Add a full parser to libavcodec for AV_CODEC_ID_JPEGXL. It finds the
end of the stream in order to packetize the codec, and it looks at
the headers to set preliminary information like dimensions and pixel
format.
Note that much of this code is duplicated from avformat/jpegxl_probe.c,
but that code will be removed and call this instead in the next commit.
Signed-off-by: Leo Izen <leo.izen@gmail.com>
Before this commit, the decoder erroneously assumes that the AVFrame
passed to the receive_frame is the same one each time. Now it keeps an
internal AVFrame to write into, and copies it over when it's done.
Signed-off-by: Leo Izen <leo.izen@gmail.com>
Fixes an error that's caused by decoding a grayscale JXL image after an
RGB image is decoded, with the same decoder instance.
Signed-off-by: Leo Izen <leo.izen@gmail.com>
It is currently probably not possible for it to be negative as
the needed 2Mb input buf size is not achievable. But it is more
robust to check for it too.
If it would become negative than code like
s->samples[0][n] = s->samples[0][s->nb_samples + n];
would crash
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: shift exponent 1285 is too large for 32-bit type 'int'
Fixes: 60870/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_WAVARC_fuzzer-5332050340347904
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
Fixes: signed integer overflow: -1403461578 + -843974775 cannot be represented in type 'int'
Fixes: 60868/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_MPEG1VIDEO_fuzzer-4599793035378688
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
if !ph_deblocking_params_present_flag is true, ph_deblocking_filter_disabled_flag infered from pps
if !sh_deblocking_params_present_flag is true, sh_deblocking_filter_disabled_flag infered from ph
Failed clips:
ENT444MAINTIER_C_Sony_3.bit
ENT444HIGHTIER_D_Sony_3.bit
Signed-off-by: James Almer <jamrial@gmail.com>
if pps_alf_info_in_ph_flag is true
sh_alf_enabled_flag infered from ph
Failed clip:
LTRP_A_ERICSSON_3.bit
Signed-off-by: James Almer <jamrial@gmail.com>
if sh_picture_header_in_slice_header_flag is true
sh_lmcs_used_flag and sh_explicit_scaling_list_used_flag are infered from ph
Failed clips:
LMCS: CLM_A_KDDI_2.bit STILL444_A_KDDI_1.bit
Scaling: SCALING_B_InterDigital_1.bit SCALING_A_InterDigital_1.bit
Signed-off-by: James Almer <jamrial@gmail.com>
By default the OpenEXR decoder outputs linear light pixel data by
applying a gamma=1.0 transfer (i.e. a no-op). When it does so, it
should tag the data as linear so color-managed filters or other tools
can work with it correctly.
Signed-off-by: Leo Izen <leo.izen@gmail.com>
Also rename the contexts and the functions so their names will reflect their
intended size.
With the earlier patch this fixes the audio corruption regression caused by
6ba0aa1770.
Fixes ticket #10029.
Signed-off-by: Marton Balint <cus@passwd.hu>
Improves the audio corruption regression caused by
6ba0aa1770 reported in ticket #10029.
There is still however a noticable audio glitch, so the FFT conversion to AVTX
probably also needs some modifications.
Signed-off-by: Marton Balint <cus@passwd.hu>
This patch changes the return instruction in the tr_32x4 macro from
BR to RET.
Function returns should always use the RET instruction instead of BR,
to avoid interfering with branch prediction.
On devices that support BTI, this is observeable as a landing pad is
required when branching with BR. The change fixes
fate-hevc-hdr-vivid-metadata when on hardware with BTI support.
Signed-off-by: Casey Smalley <casey.smalley@arm.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
This commit is the AVHWAccel analogue of commit
20f9727018: It moves the private fields
of AVHWAccel to a new struct FFHWAccel extending AVHWAccel
in an internal header (namely hwaccel_internal.h).
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
All usages of ff_hwaccel_frame_priv_alloc() have the same pattern:
Check for whether a hwaccel is in use; check whether it needs
private frame-specific data; allocate the AVBuffer and set
it.
This commit modifies ff_hwaccel_frame_priv_alloc() to perform
this task on its own.
(It also seems that the H.264 decoder did not perform proper
cleanup in case the buffer could not be allocated. This has been
changed.)
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
libavcodec/hwconfig.h currently contains HWACCEL_CAP_* flags
as well as the definition of AVCodecHWConfigInternal and some
macros to create them.
The users of these two are nearly disjoint: The flags are used
by files providing AVHWAccels whereas AVCodecHWConfigInternal
is used by files providing codecs (for FFCodec.hw_configs).
This patch therefore moves these flags to a new file hwaccel_internal.h.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
Code should make more sense now
Fixes: out of array access
Fixes: 58299/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000_fuzzer-6627570448465920
Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Reviewed-by: Pierre-Anthony Lemieux <pal@sandflow.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
