The browser starts up and never loads anything. No sidebar. No about:blank. No Mozilla QA Smoketests. Manually entering a url results in a crash. *Build Macintosh 1999052715-M6 the Talkback Incident is 9120346 here's is the stack trace: Call Stack: (Signature = 0x00000008 86d9bb9e) 0x00000008 BrowserAppCoreLoadUrl() [nsJSBrowserAppCore.cpp, line 254] js_Invoke() [jsinterp.c, line 649] js_Interpret() [jsinterp.c, line 2197] js_Invoke() [jsinterp.c, line 666] js_Interpret() [jsinterp.c, line 2197] js_Invoke() [jsinterp.c, line 666] And the MacsBug trace just for kicks: Calling chain using A6/R1 links Back chain ISA Caller 00000000 PPC 06A0DDC8 01B4B000 PPC 06A0CEE0 main+00774 01B4AE60 PPC 05D9C548 nsAppShellService::Run()+00018 01B4AE20 PPC 05A9594C nsAppShell::Run()+00038 01B4ADA0 PPC 05A962C8 nsMacMessagePump::DoMessagePump()+0003C 01B4AD50 PPC 05A9644C nsMacMessagePump::DispatchEvent(int, EventRecord*)+00048 01B4AD00 PPC 05A96CF4 nsMacMessagePump::DoKey(EventRecord&)+00088 01B4ACC0 PPC 05A9703C nsMacMessagePump::DispatchOSEventToRaptor(EventRecord&, GrafPort *)+00044 01B4AC70 PPC 05A912C0 nsMacMessageSink::DispatchOSEvent(EventRecord&, GrafPort*)+00038 01B4AC30 PPC 05A8CDE0 nsMacWindow::HandleOSEvent(EventRecord&)+00020 01B4ABD0 PPC 05A8D1A0 nsMacEventHandler::HandleOSEvent(EventRecord&)+0003C 01B4AB90 PPC 05A8DCA0 nsMacEventHandler::HandleKeyEvent(EventRecord&)+00180 01B4AAF0 PPC 05A7BC44 nsTextWidget::DispatchWindowEvent(nsGUIEvent&)+00090 01B4A9E0 PPC 05A77844 nsWindow::DispatchWindowEvent(nsGUIEvent&)+00018 01B4A9A0 PPC 05A77770 nsWindow::DispatchEvent(nsGUIEvent*, nsEventStatus&)+00090 01B4A950 PPC 05A2FC5C HandleEvent(nsGUIEvent*)+00058 01B4A900 PPC 05A2D524 nsViewManager::DispatchEvent(nsGUIEvent*, nsEventStatus&)+005D0 01B4A7C0 PPC 05A3172C nsView::HandleEvent(nsGUIEvent*, unsigned int, nsEventStatus&, i nt&)+00134 01B4A740 PPC 05A317B4 nsView::HandleEvent(nsGUIEvent*, unsigned int, nsEventStatus&, i nt&)+001BC 01B4A6C0 PPC 05667E18 PresShell::HandleEvent(nsIView*, nsGUIEvent*, nsEventStatus&)+00 1F8 01B4A620 PPC 0575BDEC nsHTMLInputElement::HandleDOMEvent(nsIPresContext&, nsEvent*, ns IDOMEvent**, unsigned int, nsEventStatus&)+00010 01B4A5E0 PPC 05841B18 nsGenericElement::HandleDOMEvent(nsIPresContext&, nsEvent*, nsID