lib: Add xxhash module

Adds xxhash kernel module with xxh32 and xxh64 hashes. xxhash is an
extremely fast non-cryptographic hash algorithm for checksumming.
The zstd compression and decompression modules added in the next patch
require xxhash. I extracted it out from zstd since it is useful on its
own. I copied the code from the upstream XXHash source repository and
translated it into kernel style. I ran benchmarks and tests in the kernel
and tests in userland.

I benchmarked xxhash as a special character device. I ran in four modes,
no-op, xxh32, xxh64, and crc32. The no-op mode simply copies the data to
kernel space and ignores it. The xxh32, xxh64, and crc32 modes compute
hashes on the copied data. I also ran it with four different buffer sizes.
The benchmark file is located in the upstream zstd source repository under
`contrib/linux-kernel/xxhash_test.c` [1].

I ran the benchmarks on a Ubuntu 14.04 VM with 2 cores and 4 GiB of RAM.
The VM is running on a MacBook Pro with a 3.1 GHz Intel Core i7 processor,
16 GB of RAM, and a SSD. I benchmarked using the file `filesystem.squashfs`
from `ubuntu-16.10-desktop-amd64.iso`, which is 1,536,217,088 B large.
Run the following commands for the benchmark:

    modprobe xxhash_test
    mknod xxhash_test c 245 0
    time cp filesystem.squashfs xxhash_test

The time is reported by the time of the userland `cp`.
The GB/s is computed with

    1,536,217,008 B / time(buffer size, hash)

which includes the time to copy from userland.
The Normalized GB/s is computed with

    1,536,217,088 B / (time(buffer size, hash) - time(buffer size, none)).

| Buffer Size (B) | Hash  | Time (s) | GB/s | Adjusted GB/s |
|-----------------|-------|----------|------|---------------|
|            1024 | none  |    0.408 | 3.77 |             - |
|            1024 | xxh32 |    0.649 | 2.37 |          6.37 |
|            1024 | xxh64 |    0.542 | 2.83 |         11.46 |
|            1024 | crc32 |    1.290 | 1.19 |          1.74 |
|            4096 | none  |    0.380 | 4.04 |             - |
|            4096 | xxh32 |    0.645 | 2.38 |          5.79 |
|            4096 | xxh64 |    0.500 | 3.07 |         12.80 |
|            4096 | crc32 |    1.168 | 1.32 |          1.95 |
|            8192 | none  |    0.351 | 4.38 |             - |
|            8192 | xxh32 |    0.614 | 2.50 |          5.84 |
|            8192 | xxh64 |    0.464 | 3.31 |         13.60 |
|            8192 | crc32 |    1.163 | 1.32 |          1.89 |
|           16384 | none  |    0.346 | 4.43 |             - |
|           16384 | xxh32 |    0.590 | 2.60 |          6.30 |
|           16384 | xxh64 |    0.466 | 3.30 |         12.80 |
|           16384 | crc32 |    1.183 | 1.30 |          1.84 |

Tested in userland using the test-suite in the zstd repo under
`contrib/linux-kernel/test/XXHashUserlandTest.cpp` [2] by mocking the
kernel functions. A line in each branch of every function in `xxhash.c`
was commented out to ensure that the test-suite fails. Additionally
tested while testing zstd and with SMHasher [3].

[1] https://phabricator.intern.facebook.com/P57526246
[2] https://github.com/facebook/zstd/blob/dev/contrib/linux-kernel/test/XXHashUserlandTest.cpp
[3] https://github.com/aappleby/smhasher

zstd source repository: https://github.com/facebook/zstd
XXHash source repository: https://github.com/cyan4973/xxhash

Signed-off-by: Nick Terrell <terrelln@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>

3 files changed, 504 insertions, 0 deletions

diff --git a/lib/Kconfig b/lib/Kconfig
index 6762529..5e7541f 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -192,6 +192,9 @@ config CRC8
 	  when they need to do cyclic redundancy check according CRC8
 	  algorithm. Module will be called crc8.
 
