commit_msg
stringlengths 1
24.2k
| commit_hash
stringlengths 2
84
⌀ | project
stringlengths 2
40
| source
stringclasses 4
values | labels
int64 0
1
| repo_url
stringlengths 26
70
⌀ | commit_url
stringlengths 74
118
⌀ | commit_date
stringlengths 25
25
⌀ |
|---|---|---|---|---|---|---|---|
Revert hack which should have never been commited.
------------------------------------------------------------------------
r464 | philipjsg | 2002-05-09 03:19:15 +0200 (Thu, 09 May 2002) | 3 lines
* Add a sleep when doing the post to ffserver. Yes, this is the wrong
solution.
Originally committed as revision 11336 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 359154bc1bb9d935e25dc2a3d818152971ddd133 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/359154bc1bb9d935e25dc2a3d818152971ddd133 | 2007-12-28 03:32:32+00:00 |
Make sure rtjpeg quality is a valid value
Originally committed as revision 10444 to svn://svn.ffmpeg.org/ffmpeg/trunk
| fd065d0593747eb38112bf02e02b23cd5eb6cba2 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fd065d0593747eb38112bf02e02b23cd5eb6cba2 | 2007-09-09 09:16:47+00:00 |
Add extralibs to the dependency checker
Originally committed as revision 10532 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 4ec45764a09b78e556e10cd65f2d2f5a57900990 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/4ec45764a09b78e556e10cd65f2d2f5a57900990 | 2007-09-18 19:07:52+00:00 |
Check for getrusage
Originally committed as revision 9627 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 66be5b4536f9edebbbfe8146d44767e4f93d15e4 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/66be5b4536f9edebbbfe8146d44767e4f93d15e4 | 2007-07-13 16:04:00+00:00 |
10l, add check forgotten in last commit
Originally committed as revision 10442 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 048905a484cde2abc051ed965d5ac6b973acbc79 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/048905a484cde2abc051ed965d5ac6b973acbc79 | 2007-09-09 08:52:53+00:00 |
Add checking for codecid to has_codec_parameters().
patch by Ronald S. Bultje, rsbultje gmail com
Date: Sat, 29 Sep 2007 09:25:52 -0400
Subject: [FFmpeg-devel] [PATCH] has_codec_parameters() addition
Originally committed as revision 10636 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 3303926c2f06841270281e7f5210c0c94292e089 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/3303926c2f06841270281e7f5210c0c94292e089 | 2007-10-01 22:11:10+00:00 |
re pipeline loop, to eliminate extra chroma reads
Originally committed as revision 23602 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
| e9d4375f1a8760ca3b82321d53f26528c5e66a81 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/e9d4375f1a8760ca3b82321d53f26528c5e66a81 | 2007-06-22 00:12:44+00:00 |
gpg and passwords rule
Originally committed as revision 8414 to svn://svn.ffmpeg.org/ffmpeg/trunk
| a7a113157a5a593f219afd0ed677f24e88e04693 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a7a113157a5a593f219afd0ed677f24e88e04693 | 2007-03-15 12:54:48+00:00 |
Group benchmark items in the patch submission checklist together.
Originally committed as revision 8887 to svn://svn.ffmpeg.org/ffmpeg/trunk
| b0fee7c53c326a28f42905acd5e0ed137389a713 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b0fee7c53c326a28f42905acd5e0ed137389a713 | 2007-05-04 19:37:01+00:00 |
check for no frame and prevent segfault
Originally committed as revision 8211 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 56c70e1d3a1cde2f7490780a4924e6b6e87e4c47 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/56c70e1d3a1cde2f7490780a4924e6b6e87e4c47 | 2007-03-03 21:33:33+00:00 |
negative stride and h_size-=8 fix
Originally committed as revision 22247 to svn://svn.mplayerhq.hu/mplayer/trunk/libswscale
| 20da77449d4427a7152b80e4f9acce6a8c93ee7d | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/20da77449d4427a7152b80e4f9acce6a8c93ee7d | 2007-02-17 13:28:44+00:00 |
Add two extra needed bounds checks
Originally committed as revision 7777 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 5fe9c42cebbac9e4875f8ff30eb84ea9f8f5b407 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/5fe9c42cebbac9e4875f8ff30eb84ea9f8f5b407 | 2007-01-31 10:29:58+00:00 |
Remove useless check, patch by Benoit Fouet, benoit.fouet purplelabs com.
Originally committed as revision 7821 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 487747de57712fbe548ce6bf4ada7f7643370092 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/487747de57712fbe548ce6bf4ada7f7643370092 | 2007-02-05 09:11:10+00:00 |
factor out dequant table lookup outside loops, gives a 1-2% speed-up
patch by Andreas Öman %andreas A olebyn P nu%
original thread:
Date: Sep 24, 2007 12:59 PM
Subject: [FFmpeg-devel] [PATCH] h264: factor out dequant table lookup outside loops
Originally committed as revision 10564 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 1136d34a44bba555601dda248b2d09f6c35394d3 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/1136d34a44bba555601dda248b2d09f6c35394d3 | 2007-09-24 13:01:15+00:00 |
Use disabled() to check for GPL.