OMEvent**, unsigned int, nsEventStatus&)+000DC 01B4A580 PPC 056E7680 nsEventListenerManager::HandleEvent(nsIPresContext&, nsEvent*, n sIDOMEvent**, unsigned int, nsEventStatus&)+0091C 01B4A480 PPC 05D53AA0 nsJSEventListener::HandleEvent(nsIDOMEvent*)+00190 01B4A3E0 PPC 05DE238C JS_CallFunctionValue+00014 01B4A3A0 PPC 05DFAB9C js_CallFunctionValue+000B8 01B4A2F0 PPC 05DFA99C js_Invoke+0057C 01B4A210 PPC 05E00B38 js_Interpret+05900 01B49FD0 PPC 05DFA99C js_Invoke+0057C 01B49EF0 PPC 05E00B38 js_Interpret+05900 01B49CB0 PPC 05DFA944 js_Invoke+00524 01B49BD0 PPC 055F96FC BrowserAppCoreLoadUrl(JSContext*, JSObject*, unsigned int, long* , long*)+0008C Return addresses on the stack Stack Addr Frame Addr ISA Caller 01B49E88 PPC 05E029A4 js_LockObj+0002C 01B49E58 PPC 05E029A4 js_LockObj+0002C 01B49E28 01B49E20 PPC 05E028D0 js_UnlockScope+0003C 01B49DF0 68K 01B49E3E 01B49DE8 01B49DE0 PPC 05E028D0 js_UnlockScope+0003C 01B49DB0 68K 01B49DEE 01B49DA0 68K 01B49DEE 01B49D98 68K 05D23F62 GlobalWindowImpl::GetProperty(JSContext*, long, lon g*)+0003A 01B49D90 68K 01B49DEE 01B49D88 01B49D80 PPC 05E02240 js_GetSlotWhileLocked+00038 01B49D78 01B49D70 PPC 05E02A08 js_UnlockObj+00010 01B49D68 01B49D60 PPC 05E029A4 js_LockObj+0002C 01B49D58 01B49D50 PPC 05DE2934 JS_GetStringBytes+0000C 01B49D38 01B49D30 PPC 05E028D0 js_UnlockScope+0003C 01B49D08 01B49D00 PPC 05E26368 js_GetStringBytes+000D0 01B49CFC 68K 0120B442 01B49CC8 01B49CC0 PPC 05E260B8 js_DeflateString+00034 01B49CB8 PPC 05E00B38 js_Interpret+05900 01B49C88 68K 012EE37A 01B49C08 68K 012EE37A 01B49BF8 01B49BF0 PPC 05E0ADB0 js_DropProperty+0002C 01B49BE8 01B49BE0 PPC 05E029A4 js_LockObj+0002C 01B49BD8 01B49BD0 PPC 05DFA944 js_Invoke+00524 01B49BBC 68K 012EE37A 01B49BB8 01B49BB0 PPC 05E02A08 js_UnlockObj+00010 01B49BA8 01B49BA0 PPC 05E02808 js_LockScope1+0003C 01B49B98 01B49B90 PPC 05DDEAE0 JS_GetPrivate+00010 01B49B48 01B49B40 PPC 05E02A08 js_UnlockObj+00010 01B49B38 01B49B30 PPC 05E029A4 js_LockObj+0002C 01B49B18 01B49B10 PPC 055F96FC BrowserAppCoreLoadUrl(JSContext*, JSObject*, unsign ed int, long*, long*)+0008C 01B49AD8 PPC 05D6C4FC nsJSUtils::nsConvertJSValToString(nsString&, JSCont ext*, long)+00040 01B49A98 01B49A90 PPC 055F52B8 nsBrowserAppCore::LoadUrl(const nsString&)+000D4

