268 articles

26 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

20:16
swscale/ppc/swscale_ppc_template: Fix av_unused placement - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swscale/ppc/swscale_ppc_template: Fix av_unused placement Forgotten in d6cb0d2c2bb8469f17d59dd82c8221b98e169d1a. Reviewed-by: Sean McGovern <gseanmcgⓐgmail.com> Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswscale/ppc/swscale_ppc_template.c
14:17
avcodec/x86/qpeldsp_init: Fix compilation without external assembly - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/qpeldsp_init: Fix compilation without external assembly Broken in 2cf9e733c6a666600423a0967f23341d9f09e3c8. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/qpeldsp_init.c

25 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

17:31
ALL: move av_unused to conform with standard requirement - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

ALL: move av_unused to conform with standard requirement This is required placement by standard attribute, works the same for __attribute__((unused)). Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] compat/w32pthreads.h [DH] fftools/opt_common.c [DH] libavcodec/aacpsy.c [DH] libavcodec/cabac_functions.h [DH] libavcodec/cbs_av1.c [DH] libavcodec/dvbsubdec.c [DH] libavcodec/flacenc.c [DH] libavcodec/get_bits.h [DH] libavcodec/h264_mvpred.h [DH] libavcodec/h264pred_template.c [DH] libavcodec/h264qpel_template.c [DH] libavcodec/hqxvlc.h [DH] libavcodec/jpeg2000dec.c [DH] (...)
17:24
avutil/attributes: prefer clang attributes even in non-gnu builds - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

avutil/attributes: prefer clang attributes even in non-gnu builds In MSVC mode Clang doesn't define __GNUC__, but we can still attributes. Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libavutil/attributes.h
17:22
avutil/attributes: use standard attributes when they are available - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

avutil/attributes: use standard attributes when they are available Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libavutil/attributes.h
17:10
avutil/attributes: add AV_HAS_STD_ATTRIBUTE - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

avutil/attributes: add AV_HAS_STD_ATTRIBUTE For testing language standard attributes, for both C++ and C. Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libavutil/attributes.h
18:35
forgejo/workflows: include size and mtime in cache hash - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

forgejo/workflows: include size and mtime in cache hash In case some file has been updated. Generally fate samples shouldn't be replaced to preserve compatibility with older revisions, but before merge it may happen that files is replaced. Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] .forgejo/workflows/test.yml
16:19
configure: suppress C4267 warnings from MSVC - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

configure: suppress C4267 warnings from MSVC Suppresses implicit integer conversion narrowing warnings: warning C4267: 'initializing': conversion from 'size_t' to 'int', possible loss of data Those implicit conversions are abundant in ffmpeg's code base. Additionally equivalent warnings are not enabled for GCC/Clang by default, so they are mostly left unfixed. Suppress reports about them to reduce noise in MSVC build log. Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] (...)

24 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

23:31
avfilter/framesync: don't attempt to rescale AV_NOPTS_VALUE - James Almer git.videolan.org Git - ffmpeg.git/rss log

avfilter/framesync: don't attempt to rescale AV_NOPTS_VALUE Part of a fix for issue #20589. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavfilter/framesync.c
23:31
avfilter/vf_scale: don't attempt to rescale AV_NOPTS_VALUE - James Almer git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_scale: don't attempt to rescale AV_NOPTS_VALUE Finishes fixing issue #20589. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavfilter/vf_scale.c
10:07
avcodec/libmpeghdec: Remove redundant manual close calls - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Remove redundant manual close calls This decoder has the INIT_CLEANUP flag set. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
10:09
avcodec/libmpeghdec: Remove private class - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Remove private class This decoder doesn't have any options. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
10:10
avcodec/libmpeghdec: Don't set AVCodec.sample_fmts - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Don't set AVCodec.sample_fmts It is mostly pointless for decoders (only useful for those codecs for which a floating-point and a fixed-point decoder exist). Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
10:13
avcodec/libmpeghdec: Inline constant - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Inline constant Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
10:18
avcodec/libmpeghdec: Don't use too big buffer - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Don't use too big buffer The channel count is fixed, so we don't need to allocate space to hold samples for nonexisting channels. This also means that the maximum channel count is no longer hardcoded in the macro. Also switch to decoder_buffer_size to be samples-based. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
10:22
avcodec/libmpeghdec: Remove always-false check - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Remove always-false check The pointer to the decoder is always set after init has been called successfully. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
13:00
avcodec/libmpeghdec: Align FFCodec initializers - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Align FFCodec initializers Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
15:05
avcodec/libmpeghdec: Don't set AVCodecContext.frame_size - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Don't set AVCodecContext.frame_size It indicates a constant frame size (in samples); yet the documentation of the MPEGH_DECODER_OUTPUT_INFO structure says that it "gives information about the currently decoded audio data" and makes no guarantees about it being constant. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
16:07
avcodec/libmpeghdec: Check channel layouts generically - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/libmpeghdec: Check channel layouts generically Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/libmpeghdec.c
03:01
avcodec/tiff_common: Remove unused ff_tadd_*_metadata() funcs - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/tiff_common: Remove unused ff_tadd_*_metadata() funcs Unused since ad77345a5d14862f4701e5ad422b03b14934a5b9. Reviewed-by: Leo Izen <leo.izenⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/tiff_common.c [DH] libavcodec/tiff_common.h
03:21
Revert "lavc/libsvtav1: set packet durations" - rcombs git.videolan.org Git - ffmpeg.git/rss log

Revert "lavc/libsvtav1: set packet durations" This reverts commit 5c9b2027bc48ae5d39b0d82696895f0834788242. This doesn't actually work the way it'd appeared to in testing; the output was based on frame *encode latency*. [DH] libavcodec/libsvtav1.c

23 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

03:34
avcodec/hpel_template: Fix unintentional usage of unsigned offsets - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/hpel_template: Fix unintentional usage of unsigned offsets The value of sizeof() is of type size_t which means that an expression like src1[i * src_stride1 + 4 * (int)sizeof(pixel)] will use a very large offset if src_stride1 is sufficiently negative. It works in practice (because it is correct modulo SIZE_MAX), but UBSan treats it as error: libavcodec/hpel_template.c:104:1: runtime error: addition of unsigned offset to 0x7ffdfa0391d8 overflowed to 0x7ffdfa0391cc Fix this by casting sizeof(pixel) to int. (This has been uncovered by a checkasm test for the hpeldsp which will be (...)
04:11
avcodec/hpeldsp,_template: Use ptrdiff_t for strides - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/hpeldsp,_template: Use ptrdiff_t for strides Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/hpel_template.c [DH] libavcodec/hpeldsp.c
01:52
avcodec/x86/hpeldsp: Add SSE2 put_no_rnd size 16 versions - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp: Add SSE2 put_no_rnd size 16 versions These currently only exist as MMX and (not bitexact) MMXEXT versions. The added functions occupy 288B here. So far, they are only for the x2 and y2 (i.e. right and down, not down-right) directions. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp.asm [DH] libavcodec/x86/hpeldsp_init.c
02:15
avcodec/x86/hpeldsp: Add SSE2 avg_no_rnd size 16 versions - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp: Add SSE2 avg_no_rnd size 16 versions These currently only exist as MMX versions. The added functions occupy 320B here. So far, they are only for the x2 and y2 (i.e. right and down, not down-right) directions. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp.asm [DH] libavcodec/x86/hpeldsp_init.c
02:32
avcodec/x86/hpeldsp_init: Remove MMX(EXT) funcs overridden by SSE2 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp_init: Remove MMX(EXT) funcs overridden by SSE2 This affects the avg,put_no_rnd_pixels16_x,y2 MMX and (put-only) MMXEXT versions. Removing these functions saved 1184B here. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp_init.c
02:49
avcodec/x86/hpeldsp_init: Avoid complicating macro - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp_init: Avoid complicating macro Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp_init.c
03:01
avcodec/x86/rnd_template: Merge into hpeldsp_init.c - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/rnd_template: Merge into hpeldsp_init.c It is now only included exactly once. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp_init.c [DH] libavcodec/x86/rnd_template.c
14:05
avcodec/ohdec: Check mutex/conditions initialization - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/ohdec: Check mutex/conditions initialization Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/ohdec.c
13:52
avcodec/ohdec: Switch to RefStruct API for internal refcounting - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/ohdec: Switch to RefStruct API for internal refcounting It avoids allocations and corresponding error conditions. Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/ohdec.c
13:03
avcodec/ohdec: Release decoder on allocation failure - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/ohdec: Release decoder on allocation failure Normally, the OH_AVCodec is wrapped inside an AVBuffer to be freed in its free callback; yet when creating the AVBuffer fails, the decoder is never destroyed. Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/ohdec.c
22:10
avcodec/x86/apv_dsp: Don't export arrays unnecessarily - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/apv_dsp: Don't export arrays unnecessarily The const here is unnecessary, because everything inside SECTION_RODATA is automatically const and using it exports these objects from the object file because const is a macro in x86inc.asm. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/apv_dsp.asm
03:40
configure: Add missing dependencies for AHX decoder - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

configure: Add missing dependencies for AHX decoder Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] configure
03:42
avcodec/mpegaudiodec_float: Don't set AVCodec.sample_fmts directly - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mpegaudiodec_float: Don't set AVCodec.sample_fmts directly It is deprecated and doing so gives warnings from Clang. Use CODEC_SAMPLEFMTS instead. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mpegaudiodec_float.c
07:45
lavc/libsvtav1: set packet durations - rcombs git.videolan.org Git - ffmpeg.git/rss log

lavc/libsvtav1: set packet durations [DH] libavcodec/libsvtav1.c

22 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

12:31
MAINTAINERS: add my OpenPGP key - Sean McGovern git.videolan.org Git - ffmpeg.git/rss log

MAINTAINERS: add my OpenPGP key My key is available at: https://keys.openpgp.org/vks/v1/by-fingerprint/6D03BC603A33E6156E2E06AD8C0681756F598684 and https://keyserver.ubuntu.com/pks/lookup?op=get&search=0x6d03bc603a33e6156e2e06ad8c0681756f598684 [DH] MAINTAINERS
12:45
MAINTAINERS: add me for Linux / PowerPC architecture - Sean McGovern git.videolan.org Git - ffmpeg.git/rss log

MAINTAINERS: add me for Linux / PowerPC architecture [DH] MAINTAINERS
01:43
tests/checkasm: Add hpeldsp checkasm - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

tests/checkasm: Add hpeldsp checkasm Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] tests/checkasm/Makefile [DH] tests/checkasm/checkasm.c [DH] tests/checkasm/checkasm.h [DH] tests/checkasm/hpeldsp.c [DH] tests/fate/checkasm.mak
03:24
avcodec/x86/hpeldsp_init: Remove MMX functions overridden by MMXEXT - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp_init: Remove MMX functions overridden by MMXEXT Forgotten in a51279bbdea0d6db920d71980262bccd0ce78226 because I only looked for MMX(EXT) functions overridden by SSE2. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp_init.c
03:41
avcodec/hpeldsp: Make put_no_rnd_pixels_tab smaller - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/hpeldsp: Make put_no_rnd_pixels_tab smaller Only the blocksizes 16 and 8 are implemented, yet the motion estimation code touches the blocksize 4 entries. But really nothing touches the blocksize 2 entries, so that we can reduce the put_no_rnd_pixels_tab array size to [3][4]. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/hpeldsp.h
15:14
avcodec/aom_film_grain: fix AVCOL_*_NB range check - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avcodec/aom_film_grain: fix AVCOL_*_NB range check These should also exclude the _NB values themselves. [DH] libavcodec/aom_film_grain.c
19:00
avformat/mov: Merge tts after building index for old-style uncompressed PCM - Derek Buitenhuis git.videolan.org Git - ffmpeg.git/rss log

avformat/mov: Merge tts after building index for old-style uncompressed PCM This was a regression introduced in 292c1df7c159c8a1a7afe52613d164ff417e81ce. Since we don't know the length of the stts data until after building the index, since we're generating it, we need to merge any ctts data after, since otherwise tts_count is set to 0, and no packets will be output. We can't remove the merge entirely, because uncompressed PCM with a ctts atom is technically valid (e.g. a constant CTS offset). This fixes old-style uncompressed PCM demuxing. Fixes #11490. Signed-off-by: Derek (...)
13:34
avformat/tls_openssl: Don't call functions inside FFMIN - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/tls_openssl: Don't call functions inside FFMIN It may call the function multiple times. Reviewed-by: Timo Rothenpieler <timoⓐrothenpieler.org> Reviewed-by: Steven Liu <lingjiujiankeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/tls_openssl.c
11:28
libavutil/arm: Rename the HWCAP defines - Brad Smith git.videolan.org Git - ffmpeg.git/rss log

libavutil/arm: Rename the HWCAP defines Rename the HWCAP defines to use the same naming scheme as AArch64 and PPC. Signed-off-by: Brad Smith <bradⓐcomstyle.com> [DH] libavutil/arm/cpu.c

