2015-04-14 16:03:42 +02:00
|
|
|
/*-------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* pg_crc32c.h
|
|
|
|
* Routines for computing CRC-32C checksums.
|
|
|
|
*
|
Use Intel SSE 4.2 CRC instructions where available.
Modern x86 and x86-64 processors with SSE 4.2 support have special
instructions, crc32b and crc32q, for calculating CRC-32C. They greatly
speed up CRC calculation.
Whether the instructions can be used or not depends on the compiler and the
target architecture. If generation of SSE 4.2 instructions is allowed for
the target (-msse4.2 flag on gcc and clang), use them. If they are not
allowed by default, but the compiler supports the -msse4.2 flag to enable
them, compile just the CRC-32C function with -msse4.2 flag, and check at
runtime whether the processor we're running on supports it. If it doesn't,
fall back to the slicing-by-8 algorithm. (With the common defaults on
current operating systems, the runtime-check variant is what you get in
practice.)
Abhijit Menon-Sen, heavily modified by me, reviewed by Andres Freund.
2015-04-14 16:05:03 +02:00
|
|
|
* The speed of CRC-32C calculation has a big impact on performance, so we
|
|
|
|
* jump through some hoops to get the best implementation for each
|
|
|
|
* platform. Some CPU architectures have special instructions for speeding
|
|
|
|
* up CRC calculations (e.g. Intel SSE 4.2), on other platforms we use the
|
|
|
|
* Slicing-by-8 algorithm which uses lookup tables.
|
|
|
|
*
|
|
|
|
* The public interface consists of four macros:
|
|
|
|
*
|
|
|
|
* INIT_CRC32C(crc)
|
|
|
|
* Initialize a CRC accumulator
|
|
|
|
*
|
|
|
|
* COMP_CRC32C(crc, data, len)
|
|
|
|
* Accumulate some (more) bytes into a CRC
|
|
|
|
*
|
|
|
|
* FIN_CRC32C(crc)
|
|
|
|
* Finish a CRC calculation
|
|
|
|
*
|
|
|
|
* EQ_CRC32C(c1, c2)
|
|
|
|
* Check for equality of two CRCs.
|
2015-04-14 16:03:42 +02:00
|
|
|
*
|
|
|
|
* Portions Copyright (c) 1996-2015, PostgreSQL Global Development Group
|
|
|
|
* Portions Copyright (c) 1994, Regents of the University of California
|
|
|
|
*
|
|
|
|
* src/include/port/pg_crc32c.h
|
|
|
|
*
|
|
|
|
*-------------------------------------------------------------------------
|
|
|
|
*/
|
|
|
|
#ifndef PG_CRC32C_H
|
|
|
|
#define PG_CRC32C_H
|
|
|
|
|
|
|
|
typedef uint32 pg_crc32c;
|
|
|
|
|
Use Intel SSE 4.2 CRC instructions where available.
Modern x86 and x86-64 processors with SSE 4.2 support have special
instructions, crc32b and crc32q, for calculating CRC-32C. They greatly
speed up CRC calculation.
Whether the instructions can be used or not depends on the compiler and the
target architecture. If generation of SSE 4.2 instructions is allowed for
the target (-msse4.2 flag on gcc and clang), use them. If they are not
allowed by default, but the compiler supports the -msse4.2 flag to enable
them, compile just the CRC-32C function with -msse4.2 flag, and check at
runtime whether the processor we're running on supports it. If it doesn't,
fall back to the slicing-by-8 algorithm. (With the common defaults on
current operating systems, the runtime-check variant is what you get in
practice.)
Abhijit Menon-Sen, heavily modified by me, reviewed by Andres Freund.
2015-04-14 16:05:03 +02:00
|
|
|
/* The INIT and EQ macros are the same for all implementations. */
|
2015-04-14 16:03:42 +02:00
|
|
|
#define INIT_CRC32C(crc) ((crc) = 0xFFFFFFFF)
|
|
|
|
#define EQ_CRC32C(c1, c2) ((c1) == (c2))
|
|
|
|
|
Use Intel SSE 4.2 CRC instructions where available.
Modern x86 and x86-64 processors with SSE 4.2 support have special
instructions, crc32b and crc32q, for calculating CRC-32C. They greatly
speed up CRC calculation.
Whether the instructions can be used or not depends on the compiler and the
target architecture. If generation of SSE 4.2 instructions is allowed for
the target (-msse4.2 flag on gcc and clang), use them. If they are not
allowed by default, but the compiler supports the -msse4.2 flag to enable
them, compile just the CRC-32C function with -msse4.2 flag, and check at
runtime whether the processor we're running on supports it. If it doesn't,
fall back to the slicing-by-8 algorithm. (With the common defaults on
current operating systems, the runtime-check variant is what you get in
practice.)
