summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* tools/target_dec_fuzzer: adjust threshold for RKAMichael Niedermayer2023-09-04
| | | | | | | | Fixes: Timeout Fixes: 59349/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_RKA_fuzzer-5334280839233536 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* avcodec/tta: fix signed overflow in decorrelateMichael Niedermayer2023-09-04
| | | | | | | | Fixes: signed integer overflow: 2079654542 - -139267653 cannot be represented in type 'int' Fixes: 60811/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_TTA_fuzzer-5915858409750528 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* fate: add rka testPaul B Mahol2023-09-04
|
* fate/ffmpeg: Add bitexact flag for ffmpeg-input-r testAndreas Rheinhardt2023-09-04
| | | | | | | | Fixes the test when the non-bitexact MMXEXT versions of the hpeldsp functions are used. Reviewed-by: Anton Khirnov <anton@khirnov.net> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avcodec/osq: improve 32bit sample format decodingPaul B Mahol2023-09-04
|
* avcodec/osq: fix type of nb_samplesPaul B Mahol2023-09-04
| | | | Fixes crash caused by signed integer overflow.
* avformat/vpcc: fix VP9 metadata in FLV and RTMPAlessandro Ros2023-09-04
| | | | | | | | | | | | | | In order to send VP9 tracks with FLV or RTMP, the enhanced RTMP specification tells that VPCodecConfigurationRecord, a.k.a. vpcC ISO-BMFF box, must be inserted into a metadata message. However, the function responsible for generating vpcCs currently returns invalid boxes, that are lacking the Version and Flag fields, inherited from FullBox. For some reason, both flags were being added manually in movenc. This patch fixes the issue. Signed-off-by: Alessandro Ros <aler9.dev@gmail.com> Reviewed-by: Steven Liu <lingjiujianke@gmail.com> Signed-off-by: James Almer <jamrial@gmail.com>
* avutil/internal: Don't auto-include emms.hAndreas Rheinhardt2023-09-04
| | | | | | Instead include emms.h wherever it is needed. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* tests/checkasm/sw_scale: Avoid declare_func_emms where possibleAndreas Rheinhardt2023-09-04
| | | | | | | This makes the test stricter because it is checked that the MMX registers are not accidentally clobbered. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* tests/checkasm/llvidencdsp: Don't use declare_func_emmsAndreas Rheinhardt2023-09-04
| | | | | | | | Only sub_media_pred has an MMXEXT version, so one can use the version with the stricter check (that checks that the MMX registers have not been clobbered) for sub_left_predict. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* tests/checkasm/hevc_*: Avoid using declare_func_emms where possibleAndreas Rheinhardt2023-09-04
| | | | | | | | | Only the idct_dc and add_residual functions have MMX versions, so one can use the version with the stricter check (that checks that the MMX registers have not been clobbered) for all the other checks. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avcodec/pcm: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/af_amultiply: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/af_headphone: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/af_sofalizer: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/af_volume: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/afir_template: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | The floating point dsp code does not use MMX registers since 2718a3be1f8867fd4f6cb3f452d6917838b1ed88. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/vf_bwdif: Remove obsolete emms_c()Andreas Rheinhardt2023-09-04
| | | | | | Obsolete since ed42a51930d9cca6dfed35c4af4b5b3a3f7f6a04. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/vf_colorspace: Remove redundant emms_c()Andreas Rheinhardt2023-09-04
| | | | | | | | | | | It is not even clear what these emms_c() are supposed to achieve: create_filtergraph() (where it may be called) does not call ASM functions itself; it merely sets some function pointers. Furthermore, there are no colorspacedsp functions using MMX (checked by checkasm which does not use declare_new_emms()). Finally, checking whether to issue emms_c() is overblown anyway. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter/vf_gblur: Remove unnecessary emms_c()Andreas Rheinhardt2023-09-04
| | | | | | There is no MMX ASM code for gblur. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avfilter: Remove unnecessary emms_c for ff_scene_sad_get_fnAndreas Rheinhardt2023-09-04
| | | | | | | ff_scene_sad_get_fn() does not return functions that use MMX at all. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* tests/fate/flvenc: add hevc in enhanced flv test caseSteven Liu2023-09-04
| | | | Signed-off-by: Steven Liu <lq@chinaffmpeg.org>
* vf_libplacebo: warn on extra_opts + too old libplaceboNiklas Haas2023-09-03
| | | | Instead of silently ignoring this option as before.
* avfilter/asrc_sinc: check return code of tx_init()Paul B Mahol2023-09-03
|
* doc: mention OSQ supportPaul B Mahol2023-09-02
|
* avcodec/wavarc: fix bugs in arithmetic coding modePaul B Mahol2023-09-02
|
* avformat/wavarc: fix data end calculationPaul B Mahol2023-09-02
|
* avformat/matroska(dec|enc): Rename macro to avoid clash with winbase.hAndreas Rheinhardt2023-09-02
| | | | | | | | winbase.h defines IGNORE and is included via bzlib.h when compiling for Windows. So rename this macro to NOTHING. Also rename the muxer macro for consistency. Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avformat/internal: Don't auto-include os_support.hAndreas Rheinhardt2023-09-02
| | | | | | | | | | | | | | | | | | | | It includes various Windows-specific headers when compiling for Windows and these sometimes cause issues: E.g. winbase.h defines IGNORE, which clashes with a macro used in the Matroska muxer (since 884653ee5be03ed38db957c14fad51b300611c8c) and demuxer. This header provides fallback defines for various stuff that is mostly not used directly by (de)muxers at all: mkdir, rename, rmdir, unlink, access, poll, pollfd, nfds_t, closesocket, socklen_t, fstat, stat, lseek, SHUT_(RD|WR|RDWR) and various POLL* constants. Ergo fix this issue by not auto-including this header in lots of places via an inclusion in internal.h and instead include it everywhere where the above stuff is used (most of these translation units already included os_support.h). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avutil: add thread executorNuo Mi2023-09-02
| | | | | | | | | | The executor design pattern was introduced by java <https://docs.oracle.com/en/java/javase/20/docs/api/java.base/java/util/concurrent/Executor.html> it also adapted by python <https://docs.python.org/3/library/concurrent.futures.html> Compared to handcrafted thread pool management, it greatly simplifies the thread code. Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
* lavc/avfft: deprecate the APILynne2023-09-01
| | | | This deprecates the currently unused API.
* lavu/tx: fix scaling of R2R transformsLynne2023-09-01
| | | | Still slightly inaccurate, but it's good enough now.
* lavu/tx: improve rdft table generation precision slightlyLynne2023-09-01
|
* avfft: wrap lavu/tx instead of ff_dctLynne2023-09-01
|
* avfft: wrap lavu/tx instead of ff_rdftLynne2023-09-01
|
* avfft: wrap lavu/tx instead of ff_mdctLynne2023-09-01
|
* avfft: wrap lavu/tx instead of ff_fftLynne2023-09-01
|
* ffplay: port to lavu/txLynne2023-09-01
|
* wmavoice: convert DCT-I/DST-I to lavu/txLynne2023-09-01
| | | | | | | | This is the very last user of any lavc transform code. This also *corrects* wmavoice decoding, as the previous DCT/DST transforms were incorrect, bringing it closer to Microsoft's own wmavoice decoder.
* wmavoice: convert RDFT to lavu/txLynne2023-09-01
|
* lavu/tx: add DCT-I and DST-I transformsLynne2023-09-01
| | | | | These are true, actual DCT-I and DST-I transforms, unlike the libavcodec versions, which are plainly not.
* lavu/tx: add real to real and real to imaginary RDFT transformsLynne2023-09-01
| | | | | | | These are in-place transforms, required for DCT-I and DST-I. Templated as the mod2 variant requires minor modifications, and is required specifically for DCT-I/DST-I.
* avformat: add OSQ demuxerPaul B Mahol2023-09-01
|
* avcodec: add OSQ audio decoderPaul B Mahol2023-09-01
|
* fftools/ffmpeg_mux: replace monotonous with monotonicLeo Izen2023-09-01
| | | | | | | The word "monotonous" means "spoken in a monotone" which is not what we mean here. We mean "monotonic" i.e. nondecreasing. Signed-off-by: Leo Izen <leo.izen@gmail.com>
* fate/jpegxl_anim: add demuxer fate test for jpegxl_animLeo Izen2023-09-01
| | | | | | | | | Adds a fate test for the jpegxl_anim demuxer, that should allow testing for true positives and false positives for animated jpegxl files. Note that two of the test cases are not animated, in order to help sort out false positives. Signed-off-by: Leo Izen <leo.izen@gmail.com>
* avcodec/rka: do not output extra invalid samples in last framePaul B Mahol2023-08-30
|
* avformat/riffdec: Pass logctx as void* instead of AVFormatContext*Andreas Rheinhardt2023-08-30
| | | | | | Also stop second-guessing error codes from ff_get_extradata(). Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avformat/matroskadec: Use named constants instead of their valueAndreas Rheinhardt2023-08-30
| | | | Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
* avformat/matroskadec: Replace switch with arrayAndreas Rheinhardt2023-08-30
| | | | | | | | | This simplification reduces codesize. (It even reduces the size of .rodata here, because the jump table used by the compiler is bigger than the actual array.) Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-23 23:28:00 +0000