Factor out some new libraries

	word64-map:
		Data.Word64Map
	network-addr:
		Network.Address
	tox-crypto:
		Crypto.Tox
	lifted-concurrent:
		Control.Concurrent.Lifted.Instrument
		Control.Concurrent.Async.Lifted.Instrument
	psq-wrap:
		Data.Wrapper.PSQInt
		Data.Wrapper.PSQ
	minmax-psq:
		Data.MinMaxPSQ
        tasks:
                Control.Concurrent.Tasks
	kad:
		Network.Kademlia
		Network.Kademlia.Bootstrap
		Network.Kademlia.Routing
		Network.Kademlia.CommonAPI
		Network.Kademlia.Persistence
		Network.Kademlia.Search

5 files changed, 724 insertions, 0 deletions

diff --git a/dht/cbits/cryptonite_bitfn.h b/dht/cbits/cryptonite_bitfn.h
new file mode 100644
index 00000000..3a00dd8a
--- /dev/null
+++ b/dht/cbits/cryptonite_bitfn.h
@@ -0,0 +1,253 @@
+/*
+ * Copyright (C) 2006-2014 Vincent Hanquez <vincent@snarc.org>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
+ * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
+ * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
+ * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
+ * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+ * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
+ * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef BITFN_H
+#define BITFN_H
+#include <stdint.h>
+
+#ifndef NO_INLINE_ASM
+/**********************************************************/
+# if (defined(__i386__))
+#  define ARCH_HAS_SWAP32
+static inline uint32_t bitfn_swap32(uint32_t a)
+{
+        asm ("bswap %0" : "=r" (a) : "0" (a));
+        return a;
+}
+/**********************************************************/
+# elif (defined(__arm__))
+#  define ARCH_HAS_SWAP32
+static inline uint32_t bitfn_swap32(uint32_t a)
+{
+        uint32_t tmp = a;
+        asm volatile ("eor %1, %0, %0, ror #16\n"
+                      "bic %1, %1, #0xff0000\n"
+                      "mov %0, %0, ror #8\n"
+                      "eor %0, %0, %1, lsr #8\n"
+                     : "=r" (a), "=r" (tmp) : "0" (a), "1" (tmp));
+        return a;
+}
+/**********************************************************/
+# elif defined(__x86_64__)
+#  define ARCH_HAS_SWAP32
+#  define ARCH_HAS_SWAP64
+static inline uint32_t bitfn_swap32(uint32_t a)
+{
+        asm ("bswap %0" : "=r" (a) : "0" (a));
+        return a;
+}
+
+static inline uint64_t bitfn_swap64(uint64_t a)
+{
+        asm ("bswap %0" : "=r" (a) : "0" (a));
+        return a;
+}
+
+# endif
+#endif /* NO_INLINE_ASM */
+/**********************************************************/
+
+#ifndef ARCH_HAS_ROL32
+static inline uint32_t rol32(uint32_t word, uint32_t shift)
+{
+        return (word << shift) | (word >> (32 - shift));
+}
+#endif
+
+#ifndef ARCH_HAS_ROR32
+static inline uint32_t ror32(uint32_t word, uint32_t shift)
+{
+        return (word >> shift) | (word << (32 - shift));
+}
+#endif
+
+#ifndef ARCH_HAS_ROL64
+static inline uint64_t rol64(uint64_t word, uint32_t shift)
+{
+        return (word << shift) | (word >> (64 - shift));
+}
+#endif
+
+#ifndef ARCH_HAS_ROR64
+static inline uint64_t ror64(uint64_t word, uint32_t shift)
+{
+        return (word >> shift) | (word << (64 - shift));
+}
+#endif
+
+#ifndef ARCH_HAS_SWAP32