+config XXHASH
+	tristate
+
 config AUDIT_GENERIC
 	bool
 	depends on AUDIT && !AUDIT_ARCH
diff --git a/lib/Makefile b/lib/Makefile
index 40c1837..d06b68a 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -102,6 +102,7 @@ obj-$(CONFIG_CRC4)	+= crc4.o
 obj-$(CONFIG_CRC7)	+= crc7.o
 obj-$(CONFIG_LIBCRC32C)	+= libcrc32c.o
 obj-$(CONFIG_CRC8)	+= crc8.o
+obj-$(CONFIG_XXHASH)	+= xxhash.o
 obj-$(CONFIG_GENERIC_ALLOCATOR) += genalloc.o
 
 obj-$(CONFIG_842_COMPRESS) += 842/
diff --git a/lib/xxhash.c b/lib/xxhash.c
new file mode 100644
index 0000000..aa61e2a
--- /dev/null
+++ b/lib/xxhash.c
@@ -0,0 +1,500 @@
+/*
+ * xxHash - Extremely Fast Hash algorithm
+ * Copyright (C) 2012-2016, Yann Collet.
+ *
+ * BSD 2-Clause License (http://www.opensource.org/licenses/bsd-license.php)
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are
+ * met:
+ *
+ *   * Redistributions of source code must retain the above copyright
+ *     notice, this list of conditions and the following disclaimer.
+ *   * 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 COPYRIGHT HOLDERS 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 COPYRIGHT
+ * OWNER 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.
+ *
+ * This program is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU General Public License version 2 as published by the
+ * Free Software Foundation. This program is dual-licensed; you may select
+ * either version 2 of the GNU General Public License ("GPL") or BSD license
+ * ("BSD").
+ *
+ * You can contact the author at:
+ * - xxHash homepage: http://cyan4973.github.io/xxHash/
+ * - xxHash source repository: https://github.com/Cyan4973/xxHash
+ */
+
+#include <asm/unaligned.h>
+#include <linux/errno.h>
+#include <linux/compiler.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include <linux/xxhash.h>
+
+/*-*************************************
+ * Macros
+ **************************************/
+#define xxh_rotl32(x, r) ((x << r) | (x >> (32 - r)))
+#define xxh_rotl64(x, r) ((x << r) | (x >> (64 - r)))
+
+#ifdef __LITTLE_ENDIAN
+# define XXH_CPU_LITTLE_ENDIAN 1
+#else
+# define XXH_CPU_LITTLE_ENDIAN 0
+#endif
+
+/*-*************************************
+ * Constants
+ **************************************/
+static const uint32_t PRIME32_1 = 2654435761U;
+static const uint32_t PRIME32_2 = 2246822519U;
+static const uint32_t PRIME32_3 = 3266489917U;
+static const uint32_t PRIME32_4 =  668265263U;
+static const uint32_t PRIME32_5 =  374761393U;
+
+static const uint64_t PRIME64_1 = 11400714785074694791ULL;
+static const uint64_t PRIME64_2 = 14029467366897019727ULL;
+static const uint64_t PRIME64_3 =  1609587929392839161ULL;
+static const uint64_t PRIME64_4 =  9650029242287828579ULL;
+static const uint64_t PRIME64_5 =  2870177450012600261ULL;
+
+/*-**************************
+ *  Utils
+ ***************************/
+void xxh32_copy_state(struct xxh32_state *dst, const struct xxh32_state *src)
+{
+	memcpy(dst, src, sizeof(*dst));
+}
+EXPORT_SYMBOL(xxh32_copy_state);
+
+void xxh64_copy_state(struct xxh64_state *dst, const struct xxh64_state *src)
+{
+	memcpy(dst, src, sizeof(*dst));
+}
+EXPORT_SYMBOL(xxh64_copy_state);
+
+/*-***************************
+ * Simple Hash Functions
+ ****************************/
+static uint32_t xxh32_round(uint32_t seed, const uint32_t input)
