Open Bug 1671082 Opened 4 years ago Updated 2 years ago

Minidump missing the crashing thread's stack

Categories

(Toolkit :: Crash Reporting, defect)

Unspecified
Linux
defect

Tracking

()

People

(Reporter: gsvelto, Unassigned)

Details

In bug 1670816 we observed a minidump with the crashing thread's stack being completely empty, in fact it appears to be missing entirely from the minidump, only the processor context was recorded. The bug has an STR so we should be able to reproduce the issue by reverting the patch and following the steps that lead to the crash.

Severity: -- → S3

We've seen more crashes with an issue like bug 1670816 after landing bug 1678152. The reason why the stacks were empty is that the stack pointer was pointing in the stack's guard page at the time of crash. The minidump writer incorrectly identified the guard page as the stack and dumped it into the minidump. I don't have access to the minidump of bug 1670816 anymore, but given the similarity in behavior I wouldn't be surprised if it was the same issue.

You need to log in before you can comment on or make changes to this bug.