Closed Bug 625560 Opened 14 years ago Closed 14 years ago

crash in [@ js_XDRFunctionObject] when calling callMessage on registered receiver

Categories

(Core :: JavaScript Engine, defect)

defect
Not set
critical

Tracking

()

RESOLVED DUPLICATE of bug 660905
Tracking Status
blocking2.0 --- -

People

(Reporter: myk, Assigned: gal)

Details

(Keywords: crash)

Crash Data

Attachments

(1 file)

When I change Add-on SDK to pass an unwrapped function to registerReceiver instead of wrapping it with the error module's "catchAndLog" wrapper, and then I run tests with e10s enabled, I get this crash in js_XDRFunctionObject: http://crash-stats.mozilla.com/report/index/013217a6-4304-473a-ae5b-341bc2110113 The crash is accompanied by the following console messages: ###!!! [Child][RPCChannel] Error: Channel error: cannot send/recv ###!!! [Child][AsyncChannel] Error: Channel error: cannot send/recv So far I haven't been able to reduce the test case further. Steps to Reproduce: 1. Install and activate the SDK. 2. Apply the attached patch. 3. Change to the packages/api-utils/ directory. 4. Do |cfx test -f test-timer.js --e10s|.
Signature js_XDRFunctionObject UUID 013217a6-4304-473a-ae5b-341bc2110113 Time 2011-01-13 10:48:07.339355 Uptime 2 Last Crash 35 seconds before submission Install Age 91596 seconds (1.1 days) since version was first installed. Product Firefox Version 4.0b10pre Build ID 20110112033217 Branch 2.0 OS Linux OS Version 0.0.0 Linux 2.6.31-22-generic #70-Ubuntu SMP Wed Dec 1 23:51:13 UTC 2010 i686 CPU x86 CPU Info GenuineIntel family 6 model 37 stepping 5 Crash Reason SIGSEGV Crash Address 0x3a User Comments Processor Notes EMCheckCompatibility False Bugzilla - Report this Crash Crashing Thread Frame Module Signature [Expand] Source 0 libxul.so js_XDRFunctionObject jsfun.cpp:1839 1 libxul.so js_XDRScript jsscript.cpp:701 2 libxul.so JS_XDRScript jsxdrapi.cpp:671 3 libxul.so js_CloneScript jsscript.cpp:1986 4 libxul.so js_CloneFunctionObject jsfun.cpp:2740 5 libxul.so CloneFunctionObject jsfun.h:464 6 libxul.so js::Interpret jsinterp.cpp:5606 7 libxul.so js::Invoke jsinterp.cpp:657 8 libxul.so js_fun_apply jsfun.cpp:2181 9 libxul.so js::Interpret jscntxtinlines.h:692 10 libxul.so js::Invoke jsinterp.cpp:657 11 libxul.so js::ExternalInvoke jsinterp.cpp:858 12 libxul.so JS_CallFunctionValue jsinterp.h:961 13 libxul.so mozilla::jetpack::JetpackActorCommon::RecvMessage JetpackActorCommon.cpp:474 14 libxul.so mozilla::jetpack::JetpackParent::AnswerCallMessage JetpackParent.cpp:242
Severity: normal → critical
Keywords: crash
Summary: crash in js_XDRFunctionObject when calling callMessage on registered receiver → crash in [@ js_XDRFunctionObject] when calling callMessage on registered receiver
Looks like a bug in the engine potentially. Why was this not triaged? Thanks for the stack timeless.
blocking2.0: --- → ?
This doesn't apply to jetpack trunk for me (code looks very different).
Ugh, I was using jetpack-sdk. Looks like trunk moved to projects/addon-sdk. Trying that.
With my patch in bug 624366 I get a shutdown leak, but no crash. I will try to see if I still crash without the patch.
Actually I can't get this to crash without the patch either. I am going to need a little help here.
Ok, I got the crash to happen now. The patch in bug 624366 indeed makes a difference. There I just get this (no crash): info: An exception occurred in the child Jetpack process. error: An exception occurred. Traceback (most recent call last): File "/home/gal/workspace/addon-sdk/packages/api-utils/data/bootstrap-remote-process.js", line 140, in null I asked Myk to confirm that the patch works for him as well. The crash is real, so we should block on this, but I think we will be able to dup this.
blocking2.0: ? → betaN+
Whiteboard: [hardblocker]
Assignee: general → gal
We are not shipping js/jetpack with FF4, so not blocking on this. Bug 629197 will disable js/jetpack.
blocking2.0: betaN+ → -
Whiteboard: [hardblocker]
Crash Signature: [@ js_XDRFunctionObject]
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: