file_integrity: use xxh64

This commit is contained in:
Marian Beermann 2017-05-31 21:47:07 +02:00
parent 23d591c1c3
commit 0221e31058
7 changed files with 997 additions and 32 deletions

View File

@ -51,3 +51,9 @@ Folding CRC32
Borg includes an extremely fast folding implementation of CRC32, Copyright 2013 Intel Corporation,
licensed under the terms of the zlib license.
xxHash
------
XXH64, a fast non-cryptographic hash algorithm. Copyright 2012-2016 Yann Collet,
licensed under a BSD 2-clause license.

View File

@ -1,6 +1,8 @@
from ..helpers import bin_to_hex
from libc.stdint cimport uint32_t
from cpython.buffer cimport PyBUF_SIMPLE, PyObject_GetBuffer, PyBuffer_Release
from cpython.bytes cimport PyBytes_FromStringAndSize
cdef extern from "crc32_dispatch.c":
@ -10,6 +12,29 @@ cdef extern from "crc32_dispatch.c":
int _have_clmul "have_clmul"()
cdef extern from "xxh64/xxhash.c":
ctypedef struct XXH64_canonical_t:
char digest[8]
ctypedef struct XXH64_state_t:
pass # opaque
ctypedef unsigned long long XXH64_hash_t
ctypedef enum XXH_errorcode:
XXH_OK,
XXH_ERROR
XXH64_hash_t XXH64 (const void* input, size_t length, unsigned long long seed);
XXH_errorcode XXH64_reset (XXH64_state_t* statePtr, unsigned long long seed);
XXH_errorcode XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length);
XXH64_hash_t XXH64_digest (const XXH64_state_t* statePtr);
void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash);
XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src);
cdef Py_buffer ro_buffer(object data) except *:
cdef Py_buffer view
PyObject_GetBuffer(data, &view, PyBUF_SIMPLE)
@ -39,3 +64,43 @@ if have_clmul:
crc32 = crc32_clmul
else:
crc32 = crc32_slice_by_8
def xxh64(data, seed=0):
cdef unsigned long long _seed = seed
cdef XXH64_hash_t hash
cdef XXH64_canonical_t digest
cdef Py_buffer data_buf = ro_buffer(data)
try:
hash = XXH64(data_buf.buf, data_buf.len, _seed)
finally:
PyBuffer_Release(&data_buf)
XXH64_canonicalFromHash(&digest, hash)
return PyBytes_FromStringAndSize(<const char*> digest.digest, 8)
cdef class StreamingXXH64:
cdef XXH64_state_t state
def __cinit__(self, seed=0):
cdef unsigned long long _seed = seed
if XXH64_reset(&self.state, _seed) != XXH_OK:
raise Exception('XXH64_reset failed')
def update(self, data):
cdef Py_buffer data_buf = ro_buffer(data)
try:
if XXH64_update(&self.state, data_buf.buf, data_buf.len) != XXH_OK:
raise Exception('XXH64_update failed')
finally:
PyBuffer_Release(&data_buf)
def digest(self):
cdef XXH64_hash_t hash
cdef XXH64_canonical_t digest
hash = XXH64_digest(&self.state)
XXH64_canonicalFromHash(&digest, hash)
return PyBytes_FromStringAndSize(<const char*> digest.digest, 8)
def hexdigest(self):
return bin_to_hex(self.digest())

