1043394 - crash in mozilla::ipc::MessageChannel::NotifyGeckoEventDispatch()

Status: RESOLVED DUPLICATE of bug 1044245

(Core :: WebRTC, defect)

Description

[Ada [:adalucinet]]

This bug was filed from the Socorro interface and is 
report bp-619b3ab4-f56e-4dd5-ae5c-0dfc82140724.
=============================================================

STR:
1. Launch Firefox and wait for openH264 plugin to install
2. Navigate to http://mozilla.github.io/webrtc-landing/pc_test_h264.html
3. Click Play on the main pc1 video (as in http://goo.gl/qDrQCC)
4. Click Start and Share Selected Devices

Regression range:
Last good revision: 5b64c42742cd (2014-07-21)
First bad revision: 63f44b4968c2 (2014-07-22)
Pushlog:
https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=5b64c42742cd&tochange=63f44b4968c2

Notes:
1. Reproducible with latest Nightly and Aurora builds.
2. Couldn't reproduce with Intel CPU, only with AMD.

Comment 1

[Georg Fritzsche [:gfritzsche]]

Alexandra, did you have any luck with the m-i regression range?

Comment 2

[Ada [:adalucinet]]

(In reply to Georg Fritzsche [:gfritzsche] from comment #1)
> Alexandra, did you have any luck with the m-i regression range?

This is what I found for (m-i) regression range:

Last good revision: 7bafd7701140
First bad revision: fee5c4bdd713
Pushlog:
https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=7bafd7701140&tochange=fee5c4bdd713

Comment 3

[Georg Fritzsche [:gfritzsche]]

(In reply to Alexandra Lucinet, QA Mentor [:adalucinet] from comment #2)
> This is what I found for (m-i) regression range:
> 
> Last good revision: 7bafd7701140
> First bad revision: fee5c4bdd713
> Pushlog:
> https://hg.mozilla.org/integration/mozilla-inbound/
> pushloghtml?fromchange=7bafd7701140&tochange=fee5c4bdd713

Hm, i don't see anything obvious in there. It might be timing sensitive?
Any chance you could use a machine to continue to a specific m-i changeset?

Comment 4

[Randell Jesup [:jesup] (needinfo me)]

So, the inbound is just a merge from m-c?  Nothing there stands out.  Also, it being AMD-only is very unusual.

Comment 5

[Ada [:adalucinet]]

I manually tested with builds from http://ftp.mozilla.org/pub/mozilla.org/firefox/tinderbox-builds/mozilla-inbound-win32/ and found:

1405989399 - Good; https://hg.mozilla.org/integration/mozilla-inbound/rev/fee5c4bdd713
1405990828 - Bad; https://hg.mozilla.org/integration/mozilla-inbound/rev/fd4c9f4cd1f0

My Windows machine is not setup to bisect further, so I hope this helps.

Comment 6

[Randell Jesup [:jesup] (needinfo me)]

Signature strongly points to bug 1044245