The issue is that we cannot rely on any context existing when we free
frames. The Vulkan functions are loaded in each context separately,
so until now, we've just been loading them on every frame's destruction.
Rather than do this, just save the function pointers we need in each
frame. The function pointers are guaranteed to not change and exist.
Fixes: out of array write
Found-by: dongsookim@korea.ac.kr
Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Since bfb28b5ce8 the permutation
type FF_IDCT_PERM_SIMPLE is ARCH_X86_32-only. So use this
knowledge to disable code for it when not on ARCH_X86_32.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
When using multi-threaded decoding, every decoding thread
has its own DBP consisting of H264Pictures and each of these
points to its own AVFrames. They are synced during
update_thread_context via av_frame_ref() and therefore
the threads actually decoding (as well as all the others)
must not modify any field that is copied by av_frame_ref()
after ff_thread_finish_setup().
Yet this is exactly what happens when an error occurs
during decoding and the AVFrame's decode_error_flags are updated.
Given that these errors only become apparent during decoding,
this can't be set before ff_thread_finish_setup() without
defeating the point of frame-threading; in practice,
this meant that the decoder did not set these flags correctly
in case frame-threading was in use. (This means that e.g.
the ffmpeg cli tool fails to output its "corrupt decoded frame"
message in a nondeterministic fashion.)
This commit fixes this by adding a new H264Picture field
that is actually propagated across threads; the field
is an AVBufferRef* whose data is an atomic_int; it is
atomic in order to allow multiple threads to update it
concurrently and not to provide synchronization
between the threads setting the field and the thread
ultimately returning the AVFrame.
This unfortunately has the overhead of one allocation
per H264Picture (both the original one as well as
creating a reference to an existing one), even in case
of no errors. In order to mitigate this, an AVBufferPool
has been used and only if frame-threading is actually
in use. This expense will be removed as soon as
a proper API for refcounted objects (not based upon
AVBuffer) is in place.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Add a pointer parameter that if supplied will be used to return
the updated decode_error_flags. This will allow to fix several
races when using frame-threading; these resulted from AVFrame
that the earlier code updated concurrently being used as source
in an av_frame_ref() call in the decoder's update_thread_context.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
When ov_model_const_input_by_name/ov_model_const_output_by_name
failed, input_port/output_port can be wild pointer.
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
All Vulkan HWAccels share the same boilerplate code for creating
session params and this includes a common bug: In case actually
creating the video session parameters fails, the buffer destined
to hold them leaks; in case of HEVC this is also true if
get_data_set_buf() fails.
This commit factors this code out and fixes the leak.
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The alignment in vulkan_unmap_from_drm() (formerly the clone
of vulkan_frame_free()) is nicer than the in vulkan_frame_free(),
let's preserve it.
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
The AVBuffer API uses uint8_t as base type for buffers
and therefore its free callbacks need to abide by this.
Therefore vulkan_frame_free() used an inappropriate signature
which caused casts whenever this function has been called
manually.
This commit changes this by making vulkan_frame_free()
use the proper type and a vulkan_frame_free_cb() that
is used as free callback for the AVBuffer API.
Reviewed-by: Lynne <dev@lynne.ee>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Fixes: signed integer overflow: -2147483300 - 12285 cannot be represented in type 'int'
Fixes: 59462/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_BONK_fuzzer-5714298807386112
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
jpeg2000 overrides the global lowres variable with a lowres field called reduction_factor
ffmpeg -lowres X causes the reduction_factor to be set
ffplay -lowres X causes both lowres and the reduction_factor to be set
ossfuss sets only lowres
only the ffmpeg variant works. This patch tries to make the other 2 work.
Alternative we could just error out if things are inconsistent.
More complex restructuring should be limited to the master branch
to keep this reasonably easy to backport
Fixes: out of array access
Fixes: 59672/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: left shift of negative value -1
Fixes: 59889/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_HUFFYUV_fuzzer-5472742275940352
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: out of array access
Fixes: 62113/clusterfuzz-testcase-minimized-ffmpeg_DEMUXER_fuzzer-5025082076168192
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: out of array access
Fixes: 61991/clusterfuzz-testcase-minimized-ffmpeg_dem_JPEGXL_ANIM_fuzzer-5524679648215040
Fixes: 62181/clusterfuzz-testcase-minimized-ffmpeg_dem_JPEGXL_ANIM_fuzzer-5504964305485824
Fixes: 62214/clusterfuzz-testcase-minimized-ffmpeg_dem_JPEGXL_ANIM_fuzzer-4782972823535616
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
An AVBPrint is initialized via av_bprint_init() (or
av_bprint_init_for_buffer()) which expects uninitialized
AVBPrints; it is therefore not necessary to zero them before
the actual initialization.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
This is not an error condition, but would be treated like one if the
program terminates on the next transcode loop iteration because of a
signal or keyboard input.
Fixes#10504
Tested-by: https://github.com/0Ky
For badly interleaved files, interleave packets from multiple tracks
at the demuxer level can trigger seeking back and forth, which can be
dramatically slow depending on the protocol. Demuxer level interleave
can be useless sometimes, e.g., reading mp4 via http and then
transcoding/remux to DASH. Disable this option when you don't need the
demuxer level interleave, and want to avoid the IO penalizes.
Co-authored-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com>
Since fb548fba04,
this return -1 is in a function returning enum AVPixelFormat
whose caller checks for AV_PIX_FMT_NONE for failure.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It indicates to the reader that said function does not modify
any state.
Reviewed-by: James Almer <jamrial@gmail.com>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Every code point in the BMP is representable with at most three bytes
in UTF-8 and every code point not in the BMP takes four bytes.
For each of the latter, the encoding of UTF-16 takes as many
bytes; for each of the former, it takes at most 3/2 as many.
Therefore one can decrease the size of the buffer allocated
here.
Reviewed-by: Peter Ross <pross@xvid.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Happens when length > INT_MAX / 2; use unsigned for the computation,
but restrict the value to INT_MAX, because avio_get_str16le()
accepts an int as buf_len argument. Notice that it can happen
that the string read by avio_get_str16le() is truncated in this case.
Reviewed-by: Peter Ross <pross@xvid.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
get_tag() is not designed with negative length in mind;
in this case, it will allocate a very small buffer
(LEN_PRETTY_GUID + 1) and might call avio_get_str16le()
with a negative maxlen (which relies on these parameters
to be signed).
Reviewed-by: Peter Ross <pross@xvid.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Each of the 16 bytes of a GUID is written as a two-character
hex value and three hyphens, leading to a length of 35.
GCC 13 emits a -Wformat-truncation= warning because of this.
Reviewed-by: Peter Ross <pross@xvid.org>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
It is undefined behaviour even in cases where it works
(it works because it is only a const uint8_t* vs. uint8_t* difference).
Instead add a cbuf parameter to pass a const buffer (for writing)
as well as a parameter indicating whether we are reading or writing;
retry_transfer_wrapper() itself then uses the correct function
based upon this information.
Reviewed-by: Marton Balint <cus@passwd.hu>
Reviewed-by: Tomas Härdin <git@haerdin.se>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
av_image_copy() accepts const uint8_t* const * as source;
lots of user have uint8_t* const * and therefore either
cast (the majority) or copy the array of pointers.
This commit changes this by adding a static inline wrapper
for av_image_copy() that casts between the two types
so that we do not need to add casts everywhere else.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Andreas Rheinhardt