View File

@ -0,0 +1,615 @@
/*
* xxHash - 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.
*
* You can contact the author at :
* - xxHash homepage: http://www.xxhash.com
* - xxHash source repository : https://github.com/Cyan4973/xxHash
*/
/* *************************************
* Tuning parameters
***************************************/
/*!XXH_FORCE_MEMORY_ACCESS :
* By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable.
* Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal.
* The below switch allow to select different access method for improved performance.
* Method 0 (default) : use `memcpy()`. Safe and portable.
* Method 1 : `__packed` statement. It depends on compiler extension (ie, not portable).
* This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`.
* Method 2 : direct access. This method doesn't depend on compiler but violate C standard.
* It can generate buggy code on targets which do not support unaligned memory accesses.
* But in some circumstances, it's the only known way to get the most performance (ie GCC + ARMv6)
* See http://stackoverflow.com/a/32095106/646947 for details.
* Prefer these methods in priority order (0 > 1 > 2)
*/
#ifndef XXH_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
# define XXH_FORCE_MEMORY_ACCESS 2
# elif defined(__INTEL_COMPILER) || \
(defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7S__) ))
# define XXH_FORCE_MEMORY_ACCESS 1
# endif
#endif
/*!XXH_ACCEPT_NULL_INPUT_POINTER :
* If the input pointer is a null pointer, xxHash default behavior is to trigger a memory access error, since it is a bad pointer.
* When this option is enabled, xxHash output for null input pointers will be the same as a null-length input.
* By default, this option is disabled. To enable it, uncomment below define :
*/
/* #define XXH_ACCEPT_NULL_INPUT_POINTER 1 */
/*!XXH_FORCE_NATIVE_FORMAT :
* By default, xxHash library provides endian-independant Hash values, based on little-endian convention.
* Results are therefore identical for little-endian and big-endian CPU.
* This comes at a performance cost for big-endian CPU, since some swapping is required to emulate little-endian format.
* Should endian-independance be of no importance for your application, you may set the #define below to 1,
* to improve speed for Big-endian CPU.
* This option has no impact on Little_Endian CPU.
*/
#ifndef XXH_FORCE_NATIVE_FORMAT /* can be defined externally */
# define XXH_FORCE_NATIVE_FORMAT 0
#endif
/*!XXH_FORCE_ALIGN_CHECK :
* This is a minor performance trick, only useful with lots of very small keys.
* It means : check for aligned/unaligned input.
* The check costs one initial branch per hash; set to 0 when the input data
* is guaranteed to be aligned.
*/
#ifndef XXH_FORCE_ALIGN_CHECK /* can be defined externally */
# if defined(__i386) || defined(_M_IX86) || defined(__x86_64__) || defined(_M_X64)
# define XXH_FORCE_ALIGN_CHECK 0
# else
# define XXH_FORCE_ALIGN_CHECK 1
# endif
#endif
/* *************************************
* Includes & Memory related functions
***************************************/
/* Modify the local functions below should you wish to use some other memory routines */
/* for malloc(), free() */
#include <stdlib.h>
static void* XXH_malloc(size_t s) { return malloc(s); }
static void XXH_free (void* p) { free(p); }
/* for memcpy() */
#include <string.h>
static void* XXH_memcpy(void* dest, const void* src, size_t size) { return memcpy(dest,src,size); }
#define XXH_STATIC_LINKING_ONLY
#include "xxhash.h"
/* *************************************
* Compiler Specific Options
***************************************/
#ifdef _MSC_VER /* Visual Studio */
# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */
# define FORCE_INLINE static __forceinline
#else
# if defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */
# ifdef __GNUC__
# define FORCE_INLINE static inline __attribute__((always_inline))
# else
# define FORCE_INLINE static inline
# endif
# else
# define FORCE_INLINE static
# endif /* __STDC_VERSION__ */
#endif
/* *************************************
* Basic Types
***************************************/
#ifndef MEM_MODULE
