The true problem happens in several previous get_vlc_symbol()
but checking that is more expensive (involving FFABS())
here its just a simple check between 2 variables we have.
Fixes: Assertion log >= k failed at libavcodec/golomb.h:406
Fixes: 429296194/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_FFV1_DEC_fuzzer-4691594622337024
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Add av_freep() to free new_elem to avoid memory leak.
Fixes: 3ca1e31e63 ("fftools/cmdutils: Atomically add elements to list of pointers, fix crash")
Signed-off-by: Jiasheng Jiang <jiashengjiangcool@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Don't rely on implicit zeroing.
Should fix fate failures on msvc targets.
Found-by: Kacper Michajlow <kasper93@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
scale was never initialized. av_tx_init() will use default scale if we
pass NULL.
Fixes: b3117f376d
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
This was in else branch of `#if 1` since ever. No need to keep dead code
like that, if anyone needs it they can get it from git history.
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
In OSS-Fuzz build environment flags are handled by CFLAGS and LDFLAGS.
In local build use --toolchain=clang-fuzz-asan-ubsan or similar
combination depending on requested sanitizers.
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
Nothing forbids them to be in any order the muxer desires.
Fixes demuxing heif samples generated by S1II.
Tested-by: Lynne <dev@lynne.ee>
Signed-off-by: James Almer <jamrial@gmail.com>
- Changes in mov_write_video_tag function to handle APV elementary stream
- Provided structure APVDecoderConfigurationRecord that specifies the decoder configuration information for APV video content
Co-Authored-by: James Almer <jamrial@gmail.com>
Signed-off-by: Dawid Kozinski <d.kozinski@samsung.com>
Signed-off-by: James Almer <jamrial@gmail.com>
Requested by a user. Even with autovectorization enabled, the compiler
performs a quite poor job of optimizing this function, due to not being
able to take advantage of the pmaxub + pcmpeqb trick for counting the number
of pixels less than or equal-to a threshold.
blackdetect8_c: 4625.0 ( 1.00x)
blackdetect8_avx2: 155.1 (29.83x)
blackdetect16_c: 2529.4 ( 1.00x)
blackdetect16_avx2: 163.6 (15.46x)
Can't read alignment from packet, codecpar or decoded frame linesizes, so don't
set it. My test device redmi k60 produces images in 4624x3472, not 32 aligned
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
Nikon HEIFs from a camera or NX studio include a small jpeg thumbnail in addition to
the expected HEVC thumbnails; allowing jpegs allows all thumbnails to
have an associated stream for Nikon HEIF files.
With this, Nikon HEIFs can finally be decoded without failing and the
thumbnails can be extracted into their own files.
Signed-off-by: Eric Joyner <erj@erj.cc>
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: James Almer <jamrial@gmail.com>
Prevents ffmpeg/ffprobe from erroring out when reading an HEIF that
contains multiple hvcC thumbnails (e.g. from a Nikon Z6III camera).
Before, move_read_iref_thmb() would always override the stored
thmb_item_id in the MOVContext with each new read thumbnail, causing a
stream and item_id mismatch later in mov_parse_heif_items(), resulting
in the "HEIF thumbnail doesn't reference a stream" error message.
To solve this,
- Turn thmb_item_id into an array of IDs because multiple thumbnails can
exist
- Change check in mov_parse_heif_items() to compare against all stored
thumbnail IDs to see if any item missing a stream is in the list of
thumbnail IDs.
Signed-off-by: Eric Joyner <erj@erj.cc>
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: James Almer <jamrial@gmail.com>
We need to ensure rtp sets NONBLOCK since the dtls
handshake has potentially overriden the sockets mode.
Signed-off-by: Jack Lau <jacklau1222@qq.com>
Signed-off-by: Timo Rothenpieler <timo@rothenpieler.org>
This naive hist[p[x]]++ loop suffers badly when there are large regions of
identical values in the image, because of store-to-load forwarding delay.
Splitting up the histogram into four "parallel" histograms and processing
them one at a time speeds things up significantly, about 40% on my end.
Since psadbw only exists for 8-bits, we have to emulate it for 16-bit
inputs. The simplest sequence is to use a normal subtraction, which is safe
as long as the inputs do not exceed 32767 - so limit this implementation
to 15-bit inputs and below.
For 16-bit inputs, we could in theory instead use a pminw / pmaxw to ensure
the resulting difference does not overflow, but this is slower, and also
breaks the subsequent use of pmaddwd, so I opted to skip 16-bit SIMD for
now.
scene_sad10_c: 114175.6 ( 1.00x)
scene_sad10_avx2: 9617.7 (11.87x)
scene_sad10_avx512: 5208.8 (21.92x)
scene_sad12_c: 114537.8 ( 1.00x)
scene_sad12_avx2: 9614.0 (11.91x)
scene_sad12_avx512: 5186.3 (22.08x)
scene_sad14_c: 114113.9 ( 1.00x)
scene_sad14_avx2: 9612.9 (11.87x)
scene_sad14_avx512: 5186.0 (22.00x)
scene_sad15_c: 114108.9 ( 1.00x)
scene_sad15_avx2: 9612.3 (11.87x)
scene_sad15_avx512: 5186.4 (22.00x)
scene_sad16_c: 114136.0 ( 1.00x)
Trivial to add, but a lot faster (on my machine).
scene_sad8_c: 114476.4 ( 1.00x)
scene_sad8_sse2: 8644.3 (13.24x)
scene_sad8_avx2: 4520.1 (25.33x)
scene_sad8_avx512: 3153.0 (36.31x)
the SSL_shutdown in tls_close need call the url_bio_bwrite
so we should keep udp still alive
Signed-off-by: Jack Lau <jacklau1222@qq.com>
Signed-off-by: Timo Rothenpieler <timo@rothenpieler.org>
It's possible to call linker indirectly through driver like Clang. In
which cases linker args has to be prefixed with -Wl.
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
This aligns declared function types in checkasm with real definition.
Fixes FATE: checkasm-{sw_rgb,sw_scale,sw_yuv2rgb,sw_yuv2yuv}
Fixes: runtime error: call to function <func> through pointer to incorrect function type
Fixes: c1a0e65763
Signed-off-by: Kacper Michajłow <kasper93@gmail.com>