Originally committed as revision 7744 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 47f9cc15cce4c92d44d675498384ee3b0fcdafbb | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/47f9cc15cce4c92d44d675498384ee3b0fcdafbb | 2007-01-28 16:28:03+00:00 |
100l: Initialize dc_scale with current quantizer for adv I frames
Originally committed as revision 7169 to svn://svn.ffmpeg.org/ffmpeg/trunk
| b8ce8f15a036780bd5ee655bcac881a8cd62f85a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b8ce8f15a036780bd5ee655bcac881a8cd62f85a | 2006-11-26 04:53:36+00:00 |
mpeg4probe patch by (Thijs Vermeir ; thijs vermeir barco com)
tabs removed by me
VOL vs. VOP check fixed by me
Originally committed as revision 6940 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 8f57cc5a328d492b132a3ced7d245093431f2721 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/8f57cc5a328d492b132a3ced7d245093431f2721 | 2006-11-08 01:21:23+00:00 |
seperate redo (we dont handle this possibly valid packet) from
error_redo (somethings is definitly wrong) in which case we try to resync from the last startcode
Originally committed as revision 7560 to svn://svn.ffmpeg.org/ffmpeg/trunk
| e56cfad04b73be35ea8dcb5fd5e8451f6924aff1 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/e56cfad04b73be35ea8dcb5fd5e8451f6924aff1 | 2007-01-17 10:19:10+00:00 |
Correct GET/PUT_UTF8 comment: the get/put functions might be called up
to 7 times, though only up to 4 times for valid utf-8 values.
Originally committed as revision 7124 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 40a08c7e9ab0ce2603a22ccacc02252bfa035b9f | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/40a08c7e9ab0ce2603a22ccacc02252bfa035b9f | 2006-11-19 09:43:26+00:00 |
10l inverted condition check generated an endless loop
Originally committed as revision 6722 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 595da759de5a94def0ee409a72b07d0be36f5d2a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/595da759de5a94def0ee409a72b07d0be36f5d2a | 2006-10-17 17:19:11+00:00 |
Update checksum to match latest ASF patch
Originally committed as revision 6566 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 25b6ce46ea27b787a83a5ec1f497a0363e27c6ca | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/25b6ce46ea27b787a83a5ec1f497a0363e27c6ca | 2006-10-06 09:45:41+00:00 |
add support for while() loops again ugly syntax while(condition, statements) but very simple implementation
Originally committed as revision 6814 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 45ee0e3282f9683fc1d5f1528f4edb889c4118a3 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/45ee0e3282f9683fc1d5f1528f4edb889c4118a3 | 2006-10-27 22:35:55+00:00 |
Use overlap filter instead of loop filter.
Originally committed as revision 5685 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 65c7bb9ec4f521a46577a1e87d71ad9a8deee6cd | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/65c7bb9ec4f521a46577a1e87d71ad9a8deee6cd | 2006-07-09 02:47:09+00:00 |
Add x86 conditionnal mov (CMOV) "detection" (based on the cpu passed by --cpu=xx)
Sets HAVE_CMOV define so we can protect relevant code with #ifdef HAVE_CMOV
Originally committed as revision 6742 to svn://svn.ffmpeg.org/ffmpeg/trunk
| e7768fc512cbe6bc22fd5d0bc22bbf4c95001e9a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/e7768fc512cbe6bc22fd5d0bc22bbf4c95001e9a | 2006-10-20 07:27:12+00:00 |
check for CODEC_ID_NONE instead 0, tobe compatible with movenc.c and possible CODEC_ID_NONE value change
Originally committed as revision 5043 to svn://svn.ffmpeg.org/ffmpeg/trunk
| fb7a2bf6956173eda6f9caceef8599fa4f83500d | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fb7a2bf6956173eda6f9caceef8599fa4f83500d | 2006-02-21 20:32:47+00:00 |
tinfoil patch: make sure pixel_ptr never goes negative
Originally committed as revision 4513 to svn://svn.ffmpeg.org/ffmpeg/trunk
| a06c7e07d71547f2dd46890c0dbf230d8542e8cc | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a06c7e07d71547f2dd46890c0dbf230d8542e8cc | 2005-08-13 17:46:09+00:00 |
Remove the "bf" OptionDef parameter so the equally named AVOption will take
over. Move over ffmpeg.c constraint checking, but allow value of zero to disable
B-frame.
Originally committed as revision 6296 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 315b0f974252120cfacb0346954a2d817dff279a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/315b0f974252120cfacb0346954a2d817dff279a | 2006-09-20 08:34:13+00:00 |
Make faac_decode_frame return -1 on error and not 0 which avoids
stalling and infinite loop.
Patch by Baptiste COUDURIER <baptiste coudurier at smartjog com>
Originally committed as revision 5110 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 980bbb13d653561d83619350db32ccb5e5248f95 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/980bbb13d653561d83619350db32ccb5e5248f95 | 2006-03-05 12:32:24+00:00 |
quick and dirty PSNR check for audio
Originally committed as revision 3902 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 16dd93ab6adbe1ac330212e70a0b7538b4ec571b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/16dd93ab6adbe1ac330212e70a0b7538b4ec571b | 2005-01-30 13:26:11+00:00 |
disable broken loop filter optimization
melanson please test your optimizations before commiting!