# if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# include <stdint.h>
typedef uint8_t BYTE;
typedef uint16_t U16;
typedef uint32_t U32;
typedef int32_t S32;
# else
typedef unsigned char BYTE;
typedef unsigned short U16;
typedef unsigned int U32;
typedef signed int S32;
# endif
#endif
#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==2))
/* Force direct memory access. Only works on CPU which support unaligned memory access in hardware */
static U32 XXH_read32(const void* memPtr) { return *(const U32*) memPtr; }
#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==1))
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
/* currently only defined for gcc and icc */
typedef union { U32 u32; } __attribute__((packed)) unalign;
static U32 XXH_read32(const void* ptr) { return ((const unalign*)ptr)->u32; }
#else
/* portable and safe solution. Generally efficient.
* see : http://stackoverflow.com/a/32095106/646947
*/
static U32 XXH_read32(const void* memPtr)
{
U32 val;
memcpy(&val, memPtr, sizeof(val));
return val;
}
#endif /* XXH_FORCE_DIRECT_MEMORY_ACCESS */
/* ****************************************
* Compiler-specific Functions and Macros
******************************************/
#define GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__)
/* Note : although _rotl exists for minGW (GCC under windows), performance seems poor */
#if defined(_MSC_VER)
# define XXH_rotl32(x,r) _rotl(x,r)
# define XXH_rotl64(x,r) _rotl64(x,r)
#else
# define XXH_rotl32(x,r) ((x << r) | (x >> (32 - r)))
# define XXH_rotl64(x,r) ((x << r) | (x >> (64 - r)))
#endif
#if defined(_MSC_VER) /* Visual Studio */
# define XXH_swap32 _byteswap_ulong
#elif GCC_VERSION >= 403
# define XXH_swap32 __builtin_bswap32
#else
static U32 XXH_swap32 (U32 x)
{
return ((x << 24) & 0xff000000 ) |
((x << 8) & 0x00ff0000 ) |
((x >> 8) & 0x0000ff00 ) |
((x >> 24) & 0x000000ff );
}
#endif
/* *************************************
* Architecture Macros
***************************************/
typedef enum { XXH_bigEndian=0, XXH_littleEndian=1 } XXH_endianess;
/* XXH_CPU_LITTLE_ENDIAN can be defined externally, for example on the compiler command line */
#ifndef XXH_CPU_LITTLE_ENDIAN
static const int g_one = 1;
# define XXH_CPU_LITTLE_ENDIAN (*(const char*)(&g_one))
#endif
/* ***************************
* Memory reads
*****************************/
typedef enum { XXH_aligned, XXH_unaligned } XXH_alignment;
FORCE_INLINE U32 XXH_readLE32_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
{
if (align==XXH_unaligned)
return endian==XXH_littleEndian ? XXH_read32(ptr) : XXH_swap32(XXH_read32(ptr));
else
return endian==XXH_littleEndian ? *(const U32*)ptr : XXH_swap32(*(const U32*)ptr);
}
FORCE_INLINE U32 XXH_readLE32(const void* ptr, XXH_endianess endian)
{
return XXH_readLE32_align(ptr, endian, XXH_unaligned);
}
/* *************************************
* Macros
***************************************/
#define XXH_STATIC_ASSERT(c) { enum { XXH_static_assert = 1/(int)(!!(c)) }; } /* use only *after* variable declarations */
XXH_PUBLIC_API unsigned XXH_versionNumber (void) { return XXH_VERSION_NUMBER; }
#ifndef XXH_NO_LONG_LONG
/* *******************************************************************
* 64-bits hash functions
*********************************************************************/
#define XXH_get32bits(p) XXH_readLE32_align(p, endian, align)
/*====== Memory access ======*/
#ifndef MEM_MODULE
# define MEM_MODULE
# if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) )
# include <stdint.h>
typedef uint64_t U64;
# else
typedef unsigned long long U64; /* if your compiler doesn't support unsigned long long, replace by another 64-bit type here. Note that xxhash.h will also need to be updated. */
# endif
#endif
#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==2))
/* Force direct memory access. Only works on CPU which support unaligned memory access in hardware */
static U64 XXH_read64(const void* memPtr) { return *(const U64*) memPtr; }
#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==1))
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
/* currently only defined for gcc and icc */
typedef union { U32 u32; U64 u64; } __attribute__((packed)) unalign64;