+{
+	seed += input * PRIME32_2;
+	seed = xxh_rotl32(seed, 13);
+	seed *= PRIME32_1;
+	return seed;
+}
+
+uint32_t xxh32(const void *input, const size_t len, const uint32_t seed)
+{
+	const uint8_t *p = (const uint8_t *)input;
+	const uint8_t *b_end = p + len;
+	uint32_t h32;
+
+	if (len >= 16) {
+		const uint8_t *const limit = b_end - 16;
+		uint32_t v1 = seed + PRIME32_1 + PRIME32_2;
+		uint32_t v2 = seed + PRIME32_2;
+		uint32_t v3 = seed + 0;
+		uint32_t v4 = seed - PRIME32_1;
+
+		do {
+			v1 = xxh32_round(v1, get_unaligned_le32(p));
+			p += 4;
+			v2 = xxh32_round(v2, get_unaligned_le32(p));
+			p += 4;
+			v3 = xxh32_round(v3, get_unaligned_le32(p));
+			p += 4;
+			v4 = xxh32_round(v4, get_unaligned_le32(p));
+			p += 4;
+		} while (p <= limit);
+
+		h32 = xxh_rotl32(v1, 1) + xxh_rotl32(v2, 7) +
+			xxh_rotl32(v3, 12) + xxh_rotl32(v4, 18);
+	} else {
+		h32 = seed + PRIME32_5;
+	}
+
+	h32 += (uint32_t)len;
+
+	while (p + 4 <= b_end) {
+		h32 += get_unaligned_le32(p) * PRIME32_3;
+		h32 = xxh_rotl32(h32, 17) * PRIME32_4;
+		p += 4;
+	}
+
+	while (p < b_end) {
+		h32 += (*p) * PRIME32_5;
+		h32 = xxh_rotl32(h32, 11) * PRIME32_1;
+		p++;
+	}
+
+	h32 ^= h32 >> 15;
+	h32 *= PRIME32_2;
+	h32 ^= h32 >> 13;
+	h32 *= PRIME32_3;
+	h32 ^= h32 >> 16;
+
+	return h32;
+}
+EXPORT_SYMBOL(xxh32);
+
+static uint64_t xxh64_round(uint64_t acc, const uint64_t input)
+{
+	acc += input * PRIME64_2;
+	acc = xxh_rotl64(acc, 31);
+	acc *= PRIME64_1;
+	return acc;
+}
+
+static uint64_t xxh64_merge_round(uint64_t acc, uint64_t val)
+{
+	val = xxh64_round(0, val);
+	acc ^= val;
+	acc = acc * PRIME64_1 + PRIME64_4;
+	return acc;
+}
+
+uint64_t xxh64(const void *input, const size_t len, const uint64_t seed)
+{
+	const uint8_t *p = (const uint8_t *)input;
+	const uint8_t *const b_end = p + len;
+	uint64_t h64;
+
+	if (len >= 32) {
+		const uint8_t *const limit = b_end - 32;
+		uint64_t v1 = seed + PRIME64_1 + PRIME64_2;
+		uint64_t v2 = seed + PRIME64_2;
+		uint64_t v3 = seed + 0;
+		uint64_t v4 = seed - PRIME64_1;
+
+		do {
+			v1 = xxh64_round(v1, get_unaligned_le64(p));
+			p += 8;
+			v2 = xxh64_round(v2, get_unaligned_le64(p));
+			p += 8;
+			v3 = xxh64_round(v3, get_unaligned_le64(p));
+			p += 8;
+			v4 = xxh64_round(v4, get_unaligned_le64(p));
+			p += 8;
+		} while (p <= limit);
+
+		h64 = xxh_rotl64(v1, 1) + xxh_rotl64(v2, 7) +
+			xxh_rotl64(v3, 12) + xxh_rotl64(v4, 18);
+		h64 = xxh64_merge_round(h64, v1);
+		h64 = xxh64_merge_round(h64, v2);
+		h64 = xxh64_merge_round(h64, v3);
+		h64 = xxh64_merge_round(h64, v4);
+
+	} else {
+		h64  = seed + PRIME64_5;
+	}
+
+	h64 += (uint64_t)len;
+
+	while (p + 8 <= b_end) {
+		const uint64_t k1 = xxh64_round(0, get_unaligned_le64(p));
+
+		h64 ^= k1;
+		h64 = xxh_rotl64(h64, 27) * PRIME64_1 + PRIME64_4;
+		p += 8;
+	}
+
+	if (p + 4 <= b_end) {
+		h64 ^= (uint64_t)(get_unaligned_le32(p)) * PRIME64_1;
+		h64 = xxh_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
+		p += 4;
+	}
+
+	while (p < b_end) {
+		h64 ^= (*p) * PRIME64_5;
+		h64 = xxh_rotl64(h64, 11) * PRIME64_1;
+		p++;
+	}
+
+	h64 ^= h64 >> 33;
+	h64 *= PRIME64_2;
+	h64 ^= h64 >> 29;
+	h64 *= PRIME64_3;
+	h64 ^= h64 >> 32;
+
+	return h64;
+}
+EXPORT_SYMBOL(xxh64);
+
+/*-**************************************************