21 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

20:51
avcodec/hpeldsp: Fix documentation - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/hpeldsp: Fix documentation This commit fixes two issues in the documentation: a) The documentation for put,avg_pixels_tab only mentions widths 16 and 8, although it explicitly mentions that there are four horizontal blocksizes. This part of the patch basically reverts e5771f4f37b67951485205e110f4da5e7e32ea74. b) The restrictions on height don't match the reality. While most users abide by it, some do not: i) vp56.c copies a 16x12 block. ii) indeo3 can copy an arbitrary multiple of four lines for block widths 4, 8 and 16. iii) SVQ3 can use block sizes luma block sizes 16x16, (...)
00:08
avcodec/x86/qpeldsp_init: Use SSE2 versions where possible - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/qpeldsp_init: Use SSE2 versions where possible The mc00 versions (i.e. the qdsp functions with no subpixel interpolation) are just wrappers around their fpel versions. There are SSE2 versions of these, yet the qpel code only uses the MMX(EXT) versions. This commit changes this and also removes the MMX(EXT) versions. This also allowed to remove ff_avg_pixels16_mmxext, ff_put_pixels16_mmx. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/fpel.asm [DH] libavcodec/x86/fpel.h [DH] (...)
03:28
avcodec/x86/hpeldsp_init: Remove MMX(EXT) funcs overridden by SSSE3 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp_init: Remove MMX(EXT) funcs overridden by SSSE3 SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD), so that the overwhelming majority of our users (particularly those that actually update their FFmpeg) will be using the SSSE3 versions. This commit therefore removes the MMX(EXT) functions overridden by them (which don't abide by the ABI) to get closer to a removal of emms_c. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp.asm [DH] (...)
03:55
avcodec/x86/rv40dsp_init: Remove MMX(EXT) funcs overridden by SSSE3 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/rv40dsp_init: Remove MMX(EXT) funcs overridden by SSSE3 SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD), so that the overwhelming majority of our users (particularly those that actually update their FFmpeg) will be using the SSSE3 versions. This commit therefore removes the MMX(EXT) functions overridden by them (which don't abide by the ABI) to get closer to a removal of emms_c. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp.asm [DH] libavcodec/x86/hpeldsp.h (...)
14:17
avcodec/x86/vvc/sao_10bit: Remove unused functions - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/vvc/sao_10bit: Remove unused functions Saves 65280B here. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/vvc/sao_10bit.asm
04:22
avcodec/x86/mpegvideoencdsp_init: Remove MMX, 3DNOw funcs overridden by SSSE3 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/mpegvideoencdsp_init: Remove MMX, 3DNOw funcs overridden by SSSE3 SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD), so that the overwhelming majority of our users (particularly those that actually update their FFmpeg) will be using the SSSE3 versions. This commit therefore removes the MMX and 3DNOW functions overridden by them (which don't abide by the ABI) to get closer to a removal of emms_c. Also merge the mpegvideoenc_qns_template.c file into the main file. The 3DNOW functions removed in this commit were the last in the codebase. Reviewed-by: Lynne (...)
11:12
avfilter/x86/vf_gradfun: Remove MMXEXT func overridden by SSSE3 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avfilter/x86/vf_gradfun: Remove MMXEXT func overridden by SSSE3 SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD), so that the overwhelming majority of our users (particularly those that actually update their FFmpeg) will be using the SSSE3 version of filter_line. This commit therefore removes the overridden MMXEXT version (which didn't abide by the ABI) which allows us to remove an emms_c() from vf_gradfun.c, so that users with SSSE3 no longer pay a price for the mere existence of an MMXEXT version. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas (...)
13:12
avcodec/x86/h264_qpel: Remove MMX(EXT) funcs overridden by SSSE3 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/h264_qpel: Remove MMX(EXT) funcs overridden by SSSE3 SSSE3 is already quite old (introduced 2006 for Intel, 2011 for AMD), so that the overwhelming majority of our users (particularly those that actually update their FFmpeg) will be using the SSSE3 versions. This commit therefore removes the MMX(EXT) functions overridden by them (which don't abide by the ABI) to get closer to a removal of emms_c. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/h264_qpel.c [DH] (...)
13:33
avfilter/af_whisper: fix srt index - Vittorio Palmisano git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_whisper: fix srt index The srt index should be incremented for each segment. [DH] libavfilter/af_whisper.c
13:19
avfilter/af_whisper: fix int64 printf format - Vittorio Palmisano git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_whisper: fix int64 printf format Use PRId64 for printing int64_t values in the SRT output. [DH] libavfilter/af_whisper.c
14:49
configure: strip non numeric trailer from gcc version - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

configure: strip non numeric trailer from gcc version Fixes: ../configure: 7820: [: Illegal number: 13-win32 Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] configure
03:06
libavutil/arm: Make use of elf_aux_info() on FreeBSD/OpenBSD - Brad Smith git.videolan.org Git - ffmpeg.git/rss log

libavutil/arm: Make use of elf_aux_info() on FreeBSD/OpenBSD - FreBSD/OpenBSD have elf_aux_info() on arm - Wrap AT_HWCAP as the value is different for BSD vs Linux (16 vs 25) Signed-off-by: Brad Smith <bradⓐcomstyle.com> [DH] libavutil/arm/cpu.c
11:28
avfilter/vf_colordetect: only report detected properties on EOF - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_colordetect: only report detected properties on EOF Instead of reporting them also when the filtergraph is suddenly destroyed mid-stream, e.g. during the `ffmpeg` tool's early init. [DH] libavfilter/vf_colordetect.c

20 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

22:26
avcodec/x86/hpeldsp_init: Remove MMX(EXT) functions overridden by SSE2FAST - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/hpeldsp_init: Remove MMX(EXT) functions overridden by SSE2FAST CPUs which support SSE2, but not in a fast way (so that they get the additional AV_CPU_FLAG_SSE2SLOW) are ancient nowadays (2007 and older), so ignore the distinction between the two and remove MMX and MMXEXT functions that are now overridden by SSE2 functions. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/x86/hpeldsp.asm [DH] (...)
23:18
avcodec/x86/h264_qpel: Remove MMX(EXT) functions overridden by SSE2FAST - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/h264_qpel: Remove MMX(EXT) functions overridden by SSE2FAST CPUs which support SSE2, but not in a fast way (so that they get the additional AV_CPU_FLAG_SSE2SLOW) are ancient nowadays (2007 and older), so ignore the distinction between the two and remove MMX and MMXEXT functions that are now overridden by SSE2 functions. Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] (...)
05:51
avformat/udp: fix warning about unused varible - Brad Smith git.videolan.org Git - ffmpeg.git/rss log

avformat/udp: fix warning about unused varible libavformat/udp.c:283:9: warning: unused variable 'i' [-Wunused-variable] Signed-off-by: Brad Smith <bradⓐcomstyle.com> [DH] libavformat/udp.c

19 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

15:35
configure: deprecate support for libnpp - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: deprecate support for libnpp [DH] configure [DH] libavfilter/version_major.h [DH] libavfilter/vf_scale_npp.c [DH] libavfilter/vf_sharpen_npp.c [DH] libavfilter/vf_transpose_npp.c
23:25
avformat/dhav: Fix off by length of read element error - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

avformat/dhav: Fix off by length of read element error Fixes: out of array read Fixes: BIGSLEEP-445383648 Fixes: crash_input_1/2/3.dhav Found-by: Google Big Sleep Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavformat/dhav.c
17:42
avformat/mccenc: Remove redundant check - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Remove redundant check This has already been checked in init. Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
17:53
avformat/mccenc: Deduplicate strings - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Deduplicate strings Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
17:16
avformat/mccenc: Constify read-only data - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Constify read-only data Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
17:16
avformat/mccenc: Remove redundant setting of time base - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Remove redundant setting of time base It has already been done in init. Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
16:49
avformat/mccenc: Fix assert check - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Fix assert check Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
18:36
avformat/mccenc: Avoid relocations - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Avoid relocations Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
18:00
avformat/mccenc: Add newlines to logmessages - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Add newlines to logmessages Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
17:55
avformat/mccenc: Check version-timecode compatibility earlier - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Check version-timecode compatibility earlier Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
18:56
avformat/mccenc: Hardcode codec names - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccenc: Hardcode codec names Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccenc.c
18:40
avformat/mccdec: Avoid relocations - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccdec: Avoid relocations Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccdec.c
18:56
avformat/mccdec: Constify data - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/mccdec: Constify data Reviewed-by: Jacob Lifshay <programmerjakeⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/mccdec.c

18 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

12:58
Revert "avformat/tls_openssl: add av_assert0() for tls_shared" - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

Revert "avformat/tls_openssl: add av_assert0() for tls_shared" This reverts commits fd55c4b5f72a157fbb128d0ef203e9922b53552b (for tls_openssl.c) and c74181a04b5f4e650eae662231e56518daef64d4 (for tls_gnutls.c). It is impossible for a pointer to a member of a structure to be NULL: If the containing structure exists, the member exists and can't have a NULL address; if the containing structure does not exist, then getting a pointer to the substructure via &c->tls_shared would already be undefined behavior. Reviewed-by: Kacper Michajłow <kasper93ⓐgmail.com> (...)
15:57
avcodec/h274: Make H274FilmGrainDatabase a shared object - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/h274: Make H274FilmGrainDatabase a shared object Right now, the private contexts of every decoder supporting H.274 film grain synthesis (namely H.264, HEVC and VVC) contain a H274FilmGrainDatabase; said structure is very large 700442B before this commit) and takes up the overwhelming majority of said contexts: Removing it reduces sizeof(H264Context) by 92.88%, sizeof(HEVCContext) by 97.78% and sizeof(VVCContext) by 99.86%. This is especially important for H.264 and HEVC when using frame-threading. The content of said film grain database does not depend on any input parameter; it (...)
00:25
avcodec/exif: check count in exif_decode_tag() - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

avcodec/exif: check count in exif_decode_tag() Fixes: out of array access Fixes: integer overflow Fixes: poc_heap_bof Found-by: *2ourc3 (Salim LARGO) Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavcodec/exif.c
01:54
avfilter/vsrc_gfxcapture: fix string format specifier - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: fix string format specifier Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libavfilter/vsrc_gfxcapture_winrt.cpp

17 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

00:45
avformat/movenc: clear subsample information on fragment flush - James Almer git.videolan.org Git - ffmpeg.git/rss log

avformat/movenc: clear subsample information on fragment flush Don't keep around information from a previous traf atom. Fixes issue #20492. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavformat/movenc.c [DH] libavformat/movenccenc.c [DH] libavformat/movenccenc.h
20:03
avfilter/vsrc_gfxcapture: keep cbdata object alive - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: keep cbdata object alive Depending on the threading backend the stdlib uses, creating a mutex/condvar can be quite expensive. So keep this object alive in the ctx, on which we synchronize via the uninit mutex anyway. [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
19:31
avfilter/vsrc_gfxcapture: fix leaking all callback handlers - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: fix leaking all callback handlers Fixes #20537 [DH] libavfilter/vsrc_gfxcapture_winrt.hpp
18:23
avfilter/vsrc_gfxcapture: use free threaded capture frame pool - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: use free threaded capture frame pool Apparently, using a normal frame pool in a multithreaded environment leads to strange resource leaks on shutdown, which vanish when using a free threaded pool. [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
17:13
avfilter/vsrc_gfxcapture: stop capture session before initializing capture thread... - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: stop capture session before initializing capture thread shutdown It might have things going on in the background that still need cleaned up, and this gives it a chance to enqueue them on the dispatch queue. [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
17:12
avfilter/vsrc_gfxcapture: fix re-using ret variable from outside of lambda scope - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: fix re-using ret variable from outside of lambda scope [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
16:31
avfilter/vsrc_gfxcapture: don't pass pointer to ComPtr - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: don't pass pointer to ComPtr While it does appear to work fine, with all the operator overloads, it at least has potential for surprises, so pass it by reference instead. [DH] libavfilter/vsrc_gfxcapture_winrt.cpp

16 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

12:11
avfilter/vf_idet: separate DSP parts - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idet: separate DSP parts To avoid pulling in the entire libavfilter when using the DSP functions from checkasm. The rest of the struct is not needed outside vf_idet.c and was moved there. [DH] libavfilter/Makefile [DH] libavfilter/vf_idet.c [DH] libavfilter/vf_idet.h [DH] libavfilter/vf_idetdsp.c [DH] libavfilter/vf_idetdsp.h [DH] libavfilter/x86/Makefile [DH] libavfilter/x86/vf_idet.asm [DH] libavfilter/x86/vf_idet_init.c [DH] libavfilter/x86/vf_idetdsp.asm [DH] (...)
12:17
avfilter/vf_idetdsp: pass actual bit depth - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idetdsp: pass actual bit depth More informative and IMO cleaner; some implementations may want to differentiate by exact bit depth or support 32 bit down the line. [DH] libavfilter/vf_idet.c [DH] libavfilter/vf_idetdsp.c [DH] libavfilter/vf_idetdsp.h [DH] libavfilter/x86/vf_idetdsp_init.c
12:23
avfilter/vf_idetdsp: use consistent uint8_t pointer type - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idetdsp: use consistent uint8_t pointer type Even for 16-bit DSP functions. Instead, cast the pointer inside the function. [DH] libavfilter/vf_idetdsp.c [DH] libavfilter/vf_idetdsp.h [DH] libavfilter/x86/vf_idetdsp_init.c
15:11
lavf: fix demuxing of FLAC files with id3v2 tags - Lynne git.videolan.org Git - ffmpeg.git/rss log

lavf: fix demuxing of FLAC files with id3v2 tags Due to the recent id3v2 refactor, FLAC was left out due to earlier code not checking for id3v2 presence on FLAC. Without the id3v2 data parsed, detection of FLAC and therefore demuxing fails. Fixes 9d037c54f209958d47ac376d2a9561608f98dfae [DH] libavformat/flacdec.c
16:56
avcodec/decode: sync initial_pict_type and intra_only_flag with thread worker's avctx - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/decode: sync initial_pict_type and intra_only_flag with thread worker's avctx Regression since 5acbdd2264d3b90dc11369f9e031e762f260882e, which removed setting both values from PerThreadContext. Given the pthread code calls ff_decode_receive_frame_internal() on the frame, any value set before it will be overwritten, so instead sync each thread's DecodeContext and let ff_decode_receive_frame_internal() handle these values. Fixes issue #20534. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/decode.c [DH] (...)
17:41
avcodec/h2645_sei: add missing ambient_viewing_environment copy in ff_h2645_sei_ctx_r... - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/h2645_sei: add missing ambient_viewing_environment copy in ff_h2645_sei_ctx_replace() Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/h2645_sei.c
17:43
tests/fate-run: pass threads argument to ffprobe - James Almer git.videolan.org Git - ffmpeg.git/rss log

tests/fate-run: pass threads argument to ffprobe Ensures it honors the THREADS variable if set to anything other than 1, which will be useful to detect bugs. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] tests/fate-run.sh
17:44
tests/fate/prores: add a probe test - James Almer git.videolan.org Git - ffmpeg.git/rss log

tests/fate/prores: add a probe test Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] tests/fate/prores.mak [DH] tests/ref/fate/prores-probe
13:32
avfilter/vsrc_gfxcapture: remove needless use of atomics - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: remove needless use of atomics [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
01:18
avcodec/x86/pngdsp: add missing emms at the end of add_png_paeth_prediction - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/pngdsp: add missing emms at the end of add_png_paeth_prediction Fixes unpredictable behavior with floats. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/x86/pngdsp.asm

15 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

16:02
avfilter/vf_idet: correctly reset pixdesc on format change - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idet: correctly reset pixdesc on format change [DH] libavfilter/vf_idet.c
15:56
avfilter/vf_idet: expose DSP init function internally - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idet: expose DSP init function internally For checkasm. [DH] libavfilter/vf_idet.c [DH] libavfilter/vf_idet.h
16:02
avfilter/vf_idet: correctly update dsp function on format change - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_idet: correctly update dsp function on format change Currently, this never updates on change from high bit depth to low bit depth. [DH] libavfilter/vf_idet.c
16:10
tests/checkasm: add vf_idet checkasm - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

tests/checkasm: add vf_idet checkasm [DH] tests/checkasm/Makefile [DH] tests/checkasm/checkasm.c [DH] tests/checkasm/checkasm.h [DH] tests/checkasm/vf_idet.c
15:20
avfilter/x86/vf_idetdsp: generalize 8-bit macro - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/x86/vf_idetdsp: generalize 8-bit macro This is mostly compatible with AVX as well, so turn it into a macro. [DH] libavfilter/x86/vf_idetdsp.asm
15:47
avfilter/x86/vf_idetdsp: add AVX2 and AVX512 implementations - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/x86/vf_idetdsp: add AVX2 and AVX512 implementations The only thing that changes slightly is the horizontal sum at the end. [DH] libavfilter/x86/vf_idetdsp.asm [DH] libavfilter/x86/vf_idetdsp_init.c
12:11
vp9: Add 8bpc AVX2 asm for inverse transforms - Henrik Gramner git.videolan.org Git - ffmpeg.git/rss log

vp9: Add 8bpc AVX2 asm for inverse transforms [DH] libavcodec/x86/Makefile [DH] libavcodec/x86/vp9dsp_init.c [DH] libavcodec/x86/vp9itxfm.asm [DH] libavcodec/x86/vp9itxfm_avx2.asm
12:11
vp9: Remove 8bpc AVX asm for inverse transforms - Henrik Gramner git.videolan.org Git - ffmpeg.git/rss log

vp9: Remove 8bpc AVX asm for inverse transforms There's very little performance difference vs SSE2/SSSE3 and most systems will use the AVX2 implementations anyway. This reduces code size and compilation time by a significant amount. [DH] libavcodec/x86/vp9dsp_init.c [DH] libavcodec/x86/vp9itxfm.asm
21:10
avcodec/tiffenc: support writing Orientation tag - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/tiffenc: support writing Orientation tag Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/tiff.h [DH] libavcodec/tiffenc.c
03:21
tests/fate/image: add Exif rotation metadata tests - James Almer git.videolan.org Git - ffmpeg.git/rss log

tests/fate/image: add Exif rotation metadata tests Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] tests/fate/image.mak [DH] tests/ref/fate/jpg-exif-autorotate [DH] tests/ref/fate/jpg-exif-noautorotate [DH] tests/ref/fate/jpg-exif-rotation-override
17:14
avfilter/vf_colordetect: Avoid sequentially consistent atomics - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_colordetect: Avoid sequentially consistent atomics Reviewed-by: Martin Storsjö <martinⓐmartin.st> Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavfilter/vf_colordetect.c
16:45
avfilter/vf_colordetect: Rename header to vf_colordetectdsp.h - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_colordetect: Rename header to vf_colordetectdsp.h It is more in line with our naming conventions. Reviewed-by: Martin Storsjö <martinⓐmartin.st> Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavfilter/aarch64/vf_colordetect_init.c [DH] libavfilter/vf_colordetect.c [DH] libavfilter/vf_colordetect.h [DH] libavfilter/vf_colordetectdsp.h [DH] libavfilter/x86/vf_colordetect_init.c [DH] (...)
17:33
avfilter/vf_colordetect: Move ff_color_detect_dsp_init() out of .c file - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_colordetect: Move ff_color_detect_dsp_init() out of .c file Otherwise checkasm pulls in vf_colordetect.o and then all of libavfilter. Reviewed-by: Martin Storsjö <martinⓐmartin.st> Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavfilter/vf_colordetect.c [DH] libavfilter/vf_colordetectdsp.h
22:13
avfilter/vsrc_gfxcapture: fix possible missed wakeup race in capture loop - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: fix possible missed wakeup race in capture loop [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
13:39
avformat/flvenc: fix event_flags check - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avformat/flvenc: fix event_flags check AVFormatContext.event_flags is checked against AVSTREAM_EVENT_FLAG, which belongs to AVStream.event_flags. There is no real issue since these two flags have the same value. [DH] libavformat/flvenc.c
13:15
configure: also initialize objcflags_filter early - Marvin Scholz git.videolan.org Git - ffmpeg.git/rss log

configure: also initialize objcflags_filter early Fixes an issue uncovered since 0ce413af9cc8040095a6b714600d6a8dceaca514 where flags were properly separated but without the objcflags_filter being initialized, if add_objcflags is called early as part of add_allcflags, it would lead to the append() function using an empty filter and trying to call the flag instead, leading to errors like: ./configure: line 976: -fsanitize=address: command not found ./configure: line 976: -fno-omit-frame-pointer: command not found [DH] (...)
12:17
configure: make --extra-objcflags show the right default flags - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: make --extra-objcflags show the right default flags [DH] configure
11:16
configure: Make -Werror=partial-availability apply to all languages - Martin Storsjö git.videolan.org Git - ffmpeg.git/rss log

configure: Make -Werror=partial-availability apply to all languages This was missed in 0ce413af9cc8040095a6b714600d6a8dceaca514. This fixes proper detection of Objective C APIs (that are missing) if targeting older macOS versions, such as the check for AVCaptureSession. [DH] configure
12:30
configure: pass CFLAGS to ObjC tests again - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: pass CFLAGS to ObjC tests again It was mistakenly removed in 0ce413af9cc8040095a6b714600d6a8dceaca514. This matches how the Makefiles actually compile ObjC code, and unlike C++, this seems to actually be benign and various existing setups rely on that behaviour. [DH] configure

14 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

17:13
avcodec/aarch64/vvc: Unroll vvc_bdof_grad_filter_8x_neon - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Unroll vvc_bdof_grad_filter_8x_neon Before and after: A53: apply_bdof_8_16x8_neon: 2733.1 ( 4.88x) apply_bdof_8_16x16_neon: 5458.6 ( 4.86x) apply_bdof_10_16x8_neon: 2789.8 ( 4.64x) apply_bdof_10_16x16_neon: 5523.8 ( 4.68x) apply_bdof_12_16x8_neon: 2792.8 ( 4.58x) apply_bdof_12_16x16_neon: 5519.5 ( 4.63x) apply_bdof_8_16x8_neon: 2571.8 ( 5.12x) apply_bdof_8_16x16_neon: 5173.3 ( 5.12x) apply_bdof_10_16x8_neon: 2635.1 ( 4.87x) apply_bdof_10_16x16_neon: 5243.0 ( 4.89x) apply_bdof_12_16x8_neon: 2613.0 ( 4.89x) apply_bdof_12_16x16_neon: 5231.7 ( 4.90x) A78: (...)
10:14
tests/fate/demux: add hxvs test case - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

tests/fate/demux: add hxvs test case [DH] tests/fate/demux.mak [DH] tests/ref/fate/hxvs-demux
19:38
avfilter/vsrc_gfxcapture: switch to std::thread - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: switch to std::thread Combining C++ with our w32pthreads does not work out. The C++ standard lib headers might include pthread.h, which will then define clashing types with our w32pthread header. To solve this, switch to using C++ native std::thread. [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
18:46
avfilter/vsrc_gfxcapture: MapWindowPoints can return 0 and still succeed - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: MapWindowPoints can return 0 and still succeed [DH] libavfilter/vsrc_gfxcapture_winrt.cpp
16:47
avfilter/vsrc_gfxcapture: make checkheaders happy - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_gfxcapture: make checkheaders happy [DH] libavfilter/vsrc_gfxcapture.h [DH] libavfilter/vsrc_gfxcapture_winrt.cpp [DH] libavfilter/vsrc_gfxcapture_winrt.h [DH] libavfilter/vsrc_gfxcapture_winrt.hpp
13:33
doc/filters: fix missed reference to outdated option name - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

doc/filters: fix missed reference to outdated option name [DH] doc/filters.texi

13 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

17:47
avformat: add hxvs demuxer - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avformat: add hxvs demuxer [DH] Changelog [DH] configure [DH] libavformat/Makefile [DH] libavformat/allformats.c [DH] libavformat/hxvs.c [DH] libavformat/version.h
12:00
avcodec/exif: Use av_fast_mallocz() in av_exif_clone_ifd() - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

avcodec/exif: Use av_fast_mallocz() in av_exif_clone_ifd() using fast realloc leaves the entries uninitialized and frees garbage pointers on errors Fixes: bug_triggering_file Found-by: *2ourc3, 5pider Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavcodec/exif.c
11:53
avcodec/exif: Do not leave uninitialized pointers on errors in exif_clone_entry() - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

avcodec/exif: Do not leave uninitialized pointers on errors in exif_clone_entry() No testcase, but this looks like it could free garbage pointers Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavcodec/exif.c
03:00
avcodec/vvc/data: Mark tables as hidden - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/vvc/data: Mark tables as hidden This allows compilers to optimize accesses like ff_vvc_diag_scan_x[2][2][x] by baking the offset derived from [2][2] into the relocation (so that it is performed at link-time). Reviewed-by: Wu Jianhua <toqsxwⓐoutlook.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/vvc/data.h
04:26
avcodec/rkmppdec: fix input queue overflow - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/rkmppdec: fix input queue overflow The strategy to count free slots isn't reliable. The value of INPUT_MAX_PACKETS is a hardcoded value, and MPP_DEC_GET_STREAM_COUNT doesn't always work as expected. When freeslots is nonzero, the mpp decoder still returns MPP_ERR_BUFFER_FULL. Before this patch, all packets are dropped once went into MPP_ERR_BUFFER_FULL state. Don't drop packet and return error in MPP_ERR_BUFFER_FULL state. Receive frame to allow the decoder's state machine to resume. Signed-off-by: Zhao Zhili <zhilizhaoⓐtencent.com> [DH] (...)
05:31
avcodec/rkmppdec: fix return EAGAIN during flushing - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/rkmppdec: fix return EAGAIN during flushing Signed-off-by: Zhao Zhili <zhilizhaoⓐtencent.com> [DH] libavcodec/rkmppdec.c
09:59
avcodec/rkmppdec: Close decoder generically on init failure - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/rkmppdec: Close decoder generically on init failure Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/rkmppdec.c
19:27
tests/checkasm/sw_ops: Avoid 1 << 32 - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

tests/checkasm/sw_ops: Avoid 1 << 32 It is UB. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] tests/checkasm/sw_ops.c
14:12
swscale/ops_tmpl_int: remove unused arguments from wrap read decl - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

swscale/ops_tmpl_int: remove unused arguments from wrap read decl Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libswscale/ops_backend.h [DH] libswscale/ops_tmpl_int.c
14:10
swscale/ops_chain: add type removed ff_sws_op_chain_free_cb - Kacper Michajłow git.videolan.org Git - ffmpeg.git/rss log

swscale/ops_chain: add type removed ff_sws_op_chain_free_cb to avoid pointer casting and UB of calling function with different pointer type. Signed-off-by: Kacper Michajłow <kasper93ⓐgmail.com> [DH] libswscale/ops_backend.c [DH] libswscale/ops_chain.c [DH] libswscale/ops_chain.h [DH] libswscale/x86/ops.c

12 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

07:37
swresample/swresample: Delete unused code - renjianguang git.videolan.org Git - ffmpeg.git/rss log

swresample/swresample: Delete unused code The uch option has been removed Signed-off-by: renjianguang <renjianguangⓐxiaomi.com> [DH] libswresample/swresample.c
00:38
avcodec/mjpegdec: Move initializing HpelDSPContext to mxpegdec.c - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Move initializing HpelDSPContext to mxpegdec.c Only the mxpeg decoder uses it (and the reference/bitmask feature of ff_mjpeg_decode_sos()). So only initialize it for mxpeg which allows to remove the mjpeg->hpeldsp dependency. Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] configure [DH] libavcodec/mjpegdec.c [DH] libavcodec/mxpegdec.c
01:47
avcodec/mjpegdec: Avoid using HpelDSPContext - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Avoid using HpelDSPContext It is quite big and we only need one function from it. Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mjpegdec.c [DH] libavcodec/mjpegdec.h [DH] libavcodec/mxpegdec.c
02:03
avcodec/mjpegdec: Move reference dimension check to mxpegdec.c - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Move reference dimension check to mxpegdec.c Only the mxpeg sets reference at all. Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mjpegdec.c [DH] libavcodec/mxpegdec.c
15:53
swresample/swresample_internal: Use union for float, int matrix - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swresample/swresample_internal: Use union for float, int matrix Saves 16KiB from SwrContext. (FATE would also pass if one made the double matrix part of the union, too, but I don't know whether this is truely correct, because swri_rematrix() accesses the double matrix (to check whether it is equal to 1.0) even when one of the other matrices is in use.) Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswresample/rematrix.c [DH] (...)
16:19
swresample/rematrix: Avoid allocation for native_one - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swresample/rematrix: Avoid allocation for native_one Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswresample/rematrix.c [DH] libswresample/swresample.c [DH] libswresample/swresample_internal.h [DH] libswresample/x86/rematrix_init.c
16:23
swresample/x86/rematrix_init: Avoid allocation for native_simd_one - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swresample/x86/rematrix_init: Avoid allocation for native_simd_one Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswresample/rematrix.c [DH] libswresample/swresample.c [DH] libswresample/swresample_internal.h [DH] libswresample/x86/rematrix_init.c
16:33
swresample/rematrix_template: Constify get_mix_any_func - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swresample/rematrix_template: Constify get_mix_any_func Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswresample/rematrix_template.c
20:22
swresample/resample, soxr_resample: Use designated initializers - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swresample/resample, soxr_resample: Use designated initializers Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswresample/resample.c [DH] libswresample/soxr_resample.c
17:19
forgejo: make aarch64 asm linter part of pre-commit - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

forgejo: make aarch64 asm linter part of pre-commit Needed some minor semantical changes to the shell wrapper, since pre-commit expects it to exit with an error when applying fixes. [DH] .forgejo/pre-commit/config.yaml [DH] .forgejo/workflows/lint.yml [DH] tools/check_arm_indent.sh
17:34
avcodec/bsf/noise: Avoid allocation for string - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/bsf/noise: Avoid allocation for string Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/bsf/noise.c
18:47
swscale/ops_chain: Free correct pointer on error - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swscale/ops_chain: Free correct pointer on error Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswscale/ops_chain.c
18:55
swscale/x86/ops: Fix leak - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swscale/x86/ops: Fix leak Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswscale/x86/ops.c
02:40
avcodec/Makefile: Add adpcm_vima->adpcm.o dependency - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/Makefile: Add adpcm_vima->adpcm.o dependency Forgotten in 9141fe9653dc078d81bff8308ffdd2aaaf35495c. Reviewed-by: Peter Ross <prossⓐxvid.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/Makefile

11 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

20:04
avcodec/mjpegdec: Remove pointless information from logmessage - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Remove pointless information from logmessage There is no reason to add the address of an element of MJpegDecodeContext (which makes logmessage nonreproducible). Also avoid always printing a zero. Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mjpegdec.c
20:33
avcodec/mjpegdec: Avoid superfluous secondary error message - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Avoid superfluous secondary error message Up until now, a DC error in decode_block() or decode_dc_progressive() would lead to a warning from mjpeg_decode_dc() and a (less verbose) error from the caller. Upgrade the former to an error status (all callers treat is an error) and remove the latter. Reviewed-by: Michael Niedermayer <michaelⓐniedermayer.cc> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] (...)
22:13
avcodec/mjpegdec: Avoid mixing return value and error code, avoid branch - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mjpegdec: Avoid mixing return value and error code, avoid branch mjpeg_decode_dc() currently has a special return value (0xfffff) for the case of invalid data; this value does not overlap with the ordinary return values, yet the compiler can't prove this (at least on x86 where an asm version of NEG_USR32 is used inside get_xbits()), so the non-error return value has to be checked for being the special value even if mjpeg_decode_dc() is inlined. This commit avoids this by separating error code and ordinary return value. It also means that the ljpeg functions return the proper (...)

10 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

23:39
avfilter/vf_libplacebo: use temporary params struct for per-pass params - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_libplacebo: use temporary params struct for per-pass params Instead of directly mutating `opts->params`. Avoids any possible leak of overriden params between invocations of this function, as well as the later `pl_render_image` during the linear output pass. [DH] libavfilter/vf_libplacebo.c
23:48
avfilter/vf_libplacebo: force premultiplied blending for linear texture - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_libplacebo: force premultiplied blending for linear texture Blending onto independent alpha framebuffers is not possible under the constraints of the supported blend operators. While we could handle blending premul-onto-premul, this would break if the base layer is YUV, since premultiplied alpha does not survive the (nonlinear) YUV conversion. Fortunately, blending independent-onto-premul is just as easy, and works in all cases. So just force this mode when using a linear intermediate blend texture, which is always RGBA. [DH] (...)
09:26
forgejo: Check the aarch64 assembly indentation as part of the lint job - Martin Storsjö git.videolan.org Git - ffmpeg.git/rss log

forgejo: Check the aarch64 assembly indentation as part of the lint job Alternatively, this could be a separate job, potentially keyed to only run on PRs that touch files matching */aarch64/*. But as this runs very quickly, it's probably less clutter to just bundle it here. [DH] .forgejo/workflows/lint.yml
12:29
avcodec/videotoolboxenc: ensure bitrate is set in low_delay mode - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/videotoolboxenc: ensure bitrate is set in low_delay mode VideoToolbox doesn't support automatic bitrate in low delay mode. Check bitrate and show error message so user knows what's going wrong. [DH] libavcodec/videotoolboxenc.c
01:35
forgejo/CODEOWNERS: add myself to d3d12va related files - Tong Wu git.videolan.org Git - ffmpeg.git/rss log

forgejo/CODEOWNERS: add myself to d3d12va related files Signed-off-by: Tong Wu <wutong1208ⓐoutlook.com> [DH] .forgejo/CODEOWNERS