static U64 XXH_read64(const void* ptr) { return ((const unalign64*)ptr)->u64; }
#else
/* portable and safe solution. Generally efficient.
* see : http://stackoverflow.com/a/32095106/646947
*/
static U64 XXH_read64(const void* memPtr)
{
U64 val;
memcpy(&val, memPtr, sizeof(val));
return val;
}
#endif /* XXH_FORCE_DIRECT_MEMORY_ACCESS */
#if defined(_MSC_VER) /* Visual Studio */
# define XXH_swap64 _byteswap_uint64
#elif GCC_VERSION >= 403
# define XXH_swap64 __builtin_bswap64
#else
static U64 XXH_swap64 (U64 x)
{
return ((x << 56) & 0xff00000000000000ULL) |
((x << 40) & 0x00ff000000000000ULL) |
((x << 24) & 0x0000ff0000000000ULL) |
((x << 8) & 0x000000ff00000000ULL) |
((x >> 8) & 0x00000000ff000000ULL) |
((x >> 24) & 0x0000000000ff0000ULL) |
((x >> 40) & 0x000000000000ff00ULL) |
((x >> 56) & 0x00000000000000ffULL);
}
#endif
FORCE_INLINE U64 XXH_readLE64_align(const void* ptr, XXH_endianess endian, XXH_alignment align)
{
if (align==XXH_unaligned)
return endian==XXH_littleEndian ? XXH_read64(ptr) : XXH_swap64(XXH_read64(ptr));
else
return endian==XXH_littleEndian ? *(const U64*)ptr : XXH_swap64(*(const U64*)ptr);
}
FORCE_INLINE U64 XXH_readLE64(const void* ptr, XXH_endianess endian)
{
return XXH_readLE64_align(ptr, endian, XXH_unaligned);
}
static U64 XXH_readBE64(const void* ptr)
{
return XXH_CPU_LITTLE_ENDIAN ? XXH_swap64(XXH_read64(ptr)) : XXH_read64(ptr);
}
/*====== xxh64 ======*/
static const U64 PRIME64_1 = 11400714785074694791ULL;
static const U64 PRIME64_2 = 14029467366897019727ULL;
static const U64 PRIME64_3 = 1609587929392839161ULL;
static const U64 PRIME64_4 = 9650029242287828579ULL;
static const U64 PRIME64_5 = 2870177450012600261ULL;
static U64 XXH64_round(U64 acc, U64 input)
{
acc += input * PRIME64_2;
acc = XXH_rotl64(acc, 31);
acc *= PRIME64_1;
return acc;
}
static U64 XXH64_mergeRound(U64 acc, U64 val)
{
val = XXH64_round(0, val);
acc ^= val;
acc = acc * PRIME64_1 + PRIME64_4;
return acc;
}
FORCE_INLINE U64 XXH64_endian_align(const void* input, size_t len, U64 seed, XXH_endianess endian, XXH_alignment align)
{
const BYTE* p = (const BYTE*)input;
const BYTE* const bEnd = p + len;
U64 h64;
#define XXH_get64bits(p) XXH_readLE64_align(p, endian, align)
#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
if (p==NULL) {
len=0;
bEnd=p=(const BYTE*)(size_t)32;
}
#endif
if (len>=32) {
const BYTE* const limit = bEnd - 32;
U64 v1 = seed + PRIME64_1 + PRIME64_2;
U64 v2 = seed + PRIME64_2;
U64 v3 = seed + 0;
U64 v4 = seed - PRIME64_1;
do {
v1 = XXH64_round(v1, XXH_get64bits(p)); p+=8;
v2 = XXH64_round(v2, XXH_get64bits(p)); p+=8;
v3 = XXH64_round(v3, XXH_get64bits(p)); p+=8;
v4 = XXH64_round(v4, XXH_get64bits(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_mergeRound(h64, v1);
h64 = XXH64_mergeRound(h64, v2);
h64 = XXH64_mergeRound(h64, v3);
h64 = XXH64_mergeRound(h64, v4);
} else {
h64 = seed + PRIME64_5;
}
h64 += (U64) len;
while (p+8<=bEnd) {
U64 const k1 = XXH64_round(0, XXH_get64bits(p));
h64 ^= k1;
h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
p+=8;
}
if (p+4<=bEnd) {
h64 ^= (U64)(XXH_get32bits(p)) * PRIME64_1;
h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
p+=4;
}
while (p<bEnd) {
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;
}
XXH_PUBLIC_API unsigned long long XXH64 (const void* input, size_t len, unsigned long long seed)
{
#if 0
/* Simple version, good for code maintenance, but unfortunately slow for small inputs */
XXH64_CREATESTATE_STATIC(state);
XXH64_reset(state, seed);
XXH64_update(state, input, len);
return XXH64_digest(state);
#else
XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
if (XXH_FORCE_ALIGN_CHECK) {
if ((((size_t)input) & 7)==0) { /* Input is aligned, let's leverage the speed advantage */
if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_aligned);
else
return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_aligned);
} }
if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
return XXH64_endian_align(input, len, seed, XXH_littleEndian, XXH_unaligned);
else
return XXH64_endian_align(input, len, seed, XXH_bigEndian, XXH_unaligned);
#endif
}
/*====== Hash Streaming ======*/
XXH_PUBLIC_API XXH64_state_t* XXH64_createState(void)
