Re: [FFmpeg-devel] [PATCH v2] doc/encoders: document libjxl encoder options

Pushed as ec07b1547753864d1fb47eb2747613dddba507b2 On 2022-04-27 08:30 pm, Leo Izen wrote: Add more detailed documenation for the libjxl encoder wrapper than is present currently inside libavcodec. --- doc/encoders.texi | 35 +++ 1 file changed, 35 insertions(+

[FFmpeg-devel] [PATCH v08 00/10] make QSV works with the Intel's oneVPL

The oneAPI Video Processing Library (oneVPL) is a single interface for encode, decode and video processing[1][2]. oneVPL is a successor to Intel(R) Media SDK, but removed obsolete features. Intel(R) Media SDK lifetime comes to an end now, new features for new Intel Gen platforms will be supported

[FFmpeg-devel] [PATCH v08 01/10] configure: ensure --enable-libmfx uses libmfx 1.x

Intel's oneVPL is a successor to MediaSDK, but removed some obsolete features of MediaSDK[1]. Some early versions of oneVPL still uses libmfx as library name[2], however some of obsolete features, including OPAQUE memory, multi-frame encode, user plugins and LA_EXT rate control mode etc, have been

[FFmpeg-devel] [PATCH v08 02/10] configure: fix the check for MFX_CODEC_VP9

The data structures for VP9 in mfxvp9.h is wrapped by MFX_VERSION_NEXT, which means those data structures have never been used in a public release. Actually MFX_CODEC_VP9 and other VP9 stuffs is added in mfxstructures.h. In addition, mfxdefs.h is included in mfxvp9.h, so we may use the check in thi

[FFmpeg-devel] [PATCH v08 03/10] qsv: remove mfx/ prefix from mfx headers

The following Cflags has been added to libmfx.pc, so mfx/ prefix is no longer needed when including mfx headers in FFmpeg. Cflags: -I${includedir} -I${includedir}/mfx Some old versions of libmfx have the following Cflags in libmfx.pc Cflags: -I${includedir} We may add -I${includedir}/mfx to

[FFmpeg-devel] [PATCH v08 04/10] qsv: load user plugin for MFX_VERSION < 2.0

User plugin isn't supported for MFX_VERSION >= 2.0[1][2]. This is in preparation for oneVPL Support [1]: https://spec.oneapi.io/versions/latest/elements/oneVPL/source/VPL_intel_media_sdk.html#msdk-full-name-feature-removals [2]: https://github.com/oneapi-src/oneVPL --- libavcodec/qsv.c

[FFmpeg-devel] [PATCH v08 05/10] qsv: build audio related code when MFX_VERSION < 2.0

Audio isn't supported for MFX_VERSION >= 2.0[1][2]. This is in preparation for oneVPL support [1]: https://spec.oneapi.io/versions/latest/elements/oneVPL/source/VPL_intel_media_sdk.html#msdk-full-name-feature-removals [2]: https://github.com/oneapi-src/oneVPL --- libavcodec/qsv.c | 5 +

[FFmpeg-devel] [PATCH v08 06/10] qsvenc: support multi-frame encode when MFX_VERSION < 2.0

Multi-frame encode isn't supported for MFX_VERSION >= 2.0[1][2]. This is in preparation for oneVPL support [1]: https://spec.oneapi.io/versions/latest/elements/oneVPL/source/VPL_intel_media_sdk.html#msdk-full-name-feature-removals [2]: https://github.com/oneapi-src/oneVPL --- libavcodec/qsvenc.h

[FFmpeg-devel] [PATCH v08 08/10] qsv: support OPAQUE memory when MFX_VERSION < 2.0

OPAQUE memory isn't supported for MFX_VERSION >= 2.0[1][2]. This is in preparation for oneVPL support [1]: https://spec.oneapi.io/versions/latest/elements/oneVPL/source/VPL_intel_media_sdk.html#msdk-full-name-feature-removals [2]: https://github.com/oneapi-src/oneVPL --- libavcodec/qsv.c

[FFmpeg-devel] [PATCH v08 07/10] qsvenc: support MFX_RATECONTROL_LA_EXT when MFX_VERSION < 2.0

MFX_RATECONTROL_LA_EXT isn't supported for MFX_VERSION >= 2.0[1][2]. This is in preparation for oneVPL support [1]: https://spec.oneapi.io/versions/latest/elements/oneVPL/source/VPL_intel_media_sdk.html#msdk-full-name-feature-removals [2]: https://github.com/oneapi-src/oneVPL --- libavcodec/qsve

[FFmpeg-devel] [PATCH v08 09/10] qsv: use a new method to create mfx session when using oneVPL

In oneVPL, MFXLoad() and MFXCreateSession() are required to create a workable mfx session[1] Add config filters for D3D9/D3D11 session (galinart) The default device is changed to d3d11va for oneVPL when both d3d11va and dxva2 are enabled on Microsoft Windows This is in preparation for oneVPL sup

[FFmpeg-devel] [PATCH v08 10/10] configure: add --enable-libvpl option

This allows user to build FFmpeg against Intel oneVPL. oneVPL 2.6 is the required minimum version when building Intel oneVPL code. It will fail to run configure script if both libmfx and libvpl are enabled. It is recommended to use oneVPL for new work, even for currently available hardwares [1]

Re: [FFmpeg-devel] [PATCH v7 09/10] qsv: use a new method to create mfx session when using oneVPL

On Wed, 2022-04-06 at 03:58 +, Xiang, Haihao wrote: [...] > > > > > > > > https://spec.oneapi.io/versions/latest/elements/oneVPL/source/API_ref/VPL_disp_api_struct.html#structmfx_impl_description > > > > > ) and user must select an available implementation before the creation > > > > > of

Re: [FFmpeg-devel] [PATCH] avcodec/wmadec: fix WMA gapless playback

Hi, I think this patch, which was applied on 2021-09-12 in 19802d170a304f5853d92e01d0513b9e06897d61 and is included in n5.0 and n5.0.1 causes a regression. This is the (shortened) diff for the test decoding in the patch: diff --git a/tests/ref/fate/flcl1905 b/tests/ref/fate/f

Re: [FFmpeg-devel] [PATCH v3 2/2] avformat/movenc: sidx earliest_presentation_time is applied after editlist

> On Mar 29, 2022, at 12:50 PM, Zhao Zhili wrote: > > Fix #8334 > --- > libavformat/movenc.c | 6 +- > tests/ref/fate/movenc | 10 +- > 2 files changed, 10 insertions(+), 6 deletions(-) > > diff --git a/libavformat/movenc.c b/libavformat/movenc.c > index 46d66c29c2..079fc70d4a 100

[FFmpeg-devel] [PATCH] Fix AVX-512-VNNI__hevc_qpel_filters_avx512icl_v_%1

Issue: On extremely new hardware using either IceLake or super sets of Intel IceLakes avx512 instructions, commit d4cd8830bdac3e26c8e75cd92e574c159fecc4f7 causes build issues. Specifically a NASM macro expansion of qpel_filter_v is never properly defined/initialized. The issue is the definition

Re: [FFmpeg-devel] [PATCH] avformat/avio: use av_match_name to match protocol lists

On 4/20/22 16:33, Leo Izen wrote: This swaps av_match_list to av_match_name inside ffurl_connect, which allows ALL to be treated as a valid value on the whitelist or blacklist in accordance with the behavior described in the documentation. --- libavformat/avio.c | 4 ++-- 1 file changed, 2 ins

[FFmpeg-devel] [PATCH-v2] Fix AVX-512-VNNI__hevc_qpel_filters_avx512icl_v_%_2

Issue: On extremely new hardware using either IceLake or super sets of Intel IceLakes avx512 instructions, commit d4cd8830bdac3e26c8e75cd92e574c159fecc4f7 causes build issues. Specifically a NASM macro expansion of qpel_filter_v is never properly defined/initialized. The issue is the definition

[FFmpeg-devel] [PATCH v5 1/4] doc/examples/transcode_aac: Don't ignore last encoded frame

The last encoded frame is now fetched on EOF. It was previously left in the encoder and caused a "1 frame left in queue" warning. Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 22 +++--- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/doc/ex

[FFmpeg-devel] [PATCH v5 2/4] doc/examples/transcode_aac: Set decoder packet timebase

Previously, the default timebase caused two warnings during decoding about not being able to update timestamps for skipped and discarded samples, respectively. Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 10 -- 1 file changed, 8 insertions(+), 2 deletions(-) dif

[FFmpeg-devel] [PATCH v5 3/4] doc/examples/transcode_aac: AAC encoder is no longer experimental

Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/doc/examples/transcode_aac.c b/doc/examples/transcode_aac.c index 52548f080d..909aa1cb73 100644 --- a/doc/examples/transcode_aac.c +++ b/doc/examples/transcode_aac.c @@ -211,9

[FFmpeg-devel] [PATCH v5 4/4] doc/examples/transcode_aac: Bump date

Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/examples/transcode_aac.c b/doc/examples/transcode_aac.c index 909aa1cb73..1b372782ab 100644 --- a/doc/examples/transcode_aac.c +++ b/doc/examples/transcode_aa

[FFmpeg-devel] [PATCH 1/3] lavc/aarch64: fix hevc sao band filter

The SAO band filter can be called with non-multiples of 8, we round up to the nearest multiple of 8 to account for this. Signed-off-by: J. Dekker --- libavcodec/aarch64/hevcdsp_init_aarch64.c | 10 +- libavcodec/aarch64/hevcdsp_sao_neon.S | 8 ++-- 2 files changed, 11 insertions

[FFmpeg-devel] [PATCH 2/3] lavc/aarch64: add hevc sao edge 16x16

bench on AWS Graviton: hevc_sao_edge_16x16_8_c: 1857.0 hevc_sao_edge_16x16_8_neon: 211.0 hevc_sao_edge_32x32_8_c: 7802.2 hevc_sao_edge_32x32_8_neon: 808.2 hevc_sao_edge_48x48_8_c: 16764.2 hevc_sao_edge_48x48_8_neon: 1796.5 hevc_sao_edge_64x64_8_c: 32647.5 hevc_sao_edge_64x64_8_neon: 3118.5 Signed

[FFmpeg-devel] [PATCH 3/3] lavc/aarch64: add hevc sao edge 8x8

bench on AWS Graviton: hevc_sao_edge_8x8_8_c: 516.0 hevc_sao_edge_8x8_8_neon: 81.0 Signed-off-by: J. Dekker --- libavcodec/aarch64/hevcdsp_init_aarch64.c | 3 ++ libavcodec/aarch64/hevcdsp_sao_neon.S | 51 +++ 2 files changed, 54 insertions(+) diff --git a/libavcodec/a

Re: [FFmpeg-devel] [PATCH 1/3] lavc/aarch64: fix hevc sao band filter

On 28 Apr 2022, at 15:42, J. Dekker wrote: > The SAO band filter can be called with non-multiples of 8, we round up > to the nearest multiple of 8 to account for this. Martin mentioned he wanted checkasm to check these extra cases. Still working on the best way to do this, will post soon. -- J

[FFmpeg-devel] [PATCH v6 1/4] doc/examples/transcode_aac: Don't ignore last encoded frame

The last encoded frame is now fetched on EOF. It was previously left in the encoder and caused a "1 frame left in queue" warning. Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 22 +++--- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/doc/ex

[FFmpeg-devel] [PATCH v6 2/4] doc/examples/transcode_aac: Set decoder packet timebase

Previously, the default timebase caused two warnings during decoding about not being able to update timestamps for skipped and discarded samples, respectively. Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 10 -- 1 file changed, 8 insertions(+), 2 deletions(-) dif

[FFmpeg-devel] [PATCH v6 3/4] doc/examples/transcode_aac: AAC encoder is no longer experimental

Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/doc/examples/transcode_aac.c b/doc/examples/transcode_aac.c index 52548f080d..909aa1cb73 100644 --- a/doc/examples/transcode_aac.c +++ b/doc/examples/transcode_aac.c @@ -211,9

[FFmpeg-devel] [PATCH v6 4/4] doc/examples/transcode_aac: Bump date

Signed-off-by: Andreas Unterweger --- doc/examples/transcode_aac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/examples/transcode_aac.c b/doc/examples/transcode_aac.c index 909aa1cb73..1b372782ab 100644 --- a/doc/examples/transcode_aac.c +++ b/doc/examples/transcode_aa

Re: [FFmpeg-devel] [PATCH v4 2/2] doc/examples/transcode_aac: Set decoder packet timebase

On Thu, 28 Apr 2022 at 05:23, Andreas Rheinhardt wrote: > > Andreas Unterweger: > > Previously, the default timebase caused two warnings during decoding > > about not being able to update timestamps for skipped and discarded > > samples, respectively. > > > > Signed-off-by: Andreas Unterweger > >

[FFmpeg-devel] [PATCH] Fix AVX-512-VNNI_hevc_qpel_filters_avx512icl

Issue: On extremely new hardware using either IceLake or super sets of Intel IceLakes avx512 instructions, commit d4cd8830bdac3e26c8e75cd92e574c159fecc4f7 causes build issues. Specifically a NASM macro expansion of qpel_filter_v is never properly defined/initialized. The issue is the definiti

[FFmpeg-devel] [PATCH] avcodec/h264dec: Skip late SEI

Fixes: Race condition Fixes: clusterfuzz-testcase-minimized-mediasource_MP2T_AVC_pipeline_integration_fuzzer-6282675434094592 Found-by: google ClusterFuzz Tested-by: Dan Sanders Signed-off-by: Michael Niedermayer --- libavcodec/h264dec.c | 4 1 file changed, 4 insertions(+) diff --git a/

Re: [FFmpeg-devel] [PATCH 3/3] lavc/aarch64: add hevc sao edge 8x8

On Thu, 28 Apr 2022, J. Dekker wrote: bench on AWS Graviton: hevc_sao_edge_8x8_8_c: 516.0 hevc_sao_edge_8x8_8_neon: 81.0 Signed-off-by: J. Dekker --- libavcodec/aarch64/hevcdsp_init_aarch64.c | 3 ++ libavcodec/aarch64/hevcdsp_sao_neon.S | 51 +++ 2 files changed, 54 in

Re: [FFmpeg-devel] [PATCH] avformat/avio: use av_match_name to match protocol lists

On Thu, Apr 28, 2022 at 08:43:21AM -0400, Leo Izen wrote: > On 4/20/22 16:33, Leo Izen wrote: > > This swaps av_match_list to av_match_name inside ffurl_connect, > > which allows ALL to be treated as a valid value on the whitelist > > or blacklist in accordance with the behavior described in the >

[FFmpeg-devel] [PATCH 01/13] fate/qt: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/qt.mak | 34 +- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/tests/fate/qt.mak b/tests/fate/qt.mak index 6e71f6e4d1..42e5fd9107 100644 --- a/tests/fate/qt.mak +++ b/tests/fate/qt.mak @@ -1,52 +1,52

[FFmpeg-devel] [PATCH 02/13] fate/canopus: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/canopus.mak | 9 ++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tests/fate/canopus.mak b/tests/fate/canopus.mak index 3d664ff911..4e50716813 100644 --- a/tests/fate/canopus.mak +++ b/tests/fate/canopus.mak @@ -8,7 +8,8 @@ f

