RepoMirrors/ffmpeg
1
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2024-12-29 10:52:20 +00:00
Code Issues Projects Releases Wiki Activity
b7d063951d
ffmpeg/libavutil/cpu.c

283 lines
12 KiB
C
Raw Normal View History

Move mm_support() from libavcodec to libavutil, make it a public function and rename it to av_get_cpu_flags(). Originally committed as revision 25076 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-08 15:07:14 +00:00
/*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
avutil/cpu: #define _GNU_SOURCE before including any standard headers Otherwise its effect might not work causing CPU_COUNT to not get defined. Fixes cpu count detection to actually use sched_getaffinity if available. Signed-off-by: Marton Balint <cus@passwd.hu>
2022-04-14 21:14:27 +00:00
#include "config.h"
#if HAVE_SCHED_GETAFFINITY
#ifndef _GNU_SOURCE
# define _GNU_SOURCE
#endif
#include <sched.h>
#endif
cpu: add a function for querying maximum required data alignment
2017-02-08 08:32:17 +00:00
#include <stddef.h>
Add missing #includes for *INT64_MAX and *INT64_C
2013-11-23 20:32:55 +00:00
#include <stdint.h>
avutil: fix data race in av_get_cpu_flags() Make the one-time initialization in av_get_cpu_flags() thread-safe. The static variable |cpu_flags| in libavutil/cpu.c is read and written using normal load and store operations. These are considered as data races. The fix is to use atomic load and store operations. The fix can be verified by running the libavutil/tests/cpu_init.c test program under ThreadSanitizer: ./configure --toolchain=clang-tsan make libavutil/tests/cpu_init libavutil/tests/cpu_init There should be no warnings from ThreadSanitizer. Co-author: Dmitry Vyukov of Google, who suggested the data race fix. Signed-off-by: Wan-Teh Chang <wtc@google.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-12-07 02:16:13 +00:00
#include <stdatomic.h>
Add missing #includes for *INT64_MAX and *INT64_C
2013-11-23 20:32:55 +00:00
Merge commit 'e6bff23f1e11aefb16a2b5d6ee72bf7469c5a66e' * commit 'e6bff23f1e11aefb16a2b5d6ee72bf7469c5a66e': cpu: add a function for querying maximum required data alignment Adapted to work with the arbitrary runtime cpuflag changes av_force_cpu_flags() can generate. Merged-by: James Almer <jamrial@gmail.com>
2017-09-28 01:56:53 +00:00
#include "attributes.h"
Move mm_support() from libavcodec to libavutil, make it a public function and rename it to av_get_cpu_flags(). Originally committed as revision 25076 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-08 15:07:14 +00:00
#include "cpu.h"
avutil: Move internal CPU detection function declarations to private header
2013-08-20 15:49:01 +00:00
#include "cpu_internal.h"
avutil: add av_parse_cpu_flags() function This moves the cpu flag parsing code from avconv to avutil so it can be accessed elsewhere. Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-04-25 17:06:51 +00:00
#include "opt.h"
cpu: Include common.h for av_popcount64 This fixes build failures on windows since 2a6eaeaa8. Signed-off-by: Martin Storsjö <martin@martin.st>
2013-05-24 09:24:27 +00:00
#include "common.h"
Move mm_support() from libavcodec to libavutil, make it a public function and rename it to av_get_cpu_flags(). Originally committed as revision 25076 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-08 15:07:14 +00:00
winrt: multithreading support _beginthreadex is for desktop only. CreateThread is available for windows store apps on windows (and phone) 8.1 and later. http://msdn.microsoft.com/en-us/library/ms682453%28VS.85%29.aspx Signed-off-by: Matt Oliver <protogonoi@gmail.com>
2015-09-29 10:11:03 +00:00
#if HAVE_GETPROCESSAFFINITYMASK || HAVE_WINRT
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
#include <windows.h>
#endif
#if HAVE_SYSCTL
#if HAVE_SYS_PARAM_H
#include <sys/param.h>
#endif
#include <sys/types.h>
#include <sys/sysctl.h>
#endif
avutil/cpu: Use HAVE_UNISTD_H instead of HAVE_SYSCONF for #include <unistd.h> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-07-08 00:50:24 +00:00
#if HAVE_UNISTD_H
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
#include <unistd.h>
#endif
avutil: fix data race in av_get_cpu_flags() Make the one-time initialization in av_get_cpu_flags() thread-safe. The static variable |cpu_flags| in libavutil/cpu.c is read and written using normal load and store operations. These are considered as data races. The fix is to use atomic load and store operations. The fix can be verified by running the libavutil/tests/cpu_init.c test program under ThreadSanitizer: ./configure --toolchain=clang-tsan make libavutil/tests/cpu_init libavutil/tests/cpu_init There should be no warnings from ThreadSanitizer. Co-author: Dmitry Vyukov of Google, who suggested the data race fix. Signed-off-by: Wan-Teh Chang <wtc@google.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-12-07 02:16:13 +00:00
static atomic_int cpu_flags = ATOMIC_VAR_INIT(-1);
lavu/cpu.c: Add av_force_cpu_count() to override auto-detection.
2021-06-06 13:15:00 +00:00
static atomic_int cpu_count = ATOMIC_VAR_INIT(-1);
avutil/cpu: remove the |checked| static variable Remove the |checked| variable because the invalid value of -1 for |flags| can be used to indicate the same condition. Also rename |flags| to |cpu_flags| because there are a local variable and a function parameter named |flags| in the same file. Co-author: Dmitry Vyukov of Google Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-23 19:21:56 +00:00
static int get_cpu_flags(void)
{
all: Replace if (ARCH_FOO) checks by #if ARCH_FOO This is more spec-compliant because it does not rely on dead-code elimination by the compiler. Especially MSVC has problems with this, as can be seen in https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/296373.html or https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/297022.html This commit does not eliminate every instance where we rely on dead code elimination: It only tackles branching to the initialization of arch-specific dsp code, not e.g. all uses of CONFIG_ and HAVE_ checks. But maybe it is already enough to compile FFmpeg with MSVC with whole-programm-optimizations enabled (if one does not disable too many components). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-06-12 03:51:12 +00:00
#if ARCH_MIPS
return ff_get_cpu_flags_mips();
#elif ARCH_AARCH64
return ff_get_cpu_flags_aarch64();
#elif ARCH_ARM
return ff_get_cpu_flags_arm();
#elif ARCH_PPC
return ff_get_cpu_flags_ppc();
lavu/cpu: detect RISC-V base extensions This introduces compile-time and run-time CPU detection on RISC-V. In practice, I doubt that FFmpeg will ever see a RISC-V CPU without all of I, F and D extensions, and if it does, it probably won't have run-time detection. So the flags are essentially always set. But as things stand, checkasm wants them that way. Compare the ARMV8 flag on AArch64. We are nowhere near running short on CPU flag bits.
2022-09-26 14:52:21 +00:00
#elif ARCH_RISCV
return ff_get_cpu_flags_riscv();
all: Replace if (ARCH_FOO) checks by #if ARCH_FOO This is more spec-compliant because it does not rely on dead-code elimination by the compiler. Especially MSVC has problems with this, as can be seen in https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/296373.html or https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/297022.html This commit does not eliminate every instance where we rely on dead code elimination: It only tackles branching to the initialization of arch-specific dsp code, not e.g. all uses of CONFIG_ and HAVE_ checks. But maybe it is already enough to compile FFmpeg with MSVC with whole-programm-optimizations enabled (if one does not disable too many components). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-06-12 03:51:12 +00:00
#elif ARCH_X86
return ff_get_cpu_flags_x86();
#elif ARCH_LOONGARCH
return ff_get_cpu_flags_loongarch();
#endif
avutil/cpu: remove the |checked| static variable Remove the |checked| variable because the invalid value of -1 for |flags| can be used to indicate the same condition. Also rename |flags| to |cpu_flags| because there are a local variable and a function parameter named |flags| in the same file. Co-author: Dmitry Vyukov of Google Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-23 19:21:56 +00:00
return 0;
}
cpudetect: add av_force_cpu_flags() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2011-06-02 14:09:22 +00:00
void av_force_cpu_flags(int arg){
lavu/cpu: disable MMX warning on non x86 platforms We have AV_CPU_FLAG_ARMV8 == AV_CPU_FLAG_SSE3 which causes a trigger of this MMX warning on AArch64.
