avcodec/nvenc: fix b_ref_mode capability check Turns out it's a bitfield, not straight values. Fixes #23061]]>
avformat/movenc: avoid negative cts offsets when using an edit list with CMAF output Fixes issue #23417. Signed-off-by: James Almer <[email protected]>]]>
aarch64: vp9lpf: Fix GCS violations The aarch64 VP9 loopfilters actually violate aarch64 GCS (Guarded Control Stack), even though we marked the code as GCS compliant in 846746be4b8edd66be8be2c123b0072de2636e9d. This means that builds with GCS enabled, after that commit, will crash when decoding VP9, on future hardware (or current QEMU) that supports GCS. This also goes for ffmpeg version 8.1.1 where the GCS enabling was backported. This matches the fix that was done for hevcdsp in 1f7ed8a78de1da743a359913ce05cc258a400b5d. This issue wasn't observed if running checkasm in QEMU - therefore, I thought all GCS issues had been fixed by 846746be4b8edd66be8be2c123b0072de2636e9d. (If I would have tested the full "make fate" with QEMU, the issue would have appeared though.) However with the new checkasm, some of the GCS violations do appear even in checkasm. The reason is that the checkasm vp9 test intentionally craft input pixels that attempt to trigger all the individual separate cases in each input buffer (in randomize_loopfilter_buffers). This means that the checkasm tests actually never test or exercise the early exit cases, which are the ones that violate GCS. With the new checkasm, the call to "bench_new" always test running the code at least once, even if not benchmarking. As the input buffers weren't reinitialized between the test and "bench_new", the pixel differences now differ from the initial setup, so that the code now some times (often) would end up hitting the early exit cases. Ideally, the vp9 checkasm test would be repeated to cover all cases of input buffers that allow early exits, in addition to covering the case with all different cases in one block.]]>
swscale/aarch64/yuv2rgb_neon: add BE 16bpp output formats BE counterparts to the LE paths in 2e142e52ae; pack adds rev16 before store. nv12/nv21 paths are added but bench-only (no C ref, same as 2e142e52ae). Test Name A55-gcc M1-clang A76-gcc ------------------------------------------------------------------------------------- yuv420p_rgb565be_1920_neon 15086.1 ( 3.91x) 5507.0 ( 4.34x) 19229.1 ( 2.02x) yuv420p_bgr565be_1920_neon 15291.7 ( 3.84x) 5476.9 ( 4.37x) 19229.4 ( 2.02x) yuv420p_rgb555be_1920_neon 15091.5 ( 3.67x) 5569.0 ( 3.97x) 19229.3 ( 1.90x) yuv420p_bgr555be_1920_neon 15298.6 ( 3.62x) 5600.6 ( 3.98x) 19228.8 ( 1.90x) yuv422p_rgb565be_1920_neon 16862.3 ( 4.00x) 6378.8 ( 4.64x) 22110.3 ( 2.07x) yuv422p_bgr565be_1920_neon 17139.3 ( 3.93x) 6448.1 ( 4.50x) 22104.1 ( 2.07x) yuv422p_rgb555be_1920_neon 16853.3 ( 3.98x) 6468.8 ( 4.12x) 22106.4 ( 1.98x) yuv422p_bgr555be_1920_neon 17202.2 ( 3.89x) 6467.0 ( 4.12x) 22110.2 ( 1.98x) yuva420p_rgb565be_1920_neon 15050.2 ( 3.92x) 5452.5 ( 4.39x) 19229.5 ( 2.02x) yuva420p_bgr565be_1920_neon 15346.6 ( 3.84x) 5462.4 ( 4.36x) 19228.9 ( 2.02x) yuva420p_rgb555be_1920_neon 15050.8 ( 3.69x) 5463.3 ( 3.95x) 19228.6 ( 1.90x) yuva420p_bgr555be_1920_neon 15352.8 ( 3.61x) 5543.6 ( 3.89x) 19228.6 ( 1.90x) Co-authored-by: Ramiro Polla <[email protected]> Signed-off-by: DROOdotFOO <[email protected]>]]>
swscale/yuv2rgb: add explicit BE/LE 565/555 cases ff_yuv2rgb_get_func_ptr() now returns the C reference for explicit BE/LE 16bpp formats, not only the NE alias. Signed-off-by: DROOdotFOO <[email protected]>]]>
avcodec/codec_id: add .props to AV_CODEC_ID_APPLE_APAC Signed-off-by: Romain Beauxis <[email protected]>]]>
avfilter/vf_scale_d3d11: ensure guids are defined fix vf_scale_d3d11.o : error LNK2001: unresolved external symbol IID_ID3D11VideoContext]]>
swscale/x86: use correct HOSTCC_E flag instead of CC_E HOSTCC and CC might be completely different compilers. Signed-off-by: Niklas Haas <[email protected]>]]>
swscale/uops_tmpl: move attributes before `static` keyword This fails to compile with C23 standard attributes otherwise. Technically only av_unused requires this, but move the other attributes as well for consistency / future proofing. Signed-off-by: Niklas Haas <[email protected]>]]>
lavc/vvc: Fix num_entry_points derivation when using RPR Context: 1. In the case sps_subpic_info_present=0, there is a single subpicture which includes the entire picture. 2. When sps_subpic_info_present=0, we might be using Reference Picture Resampling (RPR), in which picture sizes might differ in the PPS, rather than in the SPS. Because of 2., we can't rely on the sequence-level variables sps_subpic_width_minus1 and sps_subpic_height_minus1 to derive the picture-level variable num_entry_points, as the picture might have a different size to the picture used when deriving those sequence-level variables.]]>
avdevice/gdigrab: make overlay window layered WS_EX_LAYERED allows input events to pass through to windows beneath. WS_EX_NOACTIVATE prevents the window from stealing focus when created.]]>
avdevice/gdigrab: process window in a separate thread Move window creation and event processing to a dedicated thread. GetMessage only processes events from the calling thread's message queue. Because gdigrab_read_header and gdigrab_read_packet don't run on the same thread, the message queue was not being drained. Fixes: #11539]]>
libavfilter/libplacebo: gamma22 and gamma28 aliases]]>
libavcodec/options_table: gamma22 and gamma28 aliases]]>
avformat/isom_tags: Add support for detecting apple_apac Signed-off-by: Romain Beauxis <[email protected]>]]>
avcodec/codec_id: Add Apple Positional Audio Codec. Signed-off-by: Romain Beauxis <[email protected]>]]>
aacdec_usac: apply volume normalization settings]]>
aacdec_usac: implement basic DRC parameter decoding]]>
swscale/tests/sws_ops: fix uops leak on translate success path Fall through to the existing cleanup so uops is freed on both the success and failure paths. Signed-off-by: Michael Niedermayer <[email protected]>]]>
docs: refine issue tracker transition references]]>