[FFmpeg-devel] [PATCH v6] libavfi/dnn: add LibTorch as one of DNN backend

From: Wenbin Chen PyTorch is an open source machine learning framework that accelerates the path from research prototyping to production deployment. Official website: https://pytorch.org/. We call the C++ library of PyTorch as LibTorch, the same below. To build FFmpeg with LibTorch, please take

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/ffmpeg: add loopback decoding

On 2024-03-14 03:46 pm, Anton Khirnov wrote: Quoting Gyan Doshi (2024-03-13 08:49:52) Are there any limitations to loopback decoding e.g. would a vpx w/alpha encode be decoded back to a alpha pix fmt? I don't know how vpx alpha works, was it a separate stream in the demuxer or something lik

Re: [FFmpeg-devel] [PATCH] avcodec/mips/aaccoder_mips: Remove MIPS-specific aaccoder

Mar 15, 2024, 02:20 by andreas.rheinha...@outlook.com: > ff_aac_coder_init_mips() modifies a static const structure of > function pointers. This will crash if the binary uses relro > and is a data race in any case. > > Furthermore it points to a maintainability issue: The > AACCoefficientsEncoder

Re: [FFmpeg-devel] [PATCH] lavu/hwcontext_qsv: Join the download/upload session to the main session

On Ma, 2024-03-11 at 12:56 +0800, Xiang, Haihao wrote: > From: Haihao Xiang > > This may reduce the number of internal threads when using hwupload or > hwdownload filter. > > Signed-off-by: Haihao Xiang > --- >  libavutil/hwcontext_qsv.c | 24 >  1 file changed, 24 inser

Re: [FFmpeg-devel] [PATCH] lavc/qsvdec: Do not print warning when draining cached frames

On Ma, 2024-03-11 at 12:37 +0800, Xiang, Haihao wrote: > From: Haihao Xiang > > When all cached frames are drained, the output mfxSyncPoint pointer is > NULL and  MFX_ERR_MORE_DATA is returned, hence needn't print warning for > this expected behavior, otherwise the user might think the output fro

Re: [FFmpeg-devel] [PATCH v3] lavc/vaapi_encode: Enable block level bitrate control

On Vr, 2024-03-08 at 16:45 +0800, fei.w.wang-at-intel@ffmpeg.org wrote: > From: Fei Wang > > Signed-off-by: Fei Wang > --- >  doc/encoders.texi |  4 >  libavcodec/vaapi_encode.c | 13 - >  libavcodec/vaapi_encode.h |  9 - >  3 files changed, 24 insertions(+),

Re: [FFmpeg-devel] [PATCH] avcodec/vaapi_encode: add customized surface alignment

On Do, 2024-03-07 at 00:55 +0100, Araz Iusubov wrote: > This commit fixes issues with AMD HEVC encoding. > By default AMD hevc encoder asks for the alignment 64x16, while FFMPEG VAAPI > has 16x16. > Adding support for customzied surface size from VASurfaceAttribAlignmentSize s/customzied/customi

Re: [FFmpeg-devel] [PATCH v1] lavc/vaapi_decode: Don't update buffer number if allocataion fail

On Do, 2024-03-14 at 08:38 +0800, fei.w.wang-at-intel@ffmpeg.org wrote: > From: Fei Wang > > Signed-off-by: Fei Wang > --- >  libavcodec/vaapi_decode.c | 9 +++-- >  1 file changed, 3 insertions(+), 6 deletions(-) > > diff --git a/libavcodec/vaapi_decode.c b/libavcodec/vaapi_decode.c > i

Re: [FFmpeg-devel] [PATCH v2 2/2] lavc/vaapi_encode_h265: Set general_*_constriaint flags with profile

On Vr, 2024-03-08 at 16:47 +0800, fei.w.wang-at-intel@ffmpeg.org wrote: > From: Fei Wang > > According to Table A.2 in spec. > > Signed-off-by: Fei Wang > --- >  libavcodec/vaapi_encode_h265.c | 181 +++-- >  1 file changed, 128 insertions(+), 53 deletions(-) > >

Re: [FFmpeg-devel] [PATCH v5] libavfi/dnn: add LibTorch as one of DNN backend

> > -Original Message- > > From: ffmpeg-devel On Behalf Of > > wenbin.chen-at-intel@ffmpeg.org > > Sent: Monday, March 11, 2024 1:02 PM > > To: ffmpeg-devel@ffmpeg.org > > Subject: [FFmpeg-devel] [PATCH v5] libavfi/dnn: add LibTorch as one of DNN > > backend > > > > From: Wenbin Chen

[FFmpeg-devel] [PATCH] avcodec/mips/aaccoder_mips: Remove MIPS-specific aaccoder

ff_aac_coder_init_mips() modifies a static const structure of function pointers. This will crash if the binary uses relro and is a data race in any case. Furthermore it points to a maintainability issue: The AACCoefficientsEncoder structures have been constified in commit fd9212f2edfe9b107c3c08ba2

[FFmpeg-devel] [PATCH] avutil/hash: Avoid relocations for hash names

These strings are so short (longest takes 11B) that using pointers is wasteful. Avoiding them also moves hashdesc into .rodata (from .data.rel.ro). Signed-off-by: Andreas Rheinhardt --- libavutil/hash.c | 71 +++- 1 file changed, 40 insertions(+), 31 d

[FFmpeg-devel] [PATCH v3 7/9] avformat/pcm: factorize and improve determining the default packet size

- Remove the 1024 cap on the number of samples, for high sample rate audio it was suboptimal, calculate the low neighbour power of two for the number of samples (audio blocks) instead. - Make the function work correctly also for non-pcm codecs by using the stream bitrate to estimate the targe

[FFmpeg-devel] [PATCH 26/26] fate/wmavoice: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/wma.mak | 16 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/tests/fate/wma.mak b/tests/fate/wma.mak index 308dced9d6..ed29edae3a 100644 --- a/tests/fate/wma.mak +++ b/tests/fate/wma.mak @@ -1,14 +1,14 @@ -FATE_WMAPRO-$(cal

[FFmpeg-devel] [PATCH 25/26] fate/vqf: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/vqf.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/vqf.mak b/tests/fate/vqf.mak index df5ff44f9d..ccdbe50033 100644 --- a/tests/fate/vqf.mak +++ b/tests/fate/vqf.mak @@ -1,4 +1,4 @@ -FATE_VQF-$(call DEMDEC, VQF, TWINVQ)

[FFmpeg-devel] [PATCH 24/26] fate/voice: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/voice.mak | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fate/voice.mak b/tests/fate/voice.mak index 61a4617b44..ab34e353f7 100644 --- a/tests/fate/voice.mak +++ b/tests/fate/voice.mak @@ -33,7 +33,7 @@ fate-g723_1-dec-7: CM

[FFmpeg-devel] [PATCH 23/26] fate/vorbis: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/vorbis.mak | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/fate/vorbis.mak b/tests/fate/vorbis.mak index b9b7405e87..781430bcd3 100644 --- a/tests/fate/vorbis.mak +++ b/tests/fate/vorbis.mak @@ -94,7 +94,9 @@ fate-vorbis-

[FFmpeg-devel] [PATCH 22/26] fate/real: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/real.mak | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fate/real.mak b/tests/fate/real.mak index dd5f9c61db..13ceab7adf 100644 --- a/tests/fate/real.mak +++ b/tests/fate/real.mak @@ -9,13 +9,13 @@ fate-ra4-288: CMP = oneo

[FFmpeg-devel] [PATCH 21/26] fate/pcm: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/pcm.mak | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/fate/pcm.mak b/tests/fate/pcm.mak index 4e79fc7c41..b10233a9eb 100644 --- a/tests/fate/pcm.mak +++ b/tests/fate/pcm.mak @@ -1,10 +1,10 @@ -FATE_SAMPLES_PCM-$(call

[FFmpeg-devel] [PATCH 20/26] fate/mpc: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/mpc.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/mpc.mak b/tests/fate/mpc.mak index cde6e55177..b737da449b 100644 --- a/tests/fate/mpc.mak +++ b/tests/fate/mpc.mak @@ -4,7 +4,7 @@ fate-mpc7-demux: CMD = crc -i $(TARG

[FFmpeg-devel] [PATCH 19/26] fate/mp3: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/mp3.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/mp3.mak b/tests/fate/mp3.mak index b874e7c7a4..638ff63264 100644 --- a/tests/fate/mp3.mak +++ b/tests/fate/mp3.mak @@ -36,7 +36,7 @@ $(FATE_MP3): FUZZ = 18 fate-mp3-

[FFmpeg-devel] [PATCH 18/26] fate/mov: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/mov.mak | 12 +++- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/tests/fate/mov.mak b/tests/fate/mov.mak index 51fd5fa2f3..b54fe19620 100644 --- a/tests/fate/mov.mak +++ b/tests/fate/mov.mak @@ -10,11 +10,9 @@ FATE_MOV = fate-mov-3

[FFmpeg-devel] [PATCH 17/26] fate/monkeysaudio: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/monkeysaudio.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/monkeysaudio.mak b/tests/fate/monkeysaudio.mak index 03c646cd47..7ffe2fcdcf 100644 --- a/tests/fate/monkeysaudio.mak +++ b/tests/fate/monkeysaudio.mak @@ -19,5

[FFmpeg-devel] [PATCH 16/26] fate/lossless-audio: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/lossless-audio.mak | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/tests/fate/lossless-audio.mak b/tests/fate/lossless-audio.mak index f3aa700743..7c5af885b4 100644 --- a/tests/fate/lossless-audio.mak +++ b/tests/fate/lo

[FFmpeg-devel] [PATCH 15/26] fate/libswresample: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/libswresample.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/libswresample.mak b/tests/fate/libswresample.mak index 405328797e..a1e5ab91fa 100644 --- a/tests/fate/libswresample.mak +++ b/tests/fate/libswresample.mak @@

[FFmpeg-devel] [PATCH 14/26] fate/iamf: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/iamf.mak | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/tests/fate/iamf.mak b/tests/fate/iamf.mak index 6fe0d54762..cf56a67e6f 100644 --- a/tests/fate/iamf.mak +++ b/tests/fate/iamf.mak @@ -1,4 +1,4 @@ -FATE_IAMF += fat

[FFmpeg-devel] [PATCH 13/26] fate/gapless: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/gapless.mak | 78 +- 1 file changed, 39 insertions(+), 39 deletions(-) diff --git a/tests/fate/gapless.mak b/tests/fate/gapless.mak index 68a396e187..dfae964662 100644 --- a/tests/fate/gapless.mak +++ b/tests/fate/

[FFmpeg-devel] [PATCH 12/26] fate/ffmpeg: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/ffmpeg.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/ffmpeg.mak b/tests/fate/ffmpeg.mak index c242c4fe53..9cf05ead68 100644 --- a/tests/fate/ffmpeg.mak +++ b/tests/fate/ffmpeg.mak @@ -129,7 +129,7 @@ fate-ffmpeg-fix_su

[FFmpeg-devel] [PATCH 08/26] fate/audio: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/audio.mak | 28 +++- 1 file changed, 15 insertions(+), 13 deletions(-) diff --git a/tests/fate/audio.mak b/tests/fate/audio.mak index 65317c8d45..cfac90424b 100644 --- a/tests/fate/audio.mak +++ b/tests/fate/audio.mak @@ -5,12 +5,

[FFmpeg-devel] [PATCH 11/26] fate/filter-audio: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/filter-audio.mak | 40 ++--- 1 file changed, 20 insertions(+), 20 deletions(-) diff --git a/tests/fate/filter-audio.mak b/tests/fate/filter-audio.mak index 6ddc0baf57..3a929121a3 100644 --- a/tests/fate/filter-audio.mak ++

[FFmpeg-devel] [PATCH 07/26] fate/atrac: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/atrac.mak | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fate/atrac.mak b/tests/fate/atrac.mak index acf79a539c..1c91357559 100644 --- a/tests/fate/atrac.mak +++ b/tests/fate/atrac.mak @@ -7,7 +7,7 @@ fate-atrac1-2: CMD =

[FFmpeg-devel] [PATCH 10/26] fate/fate-container: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/lavf-container.mak | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fate/lavf-container.mak b/tests/fate/lavf-container.mak index fa7ecd2cb5..ce5fdd0f05 100644 --- a/tests/fate/lavf-container.mak +++ b/tests/fate/lavf-containe

[FFmpeg-devel] [PATCH 06/26] fate/amrwb: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/amrwb.mak | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/fate/amrwb.mak b/tests/fate/amrwb.mak index 7cd17d266e..5e0a47bb82 100644 --- a/tests/fate/amrwb.mak +++ b/tests/fate/amrwb.mak @@ -46,6 +46,6 @@ fate-amrwb-remux: CMD = t

[FFmpeg-devel] [PATCH 09/26] fate/demux: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/demux.mak | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fate/demux.mak b/tests/fate/demux.mak index ace8fa0b52..d9b9045f0b 100644 --- a/tests/fate/demux.mak +++ b/tests/fate/demux.mak @@ -79,7 +79,7 @@ fate-mlv-demux: CMD =

[FFmpeg-devel] [PATCH 05/26] fate/amrnb: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/amrnb.mak | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/fate/amrnb.mak b/tests/fate/amrnb.mak index 6f5ed413e3..3eb8b324f3 100644 --- a/tests/fate/amrnb.mak +++ b/tests/fate/amrnb.mak @@ -30,11 +30,11 @@ FATE_AMRNB += fate-

[FFmpeg-devel] [PATCH 04/26] fate/alac: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/alac.mak | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/tests/fate/alac.mak b/tests/fate/alac.mak index 5e29e90cbd..bb27e9db4e 100644 --- a/tests/fate/alac.mak +++ b/tests/fate/alac.mak @@ -20,5 +20,7 @@ fate-alac-24-%: CMD = en

[FFmpeg-devel] [PATCH 03/26] fate/adpcm: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/adpcm.mak | 44 ++-- 1 file changed, 22 insertions(+), 22 deletions(-) diff --git a/tests/fate/adpcm.mak b/tests/fate/adpcm.mak index a7d4dbcf6c..7265211f09 100644 --- a/tests/fate/adpcm.mak +++ b/tests/fate/adpcm.

