Bug 82262 - [chromium] Replicas should be included in the computed occlusion
Summary: [chromium] Replicas should be included in the computed occlusion
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Dana Jansens
URL:
Keywords:
Depends on: 81227
Blocks:
  Show dependency treegraph
 
Reported: 2012-03-26 16:57 PDT by Dana Jansens
Modified: 2012-04-13 18:58 PDT (History)
7 users (show)

See Also:


Attachments
Patch (7.64 KB, patch)
2012-03-26 20:10 PDT, Dana Jansens
no flags Details | Formatted Diff | Diff
Patch (9.32 KB, patch)
2012-04-09 19:28 PDT, Dana Jansens
no flags Details | Formatted Diff | Diff
Patch (13.73 KB, patch)
2012-04-09 20:29 PDT, Dana Jansens
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Dana Jansens 2012-03-26 16:57:17 PDT
[chromium] Replicas should be included in the computed occlusion
Comment 1 Dana Jansens 2012-03-26 20:10:28 PDT
Created attachment 133966 [details]
Patch
Comment 2 Adrienne Walker 2012-03-27 10:45:25 PDT
Comment on attachment 133966 [details]
Patch

Looks good.  Thanks as always for the tests.  Can you put up a rebaselined patch?
Comment 3 Dana Jansens 2012-03-27 10:52:31 PDT
Thanks, yup! Once bug #81227 is in I'll rebase this and throw it at EWS.
Comment 4 Dana Jansens 2012-04-09 19:28:59 PDT
Created attachment 136376 [details]
Patch

Trashing your R+ cuz I am adding hasReplica() to RenderSurfaceChromium, and cuz you had asked for a rebased patch and I hadn't done that yet.

The diff here is just adding RSChromium::hasReplica() for the templated functions in CCOcclusionTracker
Comment 5 Dana Jansens 2012-04-09 19:32:58 PDT
Comment on attachment 136376 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=136376&action=review

> Source/WebCore/platform/graphics/chromium/cc/CCOcclusionTracker.cpp:129
> +        transformedRegion.unite(intersection(surface->clipRect(), enclosedIntRect(transform.mapRect(FloatRect(rects[i])))));

Oh this is wrong, need to check clipRect().isEmpty() first for a surface.
Comment 6 Dana Jansens 2012-04-09 20:29:49 PDT
Created attachment 136386 [details]
Patch
Comment 7 Adrienne Walker 2012-04-13 18:06:04 PDT
Comment on attachment 136386 [details]
Patch

Looks good to me.
Comment 8 WebKit Review Bot 2012-04-13 18:58:40 PDT
Comment on attachment 136386 [details]
Patch

Clearing flags on attachment: 136386

Committed r114197: <http://trac.webkit.org/changeset/114197>
Comment 9 WebKit Review Bot 2012-04-13 18:58:44 PDT
All reviewed patches have been landed.  Closing bug.