September 2025 - ffmpeg-cvslog

[ffmpeg] branch master updated. 126671e730 avcodec/mfenc: add AVLowLatencyMode support
by ffmpeg-git＠ffmpeg.org 01 Sep '25

01 Sep '25

The branch, master has been updated via 126671e73065e29c9582e47bc74774af2200fc8f (commit) from f2414fb1aa0dcc620ed043201d329b3fe43a4690 (commit) - Log ----------------------------------------------------------------- commit 126671e73065e29c9582e47bc74774af2200fc8f Author: Cameron Gutman <aicommander(a)gmail.com> AuthorDate: Sun Aug 31 16:05:15 2025 -0500 Commit: James Almer <jamrial(a)gmail.com> CommitDate: Mon Sep 1 12:40:36 2025 +0000 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(a)gmail.com> diff --git a/libavcodec/mfenc.c b/libavcodec/mfenc.c index 6f79386404..541f7fb961 100644 --- a/libavcodec/mfenc.c +++ b/libavcodec/mfenc.c @@ -876,6 +876,9 @@ static int mf_encv_output_adjust(AVCodecContext *avctx, IMFMediaType *type) if (c->opt_enc_scenario >= 0) ICodecAPI_SetValue(c->codec_api, &ff_CODECAPI_AVScenarioInfo, FF_VAL_VT_UI4(c->opt_enc_scenario)); + + if (avctx->flags & AV_CODEC_FLAG_LOW_DELAY) + ICodecAPI_SetValue(c->codec_api, &ff_CODECAPI_AVLowLatencyMode, FF_VAL_VT_UI4(1)); } return 0; ----------------------------------------------------------------------- Summary of changes: libavcodec/mfenc.c | 3 +++ 1 file changed, 3 insertions(+) hooks/post-receive --

1 0

[ffmpeg] branch master updated. f2414fb1aa .forgejo/CODEOWNERS: add myself for various files
by ffmpeg-git＠ffmpeg.org 01 Sep '25

01 Sep '25

The branch, master has been updated via f2414fb1aa0dcc620ed043201d329b3fe43a4690 (commit) from 3ea6c2fe2569f3e8d6f1396982ce8bf8a3fc6e13 (commit) - Log ----------------------------------------------------------------- commit f2414fb1aa0dcc620ed043201d329b3fe43a4690 Author: Zhao Zhili <quinkblack(a)foxmail.com> AuthorDate: Sun Aug 31 23:07:09 2025 +0800 Commit: Timo Rothenpieler <timo(a)rothenpieler.org> CommitDate: Mon Sep 1 11:14:30 2025 +0000 .forgejo/CODEOWNERS: add myself for various files Signed-off-by: Zhao Zhili <zhilizhao(a)tencent.com> diff --git a/.forgejo/CODEOWNERS b/.forgejo/CODEOWNERS index d421aacb12..3b48612be7 100644 --- a/.forgejo/CODEOWNERS +++ b/.forgejo/CODEOWNERS @@ -39,9 +39,11 @@ libavcodec/.*jpegxl.* @lynne @Traneptora libavcodec/jv.* @pross libavcodec/.*jxl.* @lynne @Traneptora libavcodec/lead.* @pross +libavcodec/mediacodec* @quink libavcodec/mmvideo.* @pross libavcodec/msp2.* @pross libavcodec/mvc.* @pross +libavcodec/oh* @quink libavcodec/.*opus.* @lynne libavcodec/pictor.* @pross libavcodec/.*png.* @Traneptora @@ -75,6 +77,7 @@ libavcodec/bsf/smpte436m_to_eia608.* @programmerjake libavcodec/aarch64/.* @lynne @mstorsjo libavcodec/arm/.* @mstorsjo libavcodec/ppc/.* @sean_mcg +libavcodec/wasm/hevc/.* @quink libavcodec/x86/.* @lynne libavcodec/x86/vp8.* @rbultje libavcodec/x86/vp9.* @rbultje @@ -85,6 +88,8 @@ libavcodec/x86/vpx.* @rbultje libavfilter/af_whisper.* @vpalmisano libavfilter/colorspace.* @rbultje libavfilter/vf_colorspace.* @rbultje +libavfilter/vf_scale_vt.* @quink +libavfilter/vf_transpose_vt.* @quink libavfilter/vf_yadif.* @michaelni libavfilter/vsrc_mandelbrot.* @michaelni @@ -135,6 +140,8 @@ libavutil/.*crc.* @lynne @michaelni libavutil/.*d3d12va.* @jianhuaw libavutil/csp.* @rbultje libavutil/eval.* @michaelni +libavutil/hwcontext_oh.* @quink +libavutil/hwcontext_mediacodec.* @quink libavutil/iamf.* @jamrial libavutil/integer.* @michaelni libavutil/lfg.* @michaelni ----------------------------------------------------------------------- Summary of changes: .forgejo/CODEOWNERS | 7 +++++++ 1 file changed, 7 insertions(+) hooks/post-receive --

