Closed Bug 466867 Opened 16 years ago Closed 16 years ago

Linux Txul regression on November 13/14 looks like temp table changes?

Categories

(Toolkit :: Places, defect)

x86
Linux
defect
Not set
normal

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: johnath, Unassigned)

Details

So there's this linux perf regression on Txul around the evening of the 13th or morning of the 14th:

http://graphs.mozilla.org/#show=395002,395014,395042,395160&sel=1226613209,1226748073

And that gives us this window (ish):

http://hg.mozilla.org/mozilla-central/pushloghtml?startdate=november+13%2C+2008+6%3A00pm&enddate=november+14%2C+2008%2C+6%3A00am

There isn't much at all there that really leaps out, but the temp tables changes are one of the bigger diffs, and one of the few that aren't just tests or landings+backouts (neither of shouldn't cause any long standing problem.)

Basically, I would be happy to be told that this was filed in the wrong component, but I can't think of a better one, right now.  :)
I don't see how that changeset could the problem.  It just does some work on xpcom-shutdown (which we were already a listener for anyway).  Txul is about drawing a new window, right?
So, this has actually dropped back down below pre-regression levels.  Johnathan - possibly randomness?

http://graphs.mozilla.org/#show=395002,395014,395042,395160&sel=1226641988,1227976593
Flags: blocking1.9.1?
I don't really believe randomness, since all three boxes went up at once - but I also don't have a good strategy for going forward.
Status: NEW → RESOLVED
Closed: 16 years ago
Resolution: --- → WORKSFORME
Flags: blocking1.9.1?
You need to log in before you can comment on or make changes to this bug.