9 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

18:08
avfilter/vsrc_ddagrab: support rendering mouse cursor onto array textures - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter/vsrc_ddagrab: support rendering mouse cursor onto array textures [DH] libavfilter/vsrc_ddagrab.c
19:46
avutil/hwcontext_d3d11va: remove D3D11_BIND_RENDER_TARGET restriction for array textures - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avutil/hwcontext_d3d11va: remove D3D11_BIND_RENDER_TARGET restriction for array textures This was added in 4f78711f9c28b11dae4e4b96be46b6b4925eb8c6, with the commit message claiming that it's a Microsoft restriction that array textures with ArraySize > 2 cannot be created with D3D11_BIND_RENDER_TARGET. I was unable to find any documentation or other references on that, and a quick test also found it to not be the case. So this patch removes that restriction. This enables frame sources, like the d3d11 capture filters, to output frames in an array texture, which is neccesary to pass (...)
14:31
avutil/aes_ctr: reintroduce the block offset state - James Almer git.videolan.org Git - ffmpeg.git/rss log

avutil/aes_ctr: reintroduce the block offset state Wrongly removed in fe73b84879a560d69affca88ce21e61108e7c38d, it's required for calls with a payload smaller than a full block. Fixes issue #20474. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavutil/aes_ctr.c
14:32
avutil/tests/aes_ctr: extend the test to cover payloads smaller than a block - James Almer git.videolan.org Git - ffmpeg.git/rss log

avutil/tests/aes_ctr: extend the test to cover payloads smaller than a block Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavutil/tests/aes_ctr.c
21:10
avcodec/aarch64/vvc: Optimised version of classify function. - Georgii Zagoruiko git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Optimised version of classify function. Macbook Air (M2): vvc_alf_classify_8x8_8_c: 2.6 ( 1.00x) vvc_alf_classify_8x8_8_neon: 1.0 ( 2.47x) vvc_alf_classify_8x8_10_c: 2.7 ( 1.00x) vvc_alf_classify_8x8_10_neon: 0.9 ( 2.98x) vvc_alf_classify_8x8_12_c: 2.7 ( 1.00x) vvc_alf_classify_8x8_12_neon: 0.9 ( 2.97x) vvc_alf_classify_16x16_8_c: 7.3 ( 1.00x) vvc_alf_classify_16x16_8_neon: 3.4 ( 2.12x) vvc_alf_classify_16x16_10_c: 4.3 ( 1.00x) vvc_alf_classify_16x16_10_neon: 2.9 ( 1.47x) vvc_alf_classify_16x16_12_c: 4.3 ( 1.00x) vvc_alf_classify_16x16_12_neon: 3.0 ( 1.44x) (...)

8 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

18:56
avcodec/aarch64/vvc: Implement dmvr_v_8 - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Implement dmvr_v_8 A72 dmvr_v_8_12x20_neon: 207.0 ( 4.15x) dmvr_v_8_20x12_neon: 170.4 ( 4.37x) dmvr_v_8_20x20_neon: 273.4 ( 4.58x) A53 dmvr_v_8_12x20_neon: 450.6 ( 4.21x) dmvr_v_8_20x12_neon: 342.8 ( 3.70x) dmvr_v_8_20x20_neon: 550.9 ( 3.79x) [DH] libavcodec/aarch64/vvc/dsp_init.c [DH] libavcodec/aarch64/vvc/inter.S
08:08
avformat/tls_gnutls: add initial dtls support - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/tls_gnutls: add initial dtls support Set GNUTLS_DATAGRAM flag when is_dtls is true. Set mtu when it's specified. Modify the read/write function could use udp socket. There are more patches to make dtls really work. Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/tls_gnutls.c
08:31
avformat/tls_gnutls: add av_assert0() for tls_shared - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/tls_gnutls: add av_assert0() for tls_shared Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/tls_gnutls.c
02:19
avformat/whip: enable RTCP NACK for whip - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: enable RTCP NACK for whip Refer to RFC 4585 4.2, Add SDP "a=rtcp-fb" so the peer can request rtx through sending NACKs. Add basic parsing for NACK packet. Decrypt into a newly allocated buffer (do not overwrite whip->buf) so multiple NACKs in a bundled packet can be parsed in a loop while keeping whip->buf available for sending rtx Broken or wrong NACk are logged and skipped. Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] (...)
23:24
fftools: use fopen_utf8 to open FFREPORT - Marvin Scholz git.videolan.org Git - ffmpeg.git/rss log

fftools: use fopen_utf8 to open FFREPORT Should fix #11240 [DH] fftools/opt_common.c
12:17
fate/prores: Fix missing dependency - averne git.videolan.org Git - ffmpeg.git/rss log

fate/prores: Fix missing dependency Signed-off-by: averne <averne381ⓐgmail.com> [DH] tests/fate/prores.mak
18:07
avfilter/buffersink: add missing checks for alphamodes - James Almer git.videolan.org Git - ffmpeg.git/rss log

avfilter/buffersink: add missing checks for alphamodes Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavfilter/buffersink.c
18:08
fftools/ffplay: set the alphamodes buffersink array option instead of the binary one - James Almer git.videolan.org Git - ffmpeg.git/rss log

fftools/ffplay: set the alphamodes buffersink array option instead of the binary one And add the missing goto fail. This should ensure the alpha mode is set and remove bogus warnings printed by ffplay. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] fftools/ffplay.c
13:42
avfilter/vf_libplacebo: ensure linear tex is blittable - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/vf_libplacebo: ensure linear tex is blittable Required to clear textures to a background color. [DH] libavfilter/vf_libplacebo.c
12:24
configure: fix CUDA compilation with SDK version 13 nvcc - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: fix CUDA compilation with SDK version 13 nvcc Compute Capability 7.5 is now the lowest supported version. Fixes #20454 [DH] configure

7 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

20:03
compat/w32pthreads: fix compatibility with C++ - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

compat/w32pthreads: fix compatibility with C++ [DH] compat/w32pthreads.h

6 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

04:42
avcodec/videotoolboxenc: allow low latency RC with HEVC - Cameron Gutman git.videolan.org Git - ffmpeg.git/rss log

avcodec/videotoolboxenc: allow low latency RC with HEVC It is supported only for H.264 on Intel Macs, but it can be used with both H.264 and HEVC on Apple Silicon. Signed-off-by: Cameron Gutman <aicommanderⓐgmail.com> [DH] libavcodec/videotoolboxenc.c
02:18
configure: remove openssl version check for whip - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

configure: remove openssl version check for whip This version check for whip is unnecessary. Since several rencet patches to the configure have already added version checks for OpenSSL. Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] configure

5 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

17:52
avcodec/aarch64/vvc: Optimize dmvr_hv_10 - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Optimize dmvr_hv_10 Before and after on A53: dmvr_hv_10_12x20_neon: 1838.2 ( 3.02x) dmvr_hv_10_20x12_neon: 1330.2 ( 1.83x) dmvr_hv_10_20x20_neon: 2148.2 ( 1.85x) dmvr_hv_12_12x20_neon: 1839.2 ( 3.02x) dmvr_hv_12_20x12_neon: 1330.6 ( 1.83x) dmvr_hv_12_20x20_neon: 2147.2 ( 1.85x) dmvr_hv_10_12x20_neon: 1755.0 ( 3.17x) dmvr_hv_10_20x12_neon: 1165.8 ( 2.09x) dmvr_hv_10_20x20_neon: 1876.1 ( 2.12x) dmvr_hv_12_12x20_neon: 1754.4 ( 3.17x) dmvr_hv_12_20x12_neon: 1167.8 ( 2.09x) dmvr_hv_12_20x20_neon: 1878.8 ( 2.12x) [DH] (...)
10:55
avcodec/dvdsubdec: Don't return value != 0 on init success - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/dvdsubdec: Don't return value != 0 on init success Currently, any nonnegative return value from an init function is just treated as success and otherwise ignored; while it is not explicitly forbidden to return something else than 0 (there is no documentation whatsoever), the assumption is that they return zero on success. So change dvdsubdec.c accordingly. Also change dvdsub_parse_extradata() to return zero on success. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] (...)
17:17
avcodec/aarch64/vvc: Optimize vvc_apply_bdof_block_8x - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Optimize vvc_apply_bdof_block_8x Before and after: A53: apply_bdof_8_8x16_neon: 3320.5 ( 4.02x) apply_bdof_10_8x16_neon: 3317.8 ( 3.90x) apply_bdof_12_8x16_neon: 3303.6 ( 3.91x) apply_bdof_8_8x16_neon: 3168.1 ( 4.23x) apply_bdof_10_8x16_neon: 3127.8 ( 4.13x) apply_bdof_12_8x16_neon: 3119.3 ( 4.18x) A72: apply_bdof_8_8x16_neon: 1827.4 ( 5.02x) apply_bdof_10_8x16_neon: 1838.5 ( 4.89x) apply_bdof_12_8x16_neon: 1841.1 ( 4.83x) apply_bdof_8_8x16_neon: 1691.6 ( 5.46x) apply_bdof_10_8x16_neon: 1695.9 ( 5.23x) apply_bdof_12_8x16_neon: 1695.4 ( 5.29x) A78 (...)
20:24
avcodec/aarch64/vvc: Implement dmvr_h_8 - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Implement dmvr_h_8 A78: dmvr_h_8_12x20_neon: 76.6 ( 4.31x) dmvr_h_8_20x12_neon: 65.8 ( 3.49x) dmvr_h_8_20x20_neon: 106.6 ( 3.62x) A72: dmvr_h_8_12x20_neon: 190.6 ( 4.40x) dmvr_h_8_20x12_neon: 171.1 ( 4.31x) dmvr_h_8_20x20_neon: 275.1 ( 4.50x) [DH] libavcodec/aarch64/vvc/dsp_init.c [DH] libavcodec/aarch64/vvc/inter.S
22:49
avcodec/aarch64/vvc: Implemented dmvr_h_10 - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

avcodec/aarch64/vvc: Implemented dmvr_h_10 A78: dmvr_h_10_12x20_neon: 82.2 ( 6.49x) dmvr_h_10_20x12_neon: 69.9 ( 3.66x) dmvr_h_10_20x20_neon: 112.5 ( 3.74x) dmvr_h_12_12x20_neon: 81.4 ( 6.51x) dmvr_h_12_20x12_neon: 69.2 ( 3.74x) dmvr_h_12_20x20_neon: 110.2 ( 3.85x) A72: dmvr_h_10_12x20_neon: 234.1 ( 4.67x) dmvr_h_10_20x12_neon: 221.4 ( 3.48x) dmvr_h_10_20x20_neon: 356.9 ( 3.59x) dmvr_h_12_12x20_neon: 234.1 ( 4.67x) dmvr_h_12_20x12_neon: 221.5 ( 3.53x) dmvr_h_12_20x20_neon: 357.0 ( 3.64x) [DH] libavcodec/aarch64/vvc/dsp_init.c [DH] (...)
15:27
vc1_parser: Reset unesc_index after extracting header with complete frames - DE-AI git.videolan.org Git - ffmpeg.git/rss log

vc1_parser: Reset unesc_index after extracting header with complete frames When PARSER_FLAG_COMPLETE_FRAMES is set (e.g. in Matroska), unesc_index is not reset after the first frame, which causes subsequent frames to inherit leftover header data. This leads to incorrect keyframe marking and broken decoding. See also this bug report: https://trac.ffmpeg.org/ticket/8515 Signed-off-by: DE-AI <max.christophⓐpentaxmedical.com> [DH] libavcodec/vc1_parser.c
15:17
swscale: Disable avx2 hscale 8to15 on IceLake and below due to Intel Gather Data... - Alan Kelly git.videolan.org Git - ffmpeg.git/rss log

swscale: Disable avx2 hscale 8to15 on IceLake and below due to Intel Gather Data Sampling mitigation performance loss Intel provided a microcode update to mitigate this security vulnerability which has a huge negative performance impact on gather instructions. This means that hscale 8to15 avx2, which uses gather extensively, is no longer faster than SSSE3 on impacted CPUs. https://www.intel.com/content/www/us/en/developer/articles/technical/software-security-guidance/technical-documentation/gather-data-sampling.html Broadwell: hscale_8_to_15__fs_4_dstW_512_c: 3379.5 ( 1.00x) (...)
15:44
avcodec/decode: treat orientation 1 as valid displaymatrix - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

avcodec/decode: treat orientation 1 as valid displaymatrix Since 9dc79241d9680fc050376bb0cc37875a41d00cc9 we now always pop the orientation off of the IFD and use a display matrix instead. This means we should not produce a warning and refuse if the orientation field indicates a default orientation (i.e. 1). Signed-off-by: Leo Izen <leo.izenⓐgmail.com> Reported-by: Ramiro Polla <ramiro.pollaⓐgmail.com> [DH] libavcodec/decode.c [DH] tests/ref/fate/exif-image-jpg [DH] tests/ref/fate/exif-image-tiff [DH] (...)
08:55
configure: Improve the check for schannel DTLS constants - Martin Storsjö git.videolan.org Git - ffmpeg.git/rss log

configure: Improve the check for schannel DTLS constants This amends 307983b292adb60c82e75c8e5e41deedc89efbf0 to fix building with older versions of mingw-w64. The previously checked constant, SP_PROT_DTLS1_X_CLIENT, was added in mingw-w64 in df36f5deda23192d0ee99ffd661ea36df924e667 in 2020, and is included in released versions since v8.0.0. The new checked constant SECPKG_ATTR_DTLS_MTU was added in mingw-w64 in 0792283787cca8fc27dd38671107c791c87f4db3 in 2021, and first appeared in mingw-w64 v9.0.0. This fixes building with mingw-w64 v8, which is the version bundled in Ubuntu 22.04. [DH] (...)

