asm 1: movl <in15=int32#1,60(<out=int32#6)
+# asm 2: movl <in15=%eax,60(<out=%edi)
+movl %eax,60(%edi)
+
+# qhasm: *(uint32 *) (out + 44) = in11
+# asm 1: movl <in11=int32#2,44(<out=int32#6)
+# asm 2: movl <in11=%ecx,44(<out=%edi)
+movl %ecx,44(%edi)
+
+# qhasm: *(uint32 *) (out + 28) = in7
+# asm 1: movl <in7=int32#3,28(<out=int32#6)
+# asm 2: movl <in7=%edx,28(<out=%edi)
+movl %edx,28(%edi)
+
+# qhasm: *(uint32 *) (out + 12) = in3
+# asm 1: movl <in3=int32#4,12(<out=int32#6)
+# asm 2: movl <in3=%ebx,12(<out=%edi)
+movl %ebx,12(%edi)
+
+# qhasm:   bytes = bytes_stack
+# asm 1: movl <bytes_stack=stack32#7,>bytes=int32#1
+# asm 2: movl <bytes_stack=24(%esp),>bytes=%eax
+movl 24(%esp),%eax
+
+# qhasm:   in8 = ((uint32 *)&x2)[0]
+# asm 1: movl <x2=stack128#4,>in8=int32#2
+# asm 2: movl <x2=80(%esp),>in8=%ecx
+movl 80(%esp),%ecx
+
+# qhasm:   in9 = ((uint32 *)&x3)[1]
+# asm 1: movl 4+<x3=stack128#1,>in9=int32#3
+# asm 2: movl 4+<x3=32(%esp),>in9=%edx
+movl 4+32(%esp),%edx
+
+# qhasm:   carry? in8 += 1
+# asm 1: add  $1,<in8=int32#2
+# asm 2: add  $1,<in8=%ecx
+add  $1,%ecx
+
+# qhasm:   in9 += 0 + carry
+# asm 1: adc $0,<in9=int32#3
+# asm 2: adc $0,<in9=%edx
+adc $0,%edx
+
+# qhasm:   ((uint32 *)&x2)[0] = in8
+# asm 1: movl <in8=int32#2,>x2=stack128#4
+# asm 2: movl <in8=%ecx,>x2=80(%esp)
+movl %ecx,80(%esp)
+
+# qhasm:   ((uint32 *)&x3)[1] = in9
+# asm 1: movl <in9=int32#3,4+<x3=stack128#1
+# asm 2: movl <in9=%edx,4+<x3=32(%esp)
+movl %edx,4+32(%esp)
+
+# qhasm:                          unsigned>? unsigned<? bytes - 64
+# asm 1: cmp  $64,<bytes=int32#1
+# asm 2: cmp  $64,<bytes=%eax
+cmp  $64,%eax
+# comment:fp stack unchanged by jump
+
+# qhasm:   goto bytesatleast65 if unsigned>
+ja ._bytesatleast65
+# comment:fp stack unchanged by jump
+
+# qhasm:     goto bytesatleast64 if !unsigned<
+jae ._bytesatleast64
+
+# qhasm:       m = out
+# asm 1: mov  <out=int32#6,>m=int32#5
+# asm 2: mov  <out=%edi,>m=%esi
+mov  %edi,%esi
+
+# qhasm:       out = ctarget
+# asm 1: movl <ctarget=stack32#6,>out=int32#6
+# asm 2: movl <ctarget=20(%esp),>out=%edi
+movl 20(%esp),%edi
+
+# qhasm:       i = bytes
+# asm 1: mov  <bytes=int32#1,>i=int32#2
+# asm 2: mov  <bytes=%eax,>i=%ecx
+mov  %eax,%ecx
+
+# qhasm:       while (i) { *out++ = *m++; --i }
+rep movsb
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     bytesatleast64:
+._bytesatleast64:
+# comment:fp stack unchanged by fallthrough
+
+# qhasm:     done:
+._done:
+
+# qhasm:     eax = eax_stack
+# asm 1: movl <eax_stack=stack32#1,>eax=int32#1
+# asm 2: movl <eax_stack=0(%esp),>eax=%eax
+movl 0(%esp),%eax
+
+# qhasm:     ebx = ebx_stack
+# asm 1: movl <ebx_stack=stack32#2,>ebx=int32#4
+# asm 2: movl <ebx_stack=4(%esp),>ebx=%ebx
+movl 4(%esp),%ebx
+
+# qhasm:     esi = esi_stack
+# asm 1: movl <esi_stack=stack32#3,>esi=int32#5
+# asm 2: movl <esi_stack=8(%esp),>esi=%esi
+movl 8(%esp),%esi
+
+# qhasm:     edi = edi_stack
+# asm 1: movl <edi_stack=stack32#4,>edi=int32#6
+# asm 2: movl <edi_stack=12(%esp),>edi=%edi