+static inline uint32_t bitfn_swap32(uint32_t a)
+{
+        return (a << 24) | ((a & 0xff00) << 8) | ((a >> 8) & 0xff00) | (a >> 24);
+}
+#endif
+
+#ifndef ARCH_HAS_ARRAY_SWAP32
+static inline void array_swap32(uint32_t *d, uint32_t *s, uint32_t nb)
+{
+        while (nb--)
+                *d++ = bitfn_swap32(*s++);
+}
+#endif
+
+#ifndef ARCH_HAS_SWAP64
+static inline uint64_t bitfn_swap64(uint64_t a)
+{
+        return ((uint64_t) bitfn_swap32((uint32_t) (a >> 32))) |
+               (((uint64_t) bitfn_swap32((uint32_t) a)) << 32);
+}
+#endif
+
+#ifndef ARCH_HAS_ARRAY_SWAP64
+static inline void array_swap64(uint64_t *d, uint64_t *s, uint32_t nb)
+{
+        while (nb--)
+                *d++ = bitfn_swap64(*s++);
+}
+#endif
+
+#ifndef ARCH_HAS_MEMORY_ZERO
+static inline void memory_zero(void *ptr, uint32_t len)
+{
+        uint32_t *ptr32 = ptr;
+        uint8_t *ptr8;
+        int i;
+
+        for (i = 0; i < len / 4; i++)
+                *ptr32++ = 0;
+        if (len % 4) {
+                ptr8 = (uint8_t *) ptr32;
+                for (i = len % 4; i >= 0; i--)
+                        ptr8[i] = 0;
+        }
+}
+#endif
+
+#ifndef ARCH_HAS_ARRAY_COPY32
+static inline void array_copy32(uint32_t *d, uint32_t *s, uint32_t nb)
+{
+        while (nb--) *d++ = *s++;
+}
+#endif
+
+#ifndef ARCH_HAS_ARRAY_XOR32
+static inline void array_xor32(uint32_t *d, uint32_t *s, uint32_t nb)
+{
+        while (nb--) *d++ ^= *s++;
+}
+#endif
+
+#ifndef ARCH_HAS_ARRAY_COPY64
+static inline void array_copy64(uint64_t *d, uint64_t *s, uint32_t nb)
+{
+        while (nb--) *d++ = *s++;
+}
+#endif
+
+#ifdef __GNUC__
+#define bitfn_ntz(n) __builtin_ctz(n)
+#else
+#error "define ntz for your platform"
+#endif
+
+#ifdef __MINGW32__
+  # define LITTLE_ENDIAN 1234
+  # define BYTE_ORDER    LITTLE_ENDIAN
+#elif defined(__FreeBSD__) || defined(__DragonFly__) || defined(__NetBSD__)
+  # include <sys/endian.h>
+#elif defined(__OpenBSD__) || defined(__SVR4)
+  # include <sys/types.h>
+#elif defined(__APPLE__)
+  # include <machine/endian.h>
+#elif defined( BSD ) && ( BSD >= 199103 )
+  # include <machine/endian.h>
+#elif defined( __QNXNTO__ ) && defined( __LITTLEENDIAN__ )
+  # define LITTLE_ENDIAN 1234
+  # define BYTE_ORDER    LITTLE_ENDIAN
+#elif defined( __QNXNTO__ ) && defined( __BIGENDIAN__ )
+  # define BIG_ENDIAN 1234
+  # define BYTE_ORDER    BIG_ENDIAN
+#elif defined( _AIX )
+  # include <sys/machine.h>
+#else
+  # include <endian.h>
+#endif
+/* big endian to cpu */
+#if LITTLE_ENDIAN == BYTE_ORDER
+
+# define be32_to_cpu(a) bitfn_swap32(a)
+# define cpu_to_be32(a) bitfn_swap32(a)
+# define le32_to_cpu(a) (a)
+# define cpu_to_le32(a) (a)
+# define be64_to_cpu(a) bitfn_swap64(a)
+# define cpu_to_be64(a) bitfn_swap64(a)
+# define le64_to_cpu(a) (a)
+# define cpu_to_le64(a) (a)
+
+# define cpu_to_le32_array(d, s, l) array_copy32(d, s, l)
+# define le32_to_cpu_array(d, s, l) array_copy32(d, s, l)
+# define cpu_to_be32_array(d, s, l) array_swap32(d, s, l)
+# define be32_to_cpu_array(d, s, l) array_swap32(d, s, l)
+
+# define cpu_to_le64_array(d, s, l) array_copy64(d, s, l)
+# define le64_to_cpu_array(d, s, l) array_copy64(d, s, l)
