summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* cbs_vp9: Fix VP9 passthroughMark Thompson2025-08-0915-43/+72
| | | | | | | | | Don't overwrite the bitstream values when updating the top-level loop filter and segmentation state, instead do the update separately at the end of the frame parsing. This also reverts the change to the passthrough tests which made them have output not matching the input.
* hwcontext_vulkan: Fix buildMark Thompson2025-08-091-1/+1
|
* hwcontext_vulkan: Fix build with older Vulkan headersMark Thompson2025-08-091-1/+1
|
* doc/community.texi: fix spelling errorLeon Grutters2025-08-091-1/+1
| | | | | Fixes: 262d41c804 ("all: fix typos found by codespell") Signed-off-by: Leon Grutters <[email protected]>
* doc/protocols: add doc for dtlsJack Lau2025-08-091-0/+81
| | | | Signed-off-by: Jack Lau <[email protected]>
* avformat/tls: add new option use_srtp to control whether enable itJack Lau2025-08-093-12/+15
| | | | | | | | The SRTP profile string "SRTP_AES128_CM_SHA1_80" is only used when use_srtp is enabled. Move its declaration inside the "if (s->use_srtp)" block to limit scope Signed-off-by: Jack Lau <[email protected]>
* avformat/tls_openssl: add av_assert0() for tls_sharedJack Lau2025-08-091-0/+3
| | | | Signed-off-by: Jack Lau <[email protected]>
* avformat/tls_openssl: simplify the external_sock checkJack Lau2025-08-091-2/+2
| | | | Signed-off-by: Jack Lau <[email protected]>
* avformat/tls_openssl: cleanup the pointer name of TLSContext and TLSSharedJack Lau2025-08-091-116/+117
| | | | | | | | Pure cleanup, no functional changes Unify local pointer names to `TLSContext *c` and `TLSShared *s` to reduce confusion from inconsistent names (e.g. p, ctx, s, c) Signed-off-by: Jack Lau <[email protected]>
* forgejo/autolabeler: add some logic to also auto-remove the "New" labelTimo Rothenpieler2025-08-092-18/+59
|
* Bump versions for master after release/8.0n8.1-devMichael Niedermayer2025-08-097-7/+7
| | | | Signed-off-by: Michael Niedermayer <[email protected]>
* Bump versions for release/8.0Michael Niedermayer2025-08-097-7/+7
| | | | Signed-off-by: Michael Niedermayer <[email protected]>
* Changelog: Add 8.0 markerMichael Niedermayer2025-08-091-0/+3
| | | | Signed-off-by: Michael Niedermayer <[email protected]>
* doc/APIchanges: updateMichael Niedermayer2025-08-091-52/+52
| | | | Signed-off-by: Michael Niedermayer <[email protected]>
* avcodec/jpeg2000dec: Make sure the 4 extra bytes allocated are initializedMichael Niedermayer2025-08-091-0/+1
| | | | | | | | Fixes: use of uninitialized memory Fixes: 429130590/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_JPEG2000_DEC_fuzzer-5736930522497024 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <[email protected]>
* libavfilter: add af_whisper codeownerVittorio Palmisano2025-08-091-0/+1
|
* vulkan: temporarily disable threading for ASIC-based hwaccelsLynne2025-08-094-4/+4
| | | | | | | | | The issue is that a race condition exists between threads locking frames, which results in cyclic loops and deadlocks. Compute-based implementations are not affected. Fixes #20169
* avformat/tls: switch on peer certificate verification by defaultTimo Rothenpieler2025-08-093-1/+11
|
* forgejo/CODEOWNERS: add myself to d3d12vaWu Jianhua2025-08-081-0/+3
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* mailmap: add entry for myselfWu Jianhua2025-08-081-0/+2
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* MAINTAINERS: add myself as hwcontext_d3d12va maintainerWu Jianhua2025-08-081-0/+2
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* libavfilter: Whisper audio filterVittorio Palmisano2025-08-086-0/+579
| | | | | It adds a new audio filter for running audio transcriptions with the whisper model. Documentation and examples are included into the patch.
* CODEOWNERS: Add myself for VVCWu Jianhua2025-08-081-3/+3
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* CODEOWNERS: Add myself for VVCNuo Mi2025-08-081-3/+3
| | | | Signed-off-by: Nuo Mi <[email protected]>
* fate/vvc: add vvc-conformance-FIELD_A_4Wu Jianhua2025-08-082-0/+26
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* fate/vvc: add vvc-conformance-ACT_A_3Wu Jianhua2025-08-082-0/+71
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* fate/vvc: add vvc-conformance-10b422_L_5Wu Jianhua2025-08-082-1/+28
| | | | | | This commit added 10b422_L_5 for testing palette mode. Signed-off-by: Wu Jianhua <[email protected]>
* avcodec/vvc/dec: fix typo and also output log when the checksum is correctWu Jianhua2025-08-081-7/+5
| | | | | | It's helpful for developers and the same as the hevcdec. Signed-off-by: Wu Jianhua <[email protected]>
* avcodec/h274: fix hash verification on BEWu Jianhua2025-08-081-15/+57
| | | | Signed-off-by: Wu Jianhua <[email protected]>
* avcodec/vvc/ctu: should use the width and height of the start componentWu Jianhua2025-08-081-11/+18
| | | | | | This commit fixed decoding the DUAL_TREE_CHROMA palette coding unit Signed-off-by: Wu Jianhua <[email protected]>
* avfilter/avf_showcqt: fix unbounded index when copying to fft_dataMuhammad Faiz2025-08-081-2/+2
| | | | | | When timeclamp and/or fps are low, j can be negative. Fix Ticket11640
* forgejo/CODEOWNER: forgejo uses regex to match pathsTimo Rothenpieler2025-08-081-56/+58
|
* avutil/hwcontext_vulkan: use uint64_t specifier in an av_log() callJames Almer2025-08-081-1/+1
| | | | Signed-off-by: James Almer <[email protected]>
* avcodec/vulkan_encode_av1: use size_t specifier in an av_log() callJames Almer2025-08-081-1/+1
| | | | Signed-off-by: James Almer <[email protected]>
* avcodec/vulkan_encode_av1: remove unused variablesJames Almer2025-08-081-6/+0
| | | | Signed-off-by: James Almer <[email protected]>
* vulkan_encode_av1: properly give an FPS value for ff_av1_guess_levelLynne2025-08-081-4/+3
|
* configure: enable Vulkan VP9 decoding and AV1 encoding if the headers support itLynne2025-08-081-1/+7
|
* vulkan_decode: add ifdefs around VP9 definitions and privatize profile structLynne2025-08-082-10/+31
| | | | The struct is not referenced anywhere else.
* vulkan_av1: Fix frame threadingBenjamin Cheng2025-08-083-19/+11
| | | | | Basically do the same thing that was done for VP9, and remove the vestigial frame_id_alloc_mask in the context.
* vulkan_vp9: Fix frame threadingBenjamin Cheng2025-08-081-7/+11
| | | | | | | The original cleanup of frame_id_alloc_mask was done in free_frame_priv which may be called after the start_frame for the next few frames, causing the frame_id slots to be exhausted. Instead, decide the in-use frame_id slots by checking the frame_id present in the DPB as we need it.
* lavc/vaapi_encode_av1: Fix ref_order_hint value for second slotDavid Rosca2025-08-081-0/+3
| | | | | | | | We always use two slots, even when only one L0 reference is supported by the driver. However we still need to set the correct value for the ref_order_hint of the second slot. Fixes regression from bf9f921ef7 ("avcodec/hw_base_encode: restrict size of next_prev")
* lavu/hwcontext_vaapi: drop a redundant checknyanmisaka2025-08-081-6/+0
| | | | | | | | vaapi_get_image_format() will be called in vaapi_map_frame(). So it's a double check, drop it to avoid redundancy. Signed-off-by: nyanmisaka <[email protected]>
* lavu/hwcontext_vaapi: fix a race mapping to allow CSC to YUV420Pnyanmisaka2025-08-081-3/+11
| | | | | | | | | | | | | | | | | | There's a race condition for YUV420P when mapping from pix_fmt to VA fourcc, both I420 and YV12 could be found by pix_fmt. Currently, vaapi_get_image_format() iterates over the query results of pix_fmt and returns the first matching result in the order declared in the driver. This may result in an incorrect image_format. Now use fourcc to find the image_format. Fixes: ffmpeg -hwaccel vaapi -hwaccel_output_format vaapi -i INPUT \ -vf scale_vaapi=format=yuv420p,hwmap,format=yuv420p \ -vframes 1 -f rawvideo -y yuv420p.yuv Signed-off-by: nyanmisaka <[email protected]>
* avcodec/aac/aacdec_usac: Limit sfo from noise offset to be above -200Michael Niedermayer2025-08-081-2/+3
| | | | | | | | Fixes: out of array read Fixes: 397731127/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_LATM_fuzzer-5577772965101568 Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <[email protected]>
* avcodec/aacsbr_template: Check ilbMichael Niedermayer2025-08-081-0/+6
| | | | | | | | | | Fixes: index 50 out of bounds for type 'INTFLOAT [40][2]' Fixes: 401661737/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_LATM_fuzzer-4866055713652736 Someone knowing AAC well should review this, there is likely a nicer fix Found-by: continuous fuzzing process https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg Signed-off-by: Michael Niedermayer <[email protected]>
* vulkan_encode_av1: Fix uninitialized return value in write_extra_headersDavid Rosca2025-08-081-0/+1
|
* vulkan_encode_av1: Fix setting QPDavid Rosca2025-08-081-3/+3
| | | | Use the "qp" param instead of "global_quality" to match H264/5.
* avfilter/vf_libplaceb: use TS2T() macroNiklas Haas2025-08-081-5/+5
|
* avfilter/vf_libplacebo: use the first visible input for metadataNiklas Haas2025-08-081-9/+4
| | | | | | Instead of undconditionally using the first input. This covers the case of one layer fully obscuring another layer, in which case that should become the new "base" layer.
* avfilter/vf_libplacebo: flush render cache when input is invisibleNiklas Haas2025-08-081-1/+3
| | | | | | | | | This prevents leaking stale metadata from previous frames, for example if an overlay temporarily obscures this input and then un-obscures it again. It is worth pointing out that this does change the semantics subtly, because of the smoothing period on detected HDR metadata, but I argue that the new behavior is an improvement, as it will avoid leaking past metadata that is definitely no longer relevant after an image is unobscured.
generated by cgit v1.2.3 (git 2.25.1) at 2025-09-23 10:48:06 +0000