Closed Bug 93471 Opened 24 years ago Closed 23 years ago

Find crashes when "cancel" is clicked while searching - Trunk M097 N621 [@ nsEventStateManager::PreHandleEvent]

Categories

(Core :: DOM: UI Events & Focus Handling, defect)

x86
Linux
defect
Not set
critical

Tracking

()

RESOLVED WORKSFORME
mozilla1.0

People

(Reporter: bugs, Assigned: joki)

References

()

Details

(Keywords: crash, testcase, topcrash)

Crash Data

When searching a large file for a text string, if one clicks on "Cancel" the browser crashes. This behaviour does not occur under Windows/Solaris builds. Also, the entire application hangs when using Find, this is something I've noticed with other Mozilla functions. Does Mozilla use threading? To Reproduce: 1. CTRL-F when a large file is opened 2. type somerandomstring 3. Hit cancel while it is searching Expected Behaviour: Should've cancelled or at least continued the search.
Oh, I tested this on an 07/27 nightly, then an 08/03 build. Occurs in both, with additional fun of 08/03 having that javascript error that makes it impossible to file this bug on the main mozilla page.
Summary: Find crashes when "cancel" is clicked while searching → Find crashes when "cancel" is clicked while searching
need new netscape sherlock file for mozilla builds
Target Milestone: --- → mozilla0.9.4
Keywords: crash
Confirmed on linux 0.9.3 with talkback build. Talkback ID TB33788883Y
talkback is so cool now... 33788883 and 93471 Show Incident ID Set Bug ID Incident ID 33788883 Stack Signature nsEventStateManager::PreHandleEvent() 5752097f Bug ID 93471 Trigger Time 2001-08-06 07:30:14 Email Address con@kolivas.mine.nu User Comments Clicking on cancel while doing a text find in a large text file. Build ID 2001080104 Product ID MozillaTrunk Platform ID LinuxIntel Stack Trace nsEventStateManager::PreHandleEvent() PresShell::HandleEventInternal() PresShell::HandleEvent() nsView::HandleEvent() nsViewManager::DispatchEvent() HandleEvent() nsWidget::DispatchEvent() nsWidget::DispatchWindowEvent() nsWidget::DispatchFocus() nsWindow::DispatchDeactivateEvent() nsWindow::HandleMozAreaFocusOut() handle_mozarea_focus_out() libgtk-1.2.so.0 + 0x9576c (0x4029776c) libgtk-1.2.so.0 + 0xc8f56 (0x402caf56) libgtk-1.2.so.0 + 0xc828d (0x402ca28d) libgtk-1.2.so.0 + 0xc6035 (0x402c8035) libgtk-1.2.so.0 + 0x100af9 (0x40302af9) libgtk-1.2.so.0 + 0x109954 (0x4030b954) libgtk-1.2.so.0 + 0x9576c (0x4029776c) libgtk-1.2.so.0 + 0xc82cd (0x402ca2cd) libgtk-1.2.so.0 + 0xc6035 (0x402c8035) libgtk-1.2.so.0 + 0x100af9 (0x40302af9) libgtk-1.2.so.0 + 0x94794 (0x40296794) handle_gdk_event() libgdk-1.2.so.0 + 0x17d6f (0x4034dd6f) I gave this a spin on a Win98 2001080203 trunk build as was unable to repro. Could try on linux due to technical difficulties. Over to XPApps:GUI where these bugs belong. Matt was your comment relevant to this bug or no?
Component: Search → XP Apps: GUI Features
QA Contact: claudius → sairuh
no relevance at all. Wrong bug. This is a find bug. Should be law's bug.
Confirmed 2001083106 linux and 112 MB text/plain
Status: UNCONFIRMED → NEW
Ever confirmed: true
-> Peter
Assignee: matt → trudelle
Peter, This is a linux only bug. On window i don't crash but i also can't cancel Law might know this bug since he had the find dialog.
No, I worked on that dialog eons ago. It is completely different now. And the problem probably isn't in the dialog itself, but rather in the underlying webshell/docshell/document/text-services stuff (of which I know little or nothing). Kin would be a more likely source of help than me.
