| Summary: | Media control volume slider is opaque | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Roger Fong <roger_fong> | ||||
| Component: | Media | Assignee: | Nobody <webkit-unassigned> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | commit-queue, dino, roger_fong, simon.fraser | ||||
| Priority: | P2 | ||||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Bug Depends on: | 144505 | ||||||
| Bug Blocks: | |||||||
| Attachments: |
|
||||||
|
Description
Roger Fong
2015-04-30 14:35:04 PDT
Created attachment 252091 [details]
patch
Seeing as all the mix-blend-modes create new compositing regions on the controls anyways, the original z-index: 0 is now just hurting, not helping.
The fact that the volume slider blend doesn't work while all the other blends do probably has something to do with the fact that it's rotated off the main control bar.
Simon gave me the verbal go ahead over irc. http://trac.webkit.org/changeset/183687 It looks like this broke tests: <https://build.webkit.org/results/Apple%20Mavericks%20Debug%20WK1%20(Tests)/r183688%20(12422)/results.html>. Roger is not on IRC, rolling out. Re-opened since this is blocked by bug 144505 By removing a CSS property? That should never be the case. Sounds like a bug in the compositing code. The assertions went away after the rollout, so yes. Re-landed with test expectations: http://trac.webkit.org/changeset/183820 Bug to track failures: https://bugs.webkit.org/show_bug.cgi?id=144630 Is this really the right thing to do? We lose test coverage for all affected tests, not to mention that CrashReporter will make other tests more flaky. Also, we still assert on fast/regions/inline-block-inside-anonymous-overflow.html and compositing/regions/transform-transparent-positioned-video-inside-region.html |