[FFmpeg-devel] [PATCH 03/13] fate/dnxhd: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/dnxhd.mak | 15 +-- 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/tests/fate/dnxhd.mak b/tests/fate/dnxhd.mak index 59f3f90e91..f261766b53 100644 --- a/tests/fate/dnxhd.mak +++ b/tests/fate/dnxhd.mak @@ -1,15 +1,18 @@ -F

[FFmpeg-devel] [PATCH 04/13] fate/ea: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/ea.mak | 18 -- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/tests/fate/ea.mak b/tests/fate/ea.mak index 7667102ce7..2d1ac6e50c 100644 --- a/tests/fate/ea.mak +++ b/tests/fate/ea.mak @@ -1,25 +1,23 @@ -FATE_SAMPLES

[FFmpeg-devel] [PATCH 05/13] fate/vbn: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/vbn.mak | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/tests/fate/vbn.mak b/tests/fate/vbn.mak index 044b110ac1..6c3fe82c99 100644 --- a/tests/fate/vbn.mak +++ b/tests/fate/vbn.mak @@ -30,7 +30,9 @@ tests/data/raw-rgb24

[FFmpeg-devel] [PATCH 06/13] fate/vbn, tests/fate-run: Delete intermediate VBN-files after tests

Signed-off-by: Andreas Rheinhardt --- tests/fate-run.sh | 22 ++ tests/fate/vbn.mak | 24 ++-- 2 files changed, 20 insertions(+), 26 deletions(-) diff --git a/tests/fate-run.sh b/tests/fate-run.sh index a96ff049b0..3c62246c8a 100755 --- a/tests/fate-run.s

[FFmpeg-devel] [PATCH 07/13] fate/vpx: Remove unnecessary eval

Possible since ca96e337169093979d7c763064ad9dae12b3108c and ef8c93e2f18c624d0c266687e43ab99af7921dd3. Signed-off-by: Andreas Rheinhardt --- tests/fate/vpx.mak | 27 ++- 1 file changed, 10 insertions(+), 17 deletions(-) diff --git a/tests/fate/vpx.mak b/tests/fate/vpx.mak

[FFmpeg-devel] [PATCH 08/13] tests/Makefile: Add FRAMEMD5 auxiliary function

Similar to FRAMECRC for framecrc tests. Signed-off-by: Andreas Rheinhardt --- tests/Makefile | 4 1 file changed, 4 insertions(+) diff --git a/tests/Makefile b/tests/Makefile index 6e8fc5906e..b4252a991b 100644 --- a/tests/Makefile +++ b/tests/Makefile @@ -88,6 +88,10 @@ REMUX = $(call ALL

[FFmpeg-devel] [PATCH 09/13] fate/vpx: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/Makefile | 2 +- tests/fate/vpx.mak | 68 -- 2 files changed, 36 insertions(+), 34 deletions(-) diff --git a/tests/Makefile b/tests/Makefile index b4252a991b..5de9ad4c9f 100644 --- a/tests/Makefile +++ b

[FFmpeg-devel] [PATCH 10/13] fate/vpx: Move webm-dash-manifest tests to a new file

These tests have basically nothing to do with VPX (they do not even require the decoder). Signed-off-by: Andreas Rheinhardt --- tests/Makefile| 1 + tests/fate/vpx.mak| 20 tests/fate/webm-dash-manifest.mak | 21 + 3 f

[FFmpeg-devel] [PATCH 11/13] fate/prores: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/prores.mak | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/fate/prores.mak b/tests/fate/prores.mak index f2c11259dd..9e05b33b9f 100644 --- a/tests/fate/prores.mak +++ b/tests/fate/prores.mak @@ -6,9 +6,9 @@ FATE_PROR

