Thomas Hebb [Sun, 28 Aug 2016 16:24:55 +0000 (12:24 -0400)]
avformat/mov: aax: pass proper AVClass to av_log()
av_log() expects its first parameter to be a pointer to a struct whose
first element is a pointer to an AVClass. In what I can only imagine is
a typo, one call to av_log() in the AAX decryption code instead passes
a pointer to an AVSHA struct, which doesn't even contain a pointer as
its first element, much less a pointer to an AVClass. Change the call to
pass the current MOVContext, as surrounding calls do.
The incorrect AVClass was causing mpv to emit the warning "av_log
callback called with bad parameters" when playing an .aax file.
Signed-off-by: Thomas Hebb <tommyhebb@gmail.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
As Nvidia has put the most recent Video Codec SDK behind a double
registration wall, of which one needs manual approval of a lenghty
application, bundling this header saves everyone trying to use NVENC
from that headache.
The header is still MIT licensed and thus fine to bundle with ffmpeg.
Not bundling this header would get ffmpeg stuck at SDK v6, which is
still freely available, holding back future development of the NVENC
encoder.
Vittorio Giovara [Thu, 25 Aug 2016 23:14:58 +0000 (19:14 -0400)]
vf_colorspace: Add support for full range yuv
Whenever a full range video is input, since the YUVJ* formats are not
listed as supported for this filter, a range reduction takes place
through the auto-inserted format filter, forcing the conversion to
operate on a limited range,
However the filter handles full range videos perfectly fine, so adding
support to YUVJ* formats will allow skipping a conversion step, while
providing completely identical results.
Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com> Reviewed-by: "Ronald S. Bultje" <rsbultje@gmail.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Burt P [Mon, 22 Aug 2016 22:34:35 +0000 (17:34 -0500)]
af_hdcd: for easier maintenance alongside libhdcd
Mostly just re-arranges some code to make it easier to update this
filter and libhdcd together. filter_frame() is much simpler as a
result.
* use the HDCD detection data structure and functions from libhdcd,
moved detection code out of filter_frame()
* moved analyze_mode preparation out of filter_frame() into
hdcd_analyze_prepare(), from libhdcd
* moved some macro definitions to the top so they are all together
Derek Buitenhuis [Thu, 25 Aug 2016 15:58:14 +0000 (16:58 +0100)]
mov: Remove ancient heuristic hack
This breaks files with legitimate single-entry edit lists,
and the hack, introduced in f03a081df09f9c4798a17d7e24446ed47924b11b,
has no link to any known sample in its commit message.
Steven Liu [Thu, 25 Aug 2016 12:40:17 +0000 (05:40 -0700)]
tests/audiomatch: Fix compile warning
Making sure bestpos is initialized to zero to prevent gcc from kvetching.
It's harmless (although it's not obvious that it's harmless)
from code inspection:
tests/audiomatch.c: In function ‘main’:
tests/audiomatch.c:40: warning: ‘bestpos’ may be used uninitialized in this function
Thanks to Moritz Barsnick for first bringing this to the attention.
Signed-off-by: LiuQi <liuqi@gosun.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This also disabled the case of mixed x/ymin with tiles, the code
handles these cases inconsistent for the 2 coordinate axis and is
unlikely working correctly.
Fixes crash Fixes: poc1.exr, poc2.exr Found-by: Yaoguang Chen of Aliapy unLimit Security Team Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
memalign() is not guranteed to be compatible with free() or realloc()
and for platforms in this category we have --enable-memalign-hack
(which should be enabled automatically if such system is detected)
Trying to somehow half support systems that can free() memalign memory
but not reallocate it seems not worth the amount of work needed to
keep 2 then incompatible allocation systems and ensure their
seperation. That is unless this would affect a major platform
on which we want to avoid the memalign hack code
The warnings also cause wasted time and effort as people try
to maintain a separation that does not exist currently.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes occurance of NaN/Inf leading to assertion failures and out of array access Fixes: d1c38a09acc34845c6be3a127a5aacaf/signal_sigsegv_3982225_6121_d18bd5451d4245ee09408f04badd1b83.wmv Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
avformat/mux: Restore original ts in write_packet on error
Restore original timestamps in write_packet() if the
actual write operation was not successfull. This allows
to pass the same packet to nonblocking muxer repeatedly
without corrupting the timestamps.
Signed-off-by: Jan Sebechlebsky <sebechlebskyjan@gmail.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes out of array read Fixes: 049fdf78565f1ce5665df236d90f8657/asan_heap-oob_10a5a97_1026_42f9d4855547329560f385768de2f3fb.wtv Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Steven Liu [Sun, 21 Aug 2016 04:55:45 +0000 (12:55 +0800)]
lavf/hlsenc: add append_list flag into hlsenc
When ffmpeg exit by exception, start a new ffmpeg will
cover the old segment list, add this flag can continue
append the new segments into old hls segment list
Signed-off-by: LiuQi <liuqi@gosun.com> Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Jan Sebechlebsky [Sun, 21 Aug 2016 08:36:12 +0000 (10:36 +0200)]
avcodec/bsf: Add custom item name function for bsf list
which will
construct string description of filter chain. This is
done using lazy-initialization, so there is no overhead
if item name is never accessed.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Steven Robertson [Wed, 17 Aug 2016 07:25:47 +0000 (00:25 -0700)]
libavformat/mov: Accept known codepoints in 'colr'
This change relaxes the whitelist on reading color metadata in MOV/BMFF
containers. The whitelist on writing values is still in place.
As a consequence it also fixes an apparent bug in reading 'nclc' values.
The 'nclc' spec [1] is in harmony with ISO 23001-8 for the values it
lists, but the code getting removed was remapping 5->6 and 6->7 for
primaries, which is incorrect, and was remapping 6->5 for color matrix
("colorspace" in the code), which is equivalent but an unnecessary
inconsistency. This logic error doesn't appear in movenc.
Removing the whitelist allows proper conversion when the source codec
relies on the container for proper signaling of newer codepoints, such
as DNxHR and VP9. If converting to a codec or container that has updated
its spec to include the new codepoints, the metadata will be preserved.
If going back to MOV/BMFF, the output whitelist will still kick in, so
this won't result in out-of-spec files being created.
Fixes out of array read Fixes: f29f134ea5f5590df554a7733294a587/asan_stack-oob_309d14e_9188_ea01743d6355aff20530f3d4cdaa841a.wav Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes out of array read Fixes: 09f46aa2175cade93e3e3932646a56a9/asan_heap-oob_4a5385_2995_498f6abfdc0248288cefe5f4b7ad316c.mov Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes out of array read Fixes: 13262c363a28da8d6bdcc472aed6e9dc/asan_heap-oob_cfb5e2_3733_31cf3fcc783295c34222eb070a784f84.3gp Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>