+# define cpu_to_be64_array(d, s, l) array_swap64(d, s, l)
+# define be64_to_cpu_array(d, s, l) array_swap64(d, s, l)
+
+# define ror32_be(a, s) rol32(a, s)
+# define rol32_be(a, s) ror32(a, s)
+
+# define ARCH_IS_LITTLE_ENDIAN
+
+#elif BIG_ENDIAN == BYTE_ORDER
+
+# define be32_to_cpu(a) (a)
+# define cpu_to_be32(a) (a)
+# define be64_to_cpu(a) (a)
+# define cpu_to_be64(a) (a)
+# define le64_to_cpu(a) bitfn_swap64(a)
+# define cpu_to_le64(a) bitfn_swap64(a)
+# define le32_to_cpu(a) bitfn_swap32(a)
+# define cpu_to_le32(a) bitfn_swap32(a)
+
+# define cpu_to_le32_array(d, s, l) array_swap32(d, s, l)
+# define le32_to_cpu_array(d, s, l) array_swap32(d, s, l)
+# define cpu_to_be32_array(d, s, l) array_copy32(d, s, l)
+# define be32_to_cpu_array(d, s, l) array_copy32(d, s, l)
+
+# define cpu_to_le64_array(d, s, l) array_swap64(d, s, l)
+# define le64_to_cpu_array(d, s, l) array_swap64(d, s, l)
+# define cpu_to_be64_array(d, s, l) array_copy64(d, s, l)
+# define be64_to_cpu_array(d, s, l) array_copy64(d, s, l)
+
+# define ror32_be(a, s) ror32(a, s)
+# define rol32_be(a, s) rol32(a, s)
+
+# define ARCH_IS_BIG_ENDIAN
+
+#else
+# error "endian not supported"
+#endif
+
+#endif /* !BITFN_H */
diff --git a/dht/cbits/cryptonite_salsa.c b/dht/cbits/cryptonite_salsa.c
new file mode 100644
index 00000000..0bd96607
--- /dev/null
+++ b/dht/cbits/cryptonite_salsa.c
@@ -0,0 +1,297 @@
+/*
+ * Copyright (c) 2014-2015 Vincent Hanquez <vincent@snarc.org>
+ *
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the author nor the names of his contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+
+#include <stdint.h>
+#include <string.h>
+#include <stdio.h>
+#include "cryptonite_salsa.h"
+#include "cryptonite_bitfn.h"
+
+static const uint8_t sigma[16] = "expand 32-byte k";
+static const uint8_t tau[16] = "expand 16-byte k";
+
+#define QR(a,b,c,d) \
+        b ^= rol32(a+d, 7); \
+        c ^= rol32(b+a, 9); \
+        d ^= rol32(c+b, 13); \
+        a ^= rol32(d+c, 18);
+
+#define ALIGNED64(PTR) \
+        (((uintptr_t)(const void *)(PTR)) % 8 == 0)
+
+#define SALSA_CORE_LOOP \
+        for (i = rounds; i > 0; i -= 2) { \
+                QR (x0,x4,x8,x12); \
+                QR (x5,x9,x13,x1); \
+                QR (x10,x14,x2,x6); \
+                QR (x15,x3,x7,x11); \
+                QR (x0,x1,x2,x3); \
+                QR (x5,x6,x7,x4); \
+                QR (x10,x11,x8,x9); \
+                QR (x15,x12,x13,x14); \
+        }
+
+static inline uint32_t load32(const uint8_t *p)
+{
+        return le32_to_cpu(*((uint32_t *) p));
+}
+
+static void salsa_core(int rounds, block *out, const cryptonite_salsa_state *in)
+{
+        uint32_t x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15;
+        int i;
+
+        x0 = in->d[0]; x1 = in->d[1]; x2 = in->d[2]; x3 = in->d[3];
+        x4 = in->d[4]; x5 = in->d[5]; x6 = in->d[6]; x7 = in->d[7];
+        x8 = in->d[8]; x9 = in->d[9]; x10 = in->d[10]; x11 = in->d[11];
+        x12 = in->d[12]; x13 = in->d[13]; x14 = in->d[14]; x15 = in->d[15];
+
+        SALSA_CORE_LOOP;
+
+        x0 += in->d[0]; x1 += in->d[1]; x2 += in->d[2]; x3 += in->d[3];