4 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

17:01
vf_colorspace: Add an option to clamp trc LUT output - Drew Dunne git.videolan.org Git - ffmpeg.git/rss log

vf_colorspace: Add an option to clamp trc LUT output Add a new flag to the vf_colorspace filter which provides the user an option to clamp the linear and delinear transfer characteristics LUT values to the [0, 1] represented range. This helps constrain the potential value range when converting between colorspaces. Certain colors when going through the conversion can result in out of gamut colors after the rotation. The colorspace filter allows that with the extended range. The added clamping just keeps the colors within the [0, 1) range rather than using that extended range. I'm not (...)
18:51
configure: fix naming of some C++/cxx functions being mislabeled as cpp - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: fix naming of some C++/cxx functions being mislabeled as cpp [DH] configure
23:09
avcodec: Add av_cold to flush,init,close functions missing it - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec: Add av_cold to flush,init,close functions missing it Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/acelp_filters.c [DH] libavcodec/acelp_vectors.c [DH] libavcodec/adpcm.c [DH] libavcodec/adxdec.c [DH] libavcodec/agm.c [DH] libavcodec/apedec.c [DH] libavcodec/apv_decode.c [DH] libavcodec/arbc.c [DH] libavcodec/argo.c [DH] libavcodec/ass.c [DH] libavcodec/ass_split.c [DH] libavcodec/atrac9dec.c [DH] libavcodec/audio_frame_queue.c [DH] libavcodec/av1_parser.c [DH] libavcodec/av1dec.c [DH] libavcodec/avdct.c [DH] libavcodec/bink.c [DH] (...)
12:56
avcodec/pcm_tablegen: Fix hardcoded-tables if alaw,mulaw,vidc codecs disabled - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/pcm_tablegen: Fix hardcoded-tables if alaw,mulaw,vidc codecs disabled Since ae448e00afb43d7f72dfa9c82a4c45994e4fea6a the various tableinit functions are not compiled unconditionally any more, so that pcm_tablegen.c (which creates the hardcoded tables) needs to be updated. Reviewed-by: James Almer <jamrialⓐgmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/pcm_tablegen.c
10:29
avcodec/mpegaudiodec_template: Avoid write-only stores - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mpegaudiodec_template: Avoid write-only stores Reviewed-by: Peter Ross <prossⓐxvid.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mpegaudiodec_template.c
10:25
avcodec/mpegaudiodec_template: Mark flush functions as av_cold - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mpegaudiodec_template: Mark flush functions as av_cold Reviewed-by: Peter Ross <prossⓐxvid.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mpegaudiodec_template.c
10:07
avcodec/mpegaudiodec_template: Allocate sub-contexts jointly - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mpegaudiodec_template: Allocate sub-contexts jointly Reviewed-by: Peter Ross <prossⓐxvid.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mpegaudiodec_template.c
10:00
avcodec/mpegaudiodec_template: Don't modify AVCodecContext.priv_data - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/mpegaudiodec_template: Don't modify AVCodecContext.priv_data Reviewed-by: Peter Ross <prossⓐxvid.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/mpegaudiodec_template.c
17:19
avfilter/buffersink: remove unneeded parameter - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/buffersink: remove unneeded parameter This is a newly added field, so there's no point to try and keep backwards compatibility with an older API - newer clients should just use the new fields. [DH] libavfilter/buffersink.c
19:39
avfilter/Makefile: Add dependencies for premultiply_dynamic filter - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avfilter/Makefile: Add dependencies for premultiply_dynamic filter Reviewed-by: Timo Rothenpieler <timoⓐrothenpieler.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavfilter/Makefile
09:36
avformat/mov: fix unused variable entry in parse_exif_item - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avformat/mov: fix unused variable entry in parse_exif_item [DH] libavformat/mov.c
09:30
avcodec/pcm: fix some unused variables/functions warning - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/pcm: fix some unused variables/functions warning v is unused when ALAW/MULAW/VIDC are disabled. [DH] libavcodec/pcm.c
09:31
libavcodec/v4l2_buffers: avoid redefinition of USEC_PER_SEC macro - caifan3 git.videolan.org Git - ffmpeg.git/rss log

libavcodec/v4l2_buffers: avoid redefinition of USEC_PER_SEC macro The macro USEC_PER_SEC is already defined in system headers on some platforms,such as Nuttx nuttx/clock.h. Check if it's already defined before defining it to avoid compiler warnings. Signed-off-by: caifan3 <caifan3ⓐxiaomi.com> [DH] libavcodec/v4l2_buffers.c
20:56
Replace uxtl with umull in dmvr_hv_8 - Krzysztof Pyrkosz git.videolan.org Git - ffmpeg.git/rss log

Replace uxtl with umull in dmvr_hv_8 Before and after on A78: dmvr_hv_8_12x20_neon: 205.3 ( 5.21x) dmvr_hv_8_20x12_neon: 171.8 ( 3.15x) dmvr_hv_8_20x20_neon: 282.7 ( 3.11x) dmvr_hv_8_12x20_neon: 172.7 ( 5.58x) dmvr_hv_8_20x12_neon: 133.3 ( 3.36x) dmvr_hv_8_20x20_neon: 214.6 ( 3.40x) [DH] libavcodec/aarch64/vvc/inter.S
12:01
swscale/ops: Fix linking with x86 assembly disabled - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

swscale/ops: Fix linking with x86 assembly disabled Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libswscale/ops.c
12:16
all: Use "" instead of <> to include internal headers - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

all: Use "" instead of <> to include internal headers Reviewed-by: Niklas Haas <ffmpegⓐhaasn.dev> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] fftools/textformat/tf_mermaid.c [DH] libavfilter/vf_colordetect.h [DH] libavfilter/vf_deshake_opencl.c [DH] libswscale/ops_optimizer.c [DH] libswscale/x86/ops.c
01:21
avutil/hwcontext_drm: fix mapping when dst format is unset - Cameron Gutman git.videolan.org Git - ffmpeg.git/rss log

avutil/hwcontext_drm: fix mapping when dst format is unset av_hwframe_map() is documented to work with a blank dst frame, but hwcontext_drm currently fails if dst->format == AV_PIX_FMT_NONE. Signed-off-by: Cameron Gutman <aicommanderⓐgmail.com> [DH] libavutil/hwcontext_drm.c

3 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

22:51
avcodec/vaapi_encode: avoid potential overflow - Tong Wu git.videolan.org Git - ffmpeg.git/rss log

avcodec/vaapi_encode: avoid potential overflow The max_frame_size in byte will be filled in VAEncMiscParameterBufferMaxFrameSize which receives size in bit. This requires max_frame_size * 8. Signed-off-by: Tong Wu <wutong1208ⓐoutlook.com> [DH] libavcodec/vaapi_encode.h
23:49
avformat/whip: fix SDP ICE candidates parsing - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: fix SDP ICE candidates parsing fix issue #20407 Refer to RFC 5245 15.1, the foundation may be any string up to 32 chars. The old code could misread foundations as transport("udp"). This patch fully parse all these attr to avoid parsing error. Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/whip.c
02:43
avformat/whip: add PRIORITY and ICE_CONTROLLING for STUN binding request - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: add PRIORITY and ICE_CONTROLLING for STUN binding request fix issue #20402 See RFC 8445 7.1, The PRIORITY and ICE-CONTROLLED or ICE-CONTROLLING must be included in a Binding request. FFmpeg is always ICE Controlling role so we use ICE-CONTROLLING attr. Co-author-by: Sergio Garcia Murillo <sergio.garcia.murilloⓐgmail.com> Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/whip.c
11:38
avfilter/af_afade: support multiple inputs - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_afade: support multiple inputs Instead of just 2 files, generalize this filter to support crossfading arbitrarily many files. This makes the filter essentially operate similar to the `concat` filter, chaining multiple files one after another. Aside from just adding more input pads, this requires rewriting the activate function. Instead of a finite state machine, we keep track of the currently active input index; and advance it only once the current input is fully exhausted. This results in arguably simpler logic overall. [DH] doc/filters.texi [DH] (...)
00:53
avcodec/cbrt_tablegen: Deduplicate common code - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/cbrt_tablegen: Deduplicate common code Namely the part that creates a temporary LUT. Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/Makefile [DH] libavcodec/cbrt_data.h [DH] libavcodec/cbrt_tablegen.h [DH] libavcodec/cbrt_tablegen_common.c [DH] libavcodec/cbrt_tablegen_template.c
17:03
avutil/version: bump minor after recent change - Wu Jianhua git.videolan.org Git - ffmpeg.git/rss log

avutil/version: bump minor after recent change See c2ce387385996c4e6824116931930b08cfcaecc9 Signed-off-by: Wu Jianhua <toqsxwⓐoutlook.com> [DH] libavutil/version.h
11:03
swscale: Don't pass a concrete SwsOpPriv as parameter - Martin Storsjö git.videolan.org Git - ffmpeg.git/rss log

swscale: Don't pass a concrete SwsOpPriv as parameter This fixes the following compiler error, if compiling with MSVC for ARM (32 bit): src/libswscale/ops_chain.c(48): error C2719: 'priv': formal parameter with requested alignment of 16 won't be aligned This change shouldn't affect the performance of this operation (which in itself probably isn't relevant); instead of copying the contents of the SwsOpPriv struct from the stack as parameter, it gets copied straight from the caller function's stack frame instead. Separately from this issue, MSVC 17.8 and 17.9 end (...)
18:48
avformat/demux: Reindent after the previous commit - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/demux: Reindent after the previous commit Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/demux.c
18:46
avformat/demux: Avoid always-true branch - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/demux: Avoid always-true branch Since 9d037c54f209958d47ac376d2a9561608f98dfae id3v2_extra_meta can only be != NULL if the input format wants ID3v2 tags to be read generically. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavformat/demux.c
18:38
avformat/avformat: Make AVFMT_FLAG_ID3V2_AUTO private - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/avformat: Make AVFMT_FLAG_ID3V2_AUTO private This flag governs whether avformat_open_input() reads ID3v2 tags generically; some demuxers without this flag read these tags themselves in a non-generic way, e.g. oma. This makes this flag an implementation detail that should not be exposed to the user, i.e. an internal flag. Given that 9d037c54f209958d47ac376d2a9561608f98dfae did not bump version and added no APIchanges entry I deemded it inappropriate to bump version or add an APIchanges entry for the removal of AVFMT_FLAG_ID3V2_AUTO. Signed-off-by: Andreas Rheinhardt (...)
13:51
avformat/aviobuf: Don't pretend to support avio_context_free(NULL) - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avformat/aviobuf: Don't pretend to support avio_context_free(NULL) It makes no sense to ever call it that way given that avio_context_free() accepts a pointer to a pointer to an AVIOContext. Other double-pointer-free functions like av_freep() also do it that way (and therefore avio_context_free(NULL) still crashes even with 870cfed2317e311a71bc14773332486a162f059c). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] (...)

