eacde6ee7b
avcodec/svq3: Deduplicate allocating dummy frame
...
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:10:52 +02:00
cb54eca41a
avcodec/svq3: Improve returned error codes
...
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:10:52 +02:00
0468b866ca
avcodec/svq3: Factor out decoding extradata
...
Reduces indentation and avoids an extra variable for whether
a sequence header has been found.
It also fixes potential undefined behaviour:
NULL + 0 is undefined and happens when no extradata is available.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:10:52 +02:00
0b3dae6c63
avcodec/mpeg12enc: Combine put_bits()
...
This is a 16bit field in the spec, so using a single
put_bits() to write it is more natural.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:10:22 +02:00
f5f890fee4
avcodec/mpeg12dec: Remove nonsense comment
...
Everything in mpeg12dec.c is about decoding.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:10:05 +02:00
562192e283
avcodec/mpeg12dec: Remove redundant save_width/height
...
These have been added in 29644cb504andreas.rheinhardt@outlook.com >
2025-05-21 03:09:55 +02:00
56497671dd
avcodec/mpeg12dec: Set save_chroma_format also for VCR2
...
Otherwise the MpegEncContext would be unnecessarily reinitialized
once (this does not affect the output for an intra-only variant
like VCR2) in mpeg_decode_postinit().
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 03:09:45 +02:00
a1e4be5740
avcodec/mpegvideo_dec: Move ff_mpv_report_decode_progress() to h263dec.c
...
It is its only user. Also make it static and call it
before ff_mpeg_draw_horiz_band().
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 02:54:22 +02:00
e12cbf6095
avcodec/tests/avcodec: Check only frame-mt decoders set update_thread_ctx
...
Possible since 554b52b1b3andreas.rheinhardt@outlook.com >
2025-05-21 02:44:15 +02:00
a6f7b32085
avcodec/ratecontrol: Fix wrong sizeof expression
...
Fixes Coverity issue #1644200 .
Introduced in 89a8033fc9andreas.rheinhardt@outlook.com >
2025-05-21 02:44:06 +02:00
36af32f9e5
avcodec/tests/.gitignore: Add apv test tool
...
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 02:43:22 +02:00
bd63c374d2
avformat/avformat: Remove outdated io_close2 documentation
...
The io_close callback has been removed in
d6799ee0e4andreas.rheinhardt@outlook.com >
2025-05-21 02:43:08 +02:00
66f5323652
avcodec/rv60dec: Avoid branch when decoding cbp16
...
Reviewed-by: Peter Ross <pross@xvid.org >
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-21 02:41:55 +02:00
d5a9f7bdd4
avcodec/cuviddec: only flush cuvid when output queue is empty
2025-05-21 01:16:19 +02:00
431e2cae87
avcodec/cuviddec: print error when queueing frames fails
2025-05-21 01:15:29 +02:00
842fa198e9
hwcontext_vulkan: fix build with old Vulkan header versions
2025-05-21 03:11:07 +09:00
f414038779
configure: identify loong64 for loongarch
...
dpkg-architecture set DEB_HOST_ARCH_CPU as loong64 on loongarch.
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com >
2025-05-20 21:19:53 +08:00
1249776641
Makefile: Remove postproc from ALLFFLIBS
...
Fix
ld: warning: search path 'libpostproc' not found
Reviewed-by: Ramiro Polla <ramiro.polla@gmail.com >
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com >
2025-05-20 21:17:43 +08:00
4f7bc62c66
avformat/allformats: Move avisynth and dvdvideo under external libraries group
...
Signed-off-by: Zhao Zhili <zhilizhao@tencent.com >
2025-05-20 21:17:28 +08:00
eabb62813e
hwcontext_vulkan: only try exporting DMABUF memory on !WIN32 and only for DMABUF tiling
2025-05-20 19:53:02 +09:00
7c3c5c8052
hwcontext_vulkan: correct image transfer usage flags
...
By pure coincidence, BUFFER and IMAGE flags were equal for those
two usage types.
2025-05-20 19:53:02 +09:00
435db9bb49
vulkan: enable VK_KHR_shader_subgroup_rotate
...
Yet another thing that should've been always present.
2025-05-20 19:53:02 +09:00
7b45d9c5fd
vulkan_ffv1: pipe through slice decoding status
2025-05-20 19:53:02 +09:00
cb8f4b675d
vulkan/ffv1: unify encode and decode get/put primitives
...
This simply makes a get_rac/put_rac_internal variant that can be
reused.
2025-05-20 19:53:02 +09:00
7576410af7
ffv1enc_vulkan: implement RCT search for level >= 4
2025-05-20 19:53:01 +09:00
0156680f09
ffv1enc_vulkan: implement the cached EC writer from the decoder
...
This gives a 35% speedup on AMD and 50% on Nvidia.
2025-05-20 19:53:01 +09:00
a24ea37228
vulkan_ffv1: fix PCM + cached symbol reader
...
writeout_rgb requires that all subgroups are active.
2025-05-20 19:53:01 +09:00
f69db914ce
ffv1enc_vulkan: use ff_get_encode_buffer
...
We used to create our own buffer, but still used the DR1 flag,
which is not how it's supposed to work.
Instead, use ff_get_encode_buffer, and either host-map the buffer
before copying each slice via GPU transfers, or just copy each
slice manually if that fails or is unavailable.
2025-05-20 19:53:01 +09:00
8a2d921627
ffv1_common: minor RGB optimization
2025-05-20 19:53:01 +09:00
bd41838b60
ffv1enc_vulkan: switch to 2-line cache, unify prediction code
2025-05-20 19:53:01 +09:00
52595025c5
ffv1enc_vulkan: minor EC optimizations
2025-05-20 19:53:01 +09:00
7c0a8c07ce
ffv1enc_vulkan: unify EC code between setup and encode
2025-05-20 19:53:00 +09:00
69f83bafd1
ffv1enc_vulkan: get rid of temporary data for the setup shader
2025-05-20 19:53:00 +09:00
a4078abd73
vulkan/ffv1: synchronize get_pred implementations between encoder and decoder
2025-05-20 19:53:00 +09:00
ebbc7ff650
ffv1enc_vulkan: merge all encoder variants into one file
...
Makes it easier to work with, despite the heavy ifdeffery.
2025-05-20 19:52:55 +09:00
fd18ae88ae
avcodec/x86/vp9: Add AVX-512ICL for 16x16 and 32x32 8bpc inverse transforms
2025-05-19 15:56:27 +02:00
b6803bf104
aarch64: increase default alignment for functions and constants
...
Use 16-byte alignment (align=4) instead of 4-byte (align=2) in the function and
const macros. This improves instruction fetch and NEON load performance on
modern AArch64 CPUs.
2025-05-19 13:20:51 +02:00
c55d65ac0a
configure: correct liboapv feature support
...
Only encoding support has been added for liboapv
2025-05-19 15:21:33 +05:30
9c5ed57f94
ogg/opus: implement header packet skip in chained ogg bitstreams.
2025-05-19 07:24:05 +02:00
2fb6416dd0
ogg/flac: implement header packet skip in chained ogg bitstreams.
2025-05-19 07:24:05 +02:00
bd2dcfaed4
tests/fate/matroska: Add container cropping test
...
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com >
2025-05-19 03:21:27 +02:00
95c43c6d0e
tests/fate/pixfmt: fix definition of 16bit tests
...
No effect as is, but without this change, new additions to FATE_PIXFMT_16-*
will not work.
Signed-off-by: James Almer <jamrial@gmail.com >
2025-05-18 19:54:32 -03:00
9fadd6ddad
doc: add htmlxref.cnf
...
Silences warnings like
filters.texi:256: warning: no htmlxref.cnf entry found for `ffmpeg-utils'
Signed-off-by: James Almer <jamrial@gmail.com >
2025-05-18 19:54:09 -03:00
0021484d05
avformat/mpegts: update stream info when PMT ES stream_type changes
...
I have several .ts captures where video and audio codec changes even
though the PMT version does not change and the PIDs stay the same.
This happens during transition to/from slate (mpeg2 video and audio)
to network broadcast (hevc video and eac3 audio in private PES).
I've updated fate ts-demux expected results.
2025-05-18 08:57:31 -06:00
20502ba92a
ffmpeg: Don't print graphs if there are no outputs yet
...
Avoids writing an empty json blob in setup error cases.
2025-05-18 15:17:41 +01:00
c18d1b63ab
fftools/graphprint: Fix leak of graph section header string
2025-05-18 15:17:41 +01:00
2070cc138b
fftools/graphprint: Fix leak of graphprint object
2025-05-18 15:17:41 +01:00
06cee0c681
doc: add swscale rewrite design document
...
This should hopefully serve as a better introduction to my new swscale
redesign than hunting down random commit message monologues.
2025-05-18 15:00:58 +02:00
6072e27e9a
swscale/graph: prefer bools to ints
...
This is more consistent with the rest of the newly added code, which
universally switched to using bools for boolean values.
2025-05-18 15:00:45 +02:00
d95944786e
swscale/graph: move vshift() and shift_img() to shared header
...
I need to reuse these inside `ops.c`.
2025-05-18 14:39:57 +02:00