Originally committed as revision 4292 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 720537282b6878d64812c2d89579713e22bc167d | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/720537282b6878d64812c2d89579713e22bc167d | 2005-05-21 07:35:15+00:00 |
cut-n-paste error slipped through the cracks... thanks to DJ Hartman for
the report
Originally committed as revision 3226 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 13b4c18eb879d8a83eb47c9b7b62e387b666926e | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/13b4c18eb879d8a83eb47c9b7b62e387b666926e | 2004-06-16 02:53:12+00:00 |
checksum threshold removial
Originally committed as revision 2992 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 2b382987fdd7dd412f3e750ac44c98febb96107b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/2b382987fdd7dd412f3e750ac44c98febb96107b | 2004-04-09 16:47:22+00:00 |
avoid negative array indices
Originally committed as revision 2794 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 69f5de1855fb7a8d188f8f915887a9b00c7795f2 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/69f5de1855fb7a8d188f8f915887a9b00c7795f2 | 2004-02-18 12:49:30+00:00 |
correctly interleave packets during encoding
dts/pts fixed for streamcopy
dont use coded_frame->key_frame hack in muxers, use AVPacket.flags instead
Originally committed as revision 3171 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 3c895fc098f7637f6d5ec3a9d6766e724a8b9e41 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/3c895fc098f7637f6d5ec3a9d6766e724a8b9e41 | 2004-05-29 18:50:31+00:00 |
improving 3IV1 support (still far from perfect) note, disabled as it requires some checks in the inner loops which would slow things down a tiny bit
Originally committed as revision 1889 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 6d71b3820759235c02493ac73ef5949b653ad29b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/6d71b3820759235c02493ac73ef5949b653ad29b | 2003-05-18 00:07:17+00:00 |
Initial checkin for a cheaper text drawing function than imlib2.c.
Originally committed as revision 1577 to svn://svn.ffmpeg.org/ffmpeg/trunk
| df4f91471cd6e1d7aa3501ca0f2aee6b87e6b02c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/df4f91471cd6e1d7aa3501ca0f2aee6b87e6b02c | 2003-02-11 04:01:55+00:00 |
squashed a bunch of subtle array indexing bugs, fixed block mapping
added more error checking, added (and enabled, for the time being) a
keyframe-only mode
Originally committed as revision 1849 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 892fc83e88a20f9543c6c5be3626712be7a2e6f2 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/892fc83e88a20f9543c6c5be3626712be7a2e6f2 | 2003-05-10 21:46:17+00:00 |
Fix a potential endless loop (nb (1 * 3) / 2 == 1)
Originally committed as revision 823 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 64e105e051ca3e5088b0db64551244482b2836b4 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/64e105e051ca3e5088b0db64551244482b2836b4 | 2002-07-27 03:05:45+00:00 |
optimization tips
can allso be used as a torture test for ur spelling checker
Originally committed as revision 1444 to svn://svn.ffmpeg.org/ffmpeg/trunk
| a552591fb13988d7f46c83576fd339320d720d5c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a552591fb13988d7f46c83576fd339320d720d5c | 2003-01-11 10:55:17+00:00 |
* support for AAC audio streams via libfaad
* could be complied with runtime linking of libfaad (CONFIG_FAADBIN)
* missing/untested support for raw AAC streams (maybe there will be
just one entry for aac)
* missing configure check for libfaad/config option
Originally committed as revision 1857 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 280bd7b7de2e05b695228fbeaa795fc8e4922b6c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/280bd7b7de2e05b695228fbeaa795fc8e4922b6c | 2003-05-12 10:55:19+00:00 |
Let loops will be aligned
Originally committed as revision 213 to svn://svn.ffmpeg.org/ffmpeg/trunk
| f12e206b4737b84afa6071f4b464664a50b0c916 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/f12e206b4737b84afa6071f4b464664a50b0c916 | 2001-11-11 17:55:54+00:00 |
H261 fixing and cleaning:
-corrected wrong value in mv data
-set correct mb_type after adjusting index
-don't use H263 loop filter when the loop filter flag is set but when
using the H261 encoder
-use the same unquantizer as H263 (which is optimized btw)
-removed unused members in H261Context
patch by (Maarten Daniels <maarten.daniels >at< luc >dot< ac >dot< be>)
regression test checksum update by me
Originally committed as revision 3669 to svn://svn.ffmpeg.org/ffmpeg/trunk
| ccff9da62a833238db7a22eb39be0814f522c2c5 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/ccff9da62a833238db7a22eb39be0814f522c2c5 | 2004-11-12 01:21:34+00:00 |
lavc/snow_dwt: add struct MpegEncContext to fix headers check.
add missing struct MpegEncContext, use make checkheaders
found this warning.
Signed-off-by: Jun Zhao <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
| a919ab853efcecf89516e703df8018d0649e1143 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a919ab853efcecf89516e703df8018d0649e1143 | 2018-01-16 22:02:34+01:00 |
avcodec: v4l2_m2m: remove unnecessary timeout.
Qualcomm's db410c/db820 Venus driver currently present in mainline
kernel has a bug which mishandles the CMD_STOP requests causing the
decoder to block while draining [1].
This patch removes the workaround that was used to prevent that
situation.
Encoding/Decoding tested on db820c.
[1] on CMD_STOP, the driver is flushing all buffers and never raising
IPIPE which ends up in blocking on poll.
| 5d5de3eba4c7890c2e8077f5b4ae569671d11cf8 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/5d5de3eba4c7890c2e8077f5b4ae569671d11cf8 | 2018-01-21 00:37:35+00:00 |
vf_paletteuse: Add error checking to apply_palette
This fixes a segfault caused by passing NULL to ff_filter_frame
when an error occurs.
