Closed Bug 1141247 Opened 10 years ago Closed 10 years ago

10% WinXP cart regression on inbound (v.39) on March 05, 2015 from push 2bd95d9c12bb

Categories

(Core :: Layout, defect)

Product:
Core
Component:
Layout
Platform:
x86
Windows XP
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED WONTFIX

People

(Reporter: jmaher, Assigned: mstange)

References

Details

(Keywords: perf, regression, Whiteboard: [talos_regression])

Talos has detected a Firefox performance regression from your commit 2bd95d9c12bb in bug 913443. We need you to address this regression. This is a list of all known regressions and improvements related to your bug: http://alertmanager.allizom.org:8080/alerts.html?rev=2bd95d9c12bb&showAll=1 On the page above you can see Talos alert for each affected platform as well as a link to a graph showing the history of scores for this test. There is also a link to a treeherder page showing the Talos jobs in a pushlog format. To learn more about the regressing test, please see: https://wiki.mozilla.org/Buildbot/Talos/Tests#TART.2FCART Reproducing and debugging the regression: If you would like to re-run this Talos test on a potential fix, use try with the following syntax: try: -b o -p win32 -u none -t svgr # add "mozharness: --spsProfile" to generate profile data To run the test locally and do a more in-depth investigation, first set up a local Talos environment: https://wiki.mozilla.org/Buildbot/Talos/Running#Running_locally_-_Source_Code Then run the following command from the directory where you set up Talos: talos --develop -e <path>/firefox -a cart Making a decision: As the patch author we need your feedback to help us handle this regression. *** Please let us know your plans by Thursday, or the offending patch will be backed out! *** Our wiki page oulines the common responses and expectations: https://wiki.mozilla.org/Buildbot/Talos/RegressionBugsHandling
:mstange, can you look at this?
Flags: needinfo?(mstange)
Sure. But wow, 10%! I hadn't expected that. And it's in CART on Windows, the platform+test combination where profiling is working least reliably. :(
Assignee: nobody → mstange
Status: NEW → ASSIGNED
Flags: needinfo?(mstange)
thanks Markus! if there is anything I can do to help out, just ping me.
This time based on a more recent inbound revision that should include a profiler crash fix: https://treeherder.mozilla.org/#/jobs?repo=try&revision=0fe569437434 https://treeherder.mozilla.org/#/jobs?repo=try&revision=615d71948bf8
Unfortunately, profiling seems to have a lot of overhead on this test. It increases the numbers from ~35/~38 to ~58, and there's no real difference between builds with my patch and without it. I think I've identified two changes that should slightly improve performance, and I've pushed them to try here: https://treeherder.mozilla.org/#/jobs?repo=try&revision=7ee19959aefb Let's see whether the numbers improve.
Okay, so that didn't have any impact on the numbers. I'm about to admit defeat on this one.
Any luck on this? possibly mark this as wontfix?
Flags: needinfo?(mstange)
No luck. Yesterday I thought I had found another perf problem with bug 913443, but it turned out to be something else and couldn't have affected cart (see bug 1147083). Not being able to profile without making the regression go away makes this really tricky to track down. At this point I can just hope that I run across a similar perf problem somewhere else where it's easier to debug. I'm not going to spend more time on looking into CART specifically, so I'll mark this bug as wontfix.
Status: ASSIGNED → RESOLVED
Closed: 10 years ago
Flags: needinfo?(mstange)
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.