Repository navigation
Releases
Releases · VMAFx/ffmpeg-patches
ffmpeg-patches v0.7.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | rejected | link | vmafx |
| 0003 | avcodec/nvenc: drop hevc_nvenc SEI above a fixed size | merged | link | vmafx, pelorus |
| 0004 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | submitted | link | pelorus |
| 0005 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0006 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | rejected | link | pelorus |
| 0007 | vulkan: de-phase queue rotation and pin video pools to a single queue | merged | link | pelorus |
| 0008 | hwcontext_vulkan: import a frame again when the CUDA context changes | submitted | link | pelorus |
| 0009 | avcodec/qsvdec: discard the right frames on MP4 edit lists | unsubmitted | none | pelorus |
| 0010 | avfilter/dovi_reshape: add a Dolby Vision RPU reshaping helper | unsubmitted | none | vmafx, pelorus |
| 0011 | avfilter: add dovi_reshape filter | unsubmitted | none | vmafx, pelorus |
| 0012 | avfilter/bt2408_hable: add a Hable tone map on IPT intensity | unsubmitted | none | vmafx, pelorus |
| 0013 | avfilter/vf_tonemap: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0014 | avfilter/vf_zscale: keep the light levels when linearising | unsubmitted | none | vmafx, pelorus |
| 0015 | avfilter/vf_tonemap_opencl: declare the output matrix and range on the link | unsubmitted | none | vmafx, pelorus |
| 0016 | avfilter/vf_tonemap_opencl: zero the peak detection buffer | merged | link | vmafx, pelorus |
| 0017 | avfilter/vf_tonemap_opencl: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0018 | avfilter/vf_tonemap_vaapi: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0019 | avfilter/vf_tonemap_vulkan: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0020 | avfilter/vf_tonemap: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0021 | avfilter/vf_tonemap_vulkan: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0022 | avfilter/vf_dovi_reshape: add a profile 8.1 RPU output mode | unsubmitted | none | vmafx, pelorus |
| 0023 | avfilter/vf_hqdn3d_vulkan: add a Vulkan hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0024 | avfilter/vf_hqdn3d_cuda: add a CUDA hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0025 | avfilter/vf_hqdn3d_opencl: add an OpenCL hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0026 | avfilter/vf_hqdn3d_videotoolbox: add a Metal hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0027 | swscale: allow a same-colour YUV repack of an unconvertible colour space | unsubmitted | none | vmafx, pelorus |
| 0028 | avfilter/vf_scale: negotiate IPT-C2 for a same-colour repack | unsubmitted | none | vmafx, pelorus |
| 0029 | avfilter/dovi_reshape: share the HDR10 output properties | unsubmitted | none | vmafx, pelorus |
| 0030 | avfilter/vf_tonemap_opencl: reshape Dolby Vision profile 5 in the kernel | unsubmitted | none | vmafx, pelorus |
| 0031 | avfilter/tonemap_peak: read the Dolby Vision level 1 peak from side data | unsubmitted | none | vmafx, pelorus |
| 0032 | avfilter/vf_tonemap_opencl: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0033 | avfilter/vf_tonemap_vaapi: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0034 | avfilter/vf_tonemap_opencl: detect the bt2408_hable signal peak on the GPU | unsubmitted | none | vmafx, pelorus |
| 0035 | avutil/hwcontext_vulkan: Query export support with the real image flags | merged | link | pelorus |
| 0036 | checkasm/snowdsp: Include intmath before public headers | submitted | link | vmafx |
| 0037 | avcodec/aom_film_grain: Assume the predicted luma point bound | submitted | link | vmafx |
| 0038 | avcodec/vvc/intra_template: bound reference corner extension | submitted | link | vmafx |
| 0039 | avcodec/wmalosslessdec: Clear subframe flags as an array | submitted | link | vmafx |
| 0040 | avfilter/vf_libvmaf: accept more planar pixel formats for CUDA frames | submitted | link | vmafx |
| 0041 | avfilter/vf_libvmaf: copy all planes of CUDA hardware frames | submitted | link | vmafx |
| 0042 | avfilter/vf_libvmaf: pop the CUDA context when a plane copy fails | submitted | link | vmafx |
| 0043 | libavcodec: unified -qpfile across libx264/libsvtav1/libaom | unsubmitted | none | vmafx |
| 0044 | fftools: decode on the GPU in a loopback decoder | unsubmitted | none | vmafx |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.7.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.7.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.6.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | rejected | link | vmafx |