2 septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

07:03
fftools/opt_common: add separator line to show_filters output - zhanghongyuan git.videolan.org Git - ffmpeg.git/rss log

fftools/opt_common: add separator line to show_filters output Add a separator line to show_filters() function output to maintain consistent formatting with other show functions like show_codecs, show_formats_devices, etc. This provides uniform formatting across all command-line output functions for better readability and parsing by external tools. [DH] fftools/opt_common.c
23:34
build: link with CXX when -lstdc++ on linker commandline - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

build: link with CXX when -lstdc++ on linker commandline [DH] Makefile [DH] ffbuild/common.mak [DH] ffbuild/library.mak [DH] tests/api/Makefile [DH] tests/checkasm/Makefile
10:04
doc/examples/vaapi_encode: fix invalid check on fwrite - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

doc/examples/vaapi_encode: fix invalid check on fwrite enc_pkt->size is 0 after av_packet_unref, which makes the check invalid. Fix regression from 3e4bfff2. Co-Authored-by: Jin Bo <jinboⓐloongson.cn> Signed-off-by: Zhao Zhili <zhilizhaoⓐtencent.com> [DH] doc/examples/vaapi_encode.c
02:36
avcodec/pcm: use stricter conditional expressions for compilation - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/pcm: use stricter conditional expressions for compilation [DH] libavcodec/pcm.c [DH] libavcodec/pcm_tablegen.h
00:28
avformat/whip: make time measure more precise - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: make time measure more precise Use av_gettime_relative() replace av_gettime() to get better measure Changed the display precision from showing only integers to showing two decimal places (in milliseconds) Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/whip.c
11:02
avfilter/af_afade: don't leak memory on error - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_afade: don't leak memory on error This first frame was never cleaned up if the second input fails. [DH] libavfilter/af_afade.c
18:47
avcodec/cbrt_tablegen: Reduce size of LUT only used once - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/cbrt_tablegen: Reduce size of LUT only used once ff_cbrt_tableinit,_fixed() uses a LUT of doubles of the same size as the actual LUT to be initialized (8192 elems). Said LUT is only used to initialize another LUT, but because it is static, the dirty memory (64KiB) is not released. It is also too large to be put on the stack. This commit mitigates this: We only use a LUT for the powers of odd values, thereby halving its size. The generated LUT stays unchanged. Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas (...)
15:19
avcodec/cbrt_tablegen: Remove always-false branch - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/cbrt_tablegen: Remove always-false branch Each ff_cbrt_tableinit*() is called at most once, making this check always-false. Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Reviewed-by: Lynne <devⓐlynne.ee> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/cbrt_tablegen.h
21:06
avcodec/cbrt_tablegen: Avoid LUT only used once - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/cbrt_tablegen: Avoid LUT only used once This can be done by reusing the destination array to store a temporary LUT (with only half the amount of elements, but double the element size). This relies on certain assumptions about sizes, but they are always fulfilled for systems supported by us (sizeof(double) == 8 is needed/guaranteed since commit 3383a53e7d0abb9639c3ea3481f0eda9dca61a26). Furthermore, sizeof(uint32_t) is always >= four because CHAR_BIT is eight (in fact, sizeof(uint32_t) is four, because the exact width types don't have any padding). Reviewed-by: Zhao Zhili (...)
23:48
avcodec/pcm_tablegen: Fix CONFIG_HARDCODED_TABLES - Andreas Rheinhardt git.videolan.org Git - ffmpeg.git/rss log

avcodec/pcm_tablegen: Fix CONFIG_HARDCODED_TABLES Broken in ae448e00afb43d7f72dfa9c82a4c45994e4fea6a. Notice that config_components.h is safe to include, as it is valid for both the host and the target (in contrast to config.h). Reviewed-by: Zhao Zhili <quinkblackⓐfoxmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardtⓐoutlook.com> [DH] libavcodec/pcm_tablegen.h
12:01
.forgejo/CODEOWNERS: add myself for various files - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

.forgejo/CODEOWNERS: add myself for various files [DH] .forgejo/CODEOWNERS
11:10
checkasm: sw_ops: Avoid division by zero - Martin Storsjö git.videolan.org Git - ffmpeg.git/rss log

checkasm: sw_ops: Avoid division by zero If we're invoked with range == UINT_MAX, we end up doing "rnd() % (UINT_MAX + 1)", which is equal to "rnd() % 0". On arm (on all platforms) and on MSVC i386, this ends up crashing at runtime. This fixes the crash. [DH] tests/checkasm/sw_ops.c
04:07
swscale/ops: fix build with msvc - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

swscale/ops: fix build with msvc x86/ops.c(500): error C2099: initializer is not a constant [DH] libswscale/x86/ops.c
01:37
configure: report if unstable is enabled - James Almer git.videolan.org Git - ffmpeg.git/rss log

configure: report if unstable is enabled Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] configure
01:31
configure: fix logic for the unstable option - James Almer git.videolan.org Git - ffmpeg.git/rss log

configure: fix logic for the unstable option A --enable-* option should not have a condition where it's enabled by some other mean, as is the case here if swscale is enabled. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] configure

1er septembre

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

23:00
configure: don't use MinGW ANSI stdio when using UCRT - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: don't use MinGW ANSI stdio when using UCRT MinGWs ANSI stdio is quite slow, and when using UCRT, its extra features are not needed. The only troublesome part with ucrt printf is that is disagrees about the size of "long double", but FFmpeg does not use that anywhere, let alone prints it. [DH] configure
22:53
configure: properly split C/CXX and CPP flags - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

configure: properly split C/CXX and CPP flags Right now, CFLAGS like -std=c17 leak onto the C++ compilers commandline, leading to tons of warnings and even some errors. Undefining __STRICT_ANSI__ causes strong warnings from the stdlib headers. So only set it for C. [DH] configure [DH] ffbuild/common.mak
20:01
avfilter/af_afade: shorten crossfade on too short inputs - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_afade: shorten crossfade on too short inputs This behavior is currently completely broken, leading to an abrupt end of the first audio stream. I want to generalize this filter to multiple inputs, but having too short input files will always represent a significant problem. I considered a few approaches for how to handle this more gracefully, but most of them come with their own problems; in particular when a short input is sandwiched between two longer ones; or when there is a sequence of short inputs. In the end, it's simplest to just shorten the crossfade window. I also (...)
19:36
avfilter/af_afade: generalize pass_crossfade() signature - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_afade: generalize pass_crossfade() signature Prerequisite to an upcoming refactor. [DH] libavfilter/af_afade.c
20:29
fate/filter-audio: update acrossfade to test multiple inputs - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

fate/filter-audio: update acrossfade to test multiple inputs [DH] tests/fate/filter-audio.mak [DH] tests/ref/fate/filter-acrossfade
22:07
avcodec/librsvgdec: fix compilation with librsvg 2.50.3 - Andrey Semashev git.videolan.org Git - ffmpeg.git/rss log

avcodec/librsvgdec: fix compilation with librsvg 2.50.3 This fixes compilation with librsvg 2.50.3: error: viewport undeclared This was a regression since commit 86ed68420d3b60439d0b7767c53d0fdc1deb7277. Fixes #10722. Reviewed-by: Leo Izen <leo.izenⓐgmail.com> [DH] libavcodec/librsvgdec.c
14:39
avfilter/x86/vf_colordetect: fix alpha detect tail handling - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

avfilter/x86/vf_colordetect: fix alpha detect tail handling This wrapping logic still considered any nonzero return from the ASM function to be the overall result, but this is not true since the addition of FF_ALPHA_TRANSPARENT. Fix it by only early returning if FF_ALPHA_STRAIGHT is detected. Fixes: 9b8b78a8153f9ca94b7eb9cba87442954787912a See-Also: https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20301#issuecomment-4802 [DH] libavfilter/x86/vf_colordetect_init.c
14:00
avcodec/version_major: remove unnecessary define for an internal function - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/version_major: remove unnecessary define for an internal function It being avpriv and not public, it can be removed in the next bump regardless of when it happens. Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/exif.c [DH] libavcodec/exif_internal.h [DH] libavcodec/version_major.h
00:03
vp9: Add 8bpc intra prediction AVX2 asm - Henrik Gramner git.videolan.org Git - ffmpeg.git/rss log

vp9: Add 8bpc intra prediction AVX2 asm [DH] libavcodec/x86/vp9dsp_init.c [DH] libavcodec/x86/vp9intrapred.asm
06:55
avformat/movenc: remove mvex from final mp4 in hybrid_fragmented mode - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avformat/movenc: remove mvex from final mp4 in hybrid_fragmented mode In hybrid_fragmented mode, the first moov at start should contain mvex tag, since it's fmp4. When writing the moov at the second time, it's not fmp4 any more, so mvex should be skipped. [DH] libavformat/movenc.c [DH] tests/fate/mov.mak [DH] tests/ref/lavf/mov_hybrid_frag
01:02
forgejo/workflows: make autolabel workflow always use token from secrets - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

forgejo/workflows: make autolabel workflow always use token from secrets Using the built-in token seems to only partially work. And in the latest Forgejo release the issue forcing the use of the built-in token was fixed with: https://codeberg.org/forgejo/forgejo/pulls/9025 [DH] .forgejo/workflows/autolabel.yml

31 août

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

20:03
avformat/tls: use AV_OPT_TYPE_BOOL for some AVOptions - Marton Balint git.videolan.org Git - ffmpeg.git/rss log

avformat/tls: use AV_OPT_TYPE_BOOL for some AVOptions Signed-off-by: Marton Balint <cusⓐpasswd.hu> [DH] libavformat/tls.h
22:21
avformat/urldecode: fix decoding last char if it was percent encoded - Marton Balint git.videolan.org Git - ffmpeg.git/rss log

avformat/urldecode: fix decoding last char if it was percent encoded The length check was too strict and if the end of the string was a percent encoded sequence it did not decode correctly. Signed-off-by: Marton Balint <cusⓐpasswd.hu> [DH] libavformat/urldecode.c
17:36
avcodec/videotoolboxenc: support global_quality without qscale - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/videotoolboxenc: support global_quality without qscale [DH] libavcodec/videotoolboxenc.c
17:17
avcodec/videotoolboxenc: fix the loss of precision when calculating quality - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

avcodec/videotoolboxenc: fix the loss of precision when calculating quality [DH] libavcodec/videotoolboxenc.c
21:05
avcodec/mfenc: add AVLowLatencyMode support - Cameron Gutman git.videolan.org Git - ffmpeg.git/rss log

avcodec/mfenc: add AVLowLatencyMode support Set CODECAPI_AVLowLatencyMode when AV_CODEC_FLAG_LOW_DELAY is enabled on the AVCodecContext. AVLowLatencyMode can acheive lower latency encoding that may not be accessible using eAVScenarioInfo options alone. Signed-off-by: Cameron Gutman <aicommanderⓐgmail.com> [DH] libavcodec/mfenc.c
15:07
.forgejo/CODEOWNERS: add myself for various files - Zhao Zhili git.videolan.org Git - ffmpeg.git/rss log

