Bug 58657
| Summary: | REGRESSION (r83928 or before): Some tests failing assertions in MarkStack::internalAppend / MarkStack::drain | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Adam Roben (:aroben) <aroben> |
| Component: | JavaScriptCore | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED CONFIGURATION CHANGED | ||
| Severity: | Normal | CC: | beidson, bfulgham, ggaren, gyuyoung.kim, jberlin, oliver |
| Priority: | P2 | Keywords: | InRadar, LayoutTestFailure, MakingBotsRed, PlatformOnly |
| Version: | 528+ (Nightly build) | ||
| Hardware: | PC | ||
| OS: | Windows XP | ||
Adam Roben (:aroben)
The XP bots have seen assertion failures in MarkStack::internalAppend twice recently. The first time was in r83928:
http://build.webkit.org/results/Windows%20XP%20Debug%20%28Tests%29/r83928%20(27641)/dom/html/level1/core/hc_attrinsertbefore6-crash-log.txt
And then it happened again in r83943:
http://build.webkit.org/results/Windows%20XP%20Debug%20%28Tests%29/r83943%20(27650)/jquery/traversing-crash-log.txt
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Adam Roben (:aroben)
<rdar://problem/9290323>
Adam Roben (:aroben)
I assume that these assertion failures are leading to crashes in MarkStack::drain Release builds:
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84123%20(11782)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84120%20(11780)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84118%20(11778)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84117%20(11777)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84111%20(11772)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84107%20(11770)/inspector/extensions/extensions-resources-crash-log.txt
http://build.webkit.org/results/Windows%207%20Release%20%28Tests%29/r84098%20(11764)/inspector/extensions/extensions-resources-crash-log.txt
Adam Roben (:aroben)
Looks like ietestcenter/Javascript/10.4.2-1-1.html has been hitting this assertion on every single test run since r84264.
http://build.webkit.org/builders/Windows%20XP%20Debug%20%28Tests%29/builds/27814
Brady Eidson
Saw this in my local build of r84631 in Safari.
Jessie Berlin
A couple more instances on XP, one under markChildren under drain and one under internalAppend:
http://build.webkit.org/results/Windows%20XP%20Debug%20(Tests)/r85407%20(28237)/inspector/styles/styles-cancel-editing-crash-log.txt
http://build.webkit.org/results/Windows%20XP%20Debug%20(Tests)/r85411%20(28241)/editing/execCommand/insertImage-crash-log.txt
Jessie Berlin
I am seeing crashes on the Windows XP bots again:
http://build.webkit.org/results/Windows%20XP%20Debug%20(Tests)/r85475%20(28275)/results.html
http://build.webkit.org/results/Windows%20XP%20Debug%20(Tests)/r85472%20(28274)/results.html
http://build.webkit.org/results/Windows%20XP%20Debug%20(Tests)/r85471%20(28273)/results.html
I am going to skip the inspector/styles/styles-cancel-editing.html on Windows XP only, since this does not seem to be affecting Windows 7 (or at least the Win 7 bots appear unaffected).
Geoffrey Garen
Looks like ietestcenter/Javascript/10.4.2-1-1.html stopped hitting this assertion after r84584.
http://build.webkit.org/builders/Windows%20XP%20Debug%20%28Tests%29/builds/27929