RADV Vulkan Video Lands Intra-Refresh Encode
David Rosca who started out as a contractor for AMD working on their open-source video encode/decode capabilities for Linux formally joined the company earlier this summer. He's continued to be quite busy plumbing numerous enhancements into their accelerated video support on Linux, which includes RADV with Vulkan Video.
The latest milestone for RADV Vulkan Video to celebrate is enabling the VK_KHR_video_encode_intra_refresh extension for allowing applications to perform intra-refresh in video encode operations for better error resilience and bitrate efficiency. This intra-refresh encode extension premiered in Vulkan 1.4.321 back in July.
RADV Vulkan Video intra-refresh encode was merged for Mesa 25.3 yesterday via this merge request.
David Rosca has also been landing other Mesa video improvements for the Radeon stack too with it being particularly heavy in recent days. Another notable merge is adding decode/encode processing for the RadeonSI Gallium3D VA bits:
It's great seeing a clear recent uptick in the activity around the video acceleration support for AMD Radeon graphics within Mesa. Let's hope it continues with similar speed.
The latest milestone for RADV Vulkan Video to celebrate is enabling the VK_KHR_video_encode_intra_refresh extension for allowing applications to perform intra-refresh in video encode operations for better error resilience and bitrate efficiency. This intra-refresh encode extension premiered in Vulkan 1.4.321 back in July.
RADV Vulkan Video intra-refresh encode was merged for Mesa 25.3 yesterday via this merge request.
David Rosca has also been landing other Mesa video improvements for the Radeon stack too with it being particularly heavy in recent days. Another notable merge is adding decode/encode processing for the RadeonSI Gallium3D VA bits:
"This implements support for Decode processing allowing to perform processing operation on the decoded picture in one single call without having to use separate processing context.
This also implements the same functionality for encoding, which is useful to perform conversion from RGB to YUV in a single call, and it allows us to properly support the conversion inside encoder (eg. EFC on AMD).
For Encode processing the additional output buffer is required same as with Decode processing, but driver may not use it to perform the conversion (in case where the conversion can be done by the encoder hw). This means the contents of the additional buffer is undefined, and application should not rely on the buffer actually containing output picture of the conversion."
It's great seeing a clear recent uptick in the activity around the video acceleration support for AMD Radeon graphics within Mesa. Let's hope it continues with similar speed.