Signed-off-by: Derek Buitenhuis <[email protected]>
| 6470abc740367cc881c181db866891f8dd1d342f | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/6470abc740367cc881c181db866891f8dd1d342f | 2018-01-03 13:02:15-05:00 |
avfilter/af_asetnsamples: add missing error check
Signed-off-by: Paul B Mahol <[email protected]>
| 000836c2a98e1c6a2867dd9db0371c137609acf0 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/000836c2a98e1c6a2867dd9db0371c137609acf0 | 2017-11-21 18:29:39+01:00 |
beos/mov/adpcm patch by François Revol <revol at free dot fr>
* Some BeOS fixes:
- errno stuff
- nanosleep() replacement
- added a doc/README.beos
* mov reader still has problems with most videos (skips many chunks),
- It should now read .mov files with zlib-compressed moov headers (aka cmov)
- added SVQ1 support.
- removed mapping 'raw ' and 'yuv2' to h263, was my mistake.
- added IMA4 support. (tested)
- fixed frame rate (it reported 0 fps or something before)
- extended file probing ('wide' and 'free' atoms)
- improved .mov/.mp4 detection (or so I think)
* adpcm
* added zlib support, with header file and lib check
Originally committed as revision 780 to svn://svn.ffmpeg.org/ffmpeg/trunk
| 0147f198569a8e8d857df184d29ad22c8089b91e | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/0147f198569a8e8d857df184d29ad22c8089b91e | 2002-07-20 20:05:50+00:00 |
avformat/dashenc: Added configuration to override HTTP User-Agent
Reviewed-by: Steven Liu <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
| d24e08e978792e09d212018677d1c0b8208ecef8 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/d24e08e978792e09d212018677d1c0b8208ecef8 | 2017-11-21 03:13:03+01:00 |
avfilter/af_afir: add missing error check
Signed-off-by: Paul B Mahol <[email protected]>
| d1b284119bd5c6a52124443de2c45dbe569c25fc | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/d1b284119bd5c6a52124443de2c45dbe569c25fc | 2017-11-20 13:54:04+01:00 |
opus_pvq: do not compile encoding/decoding code if the encoder/decoder is disabled
This should save quite a bit of space if either has been disabled for size reasons.
Could just check if the encoding flag is set during runtime on every single location,
however the overhead of branch misses would somewhat decrease performance.
Signed-off-by: Rostislav Pehlivanov <[email protected]>
| 7b46add7257628bffac96d3002308d1f9e1ed172 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/7b46add7257628bffac96d3002308d1f9e1ed172 | 2017-12-04 10:51:19+00:00 |
nvdec: add frames_params support
| 7546964f96168cd6ac819ef4c3212ee586619f1a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/7546964f96168cd6ac819ef4c3212ee586619f1a | 2017-11-11 20:33:45-03:00 |
avcodec/nvdec: warn about thread count if applicable
| 538de4354dcd6c57154c5a5dec0744dcaa06b874 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/538de4354dcd6c57154c5a5dec0744dcaa06b874 | 2017-11-13 20:33:10+01:00 |
avformat/yuv4mpeg: add gray9/10/12 support
Signed-off-by: Mateusz Brzostek <[email protected]>
| a4743d2574254aa0c494b337947e8c9880c7ead7 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a4743d2574254aa0c494b337947e8c9880c7ead7 | 2017-10-26 18:32:28+02:00 |
Use ff_thread_once for fixed, float table init.
These tables are static so they should only be initialized once
instead of on every call to ff_mpadsp_init().
Signed-off-by: Dale Curtis <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
| 5eaaffaf64d1854493f0fe9ec822eed1b3cd9fe1 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/5eaaffaf64d1854493f0fe9ec822eed1b3cd9fe1 | 2017-11-18 11:57:57+01:00 |
vaapi_h264: Add missing return value check
Fixes CID 1419829.
| 03b1470088e6e4a11e5db0efd8c26d9ac8304d31 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/03b1470088e6e4a11e5db0efd8c26d9ac8304d31 | 2017-10-18 20:05:11+01:00 |
avcodec: Implement mpeg4 nvdec hwaccel
This was predictably nightmarish, given how ridiculous mpeg4 is.
I had to stare at the cuvid parser output for a long time to work
out what each field was supposed to be, and even then, I still don't
fully understand some of them. Particularly:
vop_coded: If I'm reading the decoder correctly, this flag will always
be 1 as the decoder will not pass the hwaccel any frame
where it is not 1.
divx_flags: There's obviously no documentation on what the possible
flags are. I simply observed that this is '0' for a
normal bitstream and '5' for packed b-frames.
gmc_enabled: I had a number of guesses as to what this mapped to.
I picked the condition I did based on when the cuvid
parser was setting flag.
Also note that as with the vdpau hwaccel, the decoder needs to
consume the entire frame and not the slice.
| 6b77a10e43e1a8ed8ead20f344485d400440055c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/6b77a10e43e1a8ed8ead20f344485d400440055c | 2017-11-20 07:21:41-08:00 |
configure: add missing zlib extralibs to the libmysofa check
Should fix checks with static builds.
Signed-off-by: James Almer <[email protected]>
| 6b52c0b583b84aa5290f3c8a022deefe8dc2fed1 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/6b52c0b583b84aa5290f3c8a022deefe8dc2fed1 | 2017-10-13 11:22:54-03:00 |
lavc/v4l2: Add missing header include
Fixes checkheaders.
| da53c424b96f57d7471767d99072d35efe15a8e0 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/da53c424b96f57d7471767d99072d35efe15a8e0 | 2017-09-27 23:10:03+01:00 |
avcodec/rkmppdec: check wether typo
Signed-off-by: Michael Niedermayer <[email protected]>
| 7bec3f78da2533968b7246ff222770582ab4aafb | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/7bec3f78da2533968b7246ff222770582ab4aafb | 2017-10-11 02:15:21+02:00 |
avfilter/tinterlace: Simplify checks for lowpass filtering flags
| 3af1060319b46005dbfb3b01f9104539caf30146 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/3af1060319b46005dbfb3b01f9104539caf30146 | 2017-09-18 00:15:58-03:00 |
lavc/v4l2_context: Change the type of the ioctl cmd to uint32_t.
