Comments 60 and onwards from bug 990230 appear to be unrelated to that bug, and instead appear to be webgl related. So I'm opening a new bug and I'll change the description of bug 990230 so that it doesn't match.
Given the timing and the fact that this is dying in webgl tests, I'm going to call this another regression from bug 980178.
I've backed out bug 980178, so we should see this stop soon if it was indeed at fault.
Fixed by backout.
Filed bug 1019934 for the 408431-1.html as they're clearly not the same as this bug.
Tweaking summary to stop false positives.
Summary: Intermittent ASAN "SUMMARY: AddressSanitizer: heap-use-after-free ??:0 ??" errors with unusable stacks → Intermittent ASAN failures in webgl tests caused by bug 980178
You need to log in before you can comment on or make changes to this bug.