[FFmpeg-devel] [PATCH 02/26] fate/ac3: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/ac3.mak | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/fate/ac3.mak b/tests/fate/ac3.mak index 2dfd59dfb1..1ecb5a3f54 100644 --- a/tests/fate/ac3.mak +++ b/tests/fate/ac3.mak @@ -65,17 +65,17 @@ fate-eac3-5: REF = $(S

[FFmpeg-devel] [PATCH 01/26] fate/aac: add missing aresample filter dependency

Signed-off-by: James Almer --- tests/fate/aac.mak | 10 +- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/tests/fate/aac.mak b/tests/fate/aac.mak index 4f8d1cdcea..817944773d 100644 --- a/tests/fate/aac.mak +++ b/tests/fate/aac.mak @@ -243,17 +243,17 @@ fate-aac-latm_stereo

Re: [FFmpeg-devel] [PATCH] tools/target_dec_fuzzer: force experimental flag for VVC

On Thu, Mar 14, 2024 at 12:43:00PM -0300, James Almer wrote: > Otherwise the decoder will not even open. > > Signed-off-by: James Almer > --- > tools/target_dec_fuzzer.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/tools/target_dec_fuzzer.c b/tools/target_dec_fuzzer.c > index 84b

Re: [FFmpeg-devel] [PATCH] vvcdec: Mark as experimental

On Wed, Feb 07, 2024 at 10:55:18PM +, Kieran Kunhya wrote: > On Wed, 7 Feb 2024 at 22:06, Paul B Mahol wrote: > > > On Wed, Feb 7, 2024 at 10:13 PM Kieran Kunhya wrote: > > > > > $subj > > > > > > As discussed at FOSDEM. > > > > > > > Author of this patch above is forced to FUZZ this decoder

Re: [FFmpeg-devel] [PATCH 1/1] avformat/demux: Fix accurate probing of durations in mpegts/ps

On Thu, Mar 14, 2024 at 06:57:39PM +0100, Nicolas Gaullier wrote: > Two issues affect accuracy of duration in estimate_timings_from_pts(): > - pkt->duration typically reports the duration of a single audio frame, > whereas a pes often contain several audio frames > - for video, compute_frame_durati

Re: [FFmpeg-devel] [PATCH v2 9/9] avformat/pcm: decrease target audio frame per sec to 10

On Thu, 14 Mar 2024, Paul B Mahol wrote: This is just ugly hack for workaround around broken ffmpeg threading. This actually improves single threaded performance as well, check it with ffprobe if you want. Sure, the reason I turned my attention to this is the speed loss of ffmpeg.c becaus

Re: [FFmpeg-devel] [PATCH v2 9/9] avformat/pcm: decrease target audio frame per sec to 10

On Thu, 14 Mar 2024, Tobias Rapp wrote: On 08/03/2024 01:53, Marton Balint wrote: This makes the wav and pcm demuxer demux bigger packets, which is more efficient. [...] +0, 136000, 136000, 2000,24000, f952be1e1c7d48acd73d652ef4ea3aad +0, 138000, 138000,

Re: [FFmpeg-devel] [PATCH 1/3] fate/lavf-audio: Test writing AIFF-native tags

Michael Niedermayer: > On Thu, Mar 14, 2024 at 01:35:36AM +0100, Andreas Rheinhardt wrote: >> In particular, test writing tags with odd strlen. >> (These tags are zero-padded to even size.) >> >> Signed-off-by: Andreas Rheinhardt >> --- >> tests/fate/lavf-audio.mak | 1 + >> tests/ref/lavf/aiff

Re: [FFmpeg-devel] [PATCH 1/3] fate/lavf-audio: Test writing AIFF-native tags

On Thu, Mar 14, 2024 at 01:35:36AM +0100, Andreas Rheinhardt wrote: > In particular, test writing tags with odd strlen. > (These tags are zero-padded to even size.) > > Signed-off-by: Andreas Rheinhardt > --- > tests/fate/lavf-audio.mak | 1 + > tests/ref/lavf/aiff | 4 ++-- > 2 files chan

Re: [FFmpeg-devel] [PATCH] avformat/internal: Remove declaration for ff_format_io_close_default()

Andreas Rheinhardt: > Forgotten in d6799ee0e41dee35ebf9c664173aed8e3ab24141. > > Signed-off-by: Andreas Rheinhardt > --- > libavformat/internal.h | 4 > 1 file changed, 4 deletions(-) > > diff --git a/libavformat/internal.h b/libavformat/internal.h > index 2e59036a53..e5f8337130 100644 > -

[FFmpeg-devel] [PATCH 3/3] avutil/hwcontext_d3d11va: add logging to dxgi debug interfaces

--- libavutil/hwcontext_d3d11va.c | 10 +- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/libavutil/hwcontext_d3d11va.c b/libavutil/hwcontext_d3d11va.c index 1b96c38d25..e30c8fc238 100644 --- a/libavutil/hwcontext_d3d11va.c +++ b/libavutil/hwcontext_d3d11va.c @@ -685,9 +685,1

[FFmpeg-devel] [PATCH 2/3] avutil/hwcontext_d3d11va: prefer DXGI 1.1 factory when available

A lot of modern stuff straight up fails on the old 1.0 factory, which is masked by the fact that it's only used when an explicit adapter is specified. --- libavutil/hwcontext_d3d11va.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/libavutil/hwcontext_d3d11va.c b/libavutil/

[FFmpeg-devel] [PATCH 1/3] avutil/hwcontext_d3d11va: remove check for d3d11 debug layer dll

At least on latest Win 11 and Visual Studio 2022, that DLL does not exist anymore and can't be installed via any of the usual means. However, debugging works just fine regardless, so this check makes debugging impossible. D3D11CreateDevice will fail anyway if debugging is not supported, so let's r

Re: [FFmpeg-devel] [PATCH] fftools/ffmpeg_sched: Remove set-but-unused variable

Andreas Rheinhardt: > Signed-off-by: Andreas Rheinhardt > --- > fftools/ffmpeg_sched.c | 2 -- > 1 file changed, 2 deletions(-) > > diff --git a/fftools/ffmpeg_sched.c b/fftools/ffmpeg_sched.c > index d1fb942c34..f739066921 100644 > --- a/fftools/ffmpeg_sched.c > +++ b/fftools/ffmpeg_sched.c > @

Re: [FFmpeg-devel] [PATCH] avcodec/ppc/vp8dsp_altivec: Fix out-of-bounds access

Andreas: On Tue, Mar 12, 2024 at 9:14 PM Andreas Rheinhardt wrote: > > h_subpel_filters_inner[i] and h_subpel_filters_outer[i / 2] > belong together and the former allows the range 0..6, > so the latter needs to support 0..3. But it has only three > elements. Add another one. > The value for the

Re: [FFmpeg-devel] [PATCH] avformat: enable UDP IPv6 multicast interface selection

On Thu, Mar 14, 2024 at 12:04:47PM +, Ignjatović, Lazar (RS) wrote: > localaddr option now properly works with IPv6 addresses. Properly > resolved interface index in places where default 0 interface index is > used (marked with TODO: within udp.c). Added SO_BINDTODEVICE for mcast > sockets that

Re: [FFmpeg-devel] [PATCH] avcodec/mpeg4videodec: Align idct-block appropriately

Sean McGovern: > Andreas: > > On Tue, Mar 12, 2024 at 8:53 PM Andreas Rheinhardt > wrote: >> >> It is accessed via AV_RN64() in ff_simple_idct_put_int32_10bit(). >> Should fix the UBSan failures in the mpeg4-simple-studio-profile >> test here: >> https://fate.ffmpeg.org/report.cgi?time=2024031201

Re: [FFmpeg-devel] [PATCH] avcodec/mpeg4videodec: Align idct-block appropriately

Andreas: On Tue, Mar 12, 2024 at 8:53 PM Andreas Rheinhardt wrote: > > It is accessed via AV_RN64() in ff_simple_idct_put_int32_10bit(). > Should fix the UBSan failures in the mpeg4-simple-studio-profile > test here: > https://fate.ffmpeg.org/report.cgi?time=20240312011016&slot=ppc-linux-gcc-13.2

Re: [FFmpeg-devel] [PATCH] avcodec/lossless_videoencdsp: Fix unaligned access

Andreas: On Tue, Mar 12, 2024 at 8:42 PM Andreas Rheinhardt wrote: > > HAVE_FAST_UNALIGNED being true does not imply that > one can simply read from any pointer via *(long*). > It is undefined behaviour in case the pointer is not > sufficiently aligned; and even if it is, it is (likely) > a viola

Re: [FFmpeg-devel] [PATCH] avformat/avidec: Fix integer overflow iff ULONG_MAX < INT64_MAX

Andreas: On Tue, Mar 12, 2024 at 6:57 PM Andreas Rheinhardt wrote: > > Affects many FATE-tests, see > http://fate.ffmpeg.org/report.cgi?time=20240312011016&slot=ppc-linux-gcc-13.2-ubsan-altivec-qemu > > Signed-off-by: Andreas Rheinhardt > --- > libavformat/avidec.c | 2 +- > 1 file changed, 1 i

Re: [FFmpeg-devel] [PATCH v2] avcodec/ppc/h264dsp: Fix unaligned stores

Sean McGovern: > Andreas: > > On Wed, Mar 13, 2024 at 7:31 AM Andreas Rheinhardt > wrote: >> >> Also fix an effective-type violation. >> Exposed by >> https://fate.ffmpeg.org/report.cgi?time=20240312011016&slot=ppc-linux-gcc-13.2-ubsan-altivec-qemu >> >> Signed-off-by: Andreas Rheinhardt >> ---

Re: [FFmpeg-devel] [PATCH v2] avcodec/ppc/h264dsp: Fix unaligned stores

On 3/14/2024 4:13 PM, Sean McGovern wrote: Andreas: On Wed, Mar 13, 2024 at 7:31 AM Andreas Rheinhardt wrote: Also fix an effective-type violation. Exposed by https://fate.ffmpeg.org/report.cgi?time=20240312011016&slot=ppc-linux-gcc-13.2-ubsan-altivec-qemu Signed-off-by: Andreas Rheinhardt

Re: [FFmpeg-devel] [PATCH v2] avcodec/ppc/h264dsp: Fix unaligned stores

Andreas: On Wed, Mar 13, 2024 at 7:31 AM Andreas Rheinhardt wrote: > > Also fix an effective-type violation. > Exposed by > https://fate.ffmpeg.org/report.cgi?time=20240312011016&slot=ppc-linux-gcc-13.2-ubsan-altivec-qemu > > Signed-off-by: Andreas Rheinhardt > --- > libavcodec/ppc/h264dsp.c |

[FFmpeg-devel] [PATCH 1/1] avformat/demux: Fix accurate probing of durations in mpegts/ps

Two issues affect accuracy of duration in estimate_timings_from_pts(): - pkt->duration typically reports the duration of a single audio frame, whereas a pes often contain several audio frames - for video, compute_frame_duration() use r_frame_rate which is not reliable; typically, it is the duration

[FFmpeg-devel] [PATCH 0/1] avformat/demux: Fix accurate probing of durations in mpegts/ps

ff_read_packet() is more lightweight, but it leads to important issues when looking for accurate durations. As a side effect, the code looks also simpler with regular av_read_frame() calls. 1) Updates in the fate tests do exhibit most of the results. 2) See also more directly the case of an audi

[FFmpeg-devel] [PATCH] doc/filters: Change rdiv (vf_convolution) documentation to reflect actual behavior

The documentation correctly states that the rdiv is a multiplier but incorrectly states the default behavior is to multiply by the sum of all matrix elements - it multiplies by 1/sum. This changes the documentation to match the code. Address trac #10889 --- doc/filters.texi | 2 +- 1 file chan

[FFmpeg-devel] [PATCH] tools/target_dec_fuzzer: force experimental flag for VVC

Otherwise the decoder will not even open. Signed-off-by: James Almer --- tools/target_dec_fuzzer.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/tools/target_dec_fuzzer.c b/tools/target_dec_fuzzer.c index 84b646b7f4..459b07070e 100644 --- a/tools/target_dec_fuzzer.c +++ b/tools/target_d

Re: [FFmpeg-devel] [PATCH] avformat/mxfdec: signal channel layouts using the new channel layout api

fre 2024-03-08 klockan 21:13 +0100 skrev Marton Balint: > Signed-off-by: Marton Balint > --- >  libavformat/mxfdec.c | 129 ++--- > -- >  1 file changed, 53 insertions(+), 76 deletions(-) No idea how the new API works, but it's less code which seems nice. Chang

Re: [FFmpeg-devel] [PATCH 1/4] aarch64: Fix ff_hevc_put_hevc_epel_h48_8_neon_i8mm

On Thu, 14 Mar 2024, J. Dekker wrote: Martin Storsjö writes: The first 32 elements of each row were correct, while the last 16 were scrambled. This hasn't been noticed, because the checkasm test erroneously only checked half of the output (for 8 bit functions), and apparently none of the sa

Re: [FFmpeg-devel] [PATCH v5] avcodec/libx264: fix extradata when config annexb=0

Zhao Zhili: > From: Zhao Zhili > > --- > v5: Decode chroma_format_idc directly instead of > ff_h264_decode_seq_parameter_set > v4: Fix missing SEI in set_avcc_extradata > v3: Remove unnecessary inclusion > > libavcodec/libx264.c | 162 --- > 1 file chang

Re: [FFmpeg-devel] [PATCH 1/4] aarch64: Fix ff_hevc_put_hevc_epel_h48_8_neon_i8mm

Martin Storsjö writes: > The first 32 elements of each row were correct, while the > last 16 were scrambled. > > This hasn't been noticed, because the checkasm test erroneously > only checked half of the output (for 8 bit functions), and > apparently none of the samples as part of "fate-hevc" se

[FFmpeg-devel] [PATCH] avformat: enable UDP IPv6 multicast interface selection

localaddr option now properly works with IPv6 addresses. Properly resolved interface index in places where default 0 interface index is used (marked with TODO: within udp.c). Added SO_BINDTODEVICE for mcast sockets that are used for reading from the network. Need for this arises from the fact that

Re: [FFmpeg-devel] [PATCH v5] libavfi/dnn: add LibTorch as one of DNN backend

> -Original Message- > From: ffmpeg-devel On Behalf Of > wenbin.chen-at-intel@ffmpeg.org > Sent: Monday, March 11, 2024 1:02 PM > To: ffmpeg-devel@ffmpeg.org > Subject: [FFmpeg-devel] [PATCH v5] libavfi/dnn: add LibTorch as one of DNN > backend > > From: Wenbin Chen > > PyTorch i

