Buffering more than one packet can be a huge performance improvement for
encoding files with small packets (e.g. wav) over SMB/CIFS.
Acked-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>
If flushing is not disabled, then mux.c will signal the end of the packets with
an AVIO_DATA_MARKER_FLUSH_POINT, and aviobuf will be able to decide to flush or
not based on the preferred minimum packet size set by the used protocol.
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Marton Balint <cus@passwd.hu>
This patch makes aviobuf work more like traditinal file IO, which is how people
think about it.
For example, in the past, aviobuf only flushed buffers until the current buffer
position, even if more data was written to it previously, and a backward seek
was used to reposition the IO context.
From now, aviobuf will keep track of the written data, so no explicit seek will
be required till the end of the buffer, or till the end of file before flushing.
This fixes at least one regression, fate-vsynth3-flv was broken if
flush_packets option was set to false, an explicit seek was removed in
4e3cc4bdd8.
Also from now on, if a forward seek in the write buffer were to cause a gap
between the already written data and the new file position, a flush will
happen.
The must_flush varable is also removed, which might have caused needless
flushes with multiple seeks whithin the write buffer. Since we know the amount
of data written to it, we will know when to flush.
Signed-off-by: Marton Balint <cus@passwd.hu>
Takes a raw input stream containing frames with correct timestamps but
possibly out of order and inserts additional show-existing-frame
packets to correct the ordering.
(cherry picked from commit 34e051d168)
(cherry picked from commit b43b95f478)
Also converted from bitstream to get_bits.
Before this, output bitstream filters would never see EOF and
therefore would not be able to flush any delayed packets.
(cherry picked from commit f64d1100a5)
<@jamrial> durandal_1707: 04aa09c4bc broke fate-lavf-ffm and fate-lavf-mxf
<@durandal_1707> how so?
<@jamrial> one byte changes
<@durandal_1707> jamrial: just update checksums
<@jamrial> durandal_1707: but why did they change at all? the commit you reverted didn't affect them
<@jamrial> why does reverting it affect these tests?
<@jamrial> i don't think updating the checksum without knowing what changed is a good idea
<@durandal_1707> jamrial: the lavfi core is in weird state after removal of recursive code
<@durandal_1707> jamrial: the change is that older ones would get progressive flag set and new one doesnt
<@jamrial> alright
If the first assembler to be probed is an old nasm build, X86ASM_DEPFLAGS
will be set and remain so after yasm is ultimately used as fallback.
This results in yasm being called with said nasm specific flags and failing
during actual object assembly but not with configure sanity checks.
Regression since 5cae5a1def
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: James Almer <jamrial@gmail.com>
set_ue_golomb just support 2^16 - 2 at most, becase this function call
put_bits, and put_bits just support write up to 31 bits, when write 32
bit in put_bits, it's will overwrite the bit buffer, and the default
assert level is 0, the av_assert2(n <= 31 && value < (1U << n)) in
put_bits can not be trigger runtime.
Signed-off-by: Jun Zhao <jun.zhao@intel.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: signed integer overflow: 227511904 + 1964113935 cannot be represented in type 'int'
Fixes: 2331/clusterfuzz-testcase-minimized-6182185830711296
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: signed integer overflow: 512 + 2147483146 cannot be represented in type 'int'
Fixes: 2314/clusterfuzz-testcase-minimized-4519333877252096
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: signed integer overflow: 538976288 * 32 cannot be represented in type 'int'
Fixes: 2310/clusterfuzz-testcase-minimized-4534784887881728
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes a NULL pointer derefence when ogg_init() returns a failure and
a stream's private data was not yet allocated.
This is a regression since 3c5a53cdfa
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
Fixes: runtime error: shift exponent 32 is too large for 32-bit type 'int'
Fixes: 2306/clusterfuzz-testcase-minimized-5002997392211968
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: left shift of negative value -1
Fixes: 2303/clusterfuzz-testcase-minimized-5529675273076736
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
This allows for more graceful fallback from NASM to Yasm if the available
NASM version is too old.
(Cherry-picked from libav commit adfd7892e3)
Signed-off-by: James Almer <jamrial@gmail.com>
NASM is more actively maintained and permits generating dependency information
as a sideeffect of assembling, thus cutting build times in half.
(Cherry-picked from libav commit 57b753b445)
Signed-off-by: James Almer <jamrial@gmail.com>
"nasm --version" works on nasm since around version 2.11 and as
such configure assumed it was yasm.
Based on libav commit f54037da8a by
Diego Biurrun.
Signed-off-by: James Almer <jamrial@gmail.com>
Consistently use object format names with "32" suffix and set object format
to "win64" on Windows x86_64, which fixes assembling with nasm.
(Cherry-picked from libav commit 808ef43597)
Signed-off-by: James Almer <jamrial@gmail.com>
This was added in e3e3c82555e2382125195c1ba9f34b5a43299abc, probably
as a workaround for the fact that the quant table was not permutated
and the IDCT coefficients are, meaning that you'd only get correct
reconstruction if the IDCT permutation was an identity matrix, which
happens to be the case when you use the simple IDCT. The quant table
permutation bug was fixed in 42dd1434bf6a7230e4175c08fcfabc3ba51a0463,
meaning this workaround is no longer necessary.
In practical terms, before 42dd1434bf6a7230e4175c08fcfabc3ba51a0463,
the PSNR between decodes of the fate-mdec using simple (C) or simplemmx
IDCTs was 35. After 42dd1434bf6a7230e4175c08fcfabc3ba51a0463, it's 90.
Properly use the b.eq form instead of the nonstandard form (which
both gas and newer clang accept though), and expand the register
lists that used a range (which the Xcode 6.2 clang, based on clang
3.5 svn, didn't support).
Signed-off-by: Martin Storsjö <martin@martin.st>
Properly use the b.eq/b.ge forms instead of the nonstandard forms
(which both gas and newer clang accept though), and expand the
register list that used a range (which the Xcode 6.2 clang, based
on clang 3.5 svn, didn't support).
This is cherrypicked from libav commit
a970f9de86.
Signed-off-by: Martin Storsjö <martin@martin.st>
Fixes: runtime error: left shift of negative value -1
Fixes: 2299/clusterfuzz-testcase-minimized-4843509351710720
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: signed integer overflow: 262144 * -16120 cannot be represented in type 'int'
Fixes: 2292/clusterfuzz-testcase-minimized-6156080415506432
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: runtime error: signed integer overflow: -1575818955 + -915383657 cannot be represented in type 'int'
Fixes: 2224/clusterfuzz-testcase-minimized-6208559949807616
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
The early check for inconsistent in-source vs out-of-source build
cannot generate a config.log otherwise.
(Cherry-picked from libav commit 0ee78020cd)
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Signed-off-by: James Almer <jamrial@gmail.com>
Use avci->last_pkt_props to get the side data. Using |pkt| doesn't work
when FF_API_MERGE_SD is set, as the compressed side data is expanded into
|tmp|, leaving the original |pkt| unchanged.
Signed-off-by: James Almer <jamrial@gmail.com>
Paul B Mahol
Marton Balint
Mark Thompson
Michael Niedermayer
Ronald S. Bultje
James Almer
Reino17
Rostislav Pehlivanov
jerett
Jun Zhao
Matthieu Bouron
Diego Biurrun
Memphiz
Luca Barbato
John Rummell