Fixes a warning on recent Linux:
libavcodec/v4l2_context.c: In function 'ff_v4l2_context_set_status':
libavcodec/v4l2_context.c:496:26: warning: comparison is always false due to limited range of data type
| 2cc51d5025c976aa268a854df1eec86014512c8c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/2cc51d5025c976aa268a854df1eec86014512c8c | 2017-11-05 19:25:29+01:00 |
lavc: external hardware frame pool initialization
This adds a new API, which allows the API user to query the required
AVHWFramesContext parameters. This also reduces code duplication across
the hwaccels by introducing ff_decode_get_hw_frames_ctx(), which uses
the new API function. It takes care of initializing the hw_frames_ctx
if needed, and does additional error handling and API usage checking.
Support for VDA and Cuvid missing.
Signed-off-by: Anton Khirnov <[email protected]>
| b46a77f19ddc4b2b5fa3187835ceb602a5244e24 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b46a77f19ddc4b2b5fa3187835ceb602a5244e24 | 2017-10-19 17:10:59+02:00 |
avcodec/prores_kostya: increase bits usage when alpha is used
Also fix undefined left shift of negative variable.
Signed-off-by: Paul B Mahol <[email protected]>
| 315f51128a95ca34ac3212c86b2a938330ba6b6e | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/315f51128a95ca34ac3212c86b2a938330ba6b6e | 2017-06-25 17:36:54+02:00 |
bink: Split read_dct_coeffs()
This works around type aliasing violations and related warnings.
Also add some missing error checking.
| fd92dafaff8844b5fedf94679b93d953939a7f7b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fd92dafaff8844b5fedf94679b93d953939a7f7b | 2017-07-05 15:06:41+02:00 |
avcodec/hevcdec: Check nb_sps
Signed-off-by: Michael Niedermayer <[email protected]>
| bc406744620710911de9157eafa3e61d0246566f | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/bc406744620710911de9157eafa3e61d0246566f | 2017-06-15 03:09:40+02:00 |
avcodec/hevcdec: Check beta and tc offset in hls_slice_header()
Signed-off-by: Michael Niedermayer <[email protected]>
| fdfbcbdd148765ba0fa5093d7feb944685a3e8c2 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fdfbcbdd148765ba0fa5093d7feb944685a3e8c2 | 2017-06-21 02:25:23+02:00 |
checkasm: aarch64: Specify alignment for the register_init const array
Loads from this strictly doesn't require alignment, but specify it
just for consistency with the arm version.
Signed-off-by: Martin Storsjö <[email protected]>
| d05c9cde0e87c23ca42957646bea483dfc09d6bf | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/d05c9cde0e87c23ca42957646bea483dfc09d6bf | 2017-05-15 10:19:46+03:00 |
mov: Validate cubemap layout
| 114324b9e04505c71d9a3ffd854f25062fa3678f | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/114324b9e04505c71d9a3ffd854f25062fa3678f | 2017-03-07 11:07:02-05:00 |
hwcontext: Make it easier to work with device types
Adds functions to convert to/from strings and a function to iterate
over all supported device types. Also adds a new invalid type
AV_HWDEVICE_TYPE_NONE, which acts as a sentinel value.
| b7487f4f3c39b4b202e1ea7bb2de13902f2dee45 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b7487f4f3c39b4b202e1ea7bb2de13902f2dee45 | 2017-03-20 23:15:43+00:00 |
avcodec/wrapped_avframe: allocate a buffer with padding
This ensures that the wrapped avframe will not get reallocated later, which
would invalidate internal references such as extended data.
Reviewed-by: wm4 <[email protected]>
Signed-off-by: Marton Balint <[email protected]>
| 436f00b10c062b75c7aab276c4a7d64524bd0444 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/436f00b10c062b75c7aab276c4a7d64524bd0444 | 2017-02-22 23:11:42+01:00 |
avcodec/flacdec: Check for invalid vlcs
Signed-off-by: Michael Niedermayer <[email protected]>
| fd00203554334f602ee1d7c5416ff25f356f4fb7 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fd00203554334f602ee1d7c5416ff25f356f4fb7 | 2017-02-04 02:26:37+01:00 |
Factor CHECKED out of DEBUG so it can be set seperatly
Suggested-by: Matt Wolenetz <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
| e8d4eacc07c61ae24f48451073a2620d8d257d33 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/e8d4eacc07c61ae24f48451073a2620d8d257d33 | 2017-02-19 15:39:36+01:00 |
lavf/mov.c: Avoid heap allocation wrap in mov_read_hdlr
Core of patch is from [email protected]
Reference https://crbug.com/643950
Signed-off-by: Michael Niedermayer <[email protected]>
Check value reduced as the code does not support larger lengths
| fd30e4d57fe5841385f845440688505b88c0f4a9 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fd30e4d57fe5841385f845440688505b88c0f4a9 | 2017-02-08 03:31:01+01:00 |
avutil/tile: check clock_gettime at runtime for apple platforms
clock_gettime is avalible since macOS 10.12 and iOS 10.0. Because of
weak linking, clock_gettime can be build without error with new
macOS/iOS sdk, but the symbol may not exist on the target system.
