[FFmpeg-cvslog] lavc/flacenc: add sse4 version of the 16-bit lpc encoder

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Tue Aug 12 23:22:02 2014 +0200| [0081a14e7d1e44536400df19ad6c9ea8118db2e9] | committer: Michael Niedermayer lavc/flacenc: add sse4 version of the 16-bit lpc encoder From 1.8 to 2.4 times faster. Runtime is reduced by 2 to 39

[FFmpeg-cvslog] cvcodec/flacdsp_lpc_template: osmetic fix

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Tue Aug 12 23:22:06 2014 +0200| [0a52bbdc3161cb2d0ff46da5d84be708054cf9e7] | committer: Michael Niedermayer cvcodec/flacdsp_lpc_template: osmetic fix Signed-off-by: Michael Niedermayer michae...@gmx.at http://git.videolan.org

[FFmpeg-cvslog] Replace incorrect use of multiply with multiple

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Wed Aug 27 22:39:29 2014 +0200| [8c1b942ad7b66034b030f2ab3cc5e81eccfa2b68] | committer: Lou Logan Replace incorrect use of multiply with multiple Also replace the plural form multiplies with multiples. http://git.videolan.org

[FFmpeg-cvslog] lavc/x86/v210: give cpuflag to INIT macro

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Fri Sep 5 00:13:35 2014 +0200| [46ef45ab59e416fae92d9fd27f050268eb2a495a] | committer: Michael Niedermayer lavc/x86/v210: give cpuflag to INIT macro This lets the cglobal macro automatically append a suffix to the function name

[FFmpeg-cvslog] docs: add example around the suggested commit message format

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Fri Sep 12 01:31:26 2014 +0200| [e3fb2b0eb7c5e420fc23e015c95b9c3e42abb2bf] | committer: Michael Niedermayer docs: add example around the suggested commit message format It gets rendered inline in HTML and becomes not very clear

[FFmpeg-cvslog] lavu/x86/x86inc: deprecate INIT_AVX

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Fri Sep 5 00:13:36 2014 +0200| [12120174ce8bf377ef212f38b1ecfbc17bd85147] | committer: Michael Niedermayer lavu/x86/x86inc: deprecate INIT_AVX The same can be done with INIT_XMM avx Signed-off-by: Michael Niedermayer michae

[FFmpeg-cvslog] avfilter/vf_removegrain: add x86 and x86_64 SSE2 functions

ffmpeg | branch: master | James Darnley james.darn...@gmail.com | Tue Jul 14 23:48:47 2015 +| [bff7242608409dc52bf2fd51a67bb9d5f171a0ab] | committer: Paul B Mahol avfilter/vf_removegrain: add x86 and x86_64 SSE2 functions Speed of all modes increased by a factor between 7.4 and 19.8

Re: [FFmpeg-cvslog] avfilter: add removegrain

On 2015-07-14 19:28, Robert Krüger wrote: On Sat, Jul 11, 2015 at 5:30 PM, Robert Krüger krue...@lesspain.de wrote: On Wed, Jul 8, 2015 at 6:03 PM, Paul B Mahol g...@videolan.org wrote: ffmpeg | branch: master | Paul B Mahol one...@gmail.com | Sat Jul 4 20:19:05 2015 +|

Re: [FFmpeg-cvslog] configure: Reenable colorized warnings and check for tput's existence

On 2015-09-08 05:42, Timothy Gu wrote: > ffmpeg | branch: master | Timothy Gu | Mon Sep 7 > 19:11:16 2015 -0700| [617d53f4c7e43b5df6f99b363b550ff7b0007c6e] | committer: > Timothy Gu > > configure: Reenable colorized warnings and check for tput's existence > > Untested.

