diff options
| author | halliwell <halliwell@chromium.org> | 2015-05-06 12:44:12 -0700 |
|---|---|---|
| committer | Commit bot <commit-bot@chromium.org> | 2015-05-06 19:44:43 +0000 |
| commit | 281b59f8d52b765ed94003e9d9a04fe34b36b8b7 (patch) | |
| tree | d03fd1d7558e3155f304464f1e2c5ff3da14743a /ipc/ipc_message_attachment_set.h | |
| parent | 7f772e4a852a1935068823472bb382904c6edacf (diff) | |
| download | chromium_src-281b59f8d52b765ed94003e9d9a04fe34b36b8b7.zip chromium_src-281b59f8d52b765ed94003e9d9a04fe34b36b8b7.tar.gz chromium_src-281b59f8d52b765ed94003e9d9a04fe34b36b8b7.tar.bz2 | |
Fixes sync point processing for overlays
When I tried to use overlays, I found:
1. DirectRenderer::DrawFrame calls ProcessForOverlays, removing my video quad from the quad_list of its render pass
2. GLRenderer::BeginDrawingFrame calls WaitSyncPointIfNeeded for quads in the render passes, so it omits the video quad.
3. GLRenderer::FinishDrawingFrame calls ScheduleOverlays, which creates a ScopedReadLockGL on the overlay resource, which asserts in ResourceProvider::LockForRead that the sync point has not been processed.
BUG=
Review URL: https://codereview.chromium.org/1066483004
Cr-Commit-Position: refs/heads/master@{#328586}
Diffstat (limited to 'ipc/ipc_message_attachment_set.h')
0 files changed, 0 insertions, 0 deletions