Explicitly checking the symbol is required.
https://developer.apple.com/library/content/documentation/MacOSX/Conceptual/BPFrameworks/Concepts/WeakLinking.html
Signed-off-by: Wang Bin <[email protected]>
Signed-off-by: Steven Liu <[email protected]>
| 0e8b68a2c4031e25082603ad88711be12210d41f | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/0e8b68a2c4031e25082603ad88711be12210d41f | 2017-01-09 11:52:06+08:00 |
doc: clarify option on looping infinitely in movie filter
Clarify that setting loop=0 is required to make the stream loop infinitely, rather than saying that a value "less than 1" is needed.
Signed-off-by: Lou Logan <[email protected]>
| 8addd56554eb240f38abfadb8bedf3ff8efc965a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/8addd56554eb240f38abfadb8bedf3ff8efc965a | 2017-02-03 13:08:58-09:00 |
avcodec/h264: mmxext 4:2:2 chroma intra deblock/loop filter
2.1 times faster (401 vs. 194 cycles)
| 815ea8c6ccf7a1a8154a829f948d7c98120ad084 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/815ea8c6ccf7a1a8154a829f948d7c98120ad084 | 2016-11-30 22:58:27+01:00 |
avcodec/flacdsp_template: Fix undefined shift in flac_decorrelate_indep_c
Fixes: left shift of negative value
Fixes: 668346-media
Found-by: Matt Wolenetz <[email protected]>
Signed-off-by: Michael Niedermayer <[email protected]>
| acc163c6ab52d2235767852262c64c7f6b273d1c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/acc163c6ab52d2235767852262c64c7f6b273d1c | 2016-12-03 17:25:40+01:00 |
dds: limit 4 bpp handling to AV_PIX_FMT_PAL8
This fixes NULL pointer dereferencing for formats, where frame->data[1]
is not allocated.
The problem was introduced in commit
257fbc3af4cba08ac471dab68924182160bde6fd.
Reviewed-by: Paul B Mahol <[email protected]>
Signed-off-by: Andreas Cadhalpun <[email protected]>
| 90ebf3c428352eb1d4116bf97b470ceca295d7d6 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/90ebf3c428352eb1d4116bf97b470ceca295d7d6 | 2016-11-16 19:29:45+01:00 |
configure: Simplify and fix avfoundation indev handling
Handle extralibs in the standard way, add missing pthreads dependency.
Also globally check for -fobj-arc with Objective-C compilers since
that option is useful for other Objective-C code as well.
| 29d2e03ed96d9f7b8cb2cc87dca0a63c2a6eff70 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/29d2e03ed96d9f7b8cb2cc87dca0a63c2a6eff70 | 2016-12-05 18:07:24+01:00 |
bmv: fix apparent sign error in the frame_off check
Fixes part of Ticket1373
Signed-off-by: Michael Niedermayer <[email protected]>
| debbcfae6010f027a0334d70d0dbb7ddd912ad5a | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/debbcfae6010f027a0334d70d0dbb7ddd912ad5a | 2012-06-02 04:18:13+02:00 |
avcodec/msvideo1: Check side data size before use
Fixes out of array read
Signed-off-by: Michael Niedermayer <[email protected]>
| 161ccdaa06d1d109e8f77d2535bda11ce02720f5 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/161ccdaa06d1d109e8f77d2535bda11ce02720f5 | 2016-10-31 01:08:45+01:00 |
avcodec/8bps: Check side data size before use
Fixes out of array read
Signed-off-by: Michael Niedermayer <[email protected]>
| 042faa847feea820451c474af0034fd3de9cff82 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/042faa847feea820451c474af0034fd3de9cff82 | 2016-10-30 14:15:00+01:00 |
avformat/movenc: Check frame rate in mov_write_uuidprof_tag()
Signed-off-by: Michael Niedermayer <[email protected]>
| ed2112fb36d7407d960b4f44475a700a7c44344c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/ed2112fb36d7407d960b4f44475a700a7c44344c | 2016-10-10 00:58:16+02:00 |
lavc/movtextdec.c: Avoid infinite loop on invalid data.
Signed-off-by: Sasi Inguva <[email protected]>
| 7e9e1b7070242a79fa6e3acd749d7fe76e39ea7b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/7e9e1b7070242a79fa6e3acd749d7fe76e39ea7b | 2016-09-27 19:49:02-07:00 |
af_hdcd: disable auto-convert by default
As all known valid HDCD sample formats and sample rates are now handled
by the filter, remove the scan that "invades the privacy" of the filter graph
and turn off autoconvert by default as requested by Nicolas George.
http://ffmpeg.org/pipermail/ffmpeg-devel/2016-August/197571.html
Signed-off-by: Burt P <[email protected]>
| 2c3d93648768cf2c00ea1731e096f698c0fae7f6 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/2c3d93648768cf2c00ea1731e096f698c0fae7f6 | 2016-10-05 12:52:56-05:00 |
avcodec/hevcpred: fix make checkheaders.
| 0bf85803545d14b004d86a741a1cde7fb9530c0c | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/0bf85803545d14b004d86a741a1cde7fb9530c0c | 2013-10-17 21:35:40+02:00 |
avcodec/cuvid: check for and warn about invalid pkt_timebase
| b91e0e59874d28f35a28acb4f886ab1e7d80cacb | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b91e0e59874d28f35a28acb4f886ab1e7d80cacb | 2016-09-09 17:56:07+02:00 |
af_hdcd: PE table limits
Signed-off-by: Burt P <[email protected]>
| 0e0f8859ba0af33e1145a4c4022e964011e2d75b | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/0e0f8859ba0af33e1145a4c4022e964011e2d75b | 2016-08-21 13:48:46-05:00 |
vf_colorspace: Interpret unspecified color range as limited range
This is the assumption that is made in pixel format conversion do
throughout the code (in particular swscale), and BT-specifications
mandate.