| 0003 | avcodec/nvenc: drop hevc_nvenc SEI above a fixed size | merged | link | vmafx, pelorus |
| 0004 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | submitted | link | pelorus |
| 0005 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0006 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | rejected | link | pelorus |
| 0007 | vulkan: de-phase queue rotation and pin video pools to a single queue | merged | link | pelorus |
| 0008 | hwcontext_vulkan: import a frame again when the CUDA context changes | submitted | link | pelorus |
| 0009 | avcodec/qsvdec: discard the right frames on MP4 edit lists | unsubmitted | none | pelorus |
| 0010 | avfilter/dovi_reshape: add a Dolby Vision RPU reshaping helper | unsubmitted | none | vmafx, pelorus |
| 0011 | avfilter: add dovi_reshape filter | unsubmitted | none | vmafx, pelorus |
| 0012 | avfilter/bt2408_hable: add a Hable tone map on IPT intensity | unsubmitted | none | vmafx, pelorus |
| 0013 | avfilter/vf_tonemap: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0014 | avfilter/vf_zscale: keep the light levels when linearising | unsubmitted | none | vmafx, pelorus |
| 0015 | avfilter/vf_tonemap_opencl: declare the output matrix and range on the link | unsubmitted | none | vmafx, pelorus |
| 0016 | avfilter/vf_tonemap_opencl: zero the peak detection buffer | merged | link | vmafx, pelorus |
| 0017 | avfilter/vf_tonemap_opencl: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0018 | avfilter/vf_tonemap_vaapi: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0019 | avfilter/vf_tonemap_vulkan: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0020 | avfilter/vf_tonemap: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0021 | avfilter/vf_tonemap_vulkan: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0022 | avfilter/vf_dovi_reshape: add a profile 8.1 RPU output mode | unsubmitted | none | vmafx, pelorus |
| 0023 | avfilter/vf_hqdn3d_vulkan: add a Vulkan hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0024 | avfilter/vf_hqdn3d_cuda: add a CUDA hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0025 | avfilter/vf_hqdn3d_opencl: add an OpenCL hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0026 | avfilter/vf_hqdn3d_videotoolbox: add a Metal hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0027 | swscale: allow a same-colour YUV repack of an unconvertible colour space | unsubmitted | none | vmafx, pelorus |
| 0028 | avfilter/vf_scale: negotiate IPT-C2 for a same-colour repack | unsubmitted | none | vmafx, pelorus |
| 0029 | avfilter/dovi_reshape: share the HDR10 output properties | unsubmitted | none | vmafx, pelorus |
| 0030 | avfilter/vf_tonemap_opencl: reshape Dolby Vision profile 5 in the kernel | unsubmitted | none | vmafx, pelorus |
| 0031 | avfilter/tonemap_peak: read the Dolby Vision level 1 peak from side data | unsubmitted | none | vmafx, pelorus |
| 0032 | avfilter/vf_tonemap_opencl: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0033 | avfilter/vf_tonemap_vaapi: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0034 | avfilter/vf_tonemap_opencl: detect the bt2408_hable signal peak on the GPU | unsubmitted | none | vmafx, pelorus |
| 0035 | avutil/hwcontext_vulkan: Query export support with the real image flags | merged | link | pelorus |
| 0036 | checkasm/snowdsp: Include intmath before public headers | submitted | link | vmafx |
| 0037 | avcodec/aom_film_grain: Assume the predicted luma point bound | submitted | link | vmafx |
| 0038 | avcodec/vvc/intra_template: bound reference corner extension | submitted | link | vmafx |
| 0039 | avcodec/wmalosslessdec: Clear subframe flags as an array | submitted | link | vmafx |
| 0040 | avfilter/vf_libvmaf: accept more planar pixel formats for CUDA frames | submitted | link | vmafx |
| 0041 | avfilter/vf_libvmaf: copy all planes of CUDA hardware frames | submitted | link | vmafx |
| 0042 | avfilter/vf_libvmaf: pop the CUDA context when a plane copy fails | unsubmitted | none | vmafx |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.6.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.6.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.5.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | rejected | link | vmafx |
| 0003 | avcodec/nvenc: drop hevc_nvenc SEI above a fixed size | merged | link | vmafx, pelorus |
