| Summary: | REGRESSION: Multiple load events dispatched to <body> (makes layout tests hang) | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | mitz | ||||
| Component: | New Bugs | Assignee: | Maciej Stachowiak <mjs> | ||||
| Status: | VERIFIED FIXED | ||||||
| Severity: | Blocker | CC: | darin, matt, mjs | ||||
| Priority: | P1 | Keywords: | HasReduction, Regression | ||||
| Version: | 420+ | ||||||
| Hardware: | Mac | ||||||
| OS: | OS X 10.4 | ||||||
| URL: | http://www.haaretz.co.il/?from=hasot | ||||||
| Attachments: |
|
||||||
|
Description
mitz
2006-02-05 06:55:13 PST
That would be very new - I don't see this problem with r12570. Possibly a regression from a fix in bug 7079 - it's known to hang on at least one regression test. (In reply to comment #1) > That would be very new - I don't see this problem with r12570. Possibly a > regression from a fix in bug 7079 - it's known to hang on at least one > regression test. > This could very well be it: it looks like the root cause for this bug is that <body>'s onload handler is being called multiple times. Created attachment 6273 [details]
Testcase
This demonstrates that the onload handler is being called twice.
Maciej was looking at this last night -- it's from his last check-in with "event clean-up". We'll probably need to roll that out. Alexey believes Maciej had a fix ready when he went to sleep last night. Rolling out the change because it's preventing us from running layout tests. Maciej will re-land it soon. |