media/gpu/v4l2vd: move stateless-specific parts into backend class
There are two V4L2 decode APIs: stateless and stateful. In the VDA era these were implemented as two completely separate classes (V4L2VideoDecodeAccelerator and V4L2SliceVideoDecodeAccelerator), which resulted in a lot of duplicated code and bugs due to fixes not being replicated on both classes. For the VD-based decoder, we want to avoid doing the same error and use a single V4L2 decoder, supported by different backends depending on which V4L2 API we are using. This CL is the first step towards that direction: it splits the V4L2SliceVideoDecoder into its general V4L2 part and the one that is specific to the stateless API. Bug: 1003223 Test: video_decode_accelerator_tests passing on Kevin. Change-Id: I373f3ea739b6799ecad009d053ca7e415b8484bd Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1840034 Commit-Queue: Alexandre Courbot <acourbot@chromium.org> Reviewed-by:Hirokazu Honda <hiroh@chromium.org> Reviewed-by:
David Staessens <dstaessens@chromium.org> Reviewed-by:
Chih-Yu Huang <akahuang@chromium.org> Cr-Commit-Position: refs/heads/master@{#705894}
Showing
This diff is collapsed.
This diff is collapsed.
Please register or sign in to comment