1362353 - Crash in itrp_ExecuteGlyphPgm

Status: RESOLVED WORKSFORME

(Core :: Graphics: Text, defect, P3)

Description

[Marcia Knous [:marcia]]

This bug was filed from the Socorro interface and is 
report bp-7066b367-efdc-4be2-8dd1-366490170504.
=============================================================


Seen while looking at nightly crash stats - crashes started using 20170503030212: http://bit.ly/2qHkrEx

Possible regression range based on Build ID: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=48c0fd9c9ec5d68061ea7b59358874ae8da72572&tochange=82c2d17e74ef9cdf38a5d5ac4eb3ae846ec30ba4

Crashes on other branches as well in small volume, so may not be a regression.

Comment 1

[Jeff Muizelaar [:jrmuizel]]

Perhaps OTS should be avoiding this problem?

Comment 2

[Jonathan Kew [:jfkthame]]

Possibly, though without a reproducible testcase we don't know if it involves a potentially bad downloadable font; it could easily be crashing on a locally-installed font, which OTS doesn't see.

Actually, I glanced at a bunch of the reports and it looked to me like they probably all came from the same user, which makes me suspect someone has a bad font installed locally. (Though an alternative is that a specific site this user is trying to visit has deployed a bad webfont, and we're not detecting/rejecting it.) If we knew the URL they're loading that triggers the crash, perhaps we could learn more.

Comment 3

[Jeff Muizelaar [:jrmuizel]]

The urls are a spread across different sites.

Comment 4

[Jeff Muizelaar [:jrmuizel]]

It would be interesting to know if this is limited to specific versions of dwrite

Comment 5

[Jonathan Kew [:jfkthame]]

I checked the "Modules" tab in about a dozen crash reports, and they all showed exactly the same info:

  DWrite.dll   6.2.9200.22084   17270510841B4657B73D127DD3F849331   DWrite.pdb

Comment 6

[Marcia Knous [:marcia]]

Only 14 crashes in 68 release. Closing as WFM based on insignificant volume.