Add a warning to inform the user that an automatic selection is being
made.
Signed-off-by: Vittorio Giovara <[email protected]>
Signed-off-by: Ronald S. Bultje <[email protected]>
| cb78d14cf9b5ab59b4a9177f390f5e1abff58644 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/cb78d14cf9b5ab59b4a9177f390f5e1abff58644 | 2016-09-19 08:26:49-04:00 |
configure: Check for arc4random() header definition.
Fixes Cygwin compilation.
| 79dc94a63b30369e39792e81cb032f2cf40539b4 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/79dc94a63b30369e39792e81cb032f2cf40539b4 | 2016-08-13 12:14:01+02:00 |
arm: vp9: Add NEON itxfm routines
This work is sponsored by, and copyright, Google.
For the transforms up to 8x8, we can fit all the data (including
temporaries) in registers and just do a straightforward transform
of all the data. For 16x16, we do a transform of 4x16 pixels in
4 slices, using a temporary buffer. For 32x32, we transform 4x32
pixels at a time, in two steps of 4x16 pixels each.
Examples of relative speedup compared to the C version, from checkasm:
Cortex A7 A8 A9 A53
vp9_inv_adst_adst_4x4_add_neon: 3.39 5.83 4.17 4.01
vp9_inv_adst_adst_8x8_add_neon: 3.79 4.86 4.23 3.98
vp9_inv_adst_adst_16x16_add_neon: 3.33 4.36 4.11 4.16
vp9_inv_dct_dct_4x4_add_neon: 4.06 6.16 4.59 4.46
vp9_inv_dct_dct_8x8_add_neon: 4.61 6.01 4.98 4.86
vp9_inv_dct_dct_16x16_add_neon: 3.35 3.44 3.36 3.79
vp9_inv_dct_dct_32x32_add_neon: 3.89 3.50 3.79 4.42
vp9_inv_wht_wht_4x4_add_neon: 3.22 5.13 3.53 3.77
Thus, the speedup vs C code is around 3-6x.
This is mostly marginally faster than the corresponding routines
in libvpx on most cores, tested with their 32x32 idct (compared to
vpx_idct32x32_1024_add_neon). These numbers are slightly in libvpx's
favour since their version doesn't clear the input buffer like ours
do (although the effect of that on the total runtime probably is
negligible.)
Cortex A7 A8 A9 A53
vp9_inv_dct_dct_32x32_add_neon: 18436.8 16874.1 14235.1 11988.9
libvpx vpx_idct32x32_1024_add_neon 20789.0 13344.3 15049.9 13030.5
Only on the Cortex A8, the libvpx function is faster. On the other cores,
ours is slightly faster even though ours has got source block clearing
integrated.
Signed-off-by: Martin Storsjö <[email protected]>
| a67ae67083151f2f9595a1f2d17b601da19b939e | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/a67ae67083151f2f9595a1f2d17b601da19b939e | 2016-11-11 11:09:05+02:00 |
decklink: Remove unneeded spaces in initializers
| e9659d1a0b77c65be114fe5cd4228cac3356f6eb | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/e9659d1a0b77c65be114fe5cd4228cac3356f6eb | 2016-08-02 22:46:28-07:00 |
avcodec/cfhd: Increase minimum band dimension to 3
The implementation does not currently support len=2
Fixes out of array accesses
Fixes: 29d1b3db5ba2205e82b0b3a533e057a3/asan_heap-oob_12b650c_9254_3b8c4e4d931eb2c32841c18ebb297f1d.avi
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <[email protected]>
| b8b36717217c6f45db71c77ad4e7c65521e7d9ff | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/b8b36717217c6f45db71c77ad4e7c65521e7d9ff | 2016-08-19 22:28:32+02:00 |
checkasm: arm: Ignore changes to bits 0-4 and 7 of FPSCR
These bits are set by exceptions in NEON instructions.
Also print the differing bits when FPSCR is clobbered,
and use bic instead of lsl, for clearing the topmost bits.
Signed-off-by: Martin Storsjö <[email protected]>
| 37961044c6cc633c28d87293775d8e985d156921 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/37961044c6cc633c28d87293775d8e985d156921 | 2016-07-17 21:48:17+03:00 |
h264: Log more information about invalid NALu size
| 61bd0ed781b56eea1e8e851aab34a2ee3b59fbac | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/61bd0ed781b56eea1e8e851aab34a2ee3b59fbac | 2016-07-13 18:18:46-04:00 |
dirac_vlc: prevent shifting a 0 by 64 bits when zeroing residue
Adding a check for bits == 0 would still make Coverity misdetect this,
so just revert to the normal way of setting the residue to 0.
