Closed
Bug 801336
Opened 13 years ago
Closed 13 years ago
PGO error? crash in ToBooleanSlow
Categories
(Core :: JavaScript Engine, defect)
Tracking
()
RESOLVED
DUPLICATE
of bug 791214
Tracking | Status | |
---|---|---|
firefox19 | - | --- |
People
(Reporter: alice0775, Unassigned)
References
()
Details
(Keywords: crash, regression, reproducible, Whiteboard: [js:p1])
Crash Data
This bug was filed from the Socorro interface and is
report bp-61d3c319-fda8-434a-acb0-4b0362121013 .
=============================================================
http://hg.mozilla.org/mozilla-central/rev/90857937b601
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121013030542
I noticed the crash when I test Bug 801267.
Steps to reproduce:
1. Open http://paperjs.org/examples/meta-balls/
2. Move mouse in the right pane
Updated•13 years ago
|
OS: Windows NT → Windows 7
![]() |
Reporter | |
Comment 1•13 years ago
|
||
tinder box build hourly does not crash.however, Nightly nightly build crashes in spite of the same change set .
Something wrong PGO?
Regression window(m-c)
Good(tinder box build):
http://hg.mozilla.org/mozilla-central/rev/90857937b601
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121012191638
Crash(nightly build):
http://hg.mozilla.org/mozilla-central/rev/90857937b601
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121013030542
Comment 2•13 years ago
|
||
64-bit builds don't crash either.
![]() |
Reporter | |
Comment 3•13 years ago
|
||
I cannot reproduce the crash in Nightly19.0a1 2012-10-14 and 2012-10-15 anymore.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → WORKSFORME
![]() |
Reporter | |
Updated•13 years ago
|
tracking-firefox19:
? → ---
![]() |
Reporter | |
Comment 4•13 years ago
|
||
bp-4c5cbb4f-1a4a-4930-a97f-b6c9d2121017
bp-4f8182a3-385a-43fa-ae7f-104a62121017
Browser crashes with same STR in Nightly19.0a1 2012-10-16 and 2012-10-17 again .
So I think something PGO error .
Status: RESOLVED → REOPENED
Resolution: WORKSFORME → ---
Summary: crash in ToBooleanSlow → PGO error? crash in ToBooleanSlow
![]() |
Reporter | |
Updated•13 years ago
|
Keywords: regressionwindow-wanted
![]() |
Reporter | |
Comment 5•13 years ago
|
||
Not crash;
http://hg.mozilla.org/mozilla-central/rev/8f145599e4bf
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121016010947
Crash:
http://hg.mozilla.org/mozilla-central/rev/8f145599e4bf
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121016030544
Not crash;
http://hg.mozilla.org/mozilla-central/rev/dac5700acf8b
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121016185305
Crash:
http://hg.mozilla.org/mozilla-central/rev/dac5700acf8b
Mozilla/5.0 (Windows NT 6.1; WOW64; rv:19.0) Gecko/19.0 Firefox/19.0 ID:20121017030548
![]() |
Reporter | |
Updated•13 years ago
|
tracking-firefox19:
--- → ?
Comment 6•13 years ago
|
||
Not clear that this issue is critical, but it is a reproducible regression and we should take a look.
I can reproduce this, and given the mess PGO made in js::Interpret, it is likely Yet Another PGO Bug. Unfortunately it's not as clear where we should deoptimize this time. ToBooleanSlow is probably not a good candidate since it's not optimized or inlined, but it's worth a shot. Next up might be ToBoolean or whatever is calling ToBooleanSlow.
Whiteboard: [js:p1]
Comment 9•13 years ago
|
||
Given the fact that this isn't a top crasher, untracking so that the JS team has time to investigate bugs 791214 and 791999.
![]() |
||
Updated•13 years ago
|
Status: REOPENED → RESOLVED
Closed: 13 years ago → 13 years ago
Resolution: --- → DUPLICATE
You need to log in
before you can comment on or make changes to this bug.
Description
•