Re: [FFmpeg-devel] [PATCH v3 0/5] avcodec/parser: fix fetch_timestamp in a scenario with unaligned packets

>De : Nicolas Gaullier >Envoyé : lundi 4 mars 2024 18:32 >Objet : [PATCH v3 0/5] avcodec/parser: fix fetch_timestamp in a scenario with >unaligned packets > >Updated from v2: >patch 1: fix audio case where pts=AV_NOPTS_VALUE but dts exists (thanks to >Michael) >now pass fate with --assert-l

Re: [FFmpeg-devel] [PATCH v2 9/9] avformat/pcm: decrease target audio frame per sec to 10

This is just ugly hack for workaround around broken ffmpeg threading. On Thu, Mar 14, 2024 at 9:10 AM Tobias Rapp wrote: > On 08/03/2024 01:53, Marton Balint wrote: > > > This makes the wav and pcm demuxer demux bigger packets, which is more > > efficient. > > > > Signed-off-by: Marton Balint >

Re: [FFmpeg-devel] [PATCH] avformat: enable UDP IPv6 multicast interface selection

>> +if (if_indextoname(iface, ifr.ifr_name)) { >> +if (setsockopt(udp_fd, SOL_SOCKET, SO_BINDTODEVICE, (void >> *)&ifr, sizeof(ifr)) < 0) { >> +perror("setsockopt SO_BINDTODEVICE"); > > This should use the proper av_log logging instead. Will fix and

[FFmpeg-devel] [PATCH] libavcodec/amfenc: Update AMF encoder options

The encoder options have been updated to the current version of the AMF. Signed-off-by: Araz Iusubov --- libavcodec/amfenc.c | 1 + libavcodec/amfenc.h | 4 + libavcodec/amfenc_av1.c | 154 +- libavcodec/amfenc_h264.c | 155 +

Re: [FFmpeg-devel] [PATCH 5/6] avcodec/tiff_data: Avoid relocations for TiffGeoTagNameType

Andreas Rheinhardt: > Instead store all the strings in one continugous string > (with internal \0) and use offsets to access the actual > substrings. This replaces the pointers to the strings > and therefore avoids relocations (and on x64, it actually > shrinks TiffGeoTagNameType by reusing padding

Re: [FFmpeg-devel] [FFmpeg-cvslog] fftools/ffmpeg: add loopback decoding

Quoting Gyan Doshi (2024-03-13 08:49:52) > > > On 2024-03-13 12:57 pm, Anton Khirnov wrote: > > ffmpeg | branch: master | Anton Khirnov | Mon Feb 19 > > 10:27:44 2024 +0100| [a9193f7b7d65aafa326e25571c6672636a8ee3d2] | > > committer: Anton Khirnov > > > > fftools/ffmpeg: add loopback decoding

[FFmpeg-devel] [PATCH] doc/ffmpeg: mention how to pass options to loopback decoders

--- doc/ffmpeg.texi | 14 ++ 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/doc/ffmpeg.texi b/doc/ffmpeg.texi index a38ef834e1..801c083705 100644 --- a/doc/ffmpeg.texi +++ b/doc/ffmpeg.texi @@ -229,12 +229,16 @@ successive integers starting at zero. These indices shoul

Re: [FFmpeg-devel] [PATCH] avformat: enable UDP IPv6 multicast interface selection

On 14 Mar 2024, at 10:22, Ignjatović, Lazar wrote: > avformat: enable UDP IPv6 multicast interface selection > > localaddr option now properly works with IPv6 addresses. Properly resolved > interface index in places where default 0 interface index is used (marked with > TODO: within udp.c). Adde

[FFmpeg-devel] [PATCH] avformat: enable UDP IPv6 multicast interface selection

avformat: enable UDP IPv6 multicast interface selection localaddr option now properly works with IPv6 addresses. Properly resolved interface index in places where default 0 interface index is used (marked with TODO: within udp.c). Added SO_BINDTODEVICE for mcast sockets that are used for reading f

[FFmpeg-devel] [PATCH v7 12/12] Changelog: add D3D12VA HEVC encoder changelog

From: Tong Wu Signed-off-by: Tong Wu --- Changelog | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Changelog b/Changelog index b7d7535a9e..81ee3a7837 100644 --- a/Changelog +++ b/Changelog @@ -34,7 +34,7 @@ version : - ffprobe (with -export_side_data film_grain) now prints

[FFmpeg-devel] [PATCH v7 11/12] avcodec: add D3D12VA hardware HEVC encoder

From: Tong Wu This implementation is based on D3D12 Video Encoding Spec: https://microsoft.github.io/DirectX-Specs/d3d/D3D12VideoEncoding.html Sample command line for transcoding: ffmpeg.exe -hwaccel d3d12va -hwaccel_output_format d3d12 -i input.mp4 -c:v hevc_d3d12va output.mp4 Signed-off-by: T