+        x4 += in->d[4]; x5 += in->d[5]; x6 += in->d[6]; x7 += in->d[7];
+        x8 += in->d[8]; x9 += in->d[9]; x10 += in->d[10]; x11 += in->d[11];
+        x12 += in->d[12]; x13 += in->d[13]; x14 += in->d[14]; x15 += in->d[15];
+
+        out->d[0] = cpu_to_le32(x0);
+        out->d[1] = cpu_to_le32(x1);
+        out->d[2] = cpu_to_le32(x2);
+        out->d[3] = cpu_to_le32(x3);
+        out->d[4] = cpu_to_le32(x4);
+        out->d[5] = cpu_to_le32(x5);
+        out->d[6] = cpu_to_le32(x6);
+        out->d[7] = cpu_to_le32(x7);
+        out->d[8] = cpu_to_le32(x8);
+        out->d[9] = cpu_to_le32(x9);
+        out->d[10] = cpu_to_le32(x10);
+        out->d[11] = cpu_to_le32(x11);
+        out->d[12] = cpu_to_le32(x12);
+        out->d[13] = cpu_to_le32(x13);
+        out->d[14] = cpu_to_le32(x14);
+        out->d[15] = cpu_to_le32(x15);
+}
+
+void cryptonite_salsa_core_xor(int rounds, block *out, block *in)
+{
+        uint32_t x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15;
+        int i;
+
+#define LOAD(i) (out->d[i] ^= in->d[i])
+        x0 = LOAD(0); x1 = LOAD(1); x2 = LOAD(2); x3 = LOAD(3);
+        x4 = LOAD(4); x5 = LOAD(5); x6 = LOAD(6); x7 = LOAD(7);
+        x8 = LOAD(8); x9 = LOAD(9); x10 = LOAD(10); x11 = LOAD(11);
+        x12 = LOAD(12); x13 = LOAD(13); x14 = LOAD(14); x15 = LOAD(15);
+#undef LOAD
+
+        SALSA_CORE_LOOP;
+
+        out->d[0] += x0; out->d[1] += x1; out->d[2] += x2; out->d[3] += x3;
+        out->d[4] += x4; out->d[5] += x5; out->d[6] += x6; out->d[7] += x7;
+        out->d[8] += x8; out->d[9] += x9; out->d[10] += x10; out->d[11] += x11;
+        out->d[12] += x12; out->d[13] += x13; out->d[14] += x14; out->d[15] += x15;
+}
+
+/* only 2 valid values for keylen are 256 (32) and 128 (16) */
+void cryptonite_salsa_init_core(cryptonite_salsa_state *st,
+                                uint32_t keylen, const uint8_t *key,
+                                uint32_t ivlen, const uint8_t *iv)
+{
+        const uint8_t *constants = (keylen == 32) ? sigma : tau;
+        int i;
+
+        st->d[0] = load32(constants + 0);
+        st->d[5] = load32(constants + 4);
+        st->d[10] = load32(constants + 8);
+        st->d[15] = load32(constants + 12);
+
+        st->d[1] = load32(key + 0);
+        st->d[2] = load32(key + 4);
+        st->d[3] = load32(key + 8);
+        st->d[4] = load32(key + 12);
+        /* we repeat the key on 128 bits */
+        if (keylen == 32)
+                key += 16;
+        st->d[11] = load32(key + 0);
+        st->d[12] = load32(key + 4);
+        st->d[13] = load32(key + 8);
+        st->d[14] = load32(key + 12);
+
+        st->d[9] = 0;
+        switch (ivlen) {
+        case 8:
+                st->d[6] = load32(iv + 0);
+                st->d[7] = load32(iv + 4);
+                st->d[8] = 0;
+                break;
+        case 12:
+                st->d[6] = load32(iv + 0);
+                st->d[7] = load32(iv + 4);
+                st->d[8] = load32(iv + 8);
+        default:
+                return;
+        }
+}
+
+void cryptonite_salsa_init(cryptonite_salsa_context *ctx, uint8_t nb_rounds,
+                           uint32_t keylen, const uint8_t *key,