+ * Advanced Hash Functions
+ ***************************************************/
+void xxh32_reset(struct xxh32_state *statePtr, const uint32_t seed)
+{
+	/* use a local state for memcpy() to avoid strict-aliasing warnings */
+	struct xxh32_state state;
+
+	memset(&state, 0, sizeof(state));
+	state.v1 = seed + PRIME32_1 + PRIME32_2;
+	state.v2 = seed + PRIME32_2;
+	state.v3 = seed + 0;
+	state.v4 = seed - PRIME32_1;
+	memcpy(statePtr, &state, sizeof(state));
+}
+EXPORT_SYMBOL(xxh32_reset);
+
+void xxh64_reset(struct xxh64_state *statePtr, const uint64_t seed)
+{
+	/* use a local state for memcpy() to avoid strict-aliasing warnings */
+	struct xxh64_state state;
+
+	memset(&state, 0, sizeof(state));
+	state.v1 = seed + PRIME64_1 + PRIME64_2;
+	state.v2 = seed + PRIME64_2;
+	state.v3 = seed + 0;
+	state.v4 = seed - PRIME64_1;
+	memcpy(statePtr, &state, sizeof(state));
+}
+EXPORT_SYMBOL(xxh64_reset);
+
+int xxh32_update(struct xxh32_state *state, const void *input, const size_t len)
+{
+	const uint8_t *p = (const uint8_t *)input;
+	const uint8_t *const b_end = p + len;
+
+	if (input == NULL)
+		return -EINVAL;
+
+	state->total_len_32 += (uint32_t)len;
+	state->large_len |= (len >= 16) | (state->total_len_32 >= 16);
+
+	if (state->memsize + len < 16) { /* fill in tmp buffer */
+		memcpy((uint8_t *)(state->mem32) + state->memsize, input, len);
+		state->memsize += (uint32_t)len;
+		return 0;
+	}
+
+	if (state->memsize) { /* some data left from previous update */
+		const uint32_t *p32 = state->mem32;
+
+		memcpy((uint8_t *)(state->mem32) + state->memsize, input,
+			16 - state->memsize);
+
+		state->v1 = xxh32_round(state->v1, get_unaligned_le32(p32));
+		p32++;
+		state->v2 = xxh32_round(state->v2, get_unaligned_le32(p32));
+		p32++;
+		state->v3 = xxh32_round(state->v3, get_unaligned_le32(p32));
+		p32++;
+		state->v4 = xxh32_round(state->v4, get_unaligned_le32(p32));
+		p32++;
+
+		p += 16-state->memsize;
+		state->memsize = 0;
+	}
+
+	if (p <= b_end - 16) {
+		const uint8_t *const limit = b_end - 16;
+		uint32_t v1 = state->v1;
+		uint32_t v2 = state->v2;
+		uint32_t v3 = state->v3;
+		uint32_t v4 = state->v4;
+
+		do {
+			v1 = xxh32_round(v1, get_unaligned_le32(p));
+			p += 4;
+			v2 = xxh32_round(v2, get_unaligned_le32(p));
+			p += 4;
+			v3 = xxh32_round(v3, get_unaligned_le32(p));
+			p += 4;
+			v4 = xxh32_round(v4, get_unaligned_le32(p));
+			p += 4;
+		} while (p <= limit);
+
+		state->v1 = v1;
+		state->v2 = v2;
+		state->v3 = v3;
+		state->v4 = v4;
+	}
+
+	if (p < b_end) {
+		memcpy(state->mem32, p, (size_t)(b_end-p));
+		state->memsize = (uint32_t)(b_end-p);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(xxh32_update);
+
+uint32_t xxh32_digest(const struct xxh32_state *state)
+{
+	const uint8_t *p = (const uint8_t *)state->mem32;
+	const uint8_t *const b_end = (const uint8_t *)(state->mem32) +
+		state->memsize;
+	uint32_t h32;
+
+	if (state->large_len) {
+		h32 = xxh_rotl32(state->v1, 1) + xxh_rotl32(state->v2, 7) +
+			xxh_rotl32(state->v3, 12) + xxh_rotl32(state->v4, 18);
+	} else {
+		h32 = state->v3 /* == seed */ + PRIME32_5;
+	}
+
+	h32 += state->total_len_32;
+
+	while (p + 4 <= b_end) {
+		h32 += get_unaligned_le32(p) * PRIME32_3;
+		h32 = xxh_rotl32(h32, 17) * PRIME32_4;
+		p += 4;
+	}
+