{
return (XXH64_state_t*)XXH_malloc(sizeof(XXH64_state_t));
}
XXH_PUBLIC_API XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr)
{
XXH_free(statePtr);
return XXH_OK;
}
XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* restrict dstState, const XXH64_state_t* restrict srcState)
{
memcpy(dstState, srcState, sizeof(*dstState));
}
XXH_PUBLIC_API XXH_errorcode XXH64_reset(XXH64_state_t* statePtr, unsigned long long seed)
{
XXH64_state_t state; /* using a local state to memcpy() in order to avoid strict-aliasing warnings */
memset(&state, 0, sizeof(state)-8); /* do not write into reserved, for future removal */
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));
return XXH_OK;
}
FORCE_INLINE XXH_errorcode XXH64_update_endian (XXH64_state_t* state, const void* input, size_t len, XXH_endianess endian)
{
const BYTE* p = (const BYTE*)input;
const BYTE* const bEnd = p + len;
#ifdef XXH_ACCEPT_NULL_INPUT_POINTER
if (input==NULL) return XXH_ERROR;
#endif
state->total_len += len;
if (state->memsize + len < 32) { /* fill in tmp buffer */
XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, len);
state->memsize += (U32)len;
return XXH_OK;
}
if (state->memsize) { /* tmp buffer is full */
XXH_memcpy(((BYTE*)state->mem64) + state->memsize, input, 32-state->memsize);
state->v1 = XXH64_round(state->v1, XXH_readLE64(state->mem64+0, endian));
state->v2 = XXH64_round(state->v2, XXH_readLE64(state->mem64+1, endian));
state->v3 = XXH64_round(state->v3, XXH_readLE64(state->mem64+2, endian));
state->v4 = XXH64_round(state->v4, XXH_readLE64(state->mem64+3, endian));
p += 32-state->memsize;
state->memsize = 0;
}
if (p+32 <= bEnd) {
const BYTE* const limit = bEnd - 32;
U64 v1 = state->v1;
U64 v2 = state->v2;
U64 v3 = state->v3;
U64 v4 = state->v4;
do {
v1 = XXH64_round(v1, XXH_readLE64(p, endian)); p+=8;
v2 = XXH64_round(v2, XXH_readLE64(p, endian)); p+=8;
v3 = XXH64_round(v3, XXH_readLE64(p, endian)); p+=8;
v4 = XXH64_round(v4, XXH_readLE64(p, endian)); p+=8;
} while (p<=limit);
state->v1 = v1;
state->v2 = v2;
state->v3 = v3;
state->v4 = v4;
}
if (p < bEnd) {
XXH_memcpy(state->mem64, p, (size_t)(bEnd-p));
state->memsize = (unsigned)(bEnd-p);
}
return XXH_OK;
}
XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* state_in, const void* input, size_t len)
{
XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
return XXH64_update_endian(state_in, input, len, XXH_littleEndian);
else
return XXH64_update_endian(state_in, input, len, XXH_bigEndian);
}
FORCE_INLINE U64 XXH64_digest_endian (const XXH64_state_t* state, XXH_endianess endian)
{
const BYTE * p = (const BYTE*)state->mem64;
const BYTE* const bEnd = (const BYTE*)state->mem64 + state->memsize;
U64 h64;
if (state->total_len >= 32) {
U64 const v1 = state->v1;
U64 const v2 = state->v2;
U64 const v3 = state->v3;
U64 const v4 = state->v4;
h64 = XXH_rotl64(v1, 1) + XXH_rotl64(v2, 7) + XXH_rotl64(v3, 12) + XXH_rotl64(v4, 18);
h64 = XXH64_mergeRound(h64, v1);
h64 = XXH64_mergeRound(h64, v2);
h64 = XXH64_mergeRound(h64, v3);
h64 = XXH64_mergeRound(h64, v4);
} else {
h64 = state->v3 + PRIME64_5;
}
h64 += (U64) state->total_len;
while (p+8<=bEnd) {
U64 const k1 = XXH64_round(0, XXH_readLE64(p, endian));
h64 ^= k1;
h64 = XXH_rotl64(h64,27) * PRIME64_1 + PRIME64_4;
p+=8;
}
if (p+4<=bEnd) {
h64 ^= (U64)(XXH_readLE32(p, endian)) * PRIME64_1;
h64 = XXH_rotl64(h64, 23) * PRIME64_2 + PRIME64_3;
p+=4;
}
while (p<bEnd) {
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;
}
XXH_PUBLIC_API unsigned long long XXH64_digest (const XXH64_state_t* state_in)
{
XXH_endianess endian_detected = (XXH_endianess)XXH_CPU_LITTLE_ENDIAN;
if ((endian_detected==XXH_littleEndian) || XXH_FORCE_NATIVE_FORMAT)
return XXH64_digest_endian(state_in, XXH_littleEndian);
else
return XXH64_digest_endian(state_in, XXH_bigEndian);
}
/*====== Canonical representation ======*/
XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash)
{
XXH_STATIC_ASSERT(sizeof(XXH64_canonical_t) == sizeof(XXH64_hash_t));
if (XXH_CPU_LITTLE_ENDIAN) hash = XXH_swap64(hash);
memcpy(dst, &hash, sizeof(*dst));
}
XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src)
{
return XXH_readBE64(src);
}
#endif /* XXH_NO_LONG_LONG */

