Closed Bug 844506 Opened 13 years ago Closed 12 years ago

crash in js::ion::LiveRangeAllocator

Categories

(Core :: JavaScript Engine, defect)

22 Branch
x86_64
All
defect
Not set
critical

Tracking

()

RESOLVED WORKSFORME
Tracking Status
firefox21 --- unaffected
firefox22 --- affected

People

(Reporter: ravi, Unassigned)

References

Details

(Keywords: crash, regression, reproducible)

Crash Data

This bug was filed from the Socorro interface and is report bp-6cdca583-2f88-4dda-aacc-8d4902130223 . ============================================================= The same crash happened when clicking between folders in Zimbra. I haven't been able to reproduce the specific condition for the crash, but the 3 times it was selecting different folders. Other two: https://crash-stats.mozilla.com/report/index/bp-6cdca583-2f88-4dda-aacc-8d4902130223 https://crash-stats.mozilla.com/report/index/bp-2245de2f-7186-4a61-a525-237ba2130223 Built from http://hg.mozilla.org/mozilla-central/rev/08a034e1d43a I wasn't sure if this should be filed in Core Javascript or not so am placing here for triage assistance.
Assignee: nobody → general
Component: General → JavaScript Engine
Product: Firefox → Core
OS: Mac OS X → All
Version: unspecified → Trunk
I've run into this multiple times today, usually with just a single window open with a Zimbra tab and a gmail tab. Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:22.0) Gecko/20130224 Firefox/22.0
Steps to reproduce: 1. Open Zimbra mail 2. Click on a folder 3. Click on a message in that folder 4. Click on the Inbox folder 5. Repeat steps 2-4 until crash occurs Sometimes happens the first time, sometimes it takes a few iterations.
Keywords: reproducible
Any feedback from JS guys? Can you at least reproduce this? It would help to know if this is something specific to my machine/profile or whether others see the same behavior.
Sorry, I tried the STR (and repeated 10 times) and I didn't get a crash. Maybe try it with a fresh profile and see if it still happens? It also might be related to some aspect of your Zimbra account.
Keywords: regression
Hardware: All → x86_64
Version: Trunk → 22 Branch
I can't reproduce it either. It would be great if people who can reproduce this could try a debug build (to see if it hits an assert somewhere) or get a good stack trace.
Depends on: 865373
ragavan reported the following similar stacks today, which he mentioned using facebook. not sure what the steps are, but he said it was against the apr 26 nightly on mac, just having facebook open and crashes. Frame Module Signature Source 0 XUL js::ion::LiveRangeAllocator<js::ion::LinearScanVirtualRegister>::buildLivenessIn js/src/ion/LiveRangeAllocator.h:403 1 XUL js::ion::LinearScanAllocator::go js/src/ion/LinearScan.cpp:1136 2 XUL js::ion::GenerateLIR js/src/ion/Ion.cpp:1128 bp-df31b628-2cd2-4170-a050-a88ca2130426 4/26/13 1:47 PM bp-04bf1a79-3d47-4d55-84d6-5418b2130426 4/26/13 1:38 PM bp-57488521-4ae6-4c0b-bfde-536762130426 4/26/13 1:33 PM bp-7b79e9cd-c8f1-4ae1-8d40-ee1a62130426 4/26/13 1:32 PM bp-8556f9eb-ad72-42bd-87d9-331752130426 4/26/13 1:32 PM bp-7552edf0-bbff-4977-a61f-5173d2130426 4/26/13 1:17 PM 862C519D-A322-4452-A053-733BD488E0BD 4/26/13 1:17 PM bp-5156c0a5-5ac2-4402-be42-5905f2130426 4/26/13 12:28 PM bp-aa516be3-1919-4a1e-be8c-352432130426 4/26/13 12:28 PM bp-68dffb70-61f4-44fe-960f-7c9662130426 4/26/13 12:27 PM bp-c30e02ac-3abd-43ac-b308-dd8f92130426 4/26/13 11:59 AM bp-03ae3aae-cd59-426b-ba38-781122130426 4/26/13 11:41 AM bp-b037758b-5a11-42fa-a422-d12752130426 4/26/13 9:40 AM
I've hit this a bunch after updating Nightly this morning from some build last week :/ Same stack as Ragavan. bp-daac0216-d2a1-4573-987e-dbf712130429 bp-fba6d03f-32d8-4261-95dc-74d172130429 bp-fe9def8a-239b-4499-a70e-baaa62130429 Brian, based on 865373, possibly bug 865635 triggered this?
What happens if you use a current nightly? Those build ids are from 4/26, and per bug 865373 comment 3 this crash should be fixed in builds dated 4/27 or later.
Ah, looks like my Nightly didn't update all the way but I though it had (and at least one of those crashes was from UX branch which is likely behind). Current Nightly hasn't crashed with the same session it was, so probably ok. Sorry for the confusion!
Ravi, is it still reproducible in Aurora and Nightly?
Flags: needinfo?(ravi)
I haven't run into the problem for some time now. Perhaps it was fixed at some point I guess I'll close as WORKSFORME.
Status: NEW → RESOLVED
Closed: 12 years ago
Flags: needinfo?(ravi)
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.