database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-07-27 12:41:57 +03:00
Code Activity

Use ARMv8 CRC instructions where available.

Browse Source 
ARMv8 introduced special CPU instructions for calculating CRC-32C. Use
them, when available, for speed.

Like with the similar Intel CRC instructions, several factors affect
whether the instructions can be used. The compiler intrinsics for them must
be supported by the compiler, and the instructions must be supported by the
target architecture. If the compilation target architecture does not
support the instructions, but adding "-march=armv8-a+crc" makes them
available, then we compile the code with a runtime check to determine if
the host we're running on supports them or not.

For the runtime check, use glibc getauxval() function. Unfortunately,
that's not very portable, but I couldn't find any more portable way to do
it. If getauxval() is not available, the CRC instructions will still be
used if the target architecture supports them without any additional
compiler flags, but the runtime check will not be available.

Original patch by Yuqi Gu, heavily modified by me. Reviewed by Andres
Freund, Thomas Munro.

Discussion: https://www.postgresql.org/message-id/HE1PR0801MB1323D171938EABC04FFE7FA9E3110%40HE1PR0801MB1323.eurprd08.prod.outlook.com
This commit is contained in:
Heikki Linnakangas
2018-04-04 12:22:45 +03:00
parent 638a199fa9
commit f044d71e33
11 changed files with 456 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files

194
configure vendored
View File

@ -646,6 +646,7 @@ MSGMERGE
MSGFMT_FLAGS
MSGFMT
PG_CRC32C_OBJS
CFLAGS_ARMV8_CRC32C
CFLAGS_SSE42
have_win32_dbghelp
HAVE_IPV6
@ -17254,28 +17255,175 @@ if ac_fn_c_try_compile "$LINENO"; then :
fi
rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
# Check for ARMv8 CRC Extension intrinsics to do CRC calculations.
#
# First check if __crc32c* intrinsics can be used with the default compiler
# flags. If not, check if adding -march=armv8-a+crc flag helps.
# CFLAGS_ARMV8_CRC32C is set if the extra flag is required.
{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for __crc32cb, __crc32ch, __crc32cw, and __crc32cd with CFLAGS=" >&5
$as_echo_n "checking for __crc32cb, __crc32ch, __crc32cw, and __crc32cd with CFLAGS=... " >&6; }
if ${pgac_cv_armv8_crc32c_intrinsics_+:} false; then :
$as_echo_n "(cached) " >&6
else
pgac_save_CFLAGS=$CFLAGS
CFLAGS="$pgac_save_CFLAGS "
cat confdefs.h - <<_ACEOF >conftest.$ac_ext
/* end confdefs.h. */
#include <arm_acle.h>
int
main ()
{
unsigned int crc = 0;
crc = __crc32cb(crc, 0);
crc = __crc32ch(crc, 0);
crc = __crc32cw(crc, 0);
crc = __crc32cd(crc, 0);
/* return computed value, to prevent the above being optimized away */
return crc == 0;
;
return 0;
}
_ACEOF
if ac_fn_c_try_link "$LINENO"; then :
pgac_cv_armv8_crc32c_intrinsics_=yes
else
pgac_cv_armv8_crc32c_intrinsics_=no
fi
rm -f core conftest.err conftest.$ac_objext \
conftest$ac_exeext conftest.$ac_ext
CFLAGS="$pgac_save_CFLAGS"
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $pgac_cv_armv8_crc32c_intrinsics_" >&5
$as_echo "$pgac_cv_armv8_crc32c_intrinsics_" >&6; }
if test x"$pgac_cv_armv8_crc32c_intrinsics_" = x"yes"; then
CFLAGS_ARMV8_CRC32C=""
pgac_armv8_crc32c_intrinsics=yes
fi
if test x"$pgac_armv8_crc32c_intrinsics" != x"yes"; then
{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for __crc32cb, __crc32ch, __crc32cw, and __crc32cd with CFLAGS=-march=armv8-a+crc" >&5
$as_echo_n "checking for __crc32cb, __crc32ch, __crc32cw, and __crc32cd with CFLAGS=-march=armv8-a+crc... " >&6; }
if ${pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc+:} false; then :
$as_echo_n "(cached) " >&6
else
pgac_save_CFLAGS=$CFLAGS
CFLAGS="$pgac_save_CFLAGS -march=armv8-a+crc"
cat confdefs.h - <<_ACEOF >conftest.$ac_ext
/* end confdefs.h. */
#include <arm_acle.h>
int
main ()
{
unsigned int crc = 0;
crc = __crc32cb(crc, 0);
crc = __crc32ch(crc, 0);
crc = __crc32cw(crc, 0);
crc = __crc32cd(crc, 0);
/* return computed value, to prevent the above being optimized away */
return crc == 0;
;
return 0;
}
_ACEOF
if ac_fn_c_try_link "$LINENO"; then :
pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc=yes
else
pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc=no
fi
rm -f core conftest.err conftest.$ac_objext \
conftest$ac_exeext conftest.$ac_ext
CFLAGS="$pgac_save_CFLAGS"
fi
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc" >&5
$as_echo "$pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc" >&6; }
if test x"$pgac_cv_armv8_crc32c_intrinsics__march_armv8_apcrc" = x"yes"; then
CFLAGS_ARMV8_CRC32C="-march=armv8-a+crc"
pgac_armv8_crc32c_intrinsics=yes
fi
fi
# In order to detect at runtime, if the ARM CRC Extension is available,
# we will do "getauxval(AT_HWCAP) & HWCAP_CRC32". Check if we have
# everything we need for that.
for ac_func in getauxval
do :
ac_fn_c_check_func "$LINENO" "getauxval" "ac_cv_func_getauxval"
if test "x$ac_cv_func_getauxval" = xyes; then :
cat >>confdefs.h <<_ACEOF
#define HAVE_GETAUXVAL 1
_ACEOF
fi
done
cat confdefs.h - <<_ACEOF >conftest.$ac_ext
/* end confdefs.h. */
#include <sys/auxv.h>
#include <asm/hwcap.h>
int
main ()
{
#ifndef AT_HWCAP
#error AT_HWCAP not defined
#endif
#ifndef HWCAP_CRC32
#error HWCAP_CRC32 not defined
#endif
;
return 0;
}
_ACEOF
if ac_fn_c_try_compile "$LINENO"; then :
HAVE_HWCAP_CRC32=1
fi
rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
# Select CRC-32C implementation.
#
# If we are targeting a processor that has SSE 4.2 instructions, we can use the
# special CRC instructions for calculating CRC-32C. If we're not targeting such
# a processor, but we can nevertheless produce code that uses the SSE
# intrinsics, perhaps with some extra CFLAGS, compile both implementations and
# select which one to use at runtime, depending on whether SSE 4.2 is supported
# by the processor we're running on.
# If we are targeting a processor that has Intel SSE 4.2 instructions, we can
# use the special CRC instructions for calculating CRC-32C. If we're not
# targeting such a processor, but we can nevertheless produce code that uses
# the SSE intrinsics, perhaps with some extra CFLAGS, compile both
# implementations and select which one to use at runtime, depending on whether
# SSE 4.2 is supported by the processor we're running on.
#
# Similarly, if we are targeting an ARM processor that has the CRC
# instructions that are part of the ARMv8 CRC Extension, use them. And if
# we're not targeting such a processor, but can nevertheless produce code that
# uses the CRC instructions, compile both, and select at runtime.
#
# You can override this logic by setting the appropriate USE_*_CRC32 flag to 1
# in the template or configure command line.
if test x"$USE_SSE42_CRC32C" = x"" && test x"$USE_SSE42_CRC32C_WITH_RUNTIME_CHECK" = x"" && test x"$USE_SLICING_BY_8_CRC32C" = x""; then
if test x"$USE_SLICING_BY_8_CRC32C" = x"" && test x"$USE_SSE42_CRC32C" = x"" && test x"$USE_SSE42_CRC32C_WITH_RUNTIME_CHECK" = x"" && test x"$USE_ARMV8_CRC32C" = x"" && test x"$USE_ARMV8_CRC32C_WITH_RUNTIME_CHECK" = x""; then
# Use Intel SSE 4.2 if available.
if test x"$pgac_sse42_crc32_intrinsics" = x"yes" && test x"$SSE4_2_TARGETED" = x"1" ; then
USE_SSE42_CRC32C=1
else
# the CPUID instruction is needed for the runtime check.
# Intel SSE 4.2, with runtime check? The CPUID instruction is needed for
# the runtime check.
if test x"$pgac_sse42_crc32_intrinsics" = x"yes" && (test x"$pgac_cv__get_cpuid" = x"yes" || test x"$pgac_cv__cpuid" = x"yes"); then
USE_SSE42_CRC32C_WITH_RUNTIME_CHECK=1
else
# fall back to slicing-by-8 algorithm which doesn't require any special
# CPU support.
USE_SLICING_BY_8_CRC32C=1
# Use ARM CRC Extension if available.
if test x"$pgac_armv8_crc32c_intrinsics" = x"yes" && test x"$CFLAGS_ARMV8_CRC32C" = x""; then
USE_ARMV8_CRC32C=1
else
# ARM CRC Extension, with runtime check? The getauxval() function and
# HWCAP_CRC32 are needed for the runtime check.
if test x"$pgac_armv8_crc32c_intrinsics" = x"yes" && test x"$ac_cv_func_getauxval" = x"yes" && test x"$HAVE_HWCAP_CRC32" = x"1"; then
USE_ARMV8_CRC32C_WITH_RUNTIME_CHECK=1
else
# fall back to slicing-by-8 algorithm, which doesn't require any
# special CPU support.
USE_SLICING_BY_8_CRC32C=1
fi
fi
fi
fi
fi
@ -17295,16 +17443,34 @@ else
$as_echo "#define USE_SSE42_CRC32C_WITH_RUNTIME_CHECK 1" >>confdefs.h
PG_CRC32C_OBJS="pg_crc32c_sse42.o pg_crc32c_sb8.o pg_crc32c_choose.o"
PG_CRC32C_OBJS="pg_crc32c_sse42.o pg_crc32c_sb8.o pg_crc32c_sse42_choose.o"
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: SSE 4.2 with runtime check" >&5
$as_echo "SSE 4.2 with runtime check" >&6; }
else
if test x"$USE_ARMV8_CRC32C" = x"1"; then
$as_echo "#define USE_ARMV8_CRC32C 1" >>confdefs.h
PG_CRC32C_OBJS="pg_crc32c_armv8.o"
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: ARMv8 CRC instructions" >&5
$as_echo "ARMv8 CRC instructions" >&6; }
else
if test x"$USE_ARMV8_CRC32C_WITH_RUNTIME_CHECK" = x"1"; then
$as_echo "#define USE_ARMV8_CRC32C_WITH_RUNTIME_CHECK 1" >>confdefs.h
PG_CRC32C_OBJS="pg_crc32c_armv8.o pg_crc32c_sb8.o pg_crc32c_armv8_choose.o"
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: ARMv8 CRC instructions with runtime check" >&5
$as_echo "ARMv8 CRC instructions with runtime check" >&6; }
else
$as_echo "#define USE_SLICING_BY_8_CRC32C 1" >>confdefs.h
PG_CRC32C_OBJS="pg_crc32c_sb8.o"
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: slicing-by-8" >&5
PG_CRC32C_OBJS="pg_crc32c_sb8.o"
{ $as_echo "$as_me:${as_lineno-$LINENO}: result: slicing-by-8" >&5
$as_echo "slicing-by-8" >&6; }
fi
fi
fi
fi