+movl 12(%esp),%edi
+
+# qhasm:     ebp = ebp_stack
+# asm 1: movl <ebp_stack=stack32#5,>ebp=int32#7
+# asm 2: movl <ebp_stack=16(%esp),>ebp=%ebp
+movl 16(%esp),%ebp
+
+# qhasm:     leave
+add %eax,%esp
+xor %eax,%eax
+ret
+
+# qhasm:   bytesatleast65:
+._bytesatleast65:
+
+# qhasm:   bytes -= 64
+# asm 1: sub  $64,<bytes=int32#1
+# asm 2: sub  $64,<bytes=%eax
+sub  $64,%eax
+
+# qhasm:   out += 64
+# asm 1: add  $64,<out=int32#6
+# asm 2: add  $64,<out=%edi
+add  $64,%edi
+
+# qhasm:   m += 64
+# asm 1: add  $64,<m=int32#5
+# asm 2: add  $64,<m=%esi
+add  $64,%esi
+# comment:fp stack unchanged by jump
+
+# qhasm: goto bytesbetween1and255
+jmp ._bytesbetween1and255
diff --git a/nacl/crypto_stream/try.c b/nacl/crypto_stream/try.c
new file mode 100644
index 00000000..9a36d760
--- /dev/null
+++ b/nacl/crypto_stream/try.c
@@ -0,0 +1,124 @@
+/*
+ * crypto_stream/try.c version 20090118
+ * D. J. Bernstein
+ * Public domain.
+ */
+
+#include <stdlib.h>
+#include "crypto_stream.h"
+
+extern unsigned char *alignedcalloc(unsigned long long);
+
+const char *primitiveimplementation = crypto_stream_IMPLEMENTATION;
+
+#define MAXTEST_BYTES 10000
+#define CHECKSUM_BYTES 4096
+#define TUNE_BYTES 1536
+
+static unsigned char *k;
+static unsigned char *n;
+static unsigned char *m;
+static unsigned char *c;
+static unsigned char *s;
+static unsigned char *k2;
+static unsigned char *n2;
+static unsigned char *m2;
+static unsigned char *c2;
+static unsigned char *s2;
+
+void preallocate(void)
+{
+}
+
+void allocate(void)
+{
+  k = alignedcalloc(crypto_stream_KEYBYTES);
+  n = alignedcalloc(crypto_stream_NONCEBYTES);
+  m = alignedcalloc(MAXTEST_BYTES);
+  c = alignedcalloc(MAXTEST_BYTES);
+  s = alignedcalloc(MAXTEST_BYTES);
+  k2 = alignedcalloc(crypto_stream_KEYBYTES);
+  n2 = alignedcalloc(crypto_stream_NONCEBYTES);
+  m2 = alignedcalloc(MAXTEST_BYTES);
+  c2 = alignedcalloc(MAXTEST_BYTES);
+  s2 = alignedcalloc(MAXTEST_BYTES);
+}
+
+void predoit(void)
+{
+}
+
+void doit(void)
+{
+  crypto_stream_xor(c,m,TUNE_BYTES,n,k);
+}
+
+char checksum[crypto_stream_KEYBYTES * 2 + 1];
+
+const char *checksum_compute(void)
+{
+  long long i;
+  long long j;
+
+  for (i = 0;i < CHECKSUM_BYTES;++i) {
+    long long mlen = i;
+    long long clen = i;
+    long long slen = i;
+    long long klen = crypto_stream_KEYBYTES;
+    long long nlen = crypto_stream_NONCEBYTES;
+    for (j = -16;j < 0;++j) m[j] = random();
+    for (j = -16;j < 0;++j) c[j] = random();
+    for (j = -16;j < 0;++j) s[j] = random();
+    for (j = -16;j < 0;++j) n[j] = random();
+    for (j = -16;j < 0;++j) k[j] = random();
+    for (j = mlen;j < mlen + 16;++j) m[j] = random();
+    for (j = clen;j < clen + 16;++j) c[j] = random();
+    for (j = slen;j < slen + 16;++j) s[j] = random();
+    for (j = nlen;j < nlen + 16;++j) n[j] = random();
+    for (j = klen;j < klen + 16;++j) k[j] = random();
+    for (j = -16;j < mlen + 16;++j) m2[j] = m[j];
+    for (j = -16;j < clen + 16;++j) c2[j] = c[j];
+    for (j = -16;j < slen + 16;++j) s2[j] = s[j];