View File

@ -0,0 +1,245 @@
/*
xxHash - Extremely Fast Hash algorithm
Header File
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.
You can contact the author at :
- xxHash source repository : https://github.com/Cyan4973/xxHash
*/
/* Notice extracted from xxHash homepage :
xxHash is an extremely fast Hash algorithm, running at RAM speed limits.
It also successfully passes all tests from the SMHasher suite.
Comparison (single thread, Windows Seven 32 bits, using SMHasher on a Core 2 Duo @3GHz)
Name Speed Q.Score Author
xxHash 5.4 GB/s 10
CrapWow 3.2 GB/s 2 Andrew
MumurHash 3a 2.7 GB/s 10 Austin Appleby
SpookyHash 2.0 GB/s 10 Bob Jenkins
SBox 1.4 GB/s 9 Bret Mulvey
Lookup3 1.2 GB/s 9 Bob Jenkins
SuperFastHash 1.2 GB/s 1 Paul Hsieh
CityHash64 1.05 GB/s 10 Pike & Alakuijala
FNV 0.55 GB/s 5 Fowler, Noll, Vo
CRC32 0.43 GB/s 9
MD5-32 0.33 GB/s 10 Ronald L. Rivest
SHA1-32 0.28 GB/s 10
Q.Score is a measure of quality of the hash function.
It depends on successfully passing SMHasher test set.
10 is a perfect score.
A 64-bits version, named XXH64, is available since r35.
It offers much better speed, but for 64-bits applications only.
Name Speed on 64 bits Speed on 32 bits
XXH64 13.8 GB/s 1.9 GB/s
XXH32 6.8 GB/s 6.0 GB/s
*/
#ifndef XXHASH_H_5627135585666179
#define XXHASH_H_5627135585666179 1
#define XXH_STATIC_LINKING_ONLY
#if defined (__cplusplus)
extern "C" {
#endif
/* ****************************
* Compiler specifics
******************************/
#if !(defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)) /* ! C99 */
# define restrict /* disable restrict */
#endif
/* ****************************
* Definitions
******************************/
#include <stddef.h> /* size_t */
typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode;
/* ****************************
* API modifier
******************************/
/** XXH_PRIVATE_API
* This is useful to include xxhash functions in `static` mode
* in order to inline them, and remove their symbol from the public list.
* Methodology :
* #define XXH_PRIVATE_API
* #include "xxhash.h"
* `xxhash.c` is automatically included.
* It's not useful to compile and link it as a separate module.
*/
#ifdef XXH_PRIVATE_API
# ifndef XXH_STATIC_LINKING_ONLY
# define XXH_STATIC_LINKING_ONLY
# endif
# if defined(__GNUC__)
# define XXH_PUBLIC_API static __inline __attribute__((unused))
# elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */)
# define XXH_PUBLIC_API static inline
# elif defined(_MSC_VER)
# define XXH_PUBLIC_API static __inline
# else
# define XXH_PUBLIC_API static /* this version may generate warnings for unused static functions; disable the relevant warning */
# endif
#else
# define XXH_PUBLIC_API /* do nothing */
#endif /* XXH_PRIVATE_API */
/*!XXH_NAMESPACE, aka Namespace Emulation :
If you want to include _and expose_ xxHash functions from within your own library,
but also want to avoid symbol collisions with other libraries which may also include xxHash,
you can use XXH_NAMESPACE, to automatically prefix any public symbol from xxhash library
with the value of XXH_NAMESPACE (therefore, avoid NULL and numeric values).
Note that no change is required within the calling program as long as it includes `xxhash.h` :
regular symbol name will be automatically translated by this header.
*/
#ifdef XXH_NAMESPACE
# define XXH_CAT(A,B) A##B
# define XXH_NAME2(A,B) XXH_CAT(A,B)
# define XXH_versionNumber XXH_NAME2(XXH_NAMESPACE, XXH_versionNumber)
# define XXH32 XXH_NAME2(XXH_NAMESPACE, XXH32)
# define XXH32_createState XXH_NAME2(XXH_NAMESPACE, XXH32_createState)
# define XXH32_freeState XXH_NAME2(XXH_NAMESPACE, XXH32_freeState)
# define XXH32_reset XXH_NAME2(XXH_NAMESPACE, XXH32_reset)
# define XXH32_update XXH_NAME2(XXH_NAMESPACE, XXH32_update)
# define XXH32_digest XXH_NAME2(XXH_NAMESPACE, XXH32_digest)
# define XXH32_copyState XXH_NAME2(XXH_NAMESPACE, XXH32_copyState)
# define XXH32_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH32_canonicalFromHash)
# define XXH32_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH32_hashFromCanonical)
# define XXH64 XXH_NAME2(XXH_NAMESPACE, XXH64)
# define XXH64_createState XXH_NAME2(XXH_NAMESPACE, XXH64_createState)
# define XXH64_freeState XXH_NAME2(XXH_NAMESPACE, XXH64_freeState)
# define XXH64_reset XXH_NAME2(XXH_NAMESPACE, XXH64_reset)
# define XXH64_update XXH_NAME2(XXH_NAMESPACE, XXH64_update)
# define XXH64_digest XXH_NAME2(XXH_NAMESPACE, XXH64_digest)
# define XXH64_copyState XXH_NAME2(XXH_NAMESPACE, XXH64_copyState)
# define XXH64_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH64_canonicalFromHash)
# define XXH64_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH64_hashFromCanonical)
#endif
/* *************************************
* Version
***************************************/
#define XXH_VERSION_MAJOR 0
#define XXH_VERSION_MINOR 6
#define XXH_VERSION_RELEASE 2
#define XXH_VERSION_NUMBER (XXH_VERSION_MAJOR *100*100 + XXH_VERSION_MINOR *100 + XXH_VERSION_RELEASE)
XXH_PUBLIC_API unsigned XXH_versionNumber (void);
#ifndef XXH_NO_LONG_LONG
/*-**********************************************************************
* 64-bits hash
************************************************************************/
typedef unsigned long long XXH64_hash_t;
/*! XXH64() :
Calculate the 64-bits hash of sequence of length "len" stored at memory address "input".
"seed" can be used to alter the result predictably.
This function runs faster on 64-bits systems, but slower on 32-bits systems (see benchmark).
*/
XXH_PUBLIC_API XXH64_hash_t XXH64 (const void* input, size_t length, unsigned long long seed);
/*====== Streaming ======*/
typedef struct XXH64_state_s XXH64_state_t; /* incomplete type */
XXH_PUBLIC_API XXH64_state_t* XXH64_createState(void);
XXH_PUBLIC_API XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr);
XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* restrict dst_state, const XXH64_state_t* restrict src_state);
XXH_PUBLIC_API XXH_errorcode XXH64_reset (XXH64_state_t* statePtr, unsigned long long seed);
XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length);
XXH_PUBLIC_API XXH64_hash_t XXH64_digest (const XXH64_state_t* statePtr);
/*====== Canonical representation ======*/
typedef struct { unsigned char digest[8]; } XXH64_canonical_t;
XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash);
XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src);
#endif /* XXH_NO_LONG_LONG */
#ifdef XXH_STATIC_LINKING_ONLY
/* ================================================================================================
This section contains definitions which are not guaranteed to remain stable.
They may change in future versions, becoming incompatible with a different version of the library.
They shall only be used with static linking.
Never use these definitions in association with dynamic linking !
=================================================================================================== */
/* These definitions are only meant to allow allocation of XXH state
statically, on stack, or in a struct for example.
Do not use members directly. */
struct XXH32_state_s {
unsigned total_len_32;
unsigned large_len;
unsigned v1;
unsigned v2;
unsigned v3;
unsigned v4;
unsigned mem32[4]; /* buffer defined as U32 for alignment */
unsigned memsize;
unsigned reserved; /* never read nor write, will be removed in a future version */
}; /* typedef'd to XXH32_state_t */
#ifndef XXH_NO_LONG_LONG
struct XXH64_state_s {
unsigned long long total_len;
unsigned long long v1;
unsigned long long v2;
unsigned long long v3;
unsigned long long v4;
unsigned long long mem64[4]; /* buffer defined as U64 for alignment */
unsigned memsize;
unsigned reserved[2]; /* never read nor write, will be removed in a future version */
}; /* typedef'd to XXH64_state_t */
#endif
# ifdef XXH_PRIVATE_API
# include "xxhash.c" /* include xxhash function bodies as `static`, for inlining */
# endif
#endif /* XXH_STATIC_LINKING_ONLY */
#if defined (__cplusplus)
}
#endif
#endif /* XXHASH_H_5627135585666179 */

View File

@ -6,6 +6,7 @@
from ..helpers import IntegrityError
from ..logger import create_logger
from ..algorithms.crc32 import StreamingXXH64
logger = create_logger()
@ -37,7 +38,7 @@ def fileno(self):
return self.fd.fileno()
class SHA512FileHashingWrapper(FileLikeWrapper):
class FileHashingWrapper(FileLikeWrapper):
"""
Wrapper for file-like objects that computes a hash on-the-fly while reading/writing.
@ -53,12 +54,13 @@ class SHA512FileHashingWrapper(FileLikeWrapper):
are illegal.
"""
ALGORITHM = 'SHA512'
ALGORITHM = None
FACTORY = None
def __init__(self, backing_fd, write):
self.fd = backing_fd
self.writing = write
self.hash = hashlib.new(self.ALGORITHM)
self.hash = self.FACTORY()
def __exit__(self, exc_type, exc_val, exc_tb):
if exc_type is None:
@ -99,6 +101,22 @@ def hash_length(self, seek_to_end=False):
self.hash.update(str(self.tell()).encode())
class SHA512FileHashingWrapper(FileHashingWrapper):
ALGORITHM = 'SHA512'
FACTORY = hashlib.sha512
class XXH64FileHashingWrapper(FileHashingWrapper):
ALGORITHM = 'XXH64'
FACTORY = StreamingXXH64
SUPPORTED_ALGORITHMS = {
SHA512FileHashingWrapper.ALGORITHM: SHA512FileHashingWrapper,
XXH64FileHashingWrapper.ALGORITHM: XXH64FileHashingWrapper,
}
class FileIntegrityError(IntegrityError):
"""File failed integrity check: {}"""
@ -109,18 +127,26 @@ def __init__(self, path, write, filename=None, override_fd=None, integrity_data=
self.writing = write
mode = 'wb' if write else 'rb'
self.file_fd = override_fd or open(path, mode)
self.digests = {}
self.fd = self.hasher = SHA512FileHashingWrapper(backing_fd=self.file_fd, write=write)
hash_cls = XXH64FileHashingWrapper
self.hash_filename(filename)
if not write:
algorithm_and_digests = self.load_integrity_data(path, integrity_data)
if algorithm_and_digests:
algorithm, self.digests = algorithm_and_digests
hash_cls = SUPPORTED_ALGORITHMS[algorithm]
if write or not integrity_data:
self.digests = {}
else:
self.digests = self.parse_integrity_data(path, integrity_data, self.hasher)
# TODO: When we're reading but don't have any digests, i.e. no integrity file existed,
# TODO: then we could just short-circuit.
self.fd = self.hasher = hash_cls(backing_fd=self.file_fd, write=write)
self.hash_filename(filename)
def load_integrity_data(self, path, integrity_data):
if integrity_data is not None:
return self.parse_integrity_data(path, integrity_data)
def hash_filename(self, filename=None):
# Hash the name of the file, but only the basename, ie. not the path.
# In Borg the name itself encodes the context (eg. index.N, cache, files),
@ -133,18 +159,18 @@ def hash_filename(self, filename=None):
self.hasher.update(filename.encode())
@classmethod
def parse_integrity_data(cls, path: str, data: str, hasher: SHA512FileHashingWrapper):
def parse_integrity_data(cls, path: str, data: str):
try:
integrity_data = json.loads(data)
# Provisions for agility now, implementation later, but make sure the on-disk joint is oiled.
algorithm = integrity_data['algorithm']
if algorithm != hasher.ALGORITHM:
if algorithm not in SUPPORTED_ALGORITHMS:
logger.warning('Cannot verify integrity of %s: Unknown algorithm %r', path, algorithm)
return
digests = integrity_data['digests']
# Require at least presence of the final digest
digests['final']
return digests
return algorithm, digests
except (ValueError, TypeError, KeyError) as e:
logger.warning('Could not parse integrity data for %s: %s', path, e)
raise FileIntegrityError(path)
@ -186,18 +212,20 @@ def __init__(self, path, write, filename=None, override_fd=None):
filename = filename or os.path.basename(path)
output_dir = os.path.dirname(path)
self.output_integrity_file = self.integrity_file_path(os.path.join(output_dir, filename))
if not write:
self.digests = self.read_integrity_file(self.path, self.hasher)
def load_integrity_data(self, path, integrity_data):
assert not integrity_data, 'Cannot pass explicit integrity_data to DetachedIntegrityCheckedFile'
return self.read_integrity_file(self.path)
@staticmethod
def integrity_file_path(path):
return path + '.integrity'
@classmethod
def read_integrity_file(cls, path, hasher):
def read_integrity_file(cls, path):
try:
with open(cls.integrity_file_path(path), 'r') as fd:
return cls.parse_integrity_data(path, fd.read(), hasher)
return cls.parse_integrity_data(path, fd.read())
except FileNotFoundError:
logger.info('No integrity file found for %s', path)
except OSError as e:

View File

@ -1,9 +1,11 @@
import os
import zlib
from binascii import unhexlify
import pytest
from ..algorithms import crc32
from ..helpers import bin_to_hex
crc32_implementations = [crc32.crc32_slice_by_8]
if crc32.have_clmul:
@ -19,3 +21,19 @@ def test_crc32(implementation):
for i in range(0, 256):
d = data[:i]
assert zlib.crc32(d, initial_crc) == implementation(d, initial_crc)
def test_xxh64():
assert bin_to_hex(crc32.xxh64(b'test', 123)) == '2b81b9401bef86cf'
assert bin_to_hex(crc32.xxh64(b'test')) == '4fdcca5ddb678139'
assert bin_to_hex(crc32.xxh64(unhexlify('6f663f01c118abdea553373d5eae44e7dac3b6829b46b9bbeff202b6c592c22d724'
'fb3d25a347cca6c5b8f20d567e4bb04b9cfa85d17f691590f9a9d32e8ccc9102e9d'
'cf8a7e6716280cd642ce48d03fdf114c9f57c20d9472bb0f81c147645e6fa3d331'))) == \
'35d5d2f545d9511a'
def test_streaming_xxh64():
hasher = crc32.StreamingXXH64(123)
hasher.update(b'te')
hasher.update(b'st')
assert bin_to_hex(hasher.digest()) == hasher.hexdigest() == '2b81b9401bef86cf'

View File

@ -8,23 +8,20 @@ class TestReadIntegrityFile:
def test_no_integrity(self, tmpdir):
protected_file = tmpdir.join('file')
protected_file.write('1234')
assert DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file), None) is None
assert DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file)) is None
def test_truncated_integrity(self, tmpdir):
protected_file = tmpdir.join('file')
protected_file.write('1234')
tmpdir.join('file.integrity').write('')
with pytest.raises(FileIntegrityError):
DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file), None)
DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file))
def test_unknown_algorithm(self, tmpdir):
class SomeHasher:
ALGORITHM = 'HMAC_FOOHASH9000'
protected_file = tmpdir.join('file')
protected_file.write('1234')
tmpdir.join('file.integrity').write('{"algorithm": "HMAC_SERIOUSHASH", "digests": "1234"}')
assert DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file), SomeHasher()) is None
assert DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file)) is None
@pytest.mark.parametrize('json', (
'{"ALGORITHM": "HMAC_SERIOUSHASH", "digests": "1234"}',
@ -38,16 +35,7 @@ def test_malformed(self, tmpdir, json):
protected_file.write('1234')
tmpdir.join('file.integrity').write(json)
with pytest.raises(FileIntegrityError):
DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file), None)
def test_valid(self, tmpdir):
class SomeHasher:
ALGORITHM = 'HMAC_FOO1'
protected_file = tmpdir.join('file')
protected_file.write('1234')
tmpdir.join('file.integrity').write('{"algorithm": "HMAC_FOO1", "digests": {"final": "1234"}}')
assert DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file), SomeHasher()) == {'final': '1234'}
DetachedIntegrityCheckedFile.read_integrity_file(str(protected_file))
class TestDetachedIntegrityCheckedFile: