[FFmpeg-cvslog] qsv: Skip the packet if decoding failure

ffmpeg | branch: master | Ruiling, Song | Wed Jan 24 10:14:51 2018 +0800| [559370f2c45110afd8308eec7194437736c323d4] | committer: Luca Barbato qsv: Skip the packet if decoding failure MediaSDK may fail to decode some frame, just skip it. Otherwise, it will keep decoding the failure packet

[FFmpeg-cvslog] lavc/qsv: default la_ds to MFX_LOOKAHEAD_DS_UNKNOWN

ffmpeg | branch: master | Ruiling Song | Sun Feb 11 15:43:45 2018 +0800| [9b09792c90b580842157ca8ce534be434725a841] | committer: Maxym Dmytrychenko lavc/qsv: default la_ds to MFX_LOOKAHEAD_DS_UNKNOWN MFX_LOOKAHEAD_DS_UNKNOWN means auto. -1 is not a valid value. Signed-off-by: Ruiling Song

[FFmpeg-cvslog] qsv: Default PicStruct to progressive

ffmpeg | branch: master | Ruiling Song | Mon Feb 12 16:37:16 2018 +0800| [8ca39b855a7b0e4d9f726fa9d285bc8edcb953e6] | committer: Luca Barbato qsv: Default PicStruct to progressive The PicStruct is required by MediaSDK, so give a default value. hwupload does not work without this. Signed-off

[FFmpeg-cvslog] qsv: align surface width/height to 16.

ffmpeg | branch: master | Ruiling Song | Tue Mar 6 10:35:56 2018 +0800| [86499771d1228d8303c8eb6509e20c0caaa02da5] | committer: Luca Barbato qsv: align surface width/height to 16. Per MediaSDK documentation, it requires width/height to 16 alignment. Without this patch, hwupload pipeline may

[FFmpeg-cvslog] lavf: make overlay_qsv work based on framesync

ffmpeg | branch: master | Ruiling Song | Tue Apr 3 09:50:19 2018 +0800| [f3341a0452419c57faf4d28aebb24be5d41312f3] | committer: Mark Thompson lavf: make overlay_qsv work based on framesync The existing version which was cherry-picked from Libav does not work with FFmpeg framework, because

[FFmpeg-cvslog] lavf/qsv: clone the frame which may be managed by framework

ffmpeg | branch: master | Ruiling Song | Tue Apr 3 09:50:20 2018 +0800| [d865783b6c8d4f96f5094ed72eff0f5a4a4908af] | committer: Mark Thompson lavf/qsv: clone the frame which may be managed by framework For filters based on framesync, the input frame was managed by framesync, so we should not

[FFmpeg-cvslog] lavf: make overlay_qsv work based on framesync

ffmpeg | branch: release/4.0 | Ruiling Song | Tue Apr 3 09:50:19 2018 +0800| [58569162c280a0765b63441001d359e62824c9de] | committer: Mark Thompson lavf: make overlay_qsv work based on framesync The existing version which was cherry-picked from Libav does not work with FFmpeg framework

[FFmpeg-cvslog] lavf/qsv: clone the frame which may be managed by framework

ffmpeg | branch: release/4.0 | Ruiling Song | Tue Apr 3 09:50:20 2018 +0800| [a768c0a3e1fa29eddc7dd348012b3093e476c94e] | committer: Mark Thompson lavf/qsv: clone the frame which may be managed by framework For filters based on framesync, the input frame was managed by framesync, so we

[FFmpeg-cvslog] lavu: add calling convention for OpenCL callback.

ffmpeg | branch: master | Ruiling Song | Wed Jun 6 23:31:44 2018 +0800| [e4006a474f689ef9534448d3382d220c62b413b6] | committer: Mark Thompson lavu: add calling convention for OpenCL callback. This fix a build error on Windows: C2440: connot convert from 'void (__cdecl *) (...)'

[FFmpeg-cvslog] lavfi: add opencl tonemap filter

ffmpeg | branch: master | Ruiling Song | Tue Jun 19 09:57:31 2018 +0800| [8b8b0e2cd26cf1f522c630859fcbcc62b6493fb9] | committer: Mark Thompson lavfi: add opencl tonemap filter This filter does HDR(HDR10/HLG) to SDR conversion with tone-mapping. An example command to use this filter with

[FFmpeg-cvslog] lavfi: make vf_colorspace use functions from colorspace.c

ffmpeg | branch: master | Ruiling Song | Tue Jun 19 09:57:32 2018 +0800| [649d7ca47780523673bbfe43407b9cb71c4ad811] | committer: Mark Thompson lavfi: make vf_colorspace use functions from colorspace.c These functions are shared among colorspace related filters. Signed-off-by: Ruiling Song

[FFmpeg-cvslog] lavfi/vf_avgblur_opencl: remove useless clFinish().

ffmpeg | branch: master | Ruiling Song | Wed Jul 4 02:16:25 2018 +0800| [923ee63d6cd140e1f957c838b1d062ca4fb73580] | committer: Mark Thompson lavfi/vf_avgblur_opencl: remove useless clFinish(). The very last clFinish() should be ok. Signed-off-by: Ruiling Song Reviewed-by: Danil Iashchenko

[FFmpeg-cvslog] lavfi/opencl: add macro for opencl error handling.

ffmpeg | branch: master | Ruiling Song | Wed Jul 4 02:16:24 2018 +0800| [48a1abed133d005aa4d7f49e4fdd5bef9aa263f5] | committer: Mark Thompson lavfi/opencl: add macro for opencl error handling. Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] doc/filters: add document for opencl filters

ffmpeg | branch: master | Ruiling Song | Mon Oct 29 13:17:59 2018 +0800| [952a299fd390ca59267c30097ce2148ca730cc17] | committer: Lou Logan doc/filters: add document for opencl filters Signed-off-by: Danil Iashchenko Signed-off-by: Ruiling Song Signed-off-by: Gyan Doshi > h

[FFmpeg-cvslog] doc/filters: add tonemap_opencl document.

ffmpeg | branch: master | Ruiling Song | Mon Oct 29 13:56:59 2018 +0800| [a587454fd07c831a6c315ae2daf1d8f66396d967] | committer: Lou Logan doc/filters: add tonemap_opencl document. Signed-off-by: Ruiling Song Signed-off-by: Gyan Doshi > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] swscale/swscale: remove useless code

ffmpeg | branch: master | Ruiling Song | Wed Apr 1 15:32:15 2020 +0800| [4700f7d6fce3ae3fcce86e069edf16b6cbae614b] | committer: Michael Niedermayer swscale/swscale: remove useless code Signed-off-by: Ruiling Song Signed-off-by: Michael Niedermayer > http://git.videolan.org/gitweb.

[FFmpeg-cvslog] mailmap: add entry for myself

ffmpeg | branch: master | Ruiling Song | Mon Jul 13 09:37:37 2020 +0800| [7772666e5bfff27c4900942e7558a2d327df8f32] | committer: Guo, Yejun mailmap: add entry for myself Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] lavf: add transpose_opencl filter

ffmpeg | branch: master | Ruiling Song | Wed Nov 28 10:27:38 2018 +0800| [416dc9a5e81729e3313bcc13aebc5faa082c63a3] | committer: Mark Thompson lavf: add transpose_opencl filter Signed-off-by: Ruiling Song Signed-off-by: Mark Thompson > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] MAINTAINERS: add myself for tonemap_opencl

ffmpeg | branch: master | Ruiling Song | Wed Feb 13 09:29:10 2019 +0800| [dfde416ea8d0ce9c11cb450211b4569b8217cfb3] | committer: Michael Niedermayer MAINTAINERS: add myself for tonemap_opencl Signed-off-by: Ruiling Song Signed-off-by: Michael Niedermayer > http://git.videolan.

[FFmpeg-cvslog] lavfi/tonemap_opencl: reuse color matrix calculation from colorspace.c

ffmpeg | branch: master | Ruiling Song | Tue Jan 22 15:01:56 2019 +0800| [8b951cd4752c8db2b4532fae9fb300d422950cdd] | committer: Ruiling Song lavfi/tonemap_opencl: reuse color matrix calculation from colorspace.c Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] lavfi/opencl: add ff_opencl_print_const_matrix_3x3()

ffmpeg | branch: master | Ruiling Song | Tue Jan 22 14:47:54 2019 +0800| [2593122a167de3294abd5b9cf04df5b8072ee3ed] | committer: Ruiling Song lavfi/opencl: add ff_opencl_print_const_matrix_3x3() This is used to print a 3x3 matrix into a part of OpenCL source code. Signed-off-by: Ruiling Song

[FFmpeg-cvslog] lavfi/colorspace: move some functions to common file

ffmpeg | branch: master | Ruiling Song | Tue Jan 22 14:27:01 2019 +0800| [d0f3798b4e7f9ec3142f74946f7de41b9e3485cb] | committer: Ruiling Song lavfi/colorspace: move some functions to common file These functions can be reused by other colorspace filters, so move them to common file. No

[FFmpeg-cvslog] lavfi/colorspace_common: add ifdef check to be more compatible.

ffmpeg | branch: master | Ruiling Song | Mon Jan 21 15:44:04 2019 +0800| [b073fb9eeae8f021a4e18886ccf73cda9f67b00c] | committer: Ruiling Song lavfi/colorspace_common: add ifdef check to be more compatible. Some filters may not need to do linearize/delinearize, thus will even not define them

[FFmpeg-cvslog] lavu/opencl: replace va_ext.h with standard name

ffmpeg | branch: master | Ruiling Song | Fri Nov 23 13:39:12 2018 +0800| [61cb505d18b8a335bd118d88c05b9daf40eb5f9b] | committer: Ruiling Song lavu/opencl: replace va_ext.h with standard name Khronos OpenCL header (https://github.com/KhronosGroup/OpenCL-Headers) uses

[FFmpeg-cvslog] lavfi/opencl: add more opencl helper macro

ffmpeg | branch: master | Ruiling Song | Fri Apr 12 16:27:54 2019 +0800| [0fc464631aafa2741e97ce00fc8f81378fcf441a] | committer: Ruiling Song lavfi/opencl: add more opencl helper macro Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] lavfi/gblur: doing several columns at the same time

ffmpeg | branch: master | Ruiling Song | Mon May 6 11:48:31 2019 +0800| [06ba4783a0ba331153b71993bc1908958d483c2d] | committer: Ruiling Song lavfi/gblur: doing several columns at the same time Instead of doing each column one by one, doing several columns together gives about 30% better

[FFmpeg-cvslog] avutil/tx: should check against (*ctx)

ffmpeg | branch: master | Ruiling Song | Thu May 16 12:47:36 2019 +0800| [65646db8e8d5aa95ef8282823fdf5ec1a5f3df69] | committer: James Almer avutil/tx: should check against (*ctx) ctx is a pointer to pointer here. Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.

[FFmpeg-cvslog] lavfi/opencl: add nlmeans_opencl filter

ffmpeg | branch: master | Ruiling Song | Fri Apr 12 16:29:03 2019 +0800| [1d74150a7dacf4912de21820ab1011b94c13eeb5] | committer: Ruiling Song lavfi/opencl: add nlmeans_opencl filter Reviewed-by: Mark Thompson Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] avfilter/vf_unsharp: enable slice threading

ffmpeg | branch: master | Ruiling Song | Tue May 7 09:46:33 2019 +0800| [94ceeba9f991ab69b192fa8527be0965de7e254b] | committer: Ruiling Song avfilter/vf_unsharp: enable slice threading benchmarking with a simple command: ffmpeg -i 1080p.mp4 -vf unsharp=la=3:ca=3 -an -f null /dev/null with

[FFmpeg-cvslog] checkasm/vf_gblur: add test for horiz_slice simd

ffmpeg | branch: master | Ruiling Song | Tue Jun 4 16:34:08 2019 +0800| [8f4963ad25ac6c71c126ac90c240b1f88ebf0312] | committer: Ruiling Song checkasm/vf_gblur: add test for horiz_slice simd Signed-off-by: Ruiling Song > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avfilter/vf_gblur: add x86 SIMD optimizations

ffmpeg | branch: master | Ruiling Song | Wed May 15 17:54:10 2019 +0800| [83f9da77684e7ea0d8e9f9712ec716424140043a] | committer: Ruiling Song avfilter/vf_gblur: add x86 SIMD optimizations The horizontal pass get ~2x performance with the patch under single thread. Tested overall performance

[FFmpeg-cvslog] avfilter/vf_convolution: add x86 SIMD for filter_3x3()

ffmpeg | branch: master | Ruiling Song | Thu Jun 27 10:07:21 2019 +0800| [98e419cbf5443b3e95d6bc7bdb807c9f54bfd6a7] | committer: Ruiling Song avfilter/vf_convolution: add x86 SIMD for filter_3x3() Tested using a simple command (apply edge enhance): ./ffmpeg_g -i ~/Downloads