+                           uint32_t ivlen, const uint8_t *iv)
+{
+        memset(ctx, 0, sizeof(*ctx));
+        ctx->nb_rounds = nb_rounds;
+        cryptonite_salsa_init_core(&ctx->st, keylen, key, ivlen, iv);
+}
+
+void cryptonite_salsa_combine(uint8_t *dst, cryptonite_salsa_context *ctx, const uint8_t *src, uint32_t bytes)
+{
+        block out;
+        cryptonite_salsa_state *st;
+        int i;
+
+        if (!bytes)
+                return;
+
+        /* xor the previous buffer first (if any) */
+        if (ctx->prev_len > 0) {
+                int to_copy = (ctx->prev_len < bytes) ? ctx->prev_len : bytes;
+                for (i = 0; i < to_copy; i++)
+                        dst[i] = src[i] ^ ctx->prev[ctx->prev_ofs+i];
+                memset(ctx->prev + ctx->prev_ofs, 0, to_copy);
+                ctx->prev_len -= to_copy;
+                ctx->prev_ofs += to_copy;
+                src += to_copy;
+                dst += to_copy;
+                bytes -= to_copy;
+        }
+
+        if (bytes == 0)
+                return;
+
+        st = &ctx->st;
+
+        /* xor new 64-bytes chunks and store the left over if any */
+        for (; bytes >= 64; bytes -= 64, src += 64, dst += 64) {
+                /* generate new chunk and update state */
+                salsa_core(ctx->nb_rounds, &out, st);
+                st->d[8] += 1;
+                if (st->d[8] == 0)
+                        st->d[9] += 1;
+
+                for (i = 0; i < 64; ++i)
+                        dst[i] = src[i] ^ out.b[i];
+        }
+
+        if (bytes > 0) {
+                /* generate new chunk and update state */
+                salsa_core(ctx->nb_rounds, &out, st);
+                st->d[8] += 1;
+                if (st->d[8] == 0)
+                        st->d[9] += 1;
+
+                /* xor as much as needed */
+                for (i = 0; i < bytes; i++)
+                        dst[i] = src[i] ^ out.b[i];
+                
+                /* copy the left over in the buffer */
+                ctx->prev_len = 64 - bytes;
+                ctx->prev_ofs = i;
+                for (; i < 64; i++) {
+                        ctx->prev[i] = out.b[i];
+                }
+        }
+}
+
+void cryptonite_salsa_generate(uint8_t *dst, cryptonite_salsa_context *ctx, uint32_t bytes)
+{
+        cryptonite_salsa_state *st;
+        block out;
+        int i;
+
+        if (!bytes)
+                return;
+
+        /* xor the previous buffer first (if any) */
+        if (ctx->prev_len > 0) {
+                int to_copy = (ctx->prev_len < bytes) ? ctx->prev_len : bytes;
+                for (i = 0; i < to_copy; i++)
+                        dst[i] = ctx->prev[ctx->prev_ofs+i];
+                memset(ctx->prev + ctx->prev_ofs, 0, to_copy);
+                ctx->prev_len -= to_copy;
+                ctx->prev_ofs += to_copy;
+                dst += to_copy;
+                bytes -= to_copy;
+        }
+
+        if (bytes == 0)
+                return;
+
+        st = &ctx->st;
+
+        if (ALIGNED64(dst)) {
+                /* xor new 64-bytes chunks and store the left over if any */
+                for (; bytes >= 64; bytes -= 64, dst += 64) {
