Featured content is also available via:

vulkan: support shader compression

12 January 16:28 - FFmpeg - Lynne

hwcontext_vulkan: enable long vector extension

12 January 16:28 - FFmpeg - Lynne

This enables us to do some neat tricks.

configure/make: support compile-time SPIR-V generation

12 January 16:28 - FFmpeg - Lynne

vulkan: add support for precompiled shaders

12 January 16:28 - FFmpeg - Lynne

vulkan: switch to static allocation for temporary descriptor data

12 January 16:28 - FFmpeg - Lynne

Simplifies management, and the hardware is limited to 4 descriptor sets and whatever bindings.

math: Sync acosh from CORE-MATH

12 January 13:34 - glibc - Adhemerval Zanella

The CORE-MATH c423b9a3 commit made atanh to use a slight different muldd_acc and polydd (which uses muldd_acc internally) compared to previous version.

math: Sync atanh from CORE-MATH

12 January 13:34 - glibc - Adhemerval Zanella

The CORE-MATH c423b9a3 commit made atanh to use a slight different muldd_acc, mulddd, and polydd (which uses muldd_acc internally) compare to asinh and acosh.

math: Sync asinh from CORE-MATH

12 January 13:34 - glibc - Adhemerval Zanella

The CORE-MATH commit cd653cf7 fixes some issues for RNDZ (below).

Linux 6.19-rc5

12 January 03:03 - Linux Kernel - Linus Torvalds

avcodec/d3d12va_encode: add Region of Interest (ROI) support

12 January 02:45 - FFmpeg - stevxiao

This commit implements ROI (Region of Interest) encoding support for D3D12VA hardware encoders, enabling spatially-adaptive quality control for H.264, HEVC, and AV1 encoders. Query for `D3D12_VIDEO_ENCODER_RATE_CONTROL_FLAG_ENABLE_DELTA_QP` support during initialization to check whether the hardware support delta QP.

avcodec/bswapdsp: improve performance by remove manually unroll

10 January 18:56 - FFmpeg - Zhao Zhili

Manually unrolling loops increases code size, which can sometimes improve performance, but more often than not, it degrades performance.

checkasm/riscv: handle other float ABIs than double

10 January 15:20 - FFmpeg - Rémi Denis-Courmont

This splits out the integer and floating point handling to add support for software, single and quad float ABI's.

avformat/tls_gnutls: implement dtls handshake logic

10 January 05:30 - FFmpeg - Jack Lau

Get and set remote addr when dtls server mode.

lavfi/nlmeans: add aarch64 neon for compute_weights_line

9 January 16:10 - FFmpeg - Jun Zhao

Implement NEON optimization for compute_weights_line. Also update the function signature to use ptrdiff_t for stack arguments (max_meaningful_diff, startx, endx).