diff options
| author | danakj@chromium.org <danakj@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2013-09-14 05:33:29 +0000 |
|---|---|---|
| committer | danakj@chromium.org <danakj@chromium.org@0039d316-1c4b-4281-b951-d872f2087c98> | 2013-09-14 05:33:29 +0000 |
| commit | ed66d1c729afdad61eb7a37e6b635a38ca5c0d94 (patch) | |
| tree | 4ea269e8595742b5f5f12f5ab60eefad69fe6c8d /extensions | |
| parent | 8fe650c2437a58ebdbf373a80a93582ff3e5ac48 (diff) | |
| download | chromium_src-ed66d1c729afdad61eb7a37e6b635a38ca5c0d94.zip chromium_src-ed66d1c729afdad61eb7a37e6b635a38ca5c0d94.tar.gz chromium_src-ed66d1c729afdad61eb7a37e6b635a38ca5c0d94.tar.bz2 | |
aura: When skipping frames, ensure next frame damages what was skipped.
In the delegated path, if we skip a frame, we are just dropping the
damage. If the resize lock times out, we can end up swapping a partial
frame that is out of sync with the rest of the visible content.
Since we always have all the quads available to us (unlike the old
double-buffered texture path), when we stop skipping a frame, just
damage the whole RenderWidgetHostView to ensure we draw the skipped
frames' damage.
Refactors ResizeLock out so that we can fake it out in tests.
Tests:
RenderWidgetHostViewAuraTest.SkippedDelegatedFrames
R=jbauman, piman
BUG=288993
Depends on: https://codereview.chromium.org/23506032/
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=223140
Review URL: https://chromiumcodereview.appspot.com/23496042
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@223228 0039d316-1c4b-4281-b951-d872f2087c98
Diffstat (limited to 'extensions')
0 files changed, 0 insertions, 0 deletions