[FFmpeg-cvslog] avcodec/v210: guard new avx2 functions from old assemblers

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Sun Jan 17 21:20:25 2016 +0100| [f59b727e2f982a22c4459c67b2f5f7f42c86d3d0] | committer: James Darnley avcodec/v210: guard new avx2 functions from old assemblers > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/v210: document the requirement for sample_factor

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 19:53:28 2016 +0100| [4c430738d9cc8a5712e681236aff4f36991fd48f] | committer: James Darnley avcodec/v210: document the requirement for sample_factor The sample factor must be the same for both 8- and 10-bit fun

[FFmpeg-cvslog] fate: add 10-bit v210 encoder tests

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 16:29:28 2016 +| [61625dcc395f98ad77e9378ec4eb48cf7c981bce] | committer: James Darnley fate: add 10-bit v210 encoder tests > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/v210: add avx2 version of the 10-bit line encoder

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 18:30:03 2016 +| [2cba1825f70c301f5513a799302f755967680651] | committer: James Darnley avcodec/v210: add avx2 version of the 10-bit line encoder Around 25% faster than the ssse3 version.

[FFmpeg-cvslog] avcodec/v210: add avx2 version of the 8-bit line encoder

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Wed Dec 30 17:20:59 2015 +| [3836f404a8c0ffd7c565260434768ef4ebba85a7] | committer: James Darnley avcodec/v210: add avx2 version of the 8-bit line encoder Around 35% faster than the avx version. Signed-off-by:

[FFmpeg-cvslog] avcodec/h264: mmxext 4:2:2 chroma deblock/loop filter

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Wed Jan 13 14:34:44 2016 +| [7042a55c55d7ac04e9a6b3cd150c85d64cedb79f] | committer: James Darnley avcodec/h264: mmxext 4:2:2 chroma deblock/loop filter 2.6 times faster (366 vs. 142 cycles) > http://git.vid

[FFmpeg-cvslog] fate: add 10-bit v210 encoder tests

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 20:35:04 2016 +0100| [883ad2c59ceea1ced5495b5ccc83695ed4bbb94b] | committer: Luca Barbato fate: add 10-bit v210 encoder tests Signed-off-by: Luca Barbato <lu_z...@gentoo.org> > http://git.videolan

[FFmpeg-cvslog] v210: Add avx2 version of the 10-bit line encoder

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 20:35:06 2016 +0100| [15ec7aa4170ed05ad1b17000ef1e3940d0a0c5e7] | committer: Luca Barbato v210: Add avx2 version of the 10-bit line encoder Around 25% faster than the ssse3 version. Signed-off-by: Luca B

[FFmpeg-cvslog] v210: Add avx2 version of the 8-bit line encoder

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Fri Jan 15 20:35:05 2016 +0100| [d29237e5578a187c5a8d91338cd70ce0fd6f6003] | committer: Luca Barbato v210: Add avx2 version of the 8-bit line encoder Around 35% faster than the avx version. Signed-off-by: Henrik Gramne

[FFmpeg-cvslog] avcodec/h264: sse2, avx h luma mbaff deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Fri Feb 10 23:17:57 2017 +0100| [533688786799b22d0711eedcfe8f84deea014f30] | committer: James Darnley avcodec/h264: sse2, avx h luma mbaff deblock/loop filter x86-64 only Yorkfield: - sse2: ~2.17x (434 vs. 200 cycles) N

[FFmpeg-cvslog] avcodec/x86: deduplicate PASS8ROWS macro

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Feb 9 22:06:17 2017 +0100| [9d815b7424b56ffcf6f4dea4fd09c53661e4133a] | committer: James Darnley avcodec/x86: deduplicate PASS8ROWS macro > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] x86util: import MOVHL macro

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Sat Feb 11 13:25:09 2017 +0100| [7627df15d411a69f236b4650e88b1ab911f38efc] | committer: James Darnley x86util: import MOVHL macro Originally committed to x264 in 1637239a by Henrik Gramner who has agreed to re-license it a

[FFmpeg-cvslog] avcodec/h264: add named parameters to x86 function

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Fri Feb 10 20:13:50 2017 +0100| [e18bc2114f3deb8ef1ab9ddaef282c8d9678669d] | committer: James Darnley avcodec/h264: add named parameters to x86 function > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/h264: enable sse2 chroma deblock/loop filter functions

ffmpeg | branch: h264_assembly | James Darnley <jdarn...@obe.tv> | Wed Feb 22 01:17:06 2017 +0100| [4deaf9e78a45e7d2ee9f8190edd16f3217c16d88] | committer: James Darnley avcodec/h264: enable sse2 chroma deblock/loop filter functions Between 1.00 and 1.16 times faster on Intel Yorkfield

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit 4:2:2 chroma h deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 14:54:11 2017 +0100| [88307b3eec018b86f6adfddeb6e9b384c95b7ca6] | committer: James Darnley avcodec/h264: add avx 8-bit 4:2:2 chroma h deblock/loop filter ~1.21x faster (68 vs. 56 cycles) compared with mmxext fu

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit chroma v intra deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 14:59:46 2017 +0100| [987ffe4b8dce88b8df12208eef788bd022a43598] | committer: James Darnley avcodec/h264: add avx 8-bit chroma v intra deblock/loop filter ~1.14x faster (90 vs 78 cycles) compared with mmxext

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit 4:2:0 chroma h deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 14:48:13 2017 +0100| [ac096fc82df6aa8190f29a51ee5b589f318cfaa8] | committer: James Darnley avcodec/h264: add avx 8-bit 4:2:0 chroma h deblock/loop filter ~1.14x faster (93 vs. 81 cycles) compared with mmxext fu

[FFmpeg-cvslog] avcodec/h264: enable sse2 chroma deblock/loop filter functions

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 22 01:17:06 2017 +0100| [33de0fee2c33c492aae96f643ed7bbaa393043dc] | committer: James Darnley avcodec/h264: enable sse2 chroma deblock/loop filter functions Between 1.00 and 1.16 times faster on Intel Yorkfield Core

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit 4:2:0 chroma h intra deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 15:03:18 2017 +0100| [0e16b3e2be3a3e20a5da94a0c48049ba6a5baacd] | committer: James Darnley avcodec/h264: add avx 8-bit 4:2:0 chroma h intra deblock/loop filter ~1.10x faster (69 vs. 63 cycles) compared to mmxext fu

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit 4:2:2 chroma h intra deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 15:05:42 2017 +0100| [cd893b9307b8c871fec067928dc613fa987e38c3] | committer: James Darnley avcodec/h264: add avx 8-bit 4:2:2 chroma h intra deblock/loop filter ~1.37x faster (147 vs. 108 cycles) compared to mmxext fu

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit chroma v deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Feb 15 14:36:20 2017 +0100| [5c56758843eb5ea8fc39177585a57606a34125bc] | committer: James Darnley avcodec/h264: add avx 8-bit chroma v deblock/loop filter ~1.24x faster (101 vs. 81 cycles) compared with mmxext function

[FFmpeg-cvslog] avcodec/h264: resolve assert being triggered when stack is not aligned

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Dec 7 21:57:15 2016 +0100| [acdd2d805d239cf8b6c1749350f0a2213150986f] | committer: James Darnley avcodec/h264: resolve assert being triggered when stack is not aligned 32-bit msvc. > http://git.videolan.org/gitweb.cgi/f

[FFmpeg-cvslog] avcodec/h264: mmx 4:2:2 idct add8 function

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Nov 28 18:22:17 2016 +0100| [1dae7ffa0b11293fa3294fc25cd38055915fcf19] | committer: James Darnley avcodec/h264: mmx 4:2:2 idct add8 function 2.87 times faster (1830 vs. 638 cycles) > http://git.videolan.org/gitweb.cgi/f

[FFmpeg-cvslog] avcodec/h264: mmxext 4:2:2 chroma intra deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Nov 28 14:26:53 2016 +0100| [815ea8c6ccf7a1a8154a829f948d7c98120ad084] | committer: James Darnley avcodec/h264: mmxext 4:2:2 chroma intra deblock/loop filter 2.1 times faster (401 vs. 194 cycles) > http://git.vid

[FFmpeg-cvslog] avcodec/h264: sse2 and avx 4:2:2 idct add8 10-bit functions

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Tue Nov 29 00:38:43 2016 +0100| [13d71c28cc44dd049d27a0a2e909add46ed1d5d0] | committer: James Darnley avcodec/h264: sse2 and avx 4:2:2 idct add8 10-bit functions Yorkfield: - sse2: - complex: 4.13x faster (1514 vs. 367

[FFmpeg-cvslog] whitespace changes after last commit

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Dec 1 02:47:24 2016 +0100| [58ca2ef62e7c3e0b9f17d7ed3e435f996aa061fe] | committer: James Darnley whitespace changes after last commit > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/h264: mmx2, sse2, avx 10-bit h chroma deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Wed Nov 30 21:17:16 2016 +0100| [add21d0bb3f5fb25fd5d1437eb746b91c2570a8f] | committer: James Darnley avcodec/h264: mmx2, sse2, avx 10-bit h chroma deblock/loop filter Yorkfield: - mmx2: 2.45x (279 vs. 114 cycles) - sse2: 3.36

[FFmpeg-cvslog] avcodec/h264: mmx2, sse2, avx 10-bit 4:2:2 h chroma deblock/loop filter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Dec 1 03:15:10 2016 +0100| [728651df06f48b49228ac5b8241b1ebe8d5b9d73] | committer: James Darnley avcodec/h264: mmx2, sse2, avx 10-bit 4:2:2 h chroma deblock/loop filter Yorkfield: - mmx2: 2.53x (504 vs. 199 cycles)

[FFmpeg-cvslog] avcodec/h264: clean up and expand x86 function definitions

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Dec 1 02:44:18 2016 +0100| [f33714a6948f9e707c00c89795222dea5d4ebe3a] | committer: James Darnley avcodec/h264: clean up and expand x86 function definitions > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86: allow future 8-bit simple idct to have "DC only hack"

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Jun 15 00:35:44 2017 +0200| [8b19467d07d5782b4140f61363f24361efb87ff6] | committer: James Darnley avcodec/x86: allow future 8-bit simple idct to have "DC only hack" Created by Ronald S. Bultje > http://git.vide

[FFmpeg-cvslog] avcodec/x86: use new x86-64 functions for -idct simple

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Jun 19 14:55:53 2017 +0200| [0c2acccd4be425d1121ec923f830c6fe99227b56] | committer: James Darnley avcodec/x86: use new x86-64 functions for -idct simple They now match according to FATE, barring any further bugs with un

[FFmpeg-cvslog] avcodec/x86: add an 8-bit simple IDCT function based on the x86-64 high depth functions

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Fri Jun 2 15:20:19 2017 +0200| [d7246ea9f229db64ed909d7446196128d6f53de0] | committer: James Darnley avcodec/x86: add an 8-bit simple IDCT function based on the x86-64 high depth functions Includes add/put functions Ro

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit h264_idct_add

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Mar 16 14:59:48 2017 +0100| [f61d454ca13f277b6ab7bbc9ebf7d26ce6d67ec6] | committer: James Darnley avcodec/h264: add avx 8-bit h264_idct_add Haswell: - 1.11x faster (522±0.4 vs. 469±1.8 decicycles) compared with mmxext Sky

[FFmpeg-cvslog] avcodec/h264: change RETs into REP_RETs where appropriate

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Feb 27 20:03:08 2017 +| [060ba9e5e3353a4c50a096f40e9c91b78145cbb8] | committer: James Darnley avcodec/h264: change RETs into REP_RETs where appropriate > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/h264: use some 3 operand forms

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Mar 2 01:48:16 2017 +0100| [b5325c6711a6789e6219f6392fd9158cb2c0fcd7] | committer: James Darnley avcodec/h264: use some 3 operand forms > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/h264: add avx 8-bit h264_idct_dc_add

