[FFmpeg-cvslog] lavc/sbrdsp: R-V V neg_odd_64

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 29 18:22:08 2023 +0200| [d06fd18f8f4c6a81ef94cbb600620d83ad51269d] | committer: Rémi Denis-Courmont lavc/sbrdsp: R-V V neg_odd_64 With 128-bit vectors, this is mostly pointless but also harmless. Performance gains should be more noticeabl

[FFmpeg-cvslog] lavc/sbrdsp: R-V V sum64x5

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 29 17:39:53 2023 +0200| [86bee424730bc0dcba0e101773e6f3c71e39bd29] | committer: Rémi Denis-Courmont lavc/sbrdsp: R-V V sum64x5 sum64x5_c: 385.0 sum64x5_rvv_f32: 116.0 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=86be

[FFmpeg-cvslog] lavc/sbrdsp: R-V V sum_square

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 29 18:10:07 2023 +0200| [b0aba7dd0c1ee34941370997fd20049150f04a36] | committer: Rémi Denis-Courmont lavc/sbrdsp: R-V V sum_square sum_square_c: 803.5 sum_square_rvv_f32: 283.2 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=comm

[FFmpeg-cvslog] avcodec/vaapi_encode: Use RefStruct pool API, stop abusing AVBuffer API

ffmpeg | branch: master | Andreas Rheinhardt | Sat Aug 6 16:36:46 2022 +0200| [92abc7266b5a7139d85699cce2b2c94c1e4ed2ed] | committer: Andreas Rheinhardt avcodec/vaapi_encode: Use RefStruct pool API, stop abusing AVBuffer API Up until now, the VAAPI encoder uses fake data with the AVBuffer-API

[FFmpeg-cvslog] avcodec/vp9: Use RefStruct-pool API for extradata

ffmpeg | branch: master | Andreas Rheinhardt | Sat Aug 6 13:59:20 2022 +0200| [8c0350f57ef4c5f7ff4ec56865029e30cbf2877e] | committer: Andreas Rheinhardt avcodec/vp9: Use RefStruct-pool API for extradata It avoids allocations and corresponding error checks. Signed-off-by: Andreas Rheinhardt

[FFmpeg-cvslog] avcodec/vp9: Join extradata buffer pools

ffmpeg | branch: master | Andreas Rheinhardt | Tue Aug 16 15:49:41 2022 +0200| [eba73142adc776d7fe0639d3cff23d878da0174a] | committer: Andreas Rheinhardt avcodec/vp9: Join extradata buffer pools Up until now each thread had its own buffer pool for extradata buffers when using frame-threading.

[FFmpeg-cvslog] avcodec/refstruct: Allow to share pools

ffmpeg | branch: master | Andreas Rheinhardt | Tue Aug 16 14:04:27 2022 +0200| [0c44f63b0282e2e830d5bf479016984cff69221a] | committer: Andreas Rheinhardt avcodec/refstruct: Allow to share pools To do this, make FFRefStructPool itself refcounted according to the RefStruct API. Signed-off-by: A

[FFmpeg-cvslog] avcodec/hevcdec: Use RefStruct-pool API instead of AVBufferPool API

ffmpeg | branch: master | Andreas Rheinhardt | Fri Aug 5 22:57:56 2022 +0200| [fd2e65871c6fdc79cba5e3eef877a81ec83d57c6] | committer: Andreas Rheinhardt avcodec/hevcdec: Use RefStruct-pool API instead of AVBufferPool API It involves less allocations and therefore has the nice property that de

[FFmpeg-cvslog] avcodec/refstruct: Allow to always return zeroed pool entries

ffmpeg | branch: master | Andreas Rheinhardt | Sat Aug 6 13:04:10 2022 +0200| [090d9956fd092023e63f61a37bde36bba55d0258] | committer: Andreas Rheinhardt avcodec/refstruct: Allow to always return zeroed pool entries This is in preparation for the following commit. Reviewed-by: Anton Khirnov

[FFmpeg-cvslog] avcodec/nvdec: Use RefStruct-pool API for decoder pool

ffmpeg | branch: master | Andreas Rheinhardt | Sat Aug 6 08:23:27 2022 +0200| [e01e30ede1d0a06df90c0c7816cb51941bcd3bc9] | committer: Andreas Rheinhardt avcodec/nvdec: Use RefStruct-pool API for decoder pool It involves less allocations, in particular no allocations after the entry has been c

[FFmpeg-cvslog] avcodec/h264dec: Use RefStruct-pool API instead of AVBufferPool API

ffmpeg | branch: master | Andreas Rheinhardt | Fri Aug 5 21:42:58 2022 +0200| [736b510fcc386df89a403a3ed11c4d1dec599cd1] | committer: Andreas Rheinhardt avcodec/h264dec: Use RefStruct-pool API instead of AVBufferPool API It involves less allocations and therefore has the nice property that de

[FFmpeg-cvslog] avcodec/refstruct: Add RefStruct pool API

ffmpeg | branch: master | Andreas Rheinhardt | Fri Aug 5 18:32:23 2022 +0200| [26c0a7321fdef730050156fcef0b986ac4373b6c] | committer: Andreas Rheinhardt avcodec/refstruct: Add RefStruct pool API Very similar to the AVBufferPool API, but with some differences: 1. Reusing an already existing en

[FFmpeg-cvslog] lavc/pixblockdsp: remove R-V V get_pixels_16

ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 29 15:07:58 2023 +0200| [92bcc6703acf346f0a5257de2410400a2e81a754] | committer: Rémi Denis-Courmont lavc/pixblockdsp: remove R-V V get_pixels_16 In the aligned case, the existing RVI assembler is actually much faster. In the unaligned cas

[FFmpeg-cvslog] lavc/jpeg2000dsp: R-V V rct_int

ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 28 21:55:16 2023 +0300| [28840cf499ad0149d690385d8d3efe15aa95] | committer: Rémi Denis-Courmont lavc/jpeg2000dsp: R-V V rct_int jpeg2000_rct_int_c: 2592.2 jpeg2000_rct_int_rvv_i32: 1154.2 > http://git.videolan.org/gitweb.cgi/ff

[FFmpeg-cvslog] lavc/jpeg2000dsp: R-V V ict_float

ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 28 19:29:12 2023 +0300| [73dea2bb917456e197cbb57b31275b6231e07acc] | committer: Rémi Denis-Courmont lavc/jpeg2000dsp: R-V V ict_float jpeg2000_ict_float_c: 3112.2 jpeg2000_ict_float_rvv_f32: 1225.0 > http://git.videolan.org/gitweb.

[FFmpeg-cvslog] lavc/jpeg2000dsp: make coefficients extern

ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 28 18:57:11 2023 +0300| [b2a441a3bed281a74b7fced77ef974897e407e32] | committer: Rémi Denis-Courmont lavc/jpeg2000dsp: make coefficients extern This is so that they can be loaded from assembler, rather than duplicated. > http://git.videol

[FFmpeg-cvslog] avcodec/vlc: Skip subtable entries in multi VLC

ffmpeg | branch: master | Michael Niedermayer | Sun Oct 22 21:20:25 2023 +0200| [356b1ba765623ec1640a366e587232f1631800f8] | committer: Michael Niedermayer avcodec/vlc: Skip subtable entries in multi VLC These entries do not correspond to VLC symbols that can be used they do corrupt various va

[FFmpeg-cvslog] avcodec/vlc: Replace mysterious max computation code in multi vlc

ffmpeg | branch: master | Michael Niedermayer | Sun Oct 22 21:36:11 2023 +0200| [8516609edde98391017fb145b4f492c01b360a03] | committer: Michael Niedermayer avcodec/vlc: Replace mysterious max computation code in multi vlc Signed-off-by: Michael Niedermayer > http://git.videolan.org/gitweb.cg

[FFmpeg-cvslog] avcodec/vlc: Pass VLC_MULTI_ELEM directly not by pointer

ffmpeg | branch: master | Michael Niedermayer | Sun Oct 22 22:14:47 2023 +0200| [a5259f326bcaf933a25df64aa49417c25990f7d3] | committer: Michael Niedermayer avcodec/vlc: Pass VLC_MULTI_ELEM directly not by pointer This makes the code more testable as uninitialized fields are 0 and not random va

[FFmpeg-cvslog] tools/target_dec_fuzzer: Adjust threshold for CSCD

ffmpeg | branch: master | Michael Niedermayer | Sat Oct 21 00:11:02 2023 +0200| [c2f2bf82c1b3987e2d1a75cc79c4b58d286a2291] | committer: Michael Niedermayer tools/target_dec_fuzzer: Adjust threshold for CSCD Fixes: Timeout Fixes: 63362/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_CSCD_fuz

[FFmpeg-cvslog] avcodec/dovi_rpu: Use 64 bit in get_us/se_coeff()

ffmpeg | branch: master | Michael Niedermayer | Fri Oct 20 23:50:07 2023 +0200| [2817efbba331ac1d3a39fbee78b480008ce20a93] | committer: Michael Niedermayer avcodec/dovi_rpu: Use 64 bit in get_us/se_coeff() Fixes: shift exponent 32 is too large for 32-bit type 'int' Fixes: 63151/clusterfuzz-te

[FFmpeg-cvslog] avcodec/apedec: Fix integer overflow in predictor_decode_stereo_3950()

ffmpeg | branch: master | Michael Niedermayer | Fri Oct 20 18:20:14 2023 +0200| [2def61778777e998e2ac538f56b8e6779b0ca8c1] | committer: Michael Niedermayer avcodec/apedec: Fix integer overflow in predictor_decode_stereo_3950() Fixes: signed integer overflow: 1900031961 + 553590817 cannot be re

[FFmpeg-cvslog] avformat/mov: Check that is_still_picture_avif has no trak based streams

ffmpeg | branch: master | Michael Niedermayer | Sun Oct 15 01:52:20 2023 +0200| [98c2711b58ce65eae02cb2ece3a664e1119fd8fe] | committer: Michael Niedermayer avformat/mov: Check that is_still_picture_avif has no trak based streams Fixes: Assertion failure in mov_read_iloc( in mov_read_iloc()) Fi

[FFmpeg-cvslog] avcodec/evc_parse: Check tid

ffmpeg | branch: master | Michael Niedermayer | Thu Oct 5 00:19:53 2023 +0200| [68cc1744db828e929b74f96478c18f1d226510be] | committer: Michael Niedermayer avcodec/evc_parse: Check tid The check is based on not infinite looping. It is likely a more strict check can be done Fixes: Infinite loo

[FFmpeg-cvslog] avcodec/evc_parse: remove pow() and log2()

ffmpeg | branch: master | Michael Niedermayer | Thu Oct 5 00:10:18 2023 +0200| [d35eecd24fb62a4060ef89dfdd1c45f5549fb71d] | committer: Michael Niedermayer avcodec/evc_parse: remove pow() and log2() The use of float based functions is both unneeded and wrong due to unpredictable rounding Sig