Signed-off-by: Rostislav Pehlivanov <[email protected]>
| f41e37b84f3d57c29d4a2a21f9337159135b981d | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/f41e37b84f3d57c29d4a2a21f9337159135b981d | 2016-07-14 01:29:34+01:00 |
aarch64: vp9: Implement NEON loop filters
This work is sponsored by, and copyright, Google.
These are ported from the ARM version; thanks to the larger
amount of registers available, we can do the loop filters with
16 pixels at a time. The implementation is fully templated, with
a single macro which can generate versions for both 8 and
16 pixels wide, for both 4, 8 and 16 pixels loop filters
(and the 4/8 mixed versions as well).
For the 8 pixel wide versions, it is pretty close in speed (the
v_4_8 and v_8_8 filters are the best examples of this; the h_4_8
and h_8_8 filters seem to get some gain in the load/transpose/store
part). For the 16 pixels wide ones, we get a speedup of around
1.2-1.4x compared to the 32 bit version.
Examples of runtimes vs the 32 bit version, on a Cortex A53:
ARM AArch64
vp9_loop_filter_h_4_8_neon: 144.0 127.2
vp9_loop_filter_h_8_8_neon: 207.0 182.5
vp9_loop_filter_h_16_8_neon: 415.0 328.7
vp9_loop_filter_h_16_16_neon: 672.0 558.6
vp9_loop_filter_mix2_h_44_16_neon: 302.0 203.5
vp9_loop_filter_mix2_h_48_16_neon: 365.0 305.2
vp9_loop_filter_mix2_h_84_16_neon: 365.0 305.2
vp9_loop_filter_mix2_h_88_16_neon: 376.0 305.2
vp9_loop_filter_mix2_v_44_16_neon: 193.2 128.2
vp9_loop_filter_mix2_v_48_16_neon: 246.7 218.4
vp9_loop_filter_mix2_v_84_16_neon: 248.0 218.5
vp9_loop_filter_mix2_v_88_16_neon: 302.0 218.2
vp9_loop_filter_v_4_8_neon: 89.0 88.7
vp9_loop_filter_v_8_8_neon: 141.0 137.7
vp9_loop_filter_v_16_8_neon: 295.0 272.7
vp9_loop_filter_v_16_16_neon: 546.0 453.7
The speedup vs C code in checkasm tests is around 2-7x, which is
pretty much the same as for the 32 bit version. Even if these functions
are faster than their 32 bit equivalent, the C version that we compare
to also became around 1.3-1.7x faster than the C version in 32 bit.
Based on START_TIMER/STOP_TIMER wrapping around a few individual
functions, the speedup vs C code is around 4-5x.
Examples of runtimes vs C on a Cortex A57 (for a slightly older version
of the patch):
A57 gcc-5.3 neon
loop_filter_h_4_8_neon: 256.6 93.4
loop_filter_h_8_8_neon: 307.3 139.1
loop_filter_h_16_8_neon: 340.1 254.1
loop_filter_h_16_16_neon: 827.0 407.9
loop_filter_mix2_h_44_16_neon: 524.5 155.4
loop_filter_mix2_h_48_16_neon: 644.5 173.3
loop_filter_mix2_h_84_16_neon: 630.5 222.0
loop_filter_mix2_h_88_16_neon: 697.3 222.0
loop_filter_mix2_v_44_16_neon: 598.5 100.6
loop_filter_mix2_v_48_16_neon: 651.5 127.0
loop_filter_mix2_v_84_16_neon: 591.5 167.1
loop_filter_mix2_v_88_16_neon: 855.1 166.7
loop_filter_v_4_8_neon: 271.7 65.3
loop_filter_v_8_8_neon: 312.5 106.9
loop_filter_v_16_8_neon: 473.3 206.5
loop_filter_v_16_16_neon: 976.1 327.8
The speed-up compared to the C functions is 2.5 to 6 and the cortex-a57
is again 30-50% faster than the cortex-a53.
This is an adapted cherry-pick from libav commits
9d2afd1eb8c5cc0633062430e66326dbf98c99e0 and
31756abe29eb039a11c59a42cb12e0cc2aef3b97.
Signed-off-by: Ronald S. Bultje <[email protected]>
| f1212e472b5f57b4f7243fc46f254647cf7e284d | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/f1212e472b5f57b4f7243fc46f254647cf7e284d | 2016-11-15 15:10:03-05:00 |
tests/checkasm: reduce cosmetic diff with libav
Chunk was not merged in ca5ec2bf51d8c4f8bb0a829d0a65c70c968888a3.
| 495a40cecb4dc167b4acc686d44bba231eca0c18 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/495a40cecb4dc167b4acc686d44bba231eca0c18 | 2016-07-13 17:11:58+02:00 |
avformat/dump: Use codec and QP limits from AVCodecContext
Fixes regression
Fixes Ticket5421
Signed-off-by: Michael Niedermayer <[email protected]>
| fc07972582acc55675a622ec6d489a7bbf559943 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fc07972582acc55675a622ec6d489a7bbf559943 | 2016-06-03 22:26:00+02:00 |
tests/checkasm/pixblockdsp: Test 8 byte aligned positions
The code is documented as to require 8byte alignment
Signed-off-by: Michael Niedermayer <[email protected]>
| fb6b6b5166cb8345caa4e7423dd7d9b97b430501 | ffmpeg | devign | 0 | https://github.com/ffmpeg/ffmpeg | https://github.com/ffmpeg/ffmpeg/commit/fb6b6b5166cb8345caa4e7423dd7d9b97b430501 | 2016-07-02 22:21:53+02:00 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.