Bug 75589

Summary: [Chromium] Do not recreate texture updater for image layer if one exists.
Product: WebKit Reporter: Daniel Sievers <sievers>
Component: New BugsAssignee: Daniel Sievers <sievers>
Status: RESOLVED FIXED    
Severity: Normal CC: cc-bugs, dglazkov, enne, jamesr, shawnsingh, vangelis, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Patch
none
Patch
none
Patch
none
Patch none

Daniel Sievers
Reported 2012-01-04 17:21:08 PST
[Chromium] Do not recreate texture updater for image layer if one exists.
Attachments
Patch (1.66 KB, patch)
2012-01-04 17:22 PST, Daniel Sievers 		no flags
DetailsFormatted DiffDiff
Patch (27.15 KB, patch)
2012-01-06 15:01 PST, Daniel Sievers 		no flags
DetailsFormatted DiffDiff
Patch (35.04 KB, patch)
2012-01-06 16:56 PST, Daniel Sievers 		no flags
DetailsFormatted DiffDiff
Patch (26.32 KB, patch)
2012-01-09 13:43 PST, Daniel Sievers 		no flags
DetailsFormatted DiffDiff
Show Obsolete (3) View All Add attachment proposed patch, testcase, etc.
Daniel Sievers
Comment 1 2012-01-04 17:22:04 PST
Created attachment 121190 [details] Patch
James Robinson
Comment 2 2012-01-04 17:23:20 PST
Comment on attachment 121190 [details] Patch Any way to test this? Perhaps a layout test that shows the image corruption?
Daniel Sievers
Comment 3 2012-01-06 15:01:30 PST
Created attachment 121502 [details] Patch
Daniel Sievers
Comment 4 2012-01-06 15:05:09 PST
I had no luck with reparenting through JavaScript, because it ended up creating a new ImageLayer in that case, which doesn't show the bug. See attached patch, growing the layer (which inserts the scrollbar layer) however shows the problem. The initial layoutController.display() before changing the layer size seems to be necessary also to make it fail. Not sure if that is the right directory for the resource .jpg. I copied it from the top-level compositing/resources dir, although I saw another test reference it from the platform dir relatively, but when I ran new-run-webkit-tests that wouldn't work to find it. (In reply to comment #2) > (From update of attachment 121190 [details]) > Any way to test this? Perhaps a layout test that shows the image corruption?
WebKit Review Bot
Comment 5 2012-01-06 16:11:37 PST
Comment on attachment 121502 [details] Patch Attachment 121502 [details] did not pass chromium-ews (chromium-xvfb): Output: http://queues.webkit.org/results/11170117 New failing tests: platform/chromium/compositing/img-layer-grow.html
Daniel Sievers
Comment 6 2012-01-06 16:56:58 PST
Created attachment 121523 [details] Patch
James Robinson
Comment 7 2012-01-06 17:57:14 PST
You should not need to copy any resources, a relative include should work fine. Are you sure you got the path right? Please use mock scrollbars or css-style them if they are going to be in the visible test output. add a layoutTestController.dumpAsText(true) to suppress the normal render tree output, it's not needed for this test.
Daniel Sievers
Comment 8 2012-01-09 13:43:02 PST
Created attachment 121721 [details] Patch
James Robinson
Comment 9 2012-01-09 15:00:22 PST
Comment on attachment 121721 [details] Patch Nice, R=me
Daniel Sievers
Comment 10 2012-01-10 11:33:17 PST
James, can you cq+? Thanks! (In reply to comment #9) > (From update of attachment 121721 [details]) > Nice, R=me
James Robinson
Comment 11 2012-01-10 11:35:01 PST
Comment on attachment 121721 [details] Patch Sure. In the future you can request this by setting the commit-queue flag to "?" or by running webkit-patch upload --request-commit
WebKit Review Bot
Comment 12 2012-01-10 11:52:53 PST
Comment on attachment 121721 [details] Patch Clearing flags on attachment: 121721 Committed r104610: <http://trac.webkit.org/changeset/104610>
WebKit Review Bot
Comment 13 2012-01-10 11:52:58 PST
All reviewed patches have been landed. Closing bug.
Shawn Singh
Comment 14 2012-01-12 10:53:18 PST
*** Bug 76185 has been marked as a duplicate of this bug. ***
Note You need to log in before you can comment on or make changes to this bug.