+    for (j = -16;j < nlen + 16;++j) n2[j] = n[j];
+    for (j = -16;j < klen + 16;++j) k2[j] = k[j];
+
+    crypto_stream_xor(c,m,mlen,n,k);
+
+    for (j = -16;j < mlen + 16;++j) if (m[j] != m2[j]) return "crypto_stream_xor overwrites m";
+    for (j = -16;j < slen + 16;++j) if (s[j] != s2[j]) return "crypto_stream_xor overwrites s";
+    for (j = -16;j < nlen + 16;++j) if (n[j] != n2[j]) return "crypto_stream_xor overwrites n";
+    for (j = -16;j < klen + 16;++j) if (k[j] != k2[j]) return "crypto_stream_xor overwrites k";
+    for (j = -16;j < 0;++j) if (c[j] != c2[j]) return "crypto_stream_xor writes before output";
+    for (j = clen;j < clen + 16;++j) if (c[j] != c2[j]) return "crypto_stream_xor writes after output";
+
+    for (j = -16;j < clen + 16;++j) c2[j] = c[j];
+
+    crypto_stream(s,slen,n,k);
+
+    for (j = -16;j < mlen + 16;++j) if (m[j] != m2[j]) return "crypto_stream overwrites m";
+    for (j = -16;j < clen + 16;++j) if (c[j] != c2[j]) return "crypto_stream overwrites c";
+    for (j = -16;j < nlen + 16;++j) if (n[j] != n2[j]) return "crypto_stream overwrites n";
+    for (j = -16;j < klen + 16;++j) if (k[j] != k2[j]) return "crypto_stream overwrites k";
+    for (j = -16;j < 0;++j) if (s[j] != s2[j]) return "crypto_stream writes before output";
+    for (j = slen;j < slen + 16;++j) if (s[j] != s2[j]) return "crypto_stream writes after output";
+
+    for (j = 0;j < mlen;++j)
+      if ((s[j] ^ m[j]) != c[j]) return "crypto_stream_xor does not match crypto_stream";
+
+    for (j = 0;j < clen;++j) k[j % klen] ^= c[j];
+    crypto_stream_xor(m,c,clen,n,k);
+    crypto_stream(s,slen,n,k);
+    for (j = 0;j < mlen;++j)
+      if ((s[j] ^ m[j]) != c[j]) return "crypto_stream_xor does not match crypto_stream";
+    for (j = 0;j < mlen;++j) n[j % nlen] ^= m[j];
+    m[mlen] = 0;
+  }
+
+  for (i = 0;i < crypto_stream_KEYBYTES;++i) {
+    checksum[2 * i] = "0123456789abcdef"[15 & (k[i] >> 4)];
+    checksum[2 * i + 1] = "0123456789abcdef"[15 & k[i]];
+  }
+  checksum[2 * i] = 0;
+
+  return 0;
+}
diff --git a/nacl/crypto_stream/wrapper-stream.cpp b/nacl/crypto_stream/wrapper-stream.cpp
new file mode 100644
index 00000000..dd10c2f6
--- /dev/null
+++ b/nacl/crypto_stream/wrapper-stream.cpp
@@ -0,0 +1,12 @@
+#include <string>
+using std::string;
+#include "crypto_stream.h"
+
+string crypto_stream(size_t clen,const string &n,const string &k)
+{
+  if (n.size() != crypto_stream_NONCEBYTES) throw "incorrect nonce length";
+  if (k.size() != crypto_stream_KEYBYTES) throw "incorrect key length";
+  unsigned char c[clen];
+  crypto_stream(c,clen,(const unsigned char *) n.c_str(),(const unsigned char *) k.c_str());
+  return string((char *) c,clen);
+}
diff --git a/nacl/crypto_stream/wrapper-xor.cpp b/nacl/crypto_stream/wrapper-xor.cpp
new file mode 100644
index 00000000..8d770d1e
--- /dev/null
+++ b/nacl/crypto_stream/wrapper-xor.cpp
@@ -0,0 +1,17 @@
+#include <string>
+using std::string;
+#include "crypto_stream.h"
+