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Mar 16 15:07:11 2017 +0100| [27460dfebc296636dec2584e0d74aaa4d48da0b9] | committer: James Darnley avcodec/h264: add avx 8-bit h264_idct_dc_add Haswell: - 1.02x faster (405±0.7 vs. 397±0.8 decicycles) compared with mmxext S

[FFmpeg-cvslog] avcodec/x86: modify simple_idct10 macros to add an action paramter

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Jun 12 12:01:21 2017 +0200| [d2597fb0c1c8e4964becf809f03abfa6596a8c14] | committer: James Darnley avcodec/x86: modify simple_idct10 macros to add an action paramter > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86: cleanup simple_idct10

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Jun 1 15:28:29 2017 +0200| [8781330d80e3cd1bd3af2a2a2c630ae26779dedd] | committer: James Darnley avcodec/x86: cleanup simple_idct10 Use named arguments for the functions so we can remove a define. The stride/linesize ar

[FFmpeg-cvslog] avcodec/mdec: override IDCT choice before initing DSP structs

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Sat Jun 17 03:13:36 2017 +0200| [9d11fedd1129565c8ba9e90b08b43e06f441b4fb] | committer: James Darnley avcodec/mdec: override IDCT choice before initing DSP structs > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86/mpegenc: check IDCT permutation type is a valid value

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Fri Jun 16 14:53:15 2017 +0200| [fa30a0a54854cd291008c065dfaf45d610e3cd04] | committer: James Darnley avcodec/x86/mpegenc: check IDCT permutation type is a valid value > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86/mpegenc: support transpose permuation type

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Fri Jun 16 15:36:16 2017 +0200| [e3db94302c795a955d97566eaba8a0b0dfef788c] | committer: James Darnley avcodec/x86/mpegenc: support transpose permuation type > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86: allow future 8-bit simple idct to use slightly different coefficients

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Jun 12 12:32:57 2017 +0100| [8221c7170317b40d9ab8ca2810cd8f5afbd77e02] | committer: James Darnley avcodec/x86: allow future 8-bit simple idct to use slightly different coefficients > http://git.videolan.org/gitweb.cgi/f

[FFmpeg-cvslog] avcodec/x86: move simple_idct to external assembly

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu May 11 02:30:26 2017 +0200| [8e89f6fd37357361e0f4db5b6f3b422ce84175b0] | committer: James Darnley avcodec/x86: move simple_idct to external assembly > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec/x86/idctdsp_init: reindent

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon May 15 13:58:18 2017 +0200| [0dea0114fb29a8c533d7c36f20df7a7440605758] | committer: James Darnley avcodec/x86/idctdsp_init: reindent > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] configure: add audio_frame_queue dependency for aptx codec

ffmpeg | branch: master | James Darnley <james.darn...@gmail.com> | Sun Nov 19 14:54:34 2017 +0100| [0b7cd29d47a396d6afe01e5b4ef8382dc0ab6701] | committer: James Darnley configure: add audio_frame_queue dependency for aptx codec > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avcodec: add stride alignment needed for AVX-512

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Nov 6 15:43:39 2017 +0100| [8f86e6623811f7713d5e72c13797e20fffb3df62] | committer: James Darnley avcodec: add stride alignment needed for AVX-512 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avutil: add alignment needed for AVX-512

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Oct 26 19:51:37 2017 +0200| [e2218ed8ce6e2f208b9279acc80b7e2b7f1eae6c] | committer: James Darnley avutil: add alignment needed for AVX-512 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avutil: detect when AVX-512 is available

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Oct 26 19:51:02 2017 +0200| [4783a01c113b727164588ada47b3145ce5d2c367] | committer: James Darnley avutil: detect when AVX-512 is available > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] configure: test whether x86 assembler supports AVX-512

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Oct 26 19:45:50 2017 +0200| [b7a3d1f249c9da9f21d10ea9137e304961da02ba] | committer: James Darnley configure: test whether x86 assembler supports AVX-512 > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] avutil: add AVX-512 flags

