Bug 295538
| Summary: | REGRESSION(287399@main): [Skia] Broke rendering of various PDFs | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Michael Catanzaro <mcatanzaro> |
| Component: | WebKitGTK | Assignee: | Nikolas Zimmermann <zimmermann> |
| Status: | RESOLVED FIXED | ||
| Severity: | Normal | CC: | bugs-noreply, mcatanzaro, plampe, webkit-bug-importer, zimmermann |
| Priority: | P2 | ||
| Version: | WebKit Nightly Build | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| See Also: | https://bugs.webkit.org/show_bug.cgi?id=283704 | ||
Michael Catanzaro
Load https://www.transitchicago.com/assets/1/6/Slow_Zone_Map_-_June_2025.pdf and scroll down to the second page. Notice the rendering is garbled and broken.
I finally bisected it; sorry it took so long. This is a regression from 287399@main "[GTK][WPE][Skia] Allow cross-thread transfer of accelerated ImageBitmap".
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Michael Catanzaro
We should probably revert 287399@main.
Michael Catanzaro
A revert actually does fix this bug, so let's do that for now.
Pawel Lampe
Yes, a revert is very good idea especially that the commit in question is just a performance improvement.
Unless already reverted, I'll revert tomorrow.
Michael Catanzaro
I prepared a revert in https://github.com/WebKit/WebKit/pull/55184 but I see I failed to link it to this bug report, oops.
Nikolas Zimmermann
I got a fix for this, will need to clean it up tomorrow and will submit a PR.
Nikolas Zimmermann
Pull request: https://github.com/WebKit/WebKit/pull/55405
Nikolas Zimmermann
Michael, I'd be glad if you could confirm that this fully fixes the regression for you as well -- the PR is still lacking a test, I'm thinking about a proper way to capture the problem in a layout test..
EWS
Committed 304687@main (48db7480703c): <https://commits.webkit.org/304687@main>
Reviewed commits have been landed. Closing PR #55405 and removing active labels.
Michael Catanzaro
Confirmed fixed, thank you!