http://crash-stats.mozilla.com/report/index/bp-df08c665-dee9-40d3-ad3a-112ac2091221 http://crash-stats.mozilla.com/report/index/bp-e4b41252-3432-4fba-92bf-536de2091221 When I installed Eventbug, the browser restarted but would crashed when it tried to load my profile. I could select the profile but not start the browser. Steps to reproduce: 1) Install Firebug 1.5X.0b8 ( http://getfirebug.com/releases/firebug/1.5X/firebug-1.5X.0b8.xpi ) 2) Install Eventbug ( http://getfirebug.com/releases/eventbug/1.5/eventbug-0.1b1.xpi ) 3) Restart the browser Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.3a1pre) Gecko/20091221 Minefield/3.7a1pre (.NET CLR 3.5.30729) ID:20091221052153

We better check FF 3.6

Actually I cannot reproduce this on FF 3.7 Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.3a1pre) Gecko/20091221 Minefield/3.7a1pre Tanner do you have other extensions installed?

Yes, a long list: Adblock Plus 1.1.2 Add to Search Bar 1.8 AutoAuth 1.3 Better Gmail 2 0.9.4 Closure Inspector 0.9 DOM Inspector 2.0.4 DownloadHelper 4.6.2 EventBug 0.1b1 [DISABLED] Firebug 1.5X.0b8 Html Validator 0.8.5.8 Microsoft .NET Framework Assistant 0.0.0 [DISABLED] Minimap Addon 0.3.13 Mozilla QA Companion 1.2.1 [DISABLED] MozMill 1.3 [DISABLED] Nightly Tester Tools 2.0.3 Page Speed 1.4 Selenium IDE 1.0.2 [DISABLED] Test Pilot 0.4 Wolfram Alpha Google 0.9 [DISABLED] XSS Me 0.4.3 Either tonight or in the morning, I will figure out which one conflicts with Eventbug.

Selenium IDE appears to be the offending add-on. Updated STR: 1) Install Firebug 1.5X.0b8 ( http://getfirebug.com/releases/firebug/1.5X/firebug-1.5X.0b8.xpi ) 2) Install Eventbug ( http://getfirebug.com/releases/eventbug/1.5/eventbug-0.1b1.xpi ) 3) Install Selenium IDE ( http://release.openqa.org/selenium-ide/1.0.2/selenium-ide-1.0.2.xpi ) 3) Restart the browser

Signature checkReportFlags UUID df08c665-dee9-40d3-ad3a-112ac2091221 Time 2009-12-21 15:34:07.264701 Uptime 2 Last Crash 20 seconds before submission Product Firefox Version 3.7a1pre Build ID 20091221052153 Branch 1.9.3 OS Windows NT OS Version 6.0.6002 Service Pack 2 CPU x86 CPU Info GenuineIntel family 6 model 23 stepping 7 Crash Reason EXCEPTION_ACCESS_VIOLATION Crash Address 0x18 User Comments Processor Notes Related Bugs Crashing Thread Frame Module Signature [Expand] Source 0 mozjs.dll checkReportFlags js/src/jscntxt.cpp:1419 1 mozjs.dll js_ReportErrorNumberVA js/src/jscntxt.cpp:1643 2 mozjs.dll JS_ReportErrorFlagsAndNumber js/src/jsapi.cpp:5601 3 mozjs.dll JSScopeProperty::set js/src/jsscope.h:857 4 mozjs.dll js_SetPropertyHelper js/src/jsobj.cpp:4713 5 mozjs.dll js_SetProperty js/src/jsobj.cpp:4826 6 mozjs.dll JS_SetPropertyById js/src/jsapi.cpp:3598 7 xul.dll XPC_SJOW_GetOrSetProperty 8 xul.dll XPC_SJOW_SetProperty js/src/xpconnect/src/XPCSafeJSObjectWrapper.cpp:582 9 mozjs.dll js_SetPropertyHelper js/src/jsobj.cpp:4811 10 mozjs.dll js_Interpret js/src/jsops.cpp:1890 11 mozjs.dll js_Invoke js/src/jsinterp.cpp:1384 12 mozjs.dll js_InternalInvoke js/src/jsinterp.cpp:1439 13 mozjs.dll JS_CallFunctionValue js/src/jsapi.cpp:5106 14 xul.dll nsJSContext::CallEventHandler dom/base/nsJSEnvironment.cpp:2164 15 xul.dll nsGlobalWindow::RunTimeout dom/base/nsGlobalWindow.cpp:8130 16 xul.dll nsGlobalWindow::TimerCallback dom/base/nsGlobalWindow.cpp:8474

igor: if you object to this patch, please write your own. this is a one off donation.

Comment on attachment 429458 [details] [diff] [review] proposal ># HG changeset patch ># Parent fadc5f2804d701ca7b4770fecae1f5e3eb6a7dab ># User timeless@mozdev.org >Bug 536306 Crash [@ checkReportFlags ] occurs when loading Eventbug and Selenium IDE >r=igor > >diff --git a/js/src/jscntxt.cpp b/js/src/jscntxt.cpp >--- a/js/src/jscntxt.cpp >+++ b/js/src/jscntxt.cpp >@@ -1419,7 +1419,10 @@ checkReportFlags(JSContext *cx, uintN *f > if (JSREPORT_IS_STRICT_MODE_ERROR(*flags)) { > /* Error in strict code; warning with strict option; okay otherwise. */ > JS_ASSERT(JS_IsRunning(cx)); >- if (js_GetTopStackFrame(cx)->script->strictModeCode) >+ JSStackFrame *fp = js_GetTopStackFrame(cx); >+ if (!fp->script) >+ return true; >+ if (fp->script->strictModeCode) The issue here is that a script calls a native function that on behalf of it do the property assignment. So to cover that case the code should use I think not js_GetTopStackFrame(cx) but rather js_GetScriptedCaller and decide strictness based on that.

The comments in the patch should explain the changes.

Comment on attachment 429622 [details] [diff] [review] fix v1 + * the nearest scripted frame is script, see bug 536306. ... is strict ?

Here is a typo fix.

Verified Fixed. Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.3a4pre) Gecko/20100317 Minefield/3.7a4pre (.NET CLR 3.5.30729) ID:20100317041424