Summary: | [GStreamer] ImageGStreamer doesn't need to hold a Cairo surface | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Philippe Normand <pnormand> | ||||
Component: | Media | Assignee: | Nobody <webkit-unassigned> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | alex | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | PC | ||||||
OS: | OS X 10.5 | ||||||
Attachments: |
|
Description
Philippe Normand
2010-08-27 02:24:54 PDT
Created attachment 65694 [details]
proposed patch
It happened again with media/video-seek-past-end-paused.html on the 30th of august: http://webkit-bots.igalia.com/amd64/svn_66395.core-when_1283190163-_-who_DumpRenderTree-_-why_6.11551.trace.html Comment on attachment 65694 [details]
proposed patch
I don't quite see in the code who will be handling the lifecycle of the surface. Can you point me to it?
Comment on attachment 65694 [details]
proposed patch
OK, I think it's in ImageCairo.cpp, the BitmapImage constructor steals the surface (I suppose this was desired, seems a bit obscure).
Thanks, see http://trac.webkit.org/changeset/66600 \m/ \m/ |