1 0

[ffmpeg] branch master updated. 3ea6c2fe25 aacencdsp: Improve consistency with assembly, for x87 math
by ffmpeg-git＠ffmpeg.org 01 Sep '25

01 Sep '25

The branch, master has been updated via 3ea6c2fe2569f3e8d6f1396982ce8bf8a3fc6e13 (commit) from 883162d5b4107dfe8c8a3efef3dd798bc4cf7a63 (commit) - Log ----------------------------------------------------------------- commit 3ea6c2fe2569f3e8d6f1396982ce8bf8a3fc6e13 Author: Martin Storsjö <martin(a)martin.st> AuthorDate: Fri Aug 22 23:26:11 2025 +0300 Commit: Martin Storsjö <martin(a)martin.st> CommitDate: Mon Sep 1 09:22:39 2025 +0000 aacencdsp: Improve consistency with assembly, for x87 math Currently, the aacencdsp checkasm tests fails for many seeds, if the C code has been built with x87 math. This happens because the excess precision of x87 math can make it end up rounding to a different integer, and the checkasm tests checks that the output integers match exactly between C and assembly. One such failing case is "tests/checkasm/checkasm --test=aacencdsp 41" when compiled with GCC. When compiled with Clang, the test seed 21 produces a failure. To avoid the issue, we need to limit the precision of intermediates to their nominal float range, matching the assembly implementations. This can be achieved when compiling with GCC, by just adding a single cast. To observe the effect of this cast, compile the following snippet, int cast(float a, float b) { return (int) #ifdef CAST (float) #endif (a + b); } with "gcc -m32 -std=c17 -O2", with/without -DCAST. For x86_64 cases (without the "-m32"), the cast doesn't make any difference on the generated code. This cast would seem to not have any effect, as a binary expression with float inputs also would have the type float. However, if compiling with GCC with -fexcess-precision=standard, the cast forces limiting the precision according to the language standard here - according to the GCC docs [1]: > When compiling C or C++, if -fexcess-precision=standard is > specified then excess precision follows the rules specified in > ISO C99 or C++; in particular, both casts and assignments cause > values to be rounded to their semantic types (whereas -ffloat-store > only affects assignments). This option is enabled by default for > C or C++ if a strict conformance option such as -std=c99 or > -std=c++17 is used. Ffmpeg's configure scripts enables -std=c17 by default. This only helps with GCC though - the cast doesn't make any difference for Clang. (Although, upstream Clang seems to default to SSE math, while Ubuntu provided Clang defaults to x87 math.) Limiting the precision with Clang would require casting to volatile float for both intermediates here - and that does have a code generation effect on all architectures. [1] https://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html diff --git a/libavcodec/aacencdsp.h b/libavcodec/aacencdsp.h index d0d86c3d70..684bbc254f 100644 --- a/libavcodec/aacencdsp.h +++ b/libavcodec/aacencdsp.h @@ -50,7 +50,7 @@ static inline void quantize_bands(int *out, const float *in, const float *scaled { for (int i = 0; i < size; i++) { float qc = scaled[i] * Q34; - int tmp = (int)FFMIN(qc + rounding, (float)maxval); + int tmp = (int)FFMIN((float)(qc + rounding), (float)maxval); if (is_signed && in[i] < 0.0f) { tmp = -tmp; } ----------------------------------------------------------------------- Summary of changes: libavcodec/aacencdsp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) hooks/post-receive --

1 0

[ffmpeg] branch master updated. 883162d5b4 forgejo/workflows: make autolabel workflow always use token from secrets
by ffmpeg-git＠ffmpeg.org 01 Sep '25

01 Sep '25

The branch, master has been updated via 883162d5b4107dfe8c8a3efef3dd798bc4cf7a63 (commit) from b2910ec92ef0f12f2cee56be331f8ae234ead4a4 (commit) - Log ----------------------------------------------------------------- commit 883162d5b4107dfe8c8a3efef3dd798bc4cf7a63 Author: Timo Rothenpieler <timo(a)rothenpieler.org> AuthorDate: Mon Sep 1 03:02:49 2025 +0200 Commit: Timo Rothenpieler <timo(a)rothenpieler.org> CommitDate: Mon Sep 1 03:02:51 2025 +0200 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 diff --git a/.forgejo/workflows/autolabel.yml b/.forgejo/workflows/autolabel.yml index 85afcdad03..cb4dac67aa 100644 --- a/.forgejo/workflows/autolabel.yml +++ b/.forgejo/workflows/autolabel.yml @@ -18,11 +18,11 @@ jobs: if: ${{ forge.event_name == 'pull_request_target' }} with: configuration-path: .forgejo/labeler/labeler.yml - repo-token: ${{ forge.event_name == 'pull_request_target' && secrets.AUTOLABELER_TOKEN || forge.token }} + repo-token: ${{ secrets.AUTOLABELER_TOKEN }} - name: Label by title-match uses: https://github.com/actions/github-script@v7 with: script: | const script = require('.forgejo/labeler/labeler.js') await script({github, context}) - github-token: ${{ forge.event_name == 'pull_request_target' && secrets.AUTOLABELER_TOKEN || forge.token }} + github-token: ${{ secrets.AUTOLABELER_TOKEN }} ----------------------------------------------------------------------- Summary of changes: .forgejo/workflows/autolabel.yml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) hooks/post-receive --

1 0

[ffmpeg] branch master updated. b2910ec92e avcodec/mfenc: fix memory leak with D3D11 input surfaces
by ffmpeg-git＠ffmpeg.org 01 Sep '25

01 Sep '25

The branch, master has been updated via b2910ec92ef0f12f2cee56be331f8ae234ead4a4 (commit) from ae448e00afb43d7f72dfa9c82a4c45994e4fea6a (commit) - Log ----------------------------------------------------------------- commit b2910ec92ef0f12f2cee56be331f8ae234ead4a4 Author: Cameron Gutman <aicommander(a)gmail.com> AuthorDate: Sun Aug 10 16:11:55 2025 -0500 Commit: Cameron Gutman <aicommander(a)gmail.com> CommitDate: Sun Aug 31 14:36:20 2025 -0500 avcodec/mfenc: fix memory leak with D3D11 input surfaces Fixes: d56522c6eb75 ("avcodec/mfenc: add support for D3D11 input surfaces") Signed-off-by: Cameron Gutman <aicommander(a)gmail.com> diff --git a/libavcodec/mfenc.c b/libavcodec/mfenc.c index 942c75cb9d..6f79386404 100644 --- a/libavcodec/mfenc.c +++ b/libavcodec/mfenc.c @@ -1366,6 +1366,9 @@ static int mf_close(AVCodecContext *avctx) if (c->async_events) IMFMediaEventGenerator_Release(c->async_events); + if (c->dxgiManager) + IMFDXGIDeviceManager_Release(c->dxgiManager); + #if !HAVE_UWP if (c->library) ff_free_mf(&c->functions, &c->mft); ----------------------------------------------------------------------- Summary of changes: libavcodec/mfenc.c | 3 +++ 1 file changed, 3 insertions(+) hooks/post-receive --

1 0