+string crypto_stream_xor(const string &m,const string &n,const string &k)
+{
+  if (n.size() != crypto_stream_NONCEBYTES) throw "incorrect nonce length";
+  if (k.size() != crypto_stream_KEYBYTES) throw "incorrect key length";
+  size_t mlen = m.size();
+  unsigned char c[mlen];
+  crypto_stream_xor(c,
+    (const unsigned char *) m.c_str(),mlen,
+    (const unsigned char *) n.c_str(),
+    (const unsigned char *) k.c_str()
+    );
+  return string((char *) c,mlen);
+}
diff --git a/nacl/crypto_stream/xsalsa20/checksum b/nacl/crypto_stream/xsalsa20/checksum
new file mode 100644
index 00000000..cae64c0d
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/checksum
@@ -0,0 +1 @@
+201bc58a96adcb6ed339ca33c188af8ca04a4ce68be1e0953309ee09a0cf8e7a
diff --git a/nacl/crypto_stream/xsalsa20/ref/api.h b/nacl/crypto_stream/xsalsa20/ref/api.h
new file mode 100644
index 00000000..6910a7dc
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/ref/api.h
@@ -0,0 +1,2 @@
+#define CRYPTO_KEYBYTES 32
+#define CRYPTO_NONCEBYTES 24
diff --git a/nacl/crypto_stream/xsalsa20/ref/implementors b/nacl/crypto_stream/xsalsa20/ref/implementors
new file mode 100644
index 00000000..f6fb3c73
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/ref/implementors
@@ -0,0 +1 @@
+Daniel J. Bernstein
diff --git a/nacl/crypto_stream/xsalsa20/ref/stream.c b/nacl/crypto_stream/xsalsa20/ref/stream.c
new file mode 100644
index 00000000..2d710709
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/ref/stream.c
@@ -0,0 +1,22 @@
+/*
+version 20080914
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_hsalsa20.h"
+#include "crypto_stream_salsa20.h"
+#include "crypto_stream.h"
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream(
+        unsigned char *c,unsigned long long clen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char subkey[32];
+  crypto_core_hsalsa20(subkey,n,k,sigma);
+  return crypto_stream_salsa20(c,clen,n + 16,subkey);
+}
diff --git a/nacl/crypto_stream/xsalsa20/ref/xor.c b/nacl/crypto_stream/xsalsa20/ref/xor.c
new file mode 100644
index 00000000..13f3134a
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/ref/xor.c
@@ -0,0 +1,23 @@
+/*
+version 20080913
+D. J. Bernstein
+Public domain.
+*/
+
+#include "crypto_core_hsalsa20.h"
+#include "crypto_stream_salsa20.h"
+#include "crypto_stream.h"
+
+static const unsigned char sigma[16] = "expand 32-byte k";
+
+int crypto_stream_xor(
+        unsigned char *c,
+  const unsigned char *m,unsigned long long mlen,
+  const unsigned char *n,
+  const unsigned char *k
+)
+{
+  unsigned char subkey[32];
+  crypto_core_hsalsa20(subkey,n,k,sigma);
+  return crypto_stream_salsa20_xor(c,m,mlen,n + 16,subkey);
+}
diff --git a/nacl/crypto_stream/xsalsa20/selected b/nacl/crypto_stream/xsalsa20/selected
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/selected
diff --git a/nacl/crypto_stream/xsalsa20/used b/nacl/crypto_stream/xsalsa20/used
new file mode 100644
index 00000000..e69de29b
--- /dev/null
+++ b/nacl/crypto_stream/xsalsa20/used