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Thu Oct 26 19:48:26 2017 +0200| [8b81eabe5789105cf4dcf223c9d758ca19278cab] | committer: James Darnley avutil: add AVX-512 flags > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=8b81eabe5789105cf4dcf223c9d758c

[FFmpeg-cvslog] checkasm: support for AVX-512 functions

ffmpeg | branch: master | James Darnley <jdarn...@obe.tv> | Mon Oct 30 13:04:59 2017 +0100| [40d4b13228bf234a2b9649674d9ed25bd020f226] | committer: James Darnley checkasm: support for AVX-512 functions > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a

[FFmpeg-cvslog] fate: add more vc2 encoder tests

ffmpeg | branch: master | James Darnley | Mon Jan 29 17:40:00 2018 +0100| [088b07427dfe05de998609fe214017ebdb703a63] | committer: James Darnley fate: add more vc2 encoder tests > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=088b07427dfe05de998609fe214017ebdb703a63 --- te

[FFmpeg-cvslog] checkasm: add test for v210dec

ffmpeg | branch: master | James Darnley | Mon Mar 4 19:50:35 2019 +0100| [76c370af64e4e2cb582c678438c91a39ee49ac2e] | committer: James Darnley checkasm: add test for v210dec > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=76c370af64e4e2cb582c678438c91a39ee49ac2e --- te

[FFmpeg-cvslog] avcodec/v210dec: move DSP function setting into dedicated function

ffmpeg | branch: master | James Darnley | Mon Mar 4 23:54:58 2019 +0100| [e2cbf24f2738e9ead5d97977e7aaf096d7954ae3] | committer: James Darnley avcodec/v210dec: move DSP function setting into dedicated function Prepare for checkasm test. > http://git.videolan.org/gitweb.cgi/ffmpeg.

[FFmpeg-cvslog] avcodec/x86/v210: fix operands of vpblendd used in new avx2 code

ffmpeg | branch: master | James Darnley | Thu May 2 21:16:58 2019 +0200| [46f1718cd97c7ecb824cff29f6bc3f311046ff8d] | committer: James Darnley avcodec/x86/v210: fix operands of vpblendd used in new avx2 code Assembly failed when using yasm rather than nasm. > http://git.videolan.

[FFmpeg-cvslog] configure: support lsan as toolchain

ffmpeg | branch: master | James Darnley | Wed Dec 7 17:04:55 2022 +0100| [f1f41d6ff5fdfc993e69972c074584efa56012ee] | committer: James Darnley configure: support lsan as toolchain > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=f1f41d6ff5fdfc993e69972c074584efa5601

[FFmpeg-cvslog] avcodec/x86/v210: add some comments to the improved avx2 function

ffmpeg | branch: master | James Darnley | Thu Dec 15 11:19:30 2022 +0100| [f30b4c2f47ab689a570a9b36b2e96e78bb462691] | committer: James Darnley avcodec/x86/v210: add some comments to the improved avx2 function > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avcodec/x86: add avx512icl function for v210dec

ffmpeg | branch: master | James Darnley | Wed Dec 14 17:16:28 2022 +0100| [6af453ca389c56cb113876628cb173577faa9464] | committer: James Darnley avcodec/x86: add avx512icl function for v210dec Ice Lake (Xeon Silver 4316): 2.01x faster (1147±36.8 vs. 571±38.2 decicycles) compared with avx2

[FFmpeg-cvslog] avcodec/x86/v210enc: change '0b' binary constant prefix to 'b' suffix

ffmpeg | branch: master | James Darnley | Sat Dec 3 16:44:24 2022 +0100| [5dfb4f9690692409a4090cb9c3b05872627b267f] | committer: James Darnley avcodec/x86/v210enc: change '0b' binary constant prefix to 'b' suffix For compatability with yasm from 0.7.0 > http://git.videolan.org/gitweb.