This is a Mac only bug. M6 of Linux and Win32 are fine.

I used lxr.mozilla.org to see the cvs log for nsJSBrowserAppCore.cpp. Saw morse@netscape.com's name next to some changes for today 5/26 (although I don't know if it's for M6 or M7). I don't know browser well enough, but I don't see anyone around at this time so I'll cc: morse on this bug too since it's a blocker bug for the browser group. Also, I'll cc: the folks who've made changes shown in the CVS change log for today (using the URL chofmann sent out): slamm, nisheeth, jj

If this is an M6 bug, it can't be my change. All my check-ins were in M7. Also there's nothing in the stacktrace that looks like any of the areas that I changed.

it would be interesting to see if viewer could load url's. that way we could establish if we are fried back in the core code, or if this is apprunner or xul.

There are two possibilties that need investigation. first is that we had to go through many gyrations to get this build on mac of the m6 branch because of updates for xpcom that landed on the trunck and the resultant changes that had to be made to the modules file and required compiler updates. In the worst case it won't be possible to do a mac build on from the branch because of all the recent upgrades. In a slightly better case it may be possible but will continue to take some more investigation to get a respin or another clean attempt and closely review that we are getting all the right branch files and using all the right tools. Another possibility is that one of the recent changes to the branch is resulting in this problem. A few of have scratch our heads but don't see any likely suspects that might result in these errors. One additional test that might help to add some more insight is to try and run viewerin this build to see if it can start and run. Is there anyone still around to try that?

There are two possibilties that need investigation. first is that we had to go through many gyrations to get this build on mac of the m6 branch because of updates for xpcom that landed on the trunck and the resultant changes that had to be made to the modules file and required compiler updates. In the worst case it won't be possible to do a mac build on from the branch because of all the recent upgrades. In a slightly better case it may be possible but will continue to take some more investigation to get a respin or another clean attempt and closely review that we are getting all the right branch files and using all the right tools. Another possibility is that one of the recent changes to the branch is resulting in this problem. A few of have scratch our heads but don't see any likely suspects that might result in these errors. One additional test that might help to add some more insight is to try and run viewerin this build to see if it can start and run. Is there anyone still around to try that?

There are two possibilties that need investigation. first is that we had to go through many gyrations to get this build on mac of the m6 branch because of updates for xpcom that landed on the trunck and the resultant changes that had to be made to the modules file and required compiler updates. In the worst case it won't be possible to do a mac build on from the branch because of all the recent upgrades. In a slightly better case it may be possible but will continue to take some more investigation to get a respin or another clean attempt and closely review that we are getting all the right branch files and using all the right tools. Another possibility is that one of the recent changes to the branch is resulting in this problem. A few of have scratch our heads but don't see any likely suspects that might result in these errors. One additional test that might help to add some more insight is to try and run viewerin this build to see if it can start and run. Is there anyone still around to try that?

There are two possibilties that need investigation. first is that we had to go through many gyrations to get this build on mac of the m6 branch because of updates for xpcom that landed on the trunck and the resultant changes that had to be made to the modules file and required compiler updates. In the worst case it won't be possible to do a mac build on from the branch because of all the recent upgrades. In a slightly better case it may be possible but will continue to take some more investigation to get a respin or another clean attempt and closely review that we are getting all the right branch files and using all the right tools. Another possibility is that one of the recent changes to the branch is resulting in this problem. A few of have scratch our heads but don't see any likely suspects that might result in these errors. One additional test that might help to add some more insight is to try and run viewerin this build to see if it can start and run. Is there anyone still around to try that?

There are two possibilties that need investigation. first is that we had to go through many gyrations to get this build on mac of the m6 branch because of updates for xpcom that landed on the trunck and the resultant changes that had to be made to the modules file and required compiler updates. In the worst case it won't be possible to do a mac build on from the branch because of all the recent upgrades. In a slightly better case it may be possible but will continue to take some more investigation to get a respin or another clean attempt and closely review that we are getting all the right branch files and using all the right tools. Another possibility is that one of the recent changes to the branch is resulting in this problem. A few of have scratch our heads but don't see any likely suspects that might result in these errors. One additional test that might help to add some more insight is to try and run viewerin this build to see if it can start and run. Is there anyone still around to try that?

looks like the crash is in BrowserAppCoreLoadUrl(). didn't we make sure the branch was working after hyatt and i found the fix for always going to the smoketest page?

this is the attempt at doing that. Its been a slow painfull process due to complications of the xpcom landing...

for the record, Viewer launches but never gets to lay out the initial page. The progress notification always says some number less than the total and then stops. Manually entering a url yields the same behavior. Pressing 'back' took me back to the never completed initial page as well(after a while). No crashes.

FIX CHECKED IN

Updating QA Contact to verify with latest build.

VERIFIED fixed for the Mac M6 build 1999060119. Not a problem for M7 builds.

Moving all Apprunner bugs past and present to Other component temporarily whilst don and I set correct component. Apprunner component will be deleted/retired shortly.