.forgejo/CODEOWNERS: add myself for various files Signed-off-by: Zhao Zhili <zhilizhaoⓐtencent.com> [DH] .forgejo/CODEOWNERS
11:18
lavc/codec_desc: add lossless flags for hevc and av1 - Gyan Doshi git.videolan.org Git - ffmpeg.git/rss log

lavc/codec_desc: add lossless flags for hevc and av1 Both codecs support a lossless mode akin to H.264/AVC [DH] libavcodec/codec_desc.c

30 août

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

11:10
lavc/cbs: Make error message more descriptive - Frank Plowman git.videolan.org Git - ffmpeg.git/rss log

lavc/cbs: Make error message more descriptive [DH] libavcodec/cbs.c
20:50
add clamping for escpase coeff - damitha git.videolan.org Git - ffmpeg.git/rss log

add clamping for escpase coeff [DH] libavcodec/aaccoder.c

29 août

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

22:45
avfilter: add gfxcapture, Windows.Graphics.Capture based window/monitor capture - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

avfilter: add gfxcapture, Windows.Graphics.Capture based window/monitor capture [DH] Changelog [DH] configure [DH] doc/filters.texi [DH] libavfilter/Makefile [DH] libavfilter/allfilters.c [DH] libavfilter/version.h [DH] libavfilter/vsrc_gfxcapture.c [DH] libavfilter/vsrc_gfxcapture.h [DH] libavfilter/vsrc_gfxcapture_shader.h [DH] libavfilter/vsrc_gfxcapture_winrt.cpp [DH] libavfilter/vsrc_gfxcapture_winrt.h
20:19
libavfilter/vf_minterpolate: Use local variable in array - Zsolt Vadasz git.videolan.org Git - ffmpeg.git/rss log

libavfilter/vf_minterpolate: Use local variable in array Signed-off-by: Zsolt Vadasz <zsolt_vadaszⓐprotonmail.com> Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavfilter/vf_minterpolate.c
20:19
libavfilter/vf_minterpolate: Hoist hot loop in `set_frame_data` - Zsolt Vadasz git.videolan.org Git - ffmpeg.git/rss log

libavfilter/vf_minterpolate: Hoist hot loop in `set_frame_data` Signed-off-by: Zsolt Vadasz <zsolt_vadaszⓐprotonmail.com> Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavfilter/vf_minterpolate.c
07:51
libavcodec/dpxenc: map AVCodecContext color transfer and color primaries fields - Peter Ross git.videolan.org Git - ffmpeg.git/rss log

libavcodec/dpxenc: map AVCodecContext color transfer and color primaries fields [DH] libavcodec/dpxenc.c [DH] tests/ref/lavf/dpx [DH] tests/ref/lavf/gbrp10le.dpx [DH] tests/ref/lavf/gbrp12le.dpx [DH] tests/ref/lavf/rgb48le.dpx [DH] tests/ref/lavf/rgb48le_10.dpx [DH] tests/ref/lavf/rgba64le.dpx
07:51
avcodec/dpx: move enum DPX_TRC and DPX_COL_SPEC to dpx.h - Peter Ross git.videolan.org Git - ffmpeg.git/rss log

avcodec/dpx: move enum DPX_TRC and DPX_COL_SPEC to dpx.h [DH] libavcodec/dpx.c [DH] libavcodec/dpx.h
10:52
.forge/CODEOWNERS: add myself to EXIF code - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

.forge/CODEOWNERS: add myself to EXIF code Now that the major EXIF overhaul has been merged, I should be suggested to review patches that modify it. Signed-off-by: Leo Izen <leo.izenⓐgmail.com> [DH] .forgejo/CODEOWNERS
09:32
avfilter/af_whisper: fix srt file format - Vittorio Palmisano git.videolan.org Git - ffmpeg.git/rss log

avfilter/af_whisper: fix srt file format The SRT file format requires commas in the time string, not periods. [DH] libavfilter/af_whisper.c
17:02
avcodec/x86/vp9dsp_init: remove preprocessor checks for some prototypes - James Almer git.videolan.org Git - ffmpeg.git/rss log

avcodec/x86/vp9dsp_init: remove preprocessor checks for some prototypes Fixes compilation with --disable-avx512icl Signed-off-by: James Almer <jamrialⓐgmail.com> [DH] libavcodec/x86/vp9dsp_init.c
15:29
avcodec/version: update the codec_ids - Nicolas Gaullier git.videolan.org Git - ffmpeg.git/rss log

avcodec/version: update the codec_ids The counters should be incremented for each new codec. Catch-up to be in sync with codec_id.h again. Signed-off-by: Nicolas Gaullier <nicolas.gaullierⓐcji.paris> [DH] libavcodec/version.c

28 août

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

06:14
prores_raw: use MKBETAG for the frame header tag - Lynne git.videolan.org Git - ffmpeg.git/rss log

prores_raw: use MKBETAG for the frame header tag Equivalent, but more explicit. All values in the header are big endian. [DH] libavcodec/prores_raw.c [DH] libavcodec/prores_raw_parser.c
06:12
prores_raw: set profile based on the codec tag - Lynne git.videolan.org Git - ffmpeg.git/rss log

prores_raw: set profile based on the codec tag This is the same as what the parser does. [DH] libavcodec/prores_raw.c
20:52
MAINTAINERS: Update the entries for AMF - Dmitrii Ovchinnikov git.videolan.org Git - ffmpeg.git/rss log

MAINTAINERS: Update the entries for AMF MAINTAINERS: Update the entries for AMF Alphabetized the records [DH] MAINTAINERS
19:56
forgejo/CODEOWNERS: add myself for relevant files - Dmitrii Ovchinnikov git.videolan.org Git - ffmpeg.git/rss log

forgejo/CODEOWNERS: add myself for relevant files [DH] .forgejo/CODEOWNERS
18:58
fftools/ffmpeg: fix gracefully shutdown - Patrick Wang git.videolan.org Git - ffmpeg.git/rss log

fftools/ffmpeg: fix gracefully shutdown d119ae2fd82a494d9430ff4d4fc262961a68c598 removed the loop-breaking condition received_sigterm. Thus, signals no longer gracefully shutdown ffmpeg. Fixes: #10834 Signed-off-by: Patrick Wang <mail6543210ⓐgmail.com> Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] fftools/ffmpeg.c
23:12
forgejo/labeler: react to 'labeled' action as well - Timo Rothenpieler git.videolan.org Git - ffmpeg.git/rss log

forgejo/labeler: react to 'labeled' action as well [DH] .forgejo/labeler/labeler.js

27 août

git.videolan.org Git - ffmpeg.git/rss log http://git.videolan.org/?p=ffmpeg.git;a=summary

15:43
avcodec/libjxldec: submit frame after file is complete - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

avcodec/libjxldec: submit frame after file is complete This commit causes the libjxl decoder wrapper to hold onto the decoded frame until the trailing metadata after the frame is possibly complete before it submits the frame. This allows EXIF and other metadata boxes that occur after the frame is fully rendered to be attached to it as side data. Signed-off-by: Leo Izen <leo.izenⓐgmail.com> [DH] libavcodec/libjxldec.c
08:39
lavf/id3v2: report incorrect BOM value on error - Lynne git.videolan.org Git - ffmpeg.git/rss log

lavf/id3v2: report incorrect BOM value on error [DH] libavformat/id3v2.c
08:24
Update libavformat/apvdec.c - dariusz-f git.videolan.org Git - ffmpeg.git/rss log

Update libavformat/apvdec.c Increase au_size limit by 4 times to fit more complex videos. Signed-off-by: dariusz-f <dariusz-fⓐnoreply.code.ffmpeg.org> [DH] libavformat/apvdec.c
15:53
avcodec/libjxldec: consume input on error - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

avcodec/libjxldec: consume input on error libjxl consumes no input if it returns an error, so this loops over and over while it spits out the same error many times. If we got an error from libjxl and consumed no input, then we instead consume the whole packet. Signed-off-by: Leo Izen <leo.izenⓐgmail.com> [DH] libavcodec/libjxldec.c
06:36
avformat/utils: add AV_FRAME_FILENAME_FLAGS_IGNORE_TRUNCATION flag - Marton Balint git.videolan.org Git - ffmpeg.git/rss log

avformat/utils: add AV_FRAME_FILENAME_FLAGS_IGNORE_TRUNCATION flag Signed-off-by: Marton Balint <cusⓐpasswd.hu> [DH] doc/APIchanges [DH] libavformat/avformat.h [DH] libavformat/utils.c [DH] libavformat/version.h
23:40
avformat/whip: fix potential 8bit overflow for profile_idc - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: fix potential 8bit overflow for profile_idc The profile contains profile_idc and constraint_set*_flag, throws away high 8 bit flags and then we get profile_idc. Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/whip.c
23:22
avformat/whip: fix 8 bits overflow and map constraint_set bits for H264 - Jack Lau git.videolan.org Git - ffmpeg.git/rss log

avformat/whip: fix 8 bits overflow and map constraint_set bits for H264 profile_iop is an 8 bits field. Previous code copied AVCodecParameters::profile (which can contain bits beyond 8 bits) into profile_iop, producing overflow and wrong values. This patch maps the constrained flags into the proper profile_iop bits (constraint_set1 / 3) Signed-off-by: Jack Lau <jacklau1222ⓐqq.com> [DH] libavformat/whip.c
10:14
swscale/graph: fix double-free when legacy pass fails initializing - Niklas Haas git.videolan.org Git - ffmpeg.git/rss log

swscale/graph: fix double-free when legacy pass fails initializing If this function returns an error after ff_sws_graph_add_pass() has been called, and the pass->free callback is therefore already set up to free the context, the graph will end up freed twice: once by the pass->free callback (during ff_sws_graph_free()), and once before that by failure path of the caller (e.g. add_legacy_sws_pass(), or init_legacy_subpass() itself for cascaded contexts.) The solution is to redefine the ownership of SwsGraph to pass clearly from the caller of add_legacy_sws_pass() to (...)
20:51
vp9: Add AVX-512ICL asm for 8bpc subpel mc - Henrik Gramner git.videolan.org Git - ffmpeg.git/rss log

vp9: Add AVX-512ICL asm for 8bpc subpel mc [DH] libavcodec/x86/vp9dsp_init.c [DH] libavcodec/x86/vp9dsp_init.h [DH] libavcodec/x86/vp9mc.asm [DH] tests/checkasm/vp9dsp.c
13:00
avcodec/exif: also copy zero termination for AV_TIFF_STRING - Michael Niedermayer git.videolan.org Git - ffmpeg.git/rss log

avcodec/exif: also copy zero termination for AV_TIFF_STRING Fixes: out of array read Fixes: 441131173/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TIFF_DEC_fuzzer-6700429212975104 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michaelⓐniedermayer.cc> [DH] libavcodec/exif.c
09:28
avcodec/libjxlenc: fix leaked EXIF ifd - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

avcodec/libjxlenc: fix leaked EXIF ifd We're missing a call to av_exif_free here. We leak the internal heap- allocated objects when &ifd goes out of scope. Signed-off-by: Leo Izen <leo.izenⓐgmail.com> [DH] libavcodec/libjxlenc.c
09:26
avcodec/libjxldec: fix leaked EXIF ifd - Leo Izen git.videolan.org Git - ffmpeg.git/rss log

avcodec/libjxldec: fix leaked EXIF ifd We're missing a call to av_exif_free here. We leak the internal heap- allocated objects when &ifd goes out of scope. Signed-off-by: Leo Izen <leo.izenⓐgmail.com> [DH] libavcodec/libjxldec.c

Articles récents

Préférences

Sources

Boussole SPIP

Sedna

26 septembre

25 septembre

24 septembre

23 septembre

22 septembre

21 septembre

20 septembre

19 septembre

18 septembre

17 septembre

16 septembre

15 septembre

14 septembre

13 septembre

12 septembre

11 septembre

10 septembre

9 septembre

8 septembre

7 septembre

6 septembre

5 septembre

4 septembre

3 septembre

2 septembre

1er septembre

31 août

30 août

29 août

28 août

27 août