[FFmpeg-cvslog] checkasm/v210dec: add extra space to the destination arrays

ffmpeg | branch: master | James Darnley | Wed Dec 21 00:36:49 2022 +0100| [eef763c7057a7f5f4b7dae7855d07b2a6da8b537] | committer: James Darnley checkasm/v210dec: add extra space to the destination arrays > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avcodec/x86/v210enc: expand and correct comments

ffmpeg | branch: master | James Darnley | Fri Nov 25 15:54:34 2022 +0100| [c67a2b14a2631040aea4dde9fe03f469f89bf7b0] | committer: James Darnley avcodec/x86/v210enc: expand and correct comments > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c67a2b14a2631040aea4dde9fe03f469f89bf

[FFmpeg-cvslog] avcodec/x86/v210enc: replace register use with named register

ffmpeg | branch: master | James Darnley | Tue Nov 8 17:51:53 2022 +0100| [bda53d2dde91b0ce5aa4c3de77b676338d3a2613] | committer: James Darnley avcodec/x86/v210enc: replace register use with named register > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] checkasm/v210enc: test the entire width of 10-bit planar input arrays

ffmpeg | branch: master | James Darnley | Tue Nov 8 17:46:49 2022 +0100| [cfd1c3c0a1bd52f3c91ffa260140ac67323bcc6d] | committer: James Darnley checkasm/v210enc: test the entire width of 10-bit planar input arrays > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avcodec/v210enc: add new 10-bit function for avx512 avx512icl

ffmpeg | branch: master | James Darnley | Tue Nov 8 17:49:05 2022 +0100| [651cb867b1eb09ef163e7635d3e13c782cf8e152] | committer: James Darnley avcodec/v210enc: add new 10-bit function for avx512 avx512icl avx512 on Skylake-X (Xeon D-2123IT): 1.19x faster (970±91.2 vs. 817±104.4 decicycles

[FFmpeg-cvslog] avcodec/x86/v210enc: remove unneeded instruction

ffmpeg | branch: master | James Darnley | Fri Nov 25 15:55:52 2022 +0100| [690b7890f0be414d6d27a952941082dc4d484d9b] | committer: James Darnley avcodec/x86/v210enc: remove unneeded instruction > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=690b7890f0be414d6d27a952941082dc4d484

[FFmpeg-cvslog] mailmap: stop git lying about who I commit things as

ffmpeg | branch: master | James Darnley | Fri Oct 28 16:04:42 2022 +0200| [f53c590f3d18799bfb85a383d0738b28fe2fd08e] | committer: James Darnley mailmap: stop git lying about who I commit things as > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] checkasm: add a verbose check function for uint32_t data

ffmpeg | branch: master | James Darnley | Wed Oct 26 17:51:36 2022 +0200| [1936c06f029bd0e793aea47767b06e72a497c268] | committer: James Darnley checkasm: add a verbose check function for uint32_t data > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] avcodec/v210enc: add new function for avx2 avx512 avx512icl

ffmpeg | branch: master | James Darnley | Tue Oct 25 16:27:38 2022 +0200| [c3d36e1b3de1e2b8a84f47d0369dcea41c8cd351] | committer: James Darnley avcodec/v210enc: add new function for avx2 avx512 avx512icl Negligible speed difference for avx2 on Zen 2 (Ryzen 5700X) and Broadwell (Xeon E5-2620

[FFmpeg-cvslog] avutil/tests/cpu: print the avx512icl flag

ffmpeg | branch: master | James Darnley | Wed Oct 26 17:40:03 2022 +0200| [0f252dfa95568f87f87ed47b63ea8ea091e76048] | committer: James Darnley avutil/tests/cpu: print the avx512icl flag > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=0f252dfa95568f87f87ed47b63ea8ea091e76

[FFmpeg-cvslog] tests: actually test yadif's 10 and 16-bit functions

ffmpeg | branch: master | James Darnley | Fri Feb 17 14:08:10 2023 +0100| [e50a02b0f674f050e8b0e6e07342e27d1fc872e0] | committer: James Darnley tests: actually test yadif's 10 and 16-bit functions > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi