]>
git.sesse.net Git - ffmpeg/log
Michael Niedermayer [Tue, 19 May 2015 11:43:18 +0000 (13:43 +0200)]
fate: Add test for -exact_rice_parameters 1
Michael Niedermayer [Tue, 19 May 2015 11:26:55 +0000 (13:26 +0200)]
avcodec/flacenc: Support calculating rice parameters exactly
Some files benefit by about 0.3% from this, and speedwise its ok
other files do not benefit and encode to the same size
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Tue, 19 May 2015 10:11:27 +0000 (12:11 +0200)]
avcodec/flacenc: Do not copy unused udata array -> 5x faster calc_rice_params()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Tue, 19 May 2015 01:23:39 +0000 (03:23 +0200)]
avcodec/golomb: Remove disabled and broken code
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Tue, 19 May 2015 01:22:33 +0000 (03:22 +0200)]
avcodec/mss2: use < 0 instead of != 0 to check for error of vlc initialization
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 18 May 2015 18:27:04 +0000 (20:27 +0200)]
ffmpeg_opt: Compensate for DTS/PTS difference in seeking when its based on DTS
Fixes Ticket4554
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Steve Borho [Sat, 16 May 2015 03:05:12 +0000 (22:05 -0500)]
libavcodec/x265: detect csps in libx265_encode_init_csp()
Without this change, if you link with an 8bit libx265 and try to specify
a 10bit input color space via:
ffmpeg -i in.mov -c:v libx265 -pix_fmt yuv420p10le out.mp4
It will error with:
Incompatible pixel format 'yuv420p10le' for codec 'libx265',
auto-selecting format 'yuv420p'
With this fix, it will learn if a 10bit libx265 is available at startup,
and thus allow 10bit input color spaces.
Reviewed-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Michael Niedermayer [Mon, 18 May 2015 12:07:48 +0000 (14:07 +0200)]
avcodec/snow_dwt: Remove ff_ prefix from ff_spatial_idwt_init() and ff_spatial_idwt_slice()
They are only used once in the file in which they are defined
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 18 May 2015 11:28:14 +0000 (13:28 +0200)]
avcodec/hevc_cabac: Rename ff_hevc_transform_skip_flag_decode() to hevc_transform_skip_flag_decode()
The function is static and used once in the file its defined in
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 18 May 2015 11:22:27 +0000 (13:22 +0200)]
avfilter/avfiltergraph: Rename ff_avfilter_graph_config_pointers to graph_config_pointers.
The function is static and only used once in the file its defined in.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 18:02:01 +0000 (20:02 +0200)]
avformat/mxfenc: Set horizontal chroma subsample value from pixel format if available
Reviewed-by: tim nicholson <nichot20@yahoo.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 17:59:18 +0000 (19:59 +0200)]
avformat/mxfenc: Set the component depth from the pixel format if available
Reviewed-by: tim nicholson <nichot20@yahoo.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Mon, 18 May 2015 00:35:12 +0000 (02:35 +0200)]
Merge remote-tracking branch 'rbultje/vp9-bugfixes'
* rbultje/vp9-bugfixes:
vp9: match another find_ref_mvs() bug in libvpx.
vp9: fix scaled motion vector clipping for sub8x8 blocks.
vp9: improve signbias check.
vp9: don't allow compound references if error_resilience is enabled.
vp9: clamp segmented lflvl before applying ref/mode deltas.
vp9: reset loopfilter mode/ref deltas on keyframe.
vp9: fix crash when playing back 440/440 content with width%64<56.
vp9: extend loopfilter workaround for vp9 h/v mix-up to work for 422.
vp9: clip motion vectors in the same way as libvpx does.
vp9: set skip flag if the block had no coded coefficients.
vp9: apply mv scaling workaround only when subsampling is enabled.
vp9: read all 4x4 blocks in sub8x8 blocks individually with scalability.
vp9: fix segmentation map referencing upon framesize change.
vp9: disable more pmulhrsw optimizations in idct16/32.
vp9: disable all pmulhrsw in 8/16 iadst x86 optimizations.