[FFmpeg-devel] [PATCH 12/13] fate/gif: Fix test requirements

Signed-off-by: Andreas Rheinhardt --- tests/fate/gif.mak | 11 +-- 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/tests/fate/gif.mak b/tests/fate/gif.mak index a5147c0e38..bf604464aa 100644 --- a/tests/fate/gif.mak +++ b/tests/fate/gif.mak @@ -13,18 +13,17 @@ fate-gif-gray:

[FFmpeg-devel] [PATCH 13/13] fate/gif: Remove nonsense requirement

It seems as if it was intended to declare fate-gif-color as prerequisite of the fate-gifenc% tests. Yet the latter do not need anything from the former, so this would be unnecessary. Furthermore, given that this line has no associated recipe, it actually cancels implicit rules for fate-gifenc% inst

Re: [FFmpeg-devel] [PATCH] avformat/avio: use av_match_name to match protocol lists

On 4/28/22 16:28, Michael Niedermayer wrote: On Thu, Apr 28, 2022 at 08:43:21AM -0400, Leo Izen wrote: On 4/20/22 16:33, Leo Izen wrote: This swaps av_match_list to av_match_name inside ffurl_connect, which allows ALL to be treated as a valid value on the whitelist or blacklist in accordance

[FFmpeg-devel] [PATCH 1/4] avfilter/vf_colormap: Properly uninit FFFrameSync, fix leak

Signed-off-by: Andreas Rheinhardt --- libavfilter/vf_colormap.c | 8 1 file changed, 8 insertions(+) diff --git a/libavfilter/vf_colormap.c b/libavfilter/vf_colormap.c index 1e82514ade..9edc89f105 100644 --- a/libavfilter/vf_colormap.c +++ b/libavfilter/vf_colormap.c @@ -537,6 +537,13 @

[FFmpeg-devel] [PATCH 2/4] avfilter/vf_colormap: Fix declaration-after-statement

Signed-off-by: Andreas Rheinhardt --- libavfilter/vf_colormap.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/libavfilter/vf_colormap.c b/libavfilter/vf_colormap.c index 9edc89f105..a0d6d7be27 100644 --- a/libavfilter/vf_colormap.c +++ b/libavfilter/vf_colormap.c @@ -129,8

[FFmpeg-devel] [PATCH 3/4] avfilter/vf_colormap: Remove unnecessary headers

In particular remove config_components.h in order to avoid unnecessary rebuilds. Signed-off-by: Andreas Rheinhardt --- libavfilter/vf_colormap.c | 5 - 1 file changed, 5 deletions(-) diff --git a/libavfilter/vf_colormap.c b/libavfilter/vf_colormap.c index a0d6d7be27..d5c5bec39c 100644 --- a

[FFmpeg-devel] [PATCH 4/4] avfilter/vf_colormap: Avoid allocation of small array

The number of elements is always two or three. Signed-off-by: Andreas Rheinhardt --- libavfilter/vf_colormap.c | 12 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/libavfilter/vf_colormap.c b/libavfilter/vf_colormap.c index d5c5bec39c..106333ced8 100644 --- a/libavfil

Re: [FFmpeg-devel] [PATCH] Fix AVX-512-VNNI_hevc_qpel_filters_avx512icl

> Felix LeClair: > Sent: 2022年4月29日 1:17 > To: ffmpeg-devel@ffmpeg.org > Subject: [FFmpeg-devel] [PATCH] Fix AVX-512-VNNI_hevc_qpel_filters_avx512icl > > This commit fixes the above by swapping lines 1796 and 1795, moving the > de

Re: [FFmpeg-devel] [PATCH v5 1/2] lavc/vaapi_encode: add support for maxframesize

On Fri, 2022-04-22 at 13:36 +0800, Fei Wang wrote: > From: Linjie Fu > > Add support for max frame size: > - max_frame_size (bytes) to indicate the max allowed size for frame. > > If the frame size exceeds the limitation, encoder will to control the frame > size by adjusting QP value. > > f