The stack points to it crashing in event code. You can't cancel in the middle of a find since the TextServices find code executes on the same thread that services the dialog events ... this means any canceling/closing of the dialog is done after the find has completed.
I forgot to add that I couldn't reproduce this crash with my 09/06/01 mozilla debug build. Is there a specific page/url that crashes all the time?
->event handling
Assignee: trudelle → joki
Component: XP Apps: GUI Features → Event Handling
QA Contact: sairuh → madhur
jay, can you search to see if this is a topcrash in talkback? per PDT
This appears to be a topcrasher with M093 (MozillaTrunk build 2001080112 for Win32 and build 2001080104 for Linux), but since then there are only 3 incidents reported in Talkback. Here is the most recent crash (which might even be a different crash than the one reported): Incident ID 35295973 Stack Signature nsEventStateManager::PreHandleEvent b83572e6 Bug ID Trigger Time 2001-09-12 11:53:37 Email Address User Comments Build ID 2001091109 Product ID MozillaTrunk Platform ID Win32 Trigger Reason Access violation Stack Trace nsEventStateManager::PreHandleEvent [d:\builds\seamonkey\mozilla\content\events\src\nsEventStateManager.cpp, line 412] PresShell::HandleEventInternal [d:\builds\seamonkey\mozilla\layout\html\base\src\nsPresShell.cpp, line 5653] PresShell::HandleEvent [d:\builds\seamonkey\mozilla\layout\html\base\src\nsPresShell.cpp, line 5583] nsView::HandleEvent [d:\builds\seamonkey\mozilla\view\src\nsView.cpp, line 377] nsView::HandleEvent [d:\builds\seamonkey\mozilla\view\src\nsView.cpp, line 350] nsViewManager::DispatchEvent [d:\builds\seamonkey\mozilla\view\src\nsViewManager.cpp, line 2058] HandleEvent [d:\builds\seamonkey\mozilla\view\src\nsView.cpp, line 68] nsWindow::DispatchEvent [d:\builds\seamonkey\mozilla\widget\src\windows\nsWindow.cpp, line 732] nsWindow::DispatchWindowEvent [d:\builds\seamonkey\mozilla\widget\src\windows\nsWindow.cpp, line 749] nsWindow::DispatchFocus [d:\builds\seamonkey\mozilla\widget\src\windows\nsWindow.cpp, line 4454] nsWindow::ProcessMessage [d:\builds\seamonkey\mozilla\widget\src\windows\nsWindow.cpp, line 3359] nsWindow::WindowProc [d:\builds\seamonkey\mozilla\widget\src\windows\nsWindow.cpp, line 997] KERNEL32.DLL + 0x3613 (0xbff63613) KERNEL32.DLL + 0x248f7 (0xbff848f7) Most of the comments from M093 data mention canceling out of a search or closing popup windows.
0.9.4 is out the door
Target Milestone: mozilla0.9.4 → mozilla0.9.5
-> 0.9.6
Target Milestone: mozilla0.9.5 → mozilla0.9.6
This continues to be a topcrasher with Mozilla 0.9.4 so adding M094 [@ nsEventStateManager::PreHandleEvent] to summary for tracking. Here is the latest info from Talkback reports: nsEventStateManager::PreHandleEvent 32 BBID range: 36080767 - 36362206 Min/Max Seconds since last crash: 89 - 516605 Min/Max Runtime: 21409 - 516605 Crash data range: 2001-09-30 to 2001-10-06 Build ID range: 2001091311 to 2001091311 Stack Trace: nsEventStateManager::PreHandleEvent() PresShell::HandleEventInternal() PresShell::HandleEvent() nsView::HandleEvent() nsView::HandleEvent() nsView::HandleEvent() nsViewManager::DispatchEvent() HandleEvent() nsWidget::DispatchEvent() nsWidget::DispatchWindowEvent() nsWidget::DispatchFocus() nsWindow::DispatchSetFocusEvent() nsWindow::SetFocus() GlobalWindowImpl::Focus() CheckForFocus() PresShell::UnsuppressAndInvalidate() PresShell::UnsuppressPainting() DocumentViewerImpl::LoadComplete() nsDocShell::EndPageLoad() nsWebShell::EndPageLoad() nsDocShell::OnStateChange() nsDocLoaderImpl::FireOnStateChange() nsDocLoaderImpl::doStopDocumentLoad() nsDocLoaderImpl::DocLoaderIsEmpty() nsDocLoaderImpl::OnStopRequest() nsLoadGroup::RemoveRequest() PresShell::RemoveDummyLayoutRequest() PresShell::DoneRemovingReflowCommands() PresShell::ProcessReflowCommands() HandlePLEvent() PL_HandleEvent() PL_ProcessPendingEvents() nsEventQueueImpl::ProcessPendingEvents() event_processor_callback() our_gdk_io_invoke() (36151912) URL: http://www.builder.com (36151912) Comments: I searched for the "Color Convertor" on Builder.comThen I didn't input a correct color code. I kept on getting alerts (up to number 8 between brackets) until finally the feedbackagent prompted.
Summary: Find crashes when "cancel" is clicked while searching → Find crashes when "cancel" is clicked while searching - M094 [@ nsEventStateManager::PreHandleEvent]
Adding topcrash keyword for talkback
Keywords: topcrash
*** Bug 109282 has been marked as a duplicate of this bug. ***
Does this occur only with plaintext, or with HTML as well? Large pages that could maybe be used as testcases: 339K HTML: http://members.kconline.com/kerr/pball.htm 634K plain text: http://www.ifarchive.org/if-archive/infocom/compilers/inform6/manuals/old/designers_manual.txt
any large text file. html was what was originally used. The problem isn't so much in find's parsing of the page, as, it seems in some thread thing when cancelling a search.
Too late for 0.9.6, this needs retargeting.
0.9.6 has passed, moving to 0.9.7. Load balancing 0.9.7 list shortly
Target Milestone: mozilla0.9.6 → mozilla0.9.7
Moving bugs from 0.9.7 for triaging in 0.9.8
Target Milestone: mozilla0.9.7 → mozilla0.9.8
Target Milestone: mozilla0.9.8 → mozilla0.9.9
Updating summary with Trunk, M097 and N621...this continues to crash with MozillaTrunk builds, Mozilla 0.9.7 and Netscpae 6.21. i don't think the stack has changed much...take a look at bug 118685 (which i think might even be a dup). Bug 101834 also has the same stack signature, but the stack looks a little different, so I'm not sure if that's a dup or not.
Summary: Find crashes when "cancel" is clicked while searching - M094 [@ nsEventStateManager::PreHandleEvent] → Find crashes when "cancel" is clicked while searching - Trunk M097 N621 [@ nsEventStateManager::PreHandleEvent]
nsbeta1+ per ADT triage
Keywords: nsbeta1+
Adding testcase keyword, since comments 20 and 21 mention a testcase that leads to this crash.
Keywords: testcase
I'm not able to reproduce this at all with build 2002022208 (RedHat 7.2, window manager is sawfish).
-> 1.0
Target Milestone: mozilla0.9.9 → mozilla1.0
Re: Bryan's comment I went back to this bug, now in 7.2 instead of 6.2 of redhat. (haven't tested it on old machine yet) As in other operating systems it seems Mozilla once more freezes completely while searching, all windows unresponsive, including, yes, the Cancel button. However, since it wasn't even redrawing when the find window, I doubt this was due to even anything like disabling cancel while searching and simply more of the one busy window locks up all of Mozilla thing. Might be sorta fixed though.
Marking this bug worksforme. There haven't been any crashes with this signature since 2/18. Please reopen if anyone can reproduce this crash with a more recent MozillaTrunk build.
Status: NEW → RESOLVED
Closed: 23 years ago
Resolution: --- → WORKSFORME
QA Contact: madhur → rakeshmishra
QA Contact: rakeshmishra → trix
Depends on: 118685
Crash Signature: [@ nsEventStateManager::PreHandleEvent]
Component: Event Handling → User events and focus handling
You need to log in before you can comment on or make changes to this bug.