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 17 May 2015 23:11:02 +0000 (01:11 +0200)]
cmdutils: Fix sign error in display matrix auto-rotation code
This makes the sample from Ticket4560 behave consistently with either branch
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 17 May 2015 22:28:57 +0000 (00:28 +0200)]
ffmpeg: Do not copy the display matrix if rotation meta-data is manually added
Fixes Ticket4560
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 17 May 2015 17:08:38 +0000 (19:08 +0200)]
avcodec/golomb: fix reading huge signed rice golomb values
No testcase is known, nor any case where such huge values would occur
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sun, 17 May 2015 17:07:17 +0000 (19:07 +0200)]
avcodec/golomb: get_ur_golomb_jpegls: Fix reading huge k values
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
MrBoogs [Tue, 21 Apr 2015 21:53:57 +0000 (15:53 -0600)]
avformat/hlsenc: do not append an endlist tag when the stream ends
Reviewed-by: Thomas Volkert <silvo@gmx.net>
MrBoogs [Tue, 21 Apr 2015 21:52:03 +0000 (15:52 -0600)]
avformat/hlsenc: Add hls flag for starting a playlist with a discontinuity on startup
Reviewed-by: Thomas Volkert <silvo@gmx.net>
MrBoogs [Tue, 21 Apr 2015 21:46:17 +0000 (15:46 -0600)]
avformat/hlsenc: Add hls flag for rounding m3u8 durations to whole numbers, problems with floats on some devices
Reviewed-by: Thomas Volkert <silvo@gmx.net>
Michael Niedermayer [Sun, 17 May 2015 11:10:33 +0000 (13:10 +0200)]
avcodec/flacdec: Attempt to auto-detect old buggy flac
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Carl Eugen Hoyos [Sun, 17 May 2015 10:58:44 +0000 (12:58 +0200)]
lavc/vc1: Never decode vc1 as gray if gray decoding was not enabled.
Carl Eugen Hoyos [Sun, 17 May 2015 10:57:27 +0000 (12:57 +0200)]
lavf/mpeg: Do not detect unknown audio in Hikvision streams as alaw.
Carl Eugen Hoyos [Sun, 17 May 2015 08:22:03 +0000 (10:22 +0200)]
lavc: Print a warning if gray decoding was requested but not enabled.
Carl Eugen Hoyos [Sat, 16 May 2015 21:33:48 +0000 (23:33 +0200)]
lavf/wav: Read files >4G if no smaller filesize was written.
Fixes second part of ticket #4543.
Carl Eugen Hoyos [Sat, 16 May 2015 21:32:56 +0000 (23:32 +0200)]
lavf/wav: Print an error if files >4G are written.
Additionally, don't write an incorrect shorter size for such files.
Fixes part of ticket #4543.
Carl Eugen Hoyos [Sat, 16 May 2015 21:32:27 +0000 (23:32 +0200)]
lavc/h263: Set color_range for gray decoding.
Carl Eugen Hoyos [Sat, 16 May 2015 21:32:14 +0000 (23:32 +0200)]
lavc/vc1: Set color_range for gray decoding.
Carl Eugen Hoyos [Sat, 16 May 2015 21:31:44 +0000 (23:31 +0200)]
lavf/mov: Tell users about the use_absolute_path option.
Fixes ticket #4539.
Carl Eugen Hoyos [Sat, 16 May 2015 20:51:16 +0000 (22:51 +0200)]
lavc/qdrw: Fix overwrite when reading invalid Quickdraw images.
Carl Eugen Hoyos [Tue, 12 May 2015 11:00:29 +0000 (13:00 +0200)]
lavc/flac: Fix encoding and decoding with high lpc.
Based on an analysis by trac user lvqcl.
Fixes ticket #4421, reported by Chase Walker.
Carl Eugen Hoyos [Tue, 12 May 2015 10:47:57 +0000 (12:47 +0200)]
lavc/flacdec: Sanitize FLACSTREAMINFO usage.
Michael Niedermayer [Sat, 16 May 2015 23:34:35 +0000 (01:34 +0200)]
avformat/avidec: add mp2 to the list of exceptions instead of generally treating dshow_block_align==1 special
Fixes Ticket4552
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 21:08:09 +0000 (23:08 +0200)]
ffmpeg: only apply last picture flush code at EOF
Fixes Ticket4562
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 22:22:09 +0000 (00:22 +0200)]
avfilter/buffersink: return EOF if closed link in av_buffersink_get_frame_flags()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 19:14:01 +0000 (21:14 +0200)]
Merge commit '
732dd658687bd9a2e5c622e38c481825b57af250 '
* commit '
732dd658687bd9a2e5c622e38c481825b57af250 ':
doc: Add $branch to FATE config template
Conflicts:
doc/fate.texi
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 17:07:31 +0000 (19:07 +0200)]
avcodec/aacenc: use < 0 instead of != 0 for error checks
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 16:37:55 +0000 (18:37 +0200)]
avcodec/j2kenc: Use ret < 0 instead of ret != 0 for error checks
This is how most code in FFmpeg checks for failures
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 16:33:51 +0000 (18:33 +0200)]
avformat/mux: use <0 instead of != 0 for error check of init_muxer()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Sat, 16 May 2015 15:59:01 +0000 (11:59 -0400)]
vp9: match another find_ref_mvs() bug in libvpx.
If we find a second non-sub8x8 motion vector for a non-first sub8x8
block, and the clamped value is identical to the first non-sub8x8
motion vector, then the resulting nearmv motion vector is forced to
zero.
Timothy Gu [Wed, 13 May 2015 23:46:47 +0000 (16:46 -0700)]
doc: Add $branch to FATE config template
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Ronald S. Bultje [Sat, 16 May 2015 12:36:19 +0000 (08:36 -0400)]
vp9: fix scaled motion vector clipping for sub8x8 blocks.
To match the obscure clipping bug behaviour in libvpx.
Michael Niedermayer [Sat, 16 May 2015 12:35:36 +0000 (14:35 +0200)]
avdevice/v4l2: Fix vertical alignment in list_formats() output
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 11:51:18 +0000 (13:51 +0200)]
avformat/mov: Fix parsing short loci
Fixes Ticket4557
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Sat, 16 May 2015 11:50:38 +0000 (13:50 +0200)]
avformat/mov: Print reason of loci parsing failure
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
anatolschwarz [Tue, 5 May 2015 13:14:35 +0000 (16:14 +0300)]
ffmpeg: Support preserving the source key frame timings
Added option "source" to force_key_frames
Ronald S. Bultje [Sat, 16 May 2015 01:14:08 +0000 (21:14 -0400)]
vp9: improve signbias check.
Otherwise it will still scale motion vectors, which leads to corrupted
prediction.
Michael Niedermayer [Fri, 15 May 2015 20:30:42 +0000 (22:30 +0200)]
ffmpeg: Remove another unneeded ost->filter_graph check
Fixes CID1241508
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 20:12:08 +0000 (22:12 +0200)]
avcodec/hevc: Fix typo in num_entry_point_offsets check
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 20:02:12 +0000 (22:02 +0200)]
avcodec/libtheoraenc: Check for av_malloc failure
Fixes CID1257799
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Fri, 15 May 2015 18:15:46 +0000 (14:15 -0400)]
vp9: don't allow compound references if error_resilience is enabled.
libvpx (probably accidentally) clears the bits if error_res is set,
along with keyframe/intraonly. This probably wasn't the intention
(since it's local data), but it's behaviour we have to copy...
Ronald S. Bultje [Fri, 15 May 2015 15:43:09 +0000 (11:43 -0400)]
vp9: clamp segmented lflvl before applying ref/mode deltas.
Ronald S. Bultje [Fri, 15 May 2015 15:42:39 +0000 (11:42 -0400)]
vp9: reset loopfilter mode/ref deltas on keyframe.
Ronald S. Bultje [Fri, 15 May 2015 02:48:59 +0000 (22:48 -0400)]
vp9: fix crash when playing back 440/440 content with width%64<56.
Ronald S. Bultje [Fri, 15 May 2015 02:40:26 +0000 (22:40 -0400)]
vp9: extend loopfilter workaround for vp9 h/v mix-up to work for 422.
Michael Niedermayer [Fri, 15 May 2015 18:19:21 +0000 (20:19 +0200)]
avformat/rtsp: print a debug level note if time parsing fails
Fixes CID733718 again
This partly reverts commit
eb7ddb5066843ab16caf33b05eafb2576648b797 .
Michael Niedermayer [Fri, 15 May 2015 18:06:24 +0000 (20:06 +0200)]
avcodec/svq1dec: Remove duplicate buf_size check
Fixes CID1297573
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 16:29:40 +0000 (18:29 +0200)]
avcodec/dcadec: Check active_bands
Fixes CID1297594 part2
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 16:04:12 +0000 (18:04 +0200)]
avcodec/dcadec: Check scale table index
Fixes CID1297594 part 1
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 15:31:58 +0000 (17:31 +0200)]
avcodec/sonic: More completely check sample_rate_index and channels
Fixes CID1271783
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 15:26:25 +0000 (17:26 +0200)]
avcodec/sonic: check memory allocations
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 15:21:10 +0000 (17:21 +0200)]
avcodec/smvjpegdec: check avcodec_decode_video2() return code
Fixes CID1271810
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 15:02:28 +0000 (17:02 +0200)]
avcodec/shorten: More complete pred_order check
Fixes CID1239055
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 14:58:51 +0000 (16:58 +0200)]
avcodec/shorten: Fix code depending on signed overflow behavior
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 14:48:31 +0000 (16:48 +0200)]
avcodec/shorten: Check skip_bytes()
Fixes CID1210526
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 14:19:53 +0000 (16:19 +0200)]
avcodec/put_bits: Remove dead code in put_bits()
Fixes CID1297574
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 13:23:32 +0000 (15:23 +0200)]
avcodec/proresdec2: Reset slice_count on deallocation
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 09:10:10 +0000 (11:10 +0200)]
avcodec/vp9: Use separate memset for counts.eob
Makes no real difference, but maybe scares coverity less (CID1297578)
Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 10:41:28 +0000 (12:41 +0200)]
ffmpeg: Print an error if avfilter_graph_queue_command() failed
Fixes CID1271809
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 10:34:20 +0000 (12:34 +0200)]
ffmpeg: remove unneeded ost->filter check
Fixes CID1241508
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 10:16:36 +0000 (12:16 +0200)]
avcodec/acelp_vectors: Assert that x is within the array in ff_set_fixed_vector()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Fri, 15 May 2015 08:42:22 +0000 (10:42 +0200)]
avcodec/vp3: Cleanup order of operations for current_run check
This shouldnt make a difference
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Fri, 15 May 2015 02:13:46 +0000 (22:13 -0400)]
vp9: clip motion vectors in the same way as libvpx does.
The practical effect of this is that the scaling will wrongly not be
applied to the interpolation edge (the 3/4 constants in this patch).
In other words, we clip to the pre-scaling interpolation, even though
these should be clipped post-scaling. The resulting out-of-frame MVs
are thus automatically clipped within the visible portion of the frame,
which is probably not the intention, but is unfortunately what libvpx
does, so we need to copy that behaviour.
Michael Niedermayer [Fri, 15 May 2015 01:05:20 +0000 (03:05 +0200)]
fate: increase the number of filtered frames to 20 for edgedetect* & hue
Suggested-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Fri, 15 May 2015 00:13:43 +0000 (20:13 -0400)]
vp9: set skip flag if the block had no coded coefficients.
This reproduces libvpx behaviour. It seems like it originally only
targeted loopfilter behaviour, but this unfortunately effects following
block contexting and thus directs bitstream sync.
Vignesh Venkatasubramanian [Thu, 14 May 2015 17:32:24 +0000 (10:32 -0700)]
lavf/webmdashenc: fix unchecked strftime
Fix unchecked strftime return value. This patch fixes Coverity
CID
1295086 .
Signed-off-by: Vignesh Venkatasubramanian <vigneshv@google.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 22:30:32 +0000 (00:30 +0200)]
avcodec/h264_slice: Fix ranges in assert
Fixes CID1297592, CID1297593
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 21:25:39 +0000 (23:25 +0200)]
avcodec/dvbsubdec: Clear w/h/size on region buffer allocation failure
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 21:16:06 +0000 (23:16 +0200)]
avcodec/diracdec: Make data_unit_size unsigned
Fixes CID1271788
with this change the value is more explicitly checked, it was fully checked
before though
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Thu, 14 May 2015 21:05:18 +0000 (17:05 -0400)]
vp9: apply mv scaling workaround only when subsampling is enabled.
Ronald S. Bultje [Thu, 14 May 2015 20:38:53 +0000 (16:38 -0400)]
vp9: read all 4x4 blocks in sub8x8 blocks individually with scalability.
Ronald S. Bultje [Thu, 14 May 2015 20:37:49 +0000 (16:37 -0400)]
vp9: fix segmentation map referencing upon framesize change.
Michael Niedermayer [Thu, 14 May 2015 19:29:19 +0000 (21:29 +0200)]
avcodec/dcadec: Check subsubframes
Fixes: CID1239152
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 18:49:25 +0000 (20:49 +0200)]
avcodec/dcadec: Check nchans
Fixes CID1239110
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Ronald S. Bultje [Thu, 14 May 2015 18:15:27 +0000 (14:15 -0400)]
vp9: disable more pmulhrsw optimizations in idct16/32.
For idct16, only when called from a adst16x16 variant, so impact is
minor. For idct32, for all, so relatively major impact.
Ronald S. Bultje [Thu, 14 May 2015 17:39:37 +0000 (13:39 -0400)]
vp9: disable all pmulhrsw in 8/16 iadst x86 optimizations.
They all overflow in various samples that are considered valid input.
wm4 [Thu, 14 May 2015 16:27:31 +0000 (18:27 +0200)]
hevc: make avcodec_decode_video2() fail if get_format() fails
Personally, I need the decoder to back out if get_format() returns no
usable pixel format. This didn't work because the error code was not
propagated down the call chain. This in turn happened because the
variable declaration removed in this patch shadowed the variable, whose
value is returned at the end of the function. Consequently, failures of
decode_nal_unit() were ignored in this place.
Reviewed-by: Andreas Cadhalpun <andreas.cadhalpun@googlemail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 16:08:33 +0000 (18:08 +0200)]
avcodec/cavsdec: Use ff_set_dimensions()
Fixes CID1239111 part2
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 16:00:09 +0000 (18:00 +0200)]
avcodec/cavsdec: Check frame_rate_code
Fixes CID1239111 part1
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 15:54:40 +0000 (17:54 +0200)]
avcodec/cavsdec: Check esc_code
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Nedeljko Babic [Thu, 14 May 2015 13:36:36 +0000 (15:36 +0200)]
libavutil/softfloat: Add test case for av_add_sf
Recently normalization (av_normalize_sf) of output was added to av_add_sf.
This normalization is used for better precision for small values and the
purpose of this (quite simple) test case is to test difference between double
and softfloat.
The values used are tailored to maximally highlighte problem with precison when
normalization is not used.
Signed-off-by: Nedeljko Babic <nedeljko.babic@imgtec.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 10:25:05 +0000 (12:25 +0200)]
avcodec/qdrw: Fix the code which asks for version 1 samples
The new code only asks for version 1 if its actually version 1 and
prints the version bytes if its something else
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 10:14:52 +0000 (12:14 +0200)]
avcodec/qdrw: another try at skipping the first 512 bytes
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Clément Bœsch [Sun, 8 Feb 2015 11:38:06 +0000 (12:38 +0100)]
avcodec/srtdec: attempt to correct SubRip positioning
The positioning was completely wrong. First, the coordinates are
expressed in ASS playback resolution (which is by default 384x288).
Secondly, the coordinates define a drawing rectangle, not a moving area.
The previous code was making subtitles move from a random position to
another random position.
Here we rescale assuming the video resolution is a DVD one (720x480). We
can't really do anything better so far, but since this positioning
information is often from a DVD rip we can consider them relatively
safe.
Clément Bœsch [Sun, 8 Feb 2015 11:32:54 +0000 (12:32 +0100)]
avcodec/ass: make default playback resolution available to decoders
Rong Yan [Thu, 14 May 2015 06:43:44 +0000 (06:43 +0000)]
avcodec/ppc/h264dsp: POWER LE support in h264_idct_dc_add_internal() fix vec_lvsl bug
We got defective video when use GCC 4.9.2 instead of GCC 4.9.1 to compile FFMEPG. And further found
that GCC 4.8 and 4.9 need patch to fix the lvsl/lvsr bug on POWER LE, and GCC 5.1 contains
the correct code since its release. The message on gcc-patches requesting approval for lvsl/lvsr
patch is at https://gcc.gnu.org/ml/gcc-patches/2014-10/msg00228.html.
The fixed code avoids using lvsl and will not depends on GCC version, also it uses less instructions on POWER LE.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Shivraj Patil [Fri, 8 May 2015 08:02:01 +0000 (13:32 +0530)]
avcodec/mips: MSA (MIPS-SIMD-Arch) optimizations for HEVC uni hv mc functions
Signed-off-by: Shivraj Patil <shivraj.patil@imgtec.com>
Reviewed-by: Nedeljko Babic <Nedeljko.Babic@imgtec.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Thu, 14 May 2015 00:23:08 +0000 (02:23 +0200)]
avformat/nutdec: Remove unused variable
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Timothy Gu [Wed, 13 May 2015 23:23:09 +0000 (16:23 -0700)]
nutdec: Remove unused label
Added in
361702660d2c37a63b7d6381d39e1e1de8405260 . Modified version that
doesn't use this label merged in
55231323b0fdc84a529418d673148cf1f3157229 ,
thus obsoleting this label.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Michael Niedermayer [Wed, 13 May 2015 23:15:20 +0000 (01:15 +0200)]
avcodec/msrledec: Ask for sample that uses a branch with a unused byte
Finding out if this is correct or buggy is easiest with a sample
Fixes: CID1297620
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>