Closed Bug 1660071 Opened 5 years ago Closed 4 years ago

Crash in [@ nsPrintJob::IsThereARangeSelection]

Categories

(Core :: Printing: Setup, defect, P2)

80 Branch
defect

Tracking

()

RESOLVED WORKSFORME
Tracking Status
firefox-esr68 --- unaffected
firefox-esr78 --- unaffected
firefox79 --- unaffected
firefox80 --- wontfix
firefox81 --- fixed

People

(Reporter: philipp, Unassigned)

Details

(Keywords: crash, regression)

Crash Data

This bug is for crash report bp-d42a316a-0a3a-44bf-9890-080060200819.

Top 10 frames of crashing thread:

0 xul.dll nsPrintJob::IsThereARangeSelection layout/printing/nsPrintJob.cpp:1123
1 xul.dll nsPrintJob::DoCommonPrint layout/printing/nsPrintJob.cpp:757
2 xul.dll nsPrintJob::CommonPrint layout/printing/nsPrintJob.cpp:609
3 xul.dll nsPrintJob::Print layout/printing/nsPrintJob.cpp:951
4 xul.dll nsDocumentViewer::Print layout/base/nsDocumentViewer.cpp:3173
5 xul.dll nsDocumentViewer::LoadComplete layout/base/nsDocumentViewer.cpp:1173
6 xul.dll nsDocShell::EndPageLoad docshell/base/nsDocShell.cpp:6172
7 xul.dll nsDocShell::OnStateChange docshell/base/nsDocShell.cpp:5499
8 xul.dll nsDocLoader::DoFireOnStateChange uriloader/base/nsDocLoader.cpp:1331
9 xul.dll nsDocLoader::doStopDocumentLoad uriloader/base/nsDocLoader.cpp:937

this content crash signature is newly appearing during the firefox 80 cycle - it's happening across platforms but in low volume till now.
i'm unsure about the exact regressor of this.

nsPrintJob::IsThereARangeSelection is removed in bug 1653334. Bob, is it possible that the crash can still happen after bug 1653334?

Severity: -- → S2
Flags: needinfo?(bobowencode)
Whiteboard: [print2020_v81]

(In reply to Ting-Yu Lin [:TYLin] (UTC-7) from comment #1)

nsPrintJob::IsThereARangeSelection is removed in bug 1653334. Bob, is it possible that the crash can still happen after bug 1653334?

Note that these crashes seem to be from Firefox 80, whereas bug 1653334 landed in the 81 nightly cycle (it's not fixed in Firefox 80).

So even if this turns out to be addressed by bug 1653334, we may need a mitigation for the 80 release cycle...

Priority: -- → P2

(In reply to Ting-Yu Lin [:TYLin] (UTC-7) from comment #1)

nsPrintJob::IsThereARangeSelection is removed in bug 1653334. Bob, is it possible that the crash can still happen after bug 1653334?

Sorry, I've been mainly out for a couple of days.
As other people have commented this can't happen now on Fx81.
Like dholbert, I do have concerns that we might see this spike up on Fx80 release though.

It looks like the window that is passed into nsPrintJob::IsThereARangeSelection doesn't have a docshell.
So, retrieving and checking that separately should be a simple fix, although I would guess that we would only consider taking this if something else drives a dot release.

Flags: needinfo?(bobowencode)

Since this is v80 only, it isn't really relevant to the v81 work so I'm going to take it off that list. We can handle this as its own separate thing if a v80 spike materializes and we want to land a ride-along for a point release.

Whiteboard: [print2020_v81]

This appears to have spiked a bit in 80, but given how close we are to the 81 release now will just keep an eye on it to close it after release.

QA Whiteboard: qa-not-actionable
You need to log in before you can comment on or make changes to this bug.