Upon some further investigation it turns out that we were hitting bug
840260, it got fixed, and now we are hitting something completely different.
The new assertion is:
F/MOZ_Assert( 642): Assertion failure: size_t(data) % sizeof(Value) ==
0, at
/builds/slave/ced-ics_a7_g-d-000000000000000/build/js/src/jsscript.cpp:1751
Full log:
https://tbpl.mozilla.org/php/getParsedLog.php?id=19742043&tree=Cedar&full=1
On 02/14/2013 03:12 PM, Andrew Halberstadt wrote:
And just as I post I see
https://bugzilla.mozilla.org/show_bug.cgi?id=840260
Please disregard.
On 02/14/2013 03:10