Certain hardware decoding APIs are not guaranteed to be thread-safe, so
having the user access decoded hardware surfaces while the decoder is
running in another thread can cause failures (this is mainly known to
happen with DXVA2).
For such hwaccels, only allow the decoding thread to run while the user
is inside a lavc decode call (avcodec_send_packet/receive_frame).
It should only be set after the decoder state has been fully initialized
for using that SPS.
Fixes possible invalid reads on get_format() failure.
CC: libav-stable@libav.org
A lot of changes happen at the same time:
- Add a framequeue fifo to AVFilterLink.
- split AVFilterLink.status into status_in and status_out: requires
changes to the few filters and programs that use it directly
(f_interleave, split, filtfmts).
- Add a field ready to AVFilterContext, marking when the filter is ready
and its activation priority.
- Add flags to mark blocked links.
- Change ff_filter_frame() to enqueue the frame.
- Change all filtering functions to update the ready field and the
blocked flags.
- Update ff_filter_graph_run_once() to use the ready field.
- buffersrc: always push the frame immediately.
This makes it possible to decode motion jpeg 2000
encoded in a transport stream without a correct PMT/PAT.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The assumption that avcodec_send_packet makes regarding decoders
consuming the entire packet is not true if the codec supports
truncated decoding mode and the truncated flag is turned on.
Steps to reproduce:
./ffmpeg_g -flags truncated \
-i "http://samples.ffmpeg.org/MPEG2/test-ebu-422.40000.pakets.ts" \
-c:v ffv1 -c:a copy -y /tmp/truncated.nut
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Without any optimization flags, MSVC does no dead code elimination (DCE) at
all, even for the most trivial cases. DCE is a prerequisite for building libav
correctly, otherwise there are undefined references to functions for other
architectures and disabled components.
-O1 is the minimal optimization flag for MSVC that does include DCE.
Clang is not able to eliminate the reference to ff_spdif_probe() when
there is a goto target in the same block and optimization is disabled.
This fixes the following build failure on OS X:
./configure --disable-everything --disable-doc \
--enable-decoder=pcm_s16le --enable-demuxer=wav \
--enable-protocol=file --disable-optimizations --cc=clang
make
...
Undefined symbols for architecture x86_64:
"_ff_spdif_probe", referenced from:
_set_spdif in libavformat.a(wavdec.o)
ld: symbol(s) not found for architecture x86_64
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Most decoders return the amount of data used.
This is more consistent
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Most decoders return the amount of data used.
This is more consistent
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Most decoders return the amount of data used.
This is more consistent
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Most decoders return the amount of data used.
This is more consistent
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
When receiving fragmented packets, the first packet declares the size,
and the later ones normally are small follow-on packets that don't repeat
the size and the other header fields. But technically, the later fragments
also can have a full header, declaring a different size than the previous
packet.
If the follow-on packet declares a larger size than the initial one, we
could end up writing outside of the allocation.
This fixes out of bounds writes.
Found-by: Paul Cher <paulcher@icloud.com>
Reviewed-by: Paul Cher <paulcher@icloud.com>
CC: libav-stable@libav.org
Signed-off-by: Martin Storsjö <martin@martin.st>
This seems to have been added by mistake in 11de006b, by not
noticing the negation for the existing condition. This block does
not contain any code that accesses the codec field in AVStream.
This function is meant to serve as a complement to compute_pkt_fields2,
which is guarded by FF_API_COMPUTE_PKT_FIELDS2 && FF_API_LAVF_AVCTX.
Signed-off-by: Martin Storsjö <martin@martin.st>
MPEG Audio frame header must be 4 bytes. If we fail to read
4 bytes bail early to avoid Use-of-uninitialized-value msan error.
Reference https://crbug.com/666874.
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The arguments of av_reduce are signed, so the cast to uint64_t is misleading.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Cadhalpun <Andreas.Cadhalpun@googlemail.com>
The former is not an official pseudo instruction although gas and llvm's
internal assembler support it. Fixes a build error with xcode 6.2
reported by Memphiz on github.
Height of canvas produced by drawtext varies depending on symbols in
text, so add example for printing separate texts aligned horizontally.
Wording suggested by Lou Logan <lou@lrcd.com>
Signed-off-by: Andrey Utkin <andrey.utkin@pb.com>
Signed-off-by: Lou Logan <lou@lrcd.com>
This improves commit 59c7022740.
In ff_thread_report_progress(), the fast code path can load
progress[field] with the relaxed memory order, and the slow code path
can store progress[field] with the release memory order. These changes
are mainly intended to avoid confusion when one inspects the source code.
They are unlikely to have measurable performance improvement.
ff_thread_report_progress() and ff_thread_await_progress() form a pair.
ff_thread_await_progress() reads progress[field] with the acquire memory
order (in the fast code path). Therefore, one expects to see
ff_thread_report_progress() write progress[field] with the matching
release memory order.
In the fast code path in ff_thread_report_progress(), the atomic load of
progress[field] doesn't need the acquire memory order because the
calling thread is trying to make the data it just decoded visible to the
other threads, rather than trying to read the data decoded by other
threads.
In ff_thread_get_buffer(), initialize progress[0] and progress[1] using
atomic_init().
Signed-off-by: Wan-Teh Chang <wtc@google.com>
Signed-off-by: Anton Khirnov <anton@khirnov.net>
This could happen when there was a frame number gap and frame threading was used.
Debugging-by: Ronald S. Bultje <rsbultje@gmail.com>
Debugging-by: Justin Ruggles <justin.ruggles@gmail.com>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
CC:libav-stable@libav.org
Signed-off-by: Anton Khirnov <anton@khirnov.net>
It is more natural for this codec and allows to avoid awkward constructs
like "consuming 0 bytes from input". Also, keep a reference to the input
packet to avoid unnecessary copying.
Anton Khirnov
Nicolas George
Paul B Mahol
Rostislav Pehlivanov
Ståle Kristoffersen
Jay Ridgeway
Pavel Koshevoy
Zhou Xiaoyong
Diego Biurrun
Steve Lhomme
Mark Harris
Michael Niedermayer
Martin Vignali
Martin Storsjö
Chris Cunningham
Andreas Cadhalpun
Janne Grunau
Andrey Utkin
Wan-Teh Chang
Derek Buitenhuis