2017-06-28 13:20:45 +00:00
if (ARCH_X86 &&
(arg & ( AV_CPU_FLAG_3DNOW |
avutil/cpu: force mmx on selection of higher x86 SIMD features Fixes various runtime failures with manually set flags that represent no existing CPU Fixes Ticket3653 Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-20 03:23:52 +00:00
AV_CPU_FLAG_3DNOWEXT |
avutil/cpu: add missing check for mmxext to av_force_cpu_flags Reviewed-by: Michael Niedermayer <michaelni@gmx.at> Signed-off-by: James Almer <jamrial@gmail.com>
2015-03-26 05:11:55 +00:00
AV_CPU_FLAG_MMXEXT |
avutil/cpu: force mmx on selection of higher x86 SIMD features Fixes various runtime failures with manually set flags that represent no existing CPU Fixes Ticket3653 Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-20 03:23:52 +00:00
AV_CPU_FLAG_SSE |
AV_CPU_FLAG_SSE2 |
AV_CPU_FLAG_SSE2SLOW |
AV_CPU_FLAG_SSE3 |
AV_CPU_FLAG_SSE3SLOW |
AV_CPU_FLAG_SSSE3 |
AV_CPU_FLAG_SSE4 |
AV_CPU_FLAG_SSE42 |
AV_CPU_FLAG_AVX |
x86/cpu: add AV_CPU_FLAG_AVXSLOW flag Reviewed-by: Michael Niedermayer <michaelni@gmx.at> Signed-off-by: James Almer <jamrial@gmail.com>
2015-05-27 06:30:20 +00:00
AV_CPU_FLAG_AVXSLOW |
avutil/cpu: force mmx on selection of higher x86 SIMD features Fixes various runtime failures with manually set flags that represent no existing CPU Fixes Ticket3653 Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-20 03:23:52 +00:00
AV_CPU_FLAG_XOP |
AV_CPU_FLAG_FMA3 |
AV_CPU_FLAG_FMA4 |
avutil: add AVX-512 flags
2017-10-26 17:48:26 +00:00
AV_CPU_FLAG_AVX2 |
AV_CPU_FLAG_AVX512 ))
avutil/cpu: force mmx on selection of higher x86 SIMD features Fixes various runtime failures with manually set flags that represent no existing CPU Fixes Ticket3653 Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-20 03:23:52 +00:00
&& !(arg & AV_CPU_FLAG_MMX)) {
av_log(NULL, AV_LOG_WARNING, "MMX implied by specified flags\n");
arg |= AV_CPU_FLAG_MMX;
}
avutil: fix data race in av_get_cpu_flags() Make the one-time initialization in av_get_cpu_flags() thread-safe. The static variable |cpu_flags| in libavutil/cpu.c is read and written using normal load and store operations. These are considered as data races. The fix is to use atomic load and store operations. The fix can be verified by running the libavutil/tests/cpu_init.c test program under ThreadSanitizer: ./configure --toolchain=clang-tsan make libavutil/tests/cpu_init libavutil/tests/cpu_init There should be no warnings from ThreadSanitizer. Co-author: Dmitry Vyukov of Google, who suggested the data race fix. Signed-off-by: Wan-Teh Chang <wtc@google.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-12-07 02:16:13 +00:00
atomic_store_explicit(&cpu_flags, arg, memory_order_relaxed);
cpudetect: add av_force_cpu_flags() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2011-06-02 14:09:22 +00:00
}
Move mm_support() from libavcodec to libavutil, make it a public function and rename it to av_get_cpu_flags(). Originally committed as revision 25076 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-08 15:07:14 +00:00
int av_get_cpu_flags(void)
{
avutil: fix data race in av_get_cpu_flags() Make the one-time initialization in av_get_cpu_flags() thread-safe. The static variable |cpu_flags| in libavutil/cpu.c is read and written using normal load and store operations. These are considered as data races. The fix is to use atomic load and store operations. The fix can be verified by running the libavutil/tests/cpu_init.c test program under ThreadSanitizer: ./configure --toolchain=clang-tsan make libavutil/tests/cpu_init libavutil/tests/cpu_init There should be no warnings from ThreadSanitizer. Co-author: Dmitry Vyukov of Google, who suggested the data race fix. Signed-off-by: Wan-Teh Chang <wtc@google.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-12-07 02:16:13 +00:00
int flags = atomic_load_explicit(&cpu_flags, memory_order_relaxed);
avutil/cpu: remove the |checked| static variable Remove the |checked| variable because the invalid value of -1 for |flags| can be used to indicate the same condition. Also rename |flags| to |cpu_flags| because there are a local variable and a function parameter named |flags| in the same file. Co-author: Dmitry Vyukov of Google Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-23 19:21:56 +00:00
if (flags == -1) {
flags = get_cpu_flags();
avutil: fix data race in av_get_cpu_flags() Make the one-time initialization in av_get_cpu_flags() thread-safe. The static variable |cpu_flags| in libavutil/cpu.c is read and written using normal load and store operations. These are considered as data races. The fix is to use atomic load and store operations. The fix can be verified by running the libavutil/tests/cpu_init.c test program under ThreadSanitizer: ./configure --toolchain=clang-tsan make libavutil/tests/cpu_init libavutil/tests/cpu_init There should be no warnings from ThreadSanitizer. Co-author: Dmitry Vyukov of Google, who suggested the data race fix. Signed-off-by: Wan-Teh Chang <wtc@google.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-12-07 02:16:13 +00:00
atomic_store_explicit(&cpu_flags, flags, memory_order_relaxed);
avutil/cpu: remove the |checked| static variable Remove the |checked| variable because the invalid value of -1 for |flags| can be used to indicate the same condition. Also rename |flags| to |cpu_flags| because there are a local variable and a function parameter named |flags| in the same file. Co-author: Dmitry Vyukov of Google Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2016-11-23 19:21:56 +00:00
}
Cache detected CPU flags Originally committed as revision 25085 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-09 18:51:49 +00:00
return flags;
Move mm_support() from libavcodec to libavutil, make it a public function and rename it to av_get_cpu_flags(). Originally committed as revision 25076 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-09-08 15:07:14 +00:00
}
cmdutils/avutil: Move cpu caps parse code into av_parse_cpu_caps() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-04-26 21:21:43 +00:00
int av_parse_cpu_caps(unsigned *flags, const char *s)
{
static const AVOption cpuflags_opts[] = {
Merge commit 'd58dd4b5b5d31cfd4092e38a5f2c894eee2ab078' * commit 'd58dd4b5b5d31cfd4092e38a5f2c894eee2ab078': avopt: Store defaults for AV_OPT_TYPE_FLAGS in the i64 union member Conflicts: libavcodec/libvpxenc.c libavcodec/options_table.h libavfilter/vf_drawtext.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 12:09:09 +00:00
{ "flags" , NULL, 0, AV_OPT_TYPE_FLAGS, { .i64 = 0 }, INT64_MIN, INT64_MAX, .unit = "flags" },
cmdutils/avutil: Move cpu caps parse code into av_parse_cpu_caps() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-04-26 21:21:43 +00:00
#if ARCH_PPC
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "altivec" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ALTIVEC }, .unit = "flags" },
cmdutils/avutil: Move cpu caps parse code into av_parse_cpu_caps() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-04-26 21:21:43 +00:00
#elif ARCH_X86
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "mmx" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MMX }, .unit = "flags" },
{ "mmx2" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MMX2 }, .unit = "flags" },
{ "mmxext" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MMX2 }, .unit = "flags" },
{ "sse" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE }, .unit = "flags" },
{ "sse2" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE2 }, .unit = "flags" },
{ "sse2slow", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE2SLOW }, .unit = "flags" },
{ "sse3" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE3 }, .unit = "flags" },
{ "sse3slow", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE3SLOW }, .unit = "flags" },
{ "ssse3" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSSE3 }, .unit = "flags" },
{ "atom" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ATOM }, .unit = "flags" },
{ "sse4.1" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE4 }, .unit = "flags" },
{ "sse4.2" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SSE42 }, .unit = "flags" },
{ "avx" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AVX }, .unit = "flags" },
x86/cpu: add AV_CPU_FLAG_AVXSLOW flag Reviewed-by: Michael Niedermayer <michaelni@gmx.at> Signed-off-by: James Almer <jamrial@gmail.com>
2015-05-27 06:30:20 +00:00
{ "avxslow" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AVXSLOW }, .unit = "flags" },
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "xop" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_XOP }, .unit = "flags" },
x86: add detection for FMA3 instruction set Based on x264 code Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-22 05:47:01 +00:00
{ "fma3" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_FMA3 }, .unit = "flags" },
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "fma4" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_FMA4 }, .unit = "flags" },
x86/cpu: add missing avx2 AVOption in av_parse_cpu_flags() Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-06 04:34:01 +00:00
{ "avx2" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AVX2 }, .unit = "flags" },
x86: add detection for Bit Manipulation Instruction sets Based on x264 code Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-22 05:47:02 +00:00
{ "bmi1" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_BMI1 }, .unit = "flags" },
{ "bmi2" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_BMI2 }, .unit = "flags" },
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "3dnow" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_3DNOW }, .unit = "flags" },
{ "3dnowext", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_3DNOWEXT }, .unit = "flags" },
cpu: add cmov ro the table, which appears to have been forgotten Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-11-17 01:04:30 +00:00
{ "cmov", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_CMOV }, .unit = "flags" },
lavu: add AESNI CPU flag
2015-10-26 07:08:58 +00:00
{ "aesni", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AESNI }, .unit = "flags" },
avutil: add AVX-512 flags
2017-10-26 17:48:26 +00:00
{ "avx512" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AVX512 }, .unit = "flags" },
avutil/cpu: add AVX512 Icelake flag Signed-off-by: Wu Jianhua <jianhua.wu@intel.com> Reviewed-by: Henrik Gramner <henrik@gramner.com> Signed-off-by: James Almer <jamrial@gmail.com>
2022-02-23 08:57:30 +00:00
{ "avx512icl", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_AVX512ICL }, .unit = "flags" },
avutil/cpu: add slowgather to av_parse_cpu_caps() Signed-off-by: James Almer <jamrial@gmail.com>
2021-12-21 20:51:03 +00:00
{ "slowgather", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SLOW_GATHER }, .unit = "flags" },
cpu: add a few cpu names as aliases for their cpu flags Help to complete this list is welcome! Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-11-17 01:08:12 +00:00
#define CPU_FLAG_P2 AV_CPU_FLAG_CMOV | AV_CPU_FLAG_MMX
#define CPU_FLAG_P3 CPU_FLAG_P2 | AV_CPU_FLAG_MMX2 | AV_CPU_FLAG_SSE
#define CPU_FLAG_P4 CPU_FLAG_P3| AV_CPU_FLAG_SSE2
{ "pentium2", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_P2 }, .unit = "flags" },
{ "pentium3", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_P3 }, .unit = "flags" },
{ "pentium4", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_P4 }, .unit = "flags" },
#define CPU_FLAG_K62 AV_CPU_FLAG_MMX | AV_CPU_FLAG_3DNOW
#define CPU_FLAG_ATHLON CPU_FLAG_K62 | AV_CPU_FLAG_CMOV | AV_CPU_FLAG_3DNOWEXT | AV_CPU_FLAG_MMX2
#define CPU_FLAG_ATHLONXP CPU_FLAG_ATHLON | AV_CPU_FLAG_SSE
#define CPU_FLAG_K8 CPU_FLAG_ATHLONXP | AV_CPU_FLAG_SSE2
{ "k6", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MMX }, .unit = "flags" },
{ "k62", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_K62 }, .unit = "flags" },
{ "athlon", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_ATHLON }, .unit = "flags" },
{ "athlonxp", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_ATHLONXP }, .unit = "flags" },
{ "k8", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = CPU_FLAG_K8 }, .unit = "flags" },
cmdutils/avutil: Move cpu caps parse code into av_parse_cpu_caps() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-04-26 21:21:43 +00:00
#elif ARCH_ARM
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "armv5te", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ARMV5TE }, .unit = "flags" },
{ "armv6", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ARMV6 }, .unit = "flags" },
{ "armv6t2", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ARMV6T2 }, .unit = "flags" },
{ "vfp", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_VFP }, .unit = "flags" },
avutil/cpu: add missing entry for vfp_vm to av_parse_cpu_caps
2016-01-02 18:07:08 +00:00
{ "vfp_vm", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_VFP_VM }, .unit = "flags" },
Merge commit '124134e42455763b28cc346fed1d07017a76e84e' * commit '124134e42455763b28cc346fed1d07017a76e84e': avopt: Store defaults for AV_OPT_TYPE_CONST in the i64 union member Conflicts: libavcodec/aacenc.c libavcodec/libopenjpegenc.c libavcodec/options_table.h libavdevice/bktr.c libavdevice/v4l2.c libavdevice/x11grab.c libavfilter/af_amix.c libavfilter/vf_drawtext.c libavformat/movenc.c libavformat/options_table.h libavutil/opt.c Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-09-05 11:58:11 +00:00
{ "vfpv3", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_VFPV3 }, .unit = "flags" },
{ "neon", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_NEON }, .unit = "flags" },
avutil: turn arm setend into a cpuflag this allows disabling and enabling it it also prevents crashes if vfpv3 and neon are disabled which previously would have enabled the flag And last but not least one can enable setend on cpus like cortex-a8 where its fast but disabled by default Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-08-07 16:34:35 +00:00
{ "setend", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_SETEND }, .unit = "flags" },
avutil/cpu: add aarch64 entries to 2nd table Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-08-06 11:59:18 +00:00
#elif ARCH_AARCH64
{ "armv8", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_ARMV8 }, .unit = "flags" },
{ "neon", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_NEON }, .unit = "flags" },
{ "vfp", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_VFP }, .unit = "flags" },
libavutil: Detect MMI and MSA flags for MIPS Add MMI & MSA runtime detection for MIPS. Basically there are two code pathes. For systems that natively support CPUCFG instruction or kernel emulated that instruction, we'll sense this feature from HWCAP and report the flags according to values grab from CPUCFG. For systems that have no CPUCFG (or not export it in HWCAP), we'll parse /proc/cpuinfo instead. Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com> Reviewed-by: Shiyou Yin <yinshiyou-hf@loongson.cn> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2020-07-18 15:35:39 +00:00
#elif ARCH_MIPS
{ "mmi", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MMI }, .unit = "flags" },
{ "msa", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_MSA }, .unit = "flags" },
avutil: [loongarch] Add support for loongarch SIMD. LSX and LASX is loongarch SIMD extention. They are enabled by default if compiler support it, and can be disabled with '--disable-lsx' '--disable-lasx'. Change-Id: Ie2608ea61dbd9b7fffadbf0ec2348bad6c124476 Reviewed-by: Shiyou Yin <yinshiyou-hf@loongson.cn> Reviewed-by: guxiwei <guxiwei-hf@loongson.cn> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2021-12-15 03:51:04 +00:00
#elif ARCH_LOONGARCH
{ "lsx", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_LSX }, .unit = "flags" },
{ "lasx", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_LASX }, .unit = "flags" },
lavu/cpu: detect RISC-V base extensions This introduces compile-time and run-time CPU detection on RISC-V. In practice, I doubt that FFmpeg will ever see a RISC-V CPU without all of I, F and D extensions, and if it does, it probably won't have run-time detection. So the flags are essentially always set. But as things stand, checkasm wants them that way. Compare the ARMV8 flag on AArch64. We are nowhere near running short on CPU flag bits.
2022-09-26 14:52:21 +00:00
#elif ARCH_RISCV
{ "rvi", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVI }, .unit = "flags" },
{ "rvf", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVF }, .unit = "flags" },
{ "rvd", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVD }, .unit = "flags" },
lavu/cpu: CPU flags for the RISC-V Vector extension RVV defines a total of 12 different extensions, including: - 5 different instruction subsets: - Zve32x: 8-, 16- and 32-bit integers, - Zve32f: Zve32x plus single precision floats, - Zve64x: Zve32x plus 64-bit integers, - Zve64f: Zve32f plus Zve64x, - Zve64d: Zve64f plus double precision floats. - 6 different vector lengths: - Zvl32b (embedded only), - Zvl64b (embedded only), - Zvl128b, - Zvl256b, - Zvl512b, - Zvl1024b, - and the V extension proper: equivalent to Zve64f and Zvl128b. In total, there are 6 different possible sets of supported instructions (including the empty set), but for convenience we allocate one bit for each type sets: up-to-32-bit ints (RVV_I32), floats (RVV_F32), 64-bit ints (RVV_I64) and doubles (RVV_F64). Whence the vector size is needed, it can be retrieved by reading the unprivileged read-only vlenb CSR. This should probably be a separate helper macro if needed at a later point.
2022-09-26 14:52:25 +00:00
{ "rvv-i32", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVV_I32 }, .unit = "flags" },
{ "rvv-f32", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVV_F32 }, .unit = "flags" },
{ "rvv-i64", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVV_I64 }, .unit = "flags" },
{ "rvv", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVV_F64 }, .unit = "flags" },
lavu/riscv: CPU flag for the Zbb extension Unfortunately, it is common, and will remain so, that the Bit manipulations are not enabled at compilation time. This is an official policy for Debian ports in general (though they do not support RISC-V officially as of yet) to stick to the minimal target baseline, which does not include the B extension or even its Zbb subset. For inline helpers (CPOP, REV8), compiler builtins (CTZ, CLZ) or even plain C code (MIN, MAX, MINU, MAXU), run-time detection seems impractical. But at least it can work for the byte-swap DSP functions.
2022-10-02 11:54:58 +00:00
{ "rvb-basic",NULL, 0, AV_OPT_TYPE_CONST, { .i64 = AV_CPU_FLAG_RVB_BASIC }, .unit = "flags" },
cmdutils/avutil: Move cpu caps parse code into av_parse_cpu_caps() Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2012-04-26 21:21:43 +00:00
#endif
{ NULL },
};
static const AVClass class = {
.class_name = "cpuflags",
.item_name = av_default_item_name,
.option = cpuflags_opts,
.version = LIBAVUTIL_VERSION_INT,
};
const AVClass *pclass = &class;
return av_opt_eval_flags(&pclass, &cpuflags_opts[0], s, flags);
}
Merge commit '2a6eaeaa85d17b27ee0dd449183ec197c35c9675' * commit '2a6eaeaa85d17b27ee0dd449183ec197c35c9675': Move get_logical_cpus() from lavc/pthread to lavu/cpu. Conflicts: doc/APIchanges libavcodec/pthread.c libavutil/cpu.c libavutil/cpu.h libavutil/version.h Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-05-24 11:13:03 +00:00
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
int av_cpu_count(void)
{
avutil/cpu: Fix race condition in av_cpu_count() av_cpu_count() intends to emit a debug message containing the number of logical cores when called the first time. The check currently works with a static volatile int; yet this does not help at all in case of concurrent accesses by multiple threads. So replace this with an atomic_int. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
2020-12-02 22:57:16 +00:00
static atomic_int printed = ATOMIC_VAR_INIT(0);
av_cpu_count: factorize "detected %d logical cores" message Also print the message just once Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-05-26 00:18:38 +00:00
cpu: Restructure code to avoid pointless ret variable indirection libavutil/cpu.c:133:9: warning: unused variable ‘ret’ [-Wunused-variable]
2013-06-01 13:38:51 +00:00
int nb_cpus = 1;
lavu/cpu.c: Add av_force_cpu_count() to override auto-detection.
2021-06-06 13:15:00 +00:00
int count = 0;
winrt: multithreading support _beginthreadex is for desktop only. CreateThread is available for windows store apps on windows (and phone) 8.1 and later. http://msdn.microsoft.com/en-us/library/ms682453%28VS.85%29.aspx Signed-off-by: Matt Oliver <protogonoi@gmail.com>
2015-09-29 10:11:03 +00:00
#if HAVE_WINRT
SYSTEM_INFO sysinfo;
#endif
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
#if HAVE_SCHED_GETAFFINITY && defined(CPU_COUNT)
cpu_set_t cpuset;
CPU_ZERO(&cpuset);
cpu: Restructure code to avoid pointless ret variable indirection libavutil/cpu.c:133:9: warning: unused variable ‘ret’ [-Wunused-variable]
2013-06-01 13:38:51 +00:00
if (!sched_getaffinity(0, sizeof(cpuset), &cpuset))
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
nb_cpus = CPU_COUNT(&cpuset);
#elif HAVE_GETPROCESSAFFINITYMASK
DWORD_PTR proc_aff, sys_aff;
cpu: Restructure code to avoid pointless ret variable indirection libavutil/cpu.c:133:9: warning: unused variable ‘ret’ [-Wunused-variable]
2013-06-01 13:38:51 +00:00
if (GetProcessAffinityMask(GetCurrentProcess(), &proc_aff, &sys_aff))
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
nb_cpus = av_popcount64(proc_aff);
avutil/cpu: Use HW_NCPUONLINE to detect # of online CPUs with OpenBSD Signed-off-by: Brad Smith <brad@comstyle.com> Signed-off-by: Marton Balint <cus@passwd.hu>
2021-04-03 18:49:20 +00:00
#elif HAVE_SYSCTL && defined(HW_NCPUONLINE)
int mib[2] = { CTL_HW, HW_NCPUONLINE };
size_t len = sizeof(nb_cpus);
if (sysctl(mib, 2, &nb_cpus, &len, NULL, 0) == -1)
nb_cpus = 0;
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
#elif HAVE_SYSCTL && defined(HW_NCPU)
int mib[2] = { CTL_HW, HW_NCPU };
size_t len = sizeof(nb_cpus);
cpu: Restructure code to avoid pointless ret variable indirection libavutil/cpu.c:133:9: warning: unused variable ‘ret’ [-Wunused-variable]
2013-06-01 13:38:51 +00:00
if (sysctl(mib, 2, &nb_cpus, &len, NULL, 0) == -1)
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
nb_cpus = 0;
#elif HAVE_SYSCONF && defined(_SC_NPROC_ONLN)
nb_cpus = sysconf(_SC_NPROC_ONLN);
#elif HAVE_SYSCONF && defined(_SC_NPROCESSORS_ONLN)
nb_cpus = sysconf(_SC_NPROCESSORS_ONLN);
winrt: multithreading support _beginthreadex is for desktop only. CreateThread is available for windows store apps on windows (and phone) 8.1 and later. http://msdn.microsoft.com/en-us/library/ms682453%28VS.85%29.aspx Signed-off-by: Matt Oliver <protogonoi@gmail.com>
2015-09-29 10:11:03 +00:00
#elif HAVE_WINRT
GetNativeSystemInfo(&sysinfo);
nb_cpus = sysinfo.dwNumberOfProcessors;
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
#endif
avutil/cpu: Fix race condition in av_cpu_count() av_cpu_count() intends to emit a debug message containing the number of logical cores when called the first time. The check currently works with a static volatile int; yet this does not help at all in case of concurrent accesses by multiple threads. So replace this with an atomic_int. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
2020-12-02 22:57:16 +00:00
if (!atomic_exchange_explicit(&printed, 1, memory_order_relaxed))
av_cpu_count: factorize "detected %d logical cores" message Also print the message just once Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-05-26 00:18:38 +00:00
av_log(NULL, AV_LOG_DEBUG, "detected %d logical cores\n", nb_cpus);
lavu/cpu.c: Add av_force_cpu_count() to override auto-detection.
2021-06-06 13:15:00 +00:00
count = atomic_load_explicit(&cpu_count, memory_order_relaxed);
if (count > 0) {
nb_cpus = count;
av_log(NULL, AV_LOG_DEBUG, "overriding to %d logical cores\n", nb_cpus);
}
Move get_logical_cpus() from lavc/pthread to lavu/cpu. It will be useful in lavfi, and could conceivably be useful to the user applications as well.
2013-05-23 06:59:07 +00:00
return nb_cpus;
}
cpu: add a function for querying maximum required data alignment
2017-02-08 08:32:17 +00:00
lavu/cpu: Use av_cpu_ prefix
2021-07-20 07:33:47 +00:00
void av_cpu_force_count(int count)
{
lavu/cpu.c: Add av_force_cpu_count() to override auto-detection.
2021-06-06 13:15:00 +00:00
atomic_store_explicit(&cpu_count, count, memory_order_relaxed);
}
cpu: add a function for querying maximum required data alignment
2017-02-08 08:32:17 +00:00
size_t av_cpu_max_align(void)
{
all: Replace if (ARCH_FOO) checks by #if ARCH_FOO This is more spec-compliant because it does not rely on dead-code elimination by the compiler. Especially MSVC has problems with this, as can be seen in https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/296373.html or https://ffmpeg.org/pipermail/ffmpeg-devel/2022-May/297022.html This commit does not eliminate every instance where we rely on dead code elimination: It only tackles branching to the initialization of arch-specific dsp code, not e.g. all uses of CONFIG_ and HAVE_ checks. But maybe it is already enough to compile FFmpeg with MSVC with whole-programm-optimizations enabled (if one does not disable too many components). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2022-06-12 03:51:12 +00:00
#if ARCH_MIPS
return ff_get_cpu_max_align_mips();
#elif ARCH_AARCH64
return ff_get_cpu_max_align_aarch64();
#elif ARCH_ARM
return ff_get_cpu_max_align_arm();
#elif ARCH_PPC
return ff_get_cpu_max_align_ppc();
#elif ARCH_X86
return ff_get_cpu_max_align_x86();
#elif ARCH_LOONGARCH
return ff_get_cpu_max_align_loongarch();
#endif
cpu: add a function for querying maximum required data alignment
2017-02-08 08:32:17 +00:00
return 8;
}
Reference in New Issue Copy Permalink