| 0004 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | submitted | link | pelorus |
| 0005 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0006 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | rejected | link | pelorus |
| 0007 | vulkan: de-phase queue rotation and pin video pools to a single queue | merged | link | pelorus |
| 0008 | hwcontext_vulkan: import a frame again when the CUDA context changes | submitted | link | pelorus |
| 0009 | avcodec/qsvdec: discard the right frames on MP4 edit lists | unsubmitted | none | pelorus |
| 0010 | avfilter/dovi_reshape: add a Dolby Vision RPU reshaping helper | unsubmitted | none | vmafx, pelorus |
| 0011 | avfilter: add dovi_reshape filter | unsubmitted | none | vmafx, pelorus |
| 0012 | avfilter/bt2408_hable: add a Hable tone map on IPT intensity | unsubmitted | none | vmafx, pelorus |
| 0013 | avfilter/vf_tonemap: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0014 | avfilter/vf_zscale: keep the light levels when linearising | unsubmitted | none | vmafx, pelorus |
| 0015 | avfilter/vf_tonemap_opencl: declare the output matrix and range on the link | unsubmitted | none | vmafx, pelorus |
| 0016 | avfilter/vf_tonemap_opencl: zero the peak detection buffer | merged | link | vmafx, pelorus |
| 0017 | avfilter/vf_tonemap_opencl: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0018 | avfilter/vf_tonemap_vaapi: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0019 | avfilter/vf_tonemap_vulkan: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0020 | avfilter/vf_tonemap: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0021 | avfilter/vf_tonemap_vulkan: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0022 | avfilter/vf_dovi_reshape: add a profile 8.1 RPU output mode | unsubmitted | none | vmafx, pelorus |
| 0023 | avfilter/vf_hqdn3d_vulkan: add a Vulkan hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0024 | avfilter/vf_hqdn3d_cuda: add a CUDA hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0025 | avfilter/vf_hqdn3d_opencl: add an OpenCL hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0026 | avfilter/vf_hqdn3d_videotoolbox: add a Metal hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0027 | swscale: allow a same-colour YUV repack of an unconvertible colour space | unsubmitted | none | vmafx, pelorus |
| 0028 | avfilter/vf_scale: negotiate IPT-C2 for a same-colour repack | unsubmitted | none | vmafx, pelorus |
| 0029 | avfilter/dovi_reshape: share the HDR10 output properties | unsubmitted | none | vmafx, pelorus |
| 0030 | avfilter/vf_tonemap_opencl: reshape Dolby Vision profile 5 in the kernel | unsubmitted | none | vmafx, pelorus |
| 0031 | avfilter/tonemap_peak: read the Dolby Vision level 1 peak from side data | unsubmitted | none | vmafx, pelorus |
| 0032 | avfilter/vf_tonemap_opencl: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0033 | avfilter/vf_tonemap_vaapi: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0034 | avfilter/vf_tonemap_opencl: detect the bt2408_hable signal peak on the GPU | unsubmitted | none | vmafx, pelorus |
| 0035 | avutil/hwcontext_vulkan: Query export support with the real image flags | merged | link | pelorus |
| 0036 | checkasm/snowdsp: Include intmath before public headers | unsubmitted | none | vmafx |
| 0037 | avcodec/aom_film_grain: Check the predicted luma point count | unsubmitted | none | vmafx |
| 0038 | avcodec/vvc/intra_template: Bound reference corner extension | unsubmitted | none | vmafx |
| 0039 | avcodec/wmalosslessdec: Clear subframe flags as an array | unsubmitted | none | vmafx |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.5.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.5.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.4.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | submitted | link | vmafx |
| 0003 | avcodec/nvenc: drop hevc_nvenc SEI above a fixed size | merged | link | vmafx, pelorus |
| 0004 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | submitted | link | pelorus |
| 0005 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0006 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | submitted | link | pelorus |
| 0007 | vulkan: de-phase queue rotation and pin video pools to a single queue | merged | link | pelorus |
| 0008 | hwcontext_vulkan: import a frame again when the CUDA context changes | submitted | link | pelorus |
| 0009 | avcodec/qsvdec: discard the right frames on MP4 edit lists | unsubmitted | none | pelorus |