Abhijit Menon-Sen, heavily modified by me, reviewed by Andres Freund.
2015-04-14 16:05:03 +02:00
|
|
|
#if defined(USE_SSE42_CRC32C)
|
|
|
|
/* Use SSE4.2 instructions. */
|
|
|
|
#define COMP_CRC32C(crc, data, len) \
|
|
|
|
((crc) = pg_comp_crc32c_sse42((crc), (data), (len)))
|
|
|
|
#define FIN_CRC32C(crc) ((crc) ^= 0xFFFFFFFF)
|
|
|
|
|
|
|
|
extern pg_crc32c pg_comp_crc32c_sse42(pg_crc32c crc, const void *data, size_t len);
|
|
|
|
|
|
|
|
#elif defined(USE_SSE42_CRC32C_WITH_RUNTIME_CHECK)
|
|
|
|
/*
|
|
|
|
* Use SSE4.2 instructions, but perform a runtime check first to check that
|
|
|
|
* they are available.
|
|
|
|
*/
|
|
|
|
#define COMP_CRC32C(crc, data, len) \
|
|
|
|
((crc) = pg_comp_crc32c((crc), (data), (len)))
|
|
|
|
#define FIN_CRC32C(crc) ((crc) ^= 0xFFFFFFFF)
|
|
|
|
|
|
|
|
extern pg_crc32c pg_comp_crc32c_sse42(pg_crc32c crc, const void *data, size_t len);
|
|
|
|
extern pg_crc32c pg_comp_crc32c_sb8(pg_crc32c crc, const void *data, size_t len);
|
|
|
|
extern pg_crc32c (*pg_comp_crc32c) (pg_crc32c crc, const void *data, size_t len);
|
|
|
|
|
|
|
|
#else
|
2015-04-14 16:03:42 +02:00
|
|
|
/*
|
|
|
|
* Use slicing-by-8 algorithm.
|
|
|
|
*
|
|
|
|
* On big-endian systems, the intermediate value is kept in reverse byte
|
|
|
|
* order, to avoid byte-swapping during the calculation. FIN_CRC32C reverses
|
|
|
|
* the bytes to the final order.
|
|
|
|
*/
|
|
|
|
#define COMP_CRC32C(crc, data, len) \
|
|
|
|
((crc) = pg_comp_crc32c_sb8((crc), (data), (len)))
|
|
|
|
#ifdef WORDS_BIGENDIAN
|
|
|
|
|
|
|
|
#ifdef HAVE__BUILTIN_BSWAP32
|
|
|
|
#define BSWAP32(x) __builtin_bswap32(x)
|
|
|
|
#else
|
|
|
|
#define BSWAP32(x) (((x << 24) & 0xff000000) | \
|
|
|
|
((x << 8) & 0x00ff0000) | \
|
|
|
|
((x >> 8) & 0x0000ff00) | \
|
|
|
|
((x >> 24) & 0x000000ff))
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#define FIN_CRC32C(crc) ((crc) = BSWAP32(crc) ^ 0xFFFFFFFF)
|
|
|
|
#else
|
|
|
|
#define FIN_CRC32C(crc) ((crc) ^= 0xFFFFFFFF)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
extern pg_crc32c pg_comp_crc32c_sb8(pg_crc32c crc, const void *data, size_t len);
|
|
|
|
|
Use Intel SSE 4.2 CRC instructions where available.
Modern x86 and x86-64 processors with SSE 4.2 support have special
instructions, crc32b and crc32q, for calculating CRC-32C. They greatly
speed up CRC calculation.
Whether the instructions can be used or not depends on the compiler and the
target architecture. If generation of SSE 4.2 instructions is allowed for
the target (-msse4.2 flag on gcc and clang), use them. If they are not
allowed by default, but the compiler supports the -msse4.2 flag to enable
them, compile just the CRC-32C function with -msse4.2 flag, and check at
runtime whether the processor we're running on supports it. If it doesn't,
fall back to the slicing-by-8 algorithm. (With the common defaults on
current operating systems, the runtime-check variant is what you get in
practice.)
Abhijit Menon-Sen, heavily modified by me, reviewed by Andres Freund.
2015-04-14 16:05:03 +02:00
|
|
|
#endif
|
|
|
|
|
2015-05-24 03:35:49 +02:00
|
|
|
#endif /* PG_CRC32C_H */
|