+                        /* generate new chunk and update state */
+                        salsa_core(ctx->nb_rounds, (block *) dst, st);
+                        st->d[8] += 1;
+                        if (st->d[8] == 0)
+                                st->d[9] += 1;
+                }
+        } else {
+                /* xor new 64-bytes chunks and store the left over if any */
+                for (; bytes >= 64; bytes -= 64, dst += 64) {
+                        /* generate new chunk and update state */
+                        salsa_core(ctx->nb_rounds, &out, st);
+                        st->d[8] += 1;
+                        if (st->d[8] == 0)
+                                st->d[9] += 1;
+
+                        for (i = 0; i < 64; ++i)
+                                dst[i] = out.b[i];
+                }
+        }
+
+        if (bytes > 0) {
+                /* generate new chunk and update state */
+                salsa_core(ctx->nb_rounds, &out, st);
+                st->d[8] += 1;
+                if (st->d[8] == 0)
+                        st->d[9] += 1;
+
+                /* xor as much as needed */
+                for (i = 0; i < bytes; i++)
+                        dst[i] = out.b[i];
+                
+                /* copy the left over in the buffer */
+                ctx->prev_len = 64 - bytes;
+                ctx->prev_ofs = i;
+                for (; i < 64; i++)
+                        ctx->prev[i] = out.b[i];
+        }
+}
+
diff --git a/dht/cbits/cryptonite_salsa.h b/dht/cbits/cryptonite_salsa.h
new file mode 100644
index 00000000..33e9cda9
--- /dev/null
+++ b/dht/cbits/cryptonite_salsa.h
@@ -0,0 +1,57 @@
+/*
+ * Copyright (c) 2014 Vincent Hanquez <vincent@snarc.org>
+ *
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the author nor the names of his contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+#ifndef CRYPTONITE_SALSA
+#define CRYPTONITE_SALSA
+
+typedef union {
+        uint64_t q[8];
+        uint32_t d[16];
+        uint8_t  b[64];
+} block;
+
+typedef block cryptonite_salsa_state;
+
+typedef struct {
+        cryptonite_salsa_state st;
+        uint8_t prev[64];
+        uint8_t prev_ofs;
+        uint8_t prev_len;
+        uint8_t nb_rounds;
+} cryptonite_salsa_context;
+
+/* for scrypt */
+void cryptonite_salsa_core_xor(int rounds, block *out, block *in);
+
+void cryptonite_salsa_init_core(cryptonite_salsa_state *st, uint32_t keylen, const uint8_t *key, uint32_t ivlen, const uint8_t *iv);
+void cryptonite_salsa_init(cryptonite_salsa_context *ctx, uint8_t nb_rounds, uint32_t keylen, const uint8_t *key, uint32_t ivlen, const uint8_t *iv);
+void cryptonite_salsa_combine(uint8_t *dst, cryptonite_salsa_context *st, const uint8_t *src, uint32_t bytes);
+void cryptonite_salsa_generate(uint8_t *dst, cryptonite_salsa_context *st, uint32_t bytes);
+
+#endif
diff --git a/dht/cbits/cryptonite_xsalsa.c b/dht/cbits/cryptonite_xsalsa.c
new file mode 100644
index 00000000..6718cd7d
--- /dev/null
+++ b/dht/cbits/cryptonite_xsalsa.c
@@ -0,0 +1,80 @@
+/*
+ * Copyright (c) 2016 Brandon Hamilton <brandon.hamilton@gmail.com>
+ * 
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the author nor the names of his contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+#include <stdint.h>
+#include <string.h>
+#include "cryptonite_xsalsa.h"
+#include "cryptonite_bitfn.h"
+
+static inline uint32_t load32(const uint8_t *p)
+{
+  return le32_to_cpu(*((uint32_t *) p));
+}
+
+/* XSalsa20 algorithm as described in https://cr.yp.to/snuffle/xsalsa-20081128.pdf */
+void cryptonite_xsalsa_init(cryptonite_salsa_context *ctx, uint8_t nb_rounds,
+                            uint32_t keylen, const uint8_t *key,
+                            uint32_t ivlen, const uint8_t *iv)
+{
+  memset(ctx, 0, sizeof(*ctx));
+  ctx->nb_rounds = nb_rounds;
+
+  /* Create initial 512-bit input block:
+       (x0, x5, x10, x15) is the Salsa20 constant
+       (x1, x2, x3, x4, x11, x12, x13, x14) is a 256-bit key
+       (x6, x7, x8, x9) is the first 128 bits of a 192-bit nonce
+  */
+  cryptonite_salsa_init_core(&ctx->st, keylen, key, 8, iv);
+  ctx->st.d[ 8] = load32(iv + 8);
+  ctx->st.d[ 9] = load32(iv + 12);
+
+  /* Compute (z0, z1, . . . , z15) = doubleround ^(r/2) (x0, x1, . . . , x15) */
+  block hSalsa;
+  memset(&hSalsa, 0, sizeof(block));
+  cryptonite_salsa_core_xor(nb_rounds, &hSalsa, &ctx->st);
+ 
+  /* Build a new 512-bit input block (x′0, x′1, . . . , x′15):
+       (x′0, x′5, x′10, x′15) is the Salsa20 constant
+       (x′1,x′2,x′3,x′4,x′11,x′12,x′13,x′14) = (z0,z5,z10,z15,z6,z7,z8,z9)
+       (x′6,x′7) is the last 64 bits of the 192-bit nonce
+       (x′8, x′9) is a 64-bit block counter.
+  */
+  ctx->st.d[ 1] = hSalsa.d[ 0] - ctx->st.d[ 0];
+  ctx->st.d[ 2] = hSalsa.d[ 5] - ctx->st.d[ 5];
+  ctx->st.d[ 3] = hSalsa.d[10] - ctx->st.d[10];
+  ctx->st.d[ 4] = hSalsa.d[15] - ctx->st.d[15];
+  ctx->st.d[11] = hSalsa.d[ 6] - ctx->st.d[ 6];
+  ctx->st.d[12] = hSalsa.d[ 7] - ctx->st.d[ 7];
+  ctx->st.d[13] = hSalsa.d[ 8] - ctx->st.d[ 8];
+  ctx->st.d[14] = hSalsa.d[ 9] - ctx->st.d[ 9];
+  ctx->st.d[ 6] = load32(iv + 16);
+  ctx->st.d[ 7] = load32(iv + 20);
+  ctx->st.d[ 8] = 0;
+  ctx->st.d[ 9] = 0;
+}
\ No newline at end of file
diff --git a/dht/cbits/cryptonite_xsalsa.h b/dht/cbits/cryptonite_xsalsa.h
new file mode 100644
index 00000000..73233cee
--- /dev/null
+++ b/dht/cbits/cryptonite_xsalsa.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (c) 2016 Brandon Hamilton <brandon.hamilton@gmail.com>
+ *
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of the author nor the names of his contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
+ * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHORS OR CONTRIBUTORS BE LIABLE
+ * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
+ * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+ * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
+ * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
+ * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
+ * SUCH DAMAGE.
+ */
+#ifndef CRYPTONITE_XSALSA
+#define CRYPTONITE_XSALSA
+
+#include "cryptonite_salsa.h"
+
+void cryptonite_xsalsa_init(cryptonite_salsa_context *ctx, uint8_t nb_rounds, uint32_t keylen, const uint8_t *key, uint32_t ivlen, const uint8_t *iv);
+
+#endif