+	while (p < b_end) {
+		h32 += (*p) * PRIME32_5;
+		h32 = xxh_rotl32(h32, 11) * PRIME32_1;
+		p++;
+	}
+
+	h32 ^= h32 >> 15;
+	h32 *= PRIME32_2;
+	h32 ^= h32 >> 13;
+	h32 *= PRIME32_3;
+	h32 ^= h32 >> 16;
+
+	return h32;
+}
+EXPORT_SYMBOL(xxh32_digest);
+
+int xxh64_update(struct xxh64_state *state, const void *input, const size_t len)
+{
+	const uint8_t *p = (const uint8_t *)input;
+	const uint8_t *const b_end = p + len;
+
+	if (input == NULL)
+		return -EINVAL;
+
+	state->total_len += len;
+
+	if (state->memsize + len < 32) { /* fill in tmp buffer */
+		memcpy(((uint8_t *)state->mem64) + state->memsize, input, len);
+		state->memsize += (uint32_t)len;
+		return 0;
+	}
+
+	if (state->memsize) { /* tmp buffer is full */
+		uint64_t *p64 = state->mem64;
+
+		memcpy(((uint8_t *)p64) + state->memsize, input,
+			32 - state->memsize);
+
+		state->v1 = xxh64_round(state->v1, get_unaligned_le64(p64));
+		p64++;
+		state->v2 = xxh64_round(state->v2, get_unaligned_le64(p64));
+		p64++;
+		state->v3 = xxh64_round(state->v3, get_unaligned_le64(p64));
+		p64++;
+		state->v4 = xxh64_round(state->v4, get_unaligned_le64(p64));
+
+		p += 32 - state->memsize;
+		state->memsize = 0;
+	}
+
+	if (p + 32 <= b_end) {
+		const uint8_t *const limit = b_end - 32;
+		uint64_t v1 = state->v1;
+		uint64_t v2 = state->v2;
+		uint64_t v3 = state->v3;
+		uint64_t v4 = state->v4;
+
+		do {
+			v1 = xxh64_round(v1, get_unaligned_le64(p));
+			p += 8;
+			v2 = xxh64_round(v2, get_unaligned_le64(p));
+			p += 8;
+			v3 = xxh64_round(v3, get_unaligned_le64(p));
+			p += 8;
+			v4 = xxh64_round(v4, get_unaligned_le64(p));
+			p += 8;
+		} while (p <= limit);
+
+		state->v1 = v1;
+		state->v2 = v2;
+		state->v3 = v3;
+		state->v4 = v4;
+	}
+
+	if (p < b_end) {
+		memcpy(state->mem64, p, (size_t)(b_end-p));
+		state->memsize = (uint32_t)(b_end - p);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(xxh64_update);
+
+uint64_t xxh64_digest(const struct xxh64_state *state)
+{
+	const uint8_t *p = (const uint8_t *)state->mem64;
+	const uint8_t *const b_end = (const uint8_t *)state->mem64 +
+		state->memsize;
+	uint64_t h64;
+
+	if (state->total_len >= 32) {
+		const uint64_t v1 = state->v1;
+		const uint64_t v2 = state->v2;
+		const uint64_t v3 = state->v3;
+		const uint64_t v4 = state->v4;
+
+		h64 = xxh_rotl64(v1, 1) + xxh_rotl64(v2, 7) +
+			xxh_rotl64(v3, 12) + xxh_rotl64(v4, 18);
+		h64 = xxh64_merge_round(h64, v1);
+		h64 = xxh64_merge_round(h64, v2);
+		h64 = xxh64_merge_round(h64, v3);
+		h64 = xxh64_merge_round(h64, v4);
+	} else {
+		h64  = state->v3 + PRIME64_5;
+	}
+
+	h64 += (uint64_t)state->total_len;
+
+	while (p + 8 <= b_end) {
+		const uint64_t k1 = xxh64_round(0, get_unaligned_le64(p));
+
+		h64 ^= k1;
+		h64 = xxh_rotl64(h64, 27) * PRIME64_1 + PRIME64_4;
+		p += 8;
+	}
+
+	if (p + 4 <= b_end) {
+		h64 ^= (uint64_t)(get_unaligned_le32(p)) * PRIME64_1;
+		h64 = xxh_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
+		p += 4;
+	}
+
+	while (p < b_end) {
+		h64 ^= (*p) * PRIME64_5;
+		h64 = xxh_rotl64(h64, 11) * PRIME64_1;
+		p++;
+	}
+
+	h64 ^= h64 >> 33;
+	h64 *= PRIME64_2;
+	h64 ^= h64 >> 29;
+	h64 *= PRIME64_3;
+	h64 ^= h64 >> 32;
+
+	return h64;
+}
+EXPORT_SYMBOL(xxh64_digest);
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_DESCRIPTION("xxHash");