| 0010 | avfilter/dovi_reshape: add a Dolby Vision RPU reshaping helper | unsubmitted | none | vmafx, pelorus |
| 0011 | avfilter: add dovi_reshape filter | unsubmitted | none | vmafx, pelorus |
| 0012 | avfilter/bt2408_hable: add a Hable tone map on IPT intensity | unsubmitted | none | vmafx, pelorus |
| 0013 | avfilter/vf_tonemap: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0014 | avfilter/vf_zscale: keep the light levels when linearising | unsubmitted | none | vmafx, pelorus |
| 0015 | avfilter/vf_tonemap_opencl: declare the output matrix and range on the link | unsubmitted | none | vmafx, pelorus |
| 0016 | avfilter/vf_tonemap_opencl: zero the peak detection buffer | merged | link | vmafx, pelorus |
| 0017 | avfilter/vf_tonemap_opencl: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0018 | avfilter/vf_tonemap_vaapi: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0019 | avfilter/vf_tonemap_vulkan: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0020 | avfilter/vf_tonemap: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0021 | avfilter/vf_tonemap_vulkan: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0022 | avfilter/vf_dovi_reshape: add a profile 8.1 RPU output mode | unsubmitted | none | vmafx, pelorus |
| 0023 | avfilter/vf_hqdn3d_vulkan: add a Vulkan hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0024 | avfilter/vf_hqdn3d_cuda: add a CUDA hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0025 | avfilter/vf_hqdn3d_opencl: add an OpenCL hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0026 | avfilter/vf_hqdn3d_videotoolbox: add a Metal hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0027 | swscale: allow a same-colour YUV repack of an unconvertible colour space | unsubmitted | none | vmafx, pelorus |
| 0028 | avfilter/vf_scale: negotiate IPT-C2 for a same-colour repack | unsubmitted | none | vmafx, pelorus |
| 0029 | avfilter/dovi_reshape: share the HDR10 output properties | unsubmitted | none | vmafx, pelorus |
| 0030 | avfilter/vf_tonemap_opencl: reshape Dolby Vision profile 5 in the kernel | unsubmitted | none | vmafx, pelorus |
| 0031 | avfilter/tonemap_peak: read the Dolby Vision level 1 peak from side data | unsubmitted | none | vmafx, pelorus |
| 0032 | avfilter/vf_tonemap_opencl: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0033 | avfilter/vf_tonemap_vaapi: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0034 | avfilter/vf_tonemap_opencl: detect the bt2408_hable signal peak on the GPU | unsubmitted | none | vmafx, pelorus |
| 0035 | avutil/hwcontext_vulkan: Query export support with the real image flags | merged | link | pelorus |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.4.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.4.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.3.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | submitted | link | vmafx |
| 0003 | avcodec/nvenc: keep hevc_nvenc user data SEI within the picture limit | submitted | link | pelorus |
| 0004 | avcodec/nvenc: drop user data SEI that NVENC would write truncated | submitted | link | vmafx, pelorus |
| 0005 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | submitted | link | pelorus |
| 0006 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0007 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | submitted | link | pelorus |
| 0008 | vulkan: de-phase queue rotation and pin video pools to a single queue | merged | link | pelorus |
| 0009 | hwcontext_vulkan: import a frame again when the CUDA context changes | submitted | link | pelorus |
| 0010 | avcodec/qsvdec: discard the right frames on MP4 edit lists | unsubmitted | none | pelorus |
| 0011 | avfilter/dovi_reshape: add a Dolby Vision RPU reshaping helper | unsubmitted | none | vmafx, pelorus |
| 0012 | avfilter: add dovi_reshape filter | unsubmitted | none | vmafx, pelorus |
| 0013 | avfilter/bt2408_hable: add a Hable tone map on IPT intensity | unsubmitted | none | vmafx, pelorus |
| 0014 | avfilter/vf_tonemap: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0015 | avfilter/vf_zscale: keep the light levels when linearising | unsubmitted | none | vmafx, pelorus |
| 0016 | avfilter/vf_tonemap_opencl: declare the output matrix and range on the link | unsubmitted | none | vmafx, pelorus |
| 0017 | avfilter/vf_tonemap_opencl: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0018 | avfilter/vf_tonemap_vaapi: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0019 | avfilter/vf_tonemap_vulkan: add the bt2408_hable tone map | unsubmitted | none | vmafx, pelorus |
| 0020 | avfilter/vf_tonemap: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0021 | avfilter/vf_tonemap_vulkan: add dynamic peak detection | unsubmitted | none | vmafx, pelorus |
| 0022 | avfilter/vf_dovi_reshape: add a profile 8.1 RPU output mode | unsubmitted | none | vmafx, pelorus |
| 0023 | avfilter/vf_hqdn3d_vulkan: add a Vulkan hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0024 | avfilter/vf_hqdn3d_cuda: add a CUDA hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0025 | avfilter/vf_hqdn3d_opencl: add an OpenCL hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0026 | avfilter/vf_hqdn3d_videotoolbox: add a Metal hqdn3d filter | unsubmitted | none | vmafx, pelorus |
| 0027 | avcodec/nvenc: move the SEI size rules into nvenc_sei.h | unsubmitted | none | pelorus |
| 0028 | swscale: allow a same-colour YUV repack of an unconvertible colour space | unsubmitted | none | vmafx, pelorus |
| 0029 | avfilter/vf_scale: negotiate IPT-C2 for a same-colour repack | unsubmitted | none | vmafx, pelorus |
| 0030 | avfilter/dovi_reshape: share the HDR10 output properties | unsubmitted | none | vmafx, pelorus |
| 0031 | avfilter/vf_tonemap_opencl: reshape Dolby Vision profile 5 in the kernel | unsubmitted | none | vmafx, pelorus |
| 0032 | avfilter/tonemap_peak: read the Dolby Vision level 1 peak from side data | unsubmitted | none | vmafx, pelorus |
| 0033 | avfilter/vf_tonemap_opencl: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
| 0034 | avfilter/vf_tonemap_vaapi: take the bt2408_hable peak from Dolby Vision level 1 | unsubmitted | none | vmafx, pelorus |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.3.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.3.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.2.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | unsubmitted | none | vmafx |
| 0003 | avcodec/nvenc: keep hevc_nvenc user data SEI within the picture limit | submitted | link | pelorus |
| 0004 | avcodec/nvenc: drop user data SEI that NVENC would write truncated | submitted | link | vmafx, pelorus |
| 0005 | hwcontext_vaapi: accept GR1616 chroma for P010 and P012 DRM import | unsubmitted | none | pelorus |
| 0006 | hwcontext_vulkan: pass the image format list to the export query | unsubmitted | none | pelorus |
| 0007 | hwcontext_vulkan: wait on the CPU before mapping RADV frames to DRM | unsubmitted | none | pelorus |
| 0008 | vulkan: de-phase queue rotation and pin video pools to a single queue | submitted | link | pelorus |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.2.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.2.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.1.0
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | unsubmitted | none | vmafx |
| 0003 | avcodec/nvenc: keep hevc_nvenc user data SEI within the picture limit | unsubmitted | link | pelorus |
| 0004 | avcodec/nvenc: drop user data SEI that NVENC would write truncated | unsubmitted | none | vmafx, pelorus |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.1.0' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.1.0.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.
ffmpeg-patches v0.1.0-rc.1
v0.1.0-rc.1
lusoris
Kilian
This tag was signed with the committer’s verified signature.
lusoris
Kilian
FFmpeg base: n9.0.2 (946fcce07b6dcd0331c8cc609192aeff5e1924f8).
| # | Patch | Status | Upstream | Needed by |
|---|---|---|---|---|
| 0001 | avutil/vulkan: don't release frames to VK_QUEUE_FAMILY_IGNORED | submitted | link | pelorus |
| 0002 | build: eliminate GCC 14 and 16 diagnostics | unsubmitted | none | vmafx |
| 0003 | avcodec/nvenc: keep hevc_nvenc user data SEI within the picture limit | unsubmitted | link | pelorus |
| 0004 | avcodec/nvenc: drop user data SEI that NVENC would write truncated | unsubmitted | none | vmafx, pelorus |
Verify the release files:
cosign verify-blob --bundle SHA256SUMS.sigstore.json \
--certificate-identity 'https://github.com/VMAFx/ffmpeg-patches/.github/workflows/release-build.yml@refs/tags/v0.1.0-rc.1' \
--certificate-oidc-issuer https://token.actions.githubusercontent.com SHA256SUMS
sha256sum --check SHA256SUMS
gh attestation verify ffmpeg-patches-v0.1.0-rc.1.tar.gz --repo VMAFx/ffmpeg-patchesApply the patches in series.txt order with git am --3way onto the base, or run scripts/series.py apply --expect-base <ffmpeg-tree> from the tarball.