[FFmpeg-devel] [PATCH v7 10/12] avutil/hwcontext_d3d12va: add Flags for resource creation

From: Tong Wu Flags field is added to support diffferent resource creation. Signed-off-by: Tong Wu --- doc/APIchanges| 3 +++ libavutil/hwcontext_d3d12va.c | 2 +- libavutil/hwcontext_d3d12va.h | 8 libavutil/version.h | 2 +- 4 files changed, 13 insertions(+

[FFmpeg-devel] [PATCH v7 09/12] avcodec/vaapi_encode: extract a free funtion to base layer

From: Tong Wu Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 11 +++ libavcodec/hw_base_encode.h | 2 ++ libavcodec/vaapi_encode.c | 6 +- 3 files changed, 14 insertions(+), 5 deletions(-) diff --git a/libavcodec/hw_base_encode.c b/libavcodec/hw_base_encode.c index d2

[FFmpeg-devel] [PATCH v7 08/12] avcodec/vaapi_encode: extract a get_recon_format function to base layer

From: Tong Wu Get constraints and set recon frame format can be shared with other HW encoder such as D3D12. Extract this part as a new function to base layer. Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 58 + libavcodec/hw_base_encode.h | 2 ++

[FFmpeg-devel] [PATCH v7 07/12] avcodec/vaapi_encode: extract gop configuration to base layer

From: Tong Wu Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 54 + libavcodec/hw_base_encode.h | 3 +++ libavcodec/vaapi_encode.c | 52 +++ 3 files changed, 61 insertions(+), 48 deletions(-) diff --git a/libavcode

[FFmpeg-devel] [PATCH v7 06/12] avcodec/vaapi_encode: extract set_output_property to base layer

From: Tong Wu Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 40 + libavcodec/hw_base_encode.h | 3 +++ libavcodec/vaapi_encode.c | 44 ++--- 3 files changed, 45 insertions(+), 42 deletions(-) diff --git a/libavcodec/

[FFmpeg-devel] [PATCH v7 05/12] avcodec/vaapi_encode: extract a close function for base layer

From: Tong Wu Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 16 libavcodec/hw_base_encode.h | 2 ++ libavcodec/vaapi_encode.c | 8 +--- 3 files changed, 19 insertions(+), 7 deletions(-) diff --git a/libavcodec/hw_base_encode.c b/libavcodec/hw_base_encode.c i

[FFmpeg-devel] [PATCH v7 04/12] avcodec/vaapi_encode: extract a init function to base layer

From: Tong Wu Move the base_ctx parameter initialization to base layer. Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.c | 33 + libavcodec/hw_base_encode.h | 2 ++ libavcodec/vaapi_encode.c | 36 3 files changed, 39

[FFmpeg-devel] [PATCH v7 03/12] avcodec/vaapi_encode: move the dpb logic from VAAPI to base layer

From: Tong Wu Move receive_packet function to base. Add *alloc, *issue, *output, *free as hardware callback. DPB management logic can be fully extracted to base layer as-is. Signed-off-by: Tong Wu --- libavcodec/Makefile | 2 +- libavcodec/hw_base_encode.c | 599 +

[FFmpeg-devel] [PATCH v7 02/12] avcodec/vaapi_encode: introduce a base layer for vaapi encode

From: Tong Wu Since VAAPI and future D3D12VA implementation may share some common parameters, a base layer encode context is introduced as vaapi context's base. Signed-off-by: Tong Wu --- libavcodec/hw_base_encode.h | 241 libavcodec/vaapi_encode.c | 392

[FFmpeg-devel] [PATCH v7 01/12] avcodec/vaapi_encode: move pic->input_surface initialization to encode_alloc

From: Tong Wu When allocating the VAAPIEncodePicture, pic->input_surface can be initialized right in the place. This movement simplifies the send_frame logic and is the preparation for moving vaapi_encode_send_frame to the base layer. Signed-off-by: Tong Wu --- libavcodec/vaapi_encode.c | 8 +

Re: [FFmpeg-devel] [PATCH v2 9/9] avformat/pcm: decrease target audio frame per sec to 10

On 08/03/2024 01:53, Marton Balint wrote: This makes the wav and pcm demuxer demux bigger packets, which is more efficient. Signed-off-by: Marton Balint --- libavformat/pcm.c | 2 +- tests/ref/acodec/pcm-s16be| 2 +- tests/ref/acodec/pcm-s

Re: [FFmpeg-devel] Add protocol for Android content providers (v2)

On Mon, Mar 04, 2024 at 09:21:43AM +0100, Matthieu Bouron wrote: > On Tue, Feb 27, 2024 at 03:50:36PM +0100, Matthieu Bouron wrote: > > Diff from initial patchset: > > - directly include libavcodec/ffjni.c from libavformat/file.c instead of > > migrating ffjni to libavutil (avpriv_*) > > - check