1359995 - Crash in nsTArray_Impl<T>::AppendElement<T> | mozilla::RestyleTracker::AddPendingRestyle

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[Marcia Knous [:marcia]]

This bug was filed from the Socorro interface and is 
report bp-7a668e94-3cec-447c-911c-8a5150170426.
=============================================================

Seen while looking at nightly crash data. Crashes started using 20170426030329: http://bit.ly/2oKq6rh

Possible regression range based on Build ID: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=a30dc237c3a600a5231f2974fc2b85dfb5513414&tochange=0f5ba06c4c5959030a05cb852656d854065e2226

Bug 1357142 touched code in this area. ni on :emilio

Comment 1

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1359995: Add back the null-check on curData.

I removed it, but seems it can be hit. It'd be nice to have a test-case where it
fails though...

Review commit: https://reviewboard.mozilla.org/r/134180/diff/#index_header
See other reviews: https://reviewboard.mozilla.org/r/134180/

Comment 2

[Emilio Cobos Álvarez (:emilio)]

Thanks for filling :)

Comment 3

[Cameron McCormack (:heycam)]

Comment on attachment 8862200 [details]
Bug 1359995: Add back the null-check on curData.

https://reviewboard.mozilla.org/r/134180/#review137092

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/5ab8cf2eca5f
Add back the null-check on curData. r=heycam

Comment 5

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/5ab8cf2eca5f

Comment 6

[Andrew McCreight [:mccr8]]

Emilio, I came across a similar looking crash signature. Is this likely to be fixed by your patch, or should I file a new bug? It also looks new as of 4-26. bp-24037bc5-87f1-478d-a863-ebcb30170426

Comment 7

[Emilio Cobos Álvarez (:emilio)]

Yeah, it's the same, this time coming from a different function in GeckoRestyleManager. I wouldn't bother filing a new bug.

Thanks for checking though!

Comment 8

[Julian Seward [:jseward]]

This is #4 topcrash in the Windows nightlies of 20170427030231, with
62 crashes from perhaps 5 or 6 different installations.

Comment 9

[Andrew McCreight [:mccr8]]

Both crash signatures haven't been seen since the 4-27 Nightly.