Summary: | [chromium] Prevent deadlock on CCVideoLayerImpl destruction | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Adrienne Walker <enne> | ||||
Component: | New Bugs | Assignee: | Adrienne Walker <enne> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | cc-bugs, enne, eric.carlson, feature-media-reviews, fischman, jamesr, nduca, scherkus, webkit.review.bot | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Attachments: |
|
Description
Adrienne Walker
2012-05-11 15:23:46 PDT
Created attachment 141510 [details]
Patch
jamesr: Your suggestion to use CCLayerTreeHost::acquireLayerTextures to block the compositor thread from using a destroyed provider worked great in the case where you had another commit and much less great when the tree got destroyed and it was unclear whether a client should clear its provider pointer or not. Comment on attachment 141510 [details]
Patch
R=me
WIN I also sense great win in this patch. Comment on attachment 141510 [details] Patch Clearing flags on attachment: 141510 Committed r116833: <http://trac.webkit.org/changeset/116833> All reviewed patches have been landed. Closing bug. |