If you think a bug might affect users in the 57 release, please set the correct tracking and status flags for Release Management.

crash in DMD child process due to timeout

NEW
Unassigned

Status

()

Core
DMD
a year ago
a year ago

People

(Reporter: mccr8, Unassigned)

Tracking

Trunk
Points:
---

Firefox Tracking Flags

(firefox51 affected)

Details

(Reporter)

Description

a year ago
baku is looking at bug 1249739, which involves a huge amount of heap unclassified, so I told him to try DMD. This was in a non-optimized debug build. Getting the log took long enough that we hit some IPC error. It would be nice if we could make IPC timeouts or whatever be more lenient in DMD builds. I'm not sure how that would work. (We worked around it by just running in non-e10s.)
jseward just worked around the exact same problem with TSan in bug 1299384. It was a one-line fix there.

Also, in general I'd recommend using opt builds with DMD... that's certainly what I do :)
> jseward just worked around the exact same problem with TSan in bug 1299384.

Sorry, that should be bug 1295911.
(Reporter)

Comment 3

a year ago
(In reply to Nicholas Nethercote [:njn] from comment #1)
> Also, in general I'd recommend using opt builds with DMD... that's certainly
> what I do :)

Yeah, baku just had his mozconfig set up to do non-opt, and I didn't manage to suggest he should change that for DMD. This is a case where we're using gigs of memory so it'll be a little slow either way, though hopefully not too bad.
You need to log in before you can comment on or make changes to this bug.