Closed
Bug 568476
Opened 15 years ago
Closed 6 years ago
[e10s-fennec] crash [@ nsGlobalWindow::SaveWindowState]
Categories
(Core :: IPC, defect, P3)
Tracking
()
RESOLVED
WONTFIX
Tracking | Status | |
---|---|---|
e10s | - | --- |
People
(Reporter: romaxa, Assigned: romaxa)
Details
(Keywords: crash)
Crash Data
Attachments
(4 files)
7.18 KB,
text/plain
|
Details | |
17.03 KB,
text/plain
|
Details | |
974 bytes,
patch
|
smaug
:
review+
|
Details | Diff | Splinter Review |
818 bytes,
patch
|
Details | Diff | Splinter Review |
xulrunner e10s 43217:dc9cc3820b02
fennec-e10s 1643:88f97721f8bb
Qt port
resize fennec to small screen 800x480
load about:license URL
when page is visible, type url bar and load another URL "about:buildconfig"
result:
Fennec crashes in nsGlobalWindow::SaveWindowState, because inner == NULL.
Assignee | ||
Comment 1•15 years ago
|
||
Assignee | ||
Comment 2•15 years ago
|
||
Severity: normal → critical
Keywords: crash
Summary: [e10s-fennec] crash nsGlobalWindow::SaveWindowState → [e10s-fennec] crash [@ nsGlobalWindow::SaveWindowState]
Comment 3•15 years ago
|
||
For the content process crash a simple null check should be ok.
Assignee | ||
Comment 4•15 years ago
|
||
Attachment #449192 -
Flags: review?(Olli.Pettay)
Comment 5•15 years ago
|
||
Comment on attachment 449192 [details] [diff] [review]
Null check.
Remove the useless assertion
NS_ASSERTION(inner, "No inner window to save");
Attachment #449192 -
Flags: review?(Olli.Pettay) → review+
Assignee | ||
Comment 6•15 years ago
|
||
Assignee | ||
Updated•15 years ago
|
Attachment #449798 -
Attachment description: NULL check no, assertion.. → to checking, NULL check no, assertion..
Assignee | ||
Updated•15 years ago
|
Keywords: checkin-needed
![]() |
||
Comment 7•15 years ago
|
||
Wait. So how are we ending up with a null inner here, exactly?
Updated•15 years ago
|
Assignee: nobody → romaxa
Assignee | ||
Comment 8•15 years ago
|
||
(In reply to comment #7)
> Wait. So how are we ending up with a null inner here, exactly?
Yes, null inner
![]() |
||
Comment 9•15 years ago
|
||
I understand that, but WHY? That shouldn't happen; that's why the assertion is there. Why is it happening? Who screwed up and how?
Assignee | ||
Comment 10•15 years ago
|
||
I started after previous merge with m-c...
I had the same question, but did not get what is going on...
there is backtrace + debug log in attachment
![]() |
||
Comment 11•15 years ago
|
||
OK. So the docshell has an mContentViewer and an mOSHE but its window has no inner? That's just broken beyond belief. The inner is clearly non-null when the previous page is loaded. Care to set a watchpoint in the debugger and tell me when it becomes null?
Comment 12•15 years ago
|
||
I'm going to clear checkin-needed since there are unanswered questions remaining.
Keywords: checkin-needed
Updated•14 years ago
|
Crash Signature: [@ nsGlobalWindow::SaveWindowState]
Updated•11 years ago
|
tracking-e10s:
--- → +
Updated•11 years ago
|
Updated•7 years ago
|
Priority: -- → P3
Comment 14•6 years ago
|
||
Closing because no crash reported since 12 weeks.
Status: NEW → RESOLVED
Closed: 6 years ago
Resolution: --- → WONTFIX
You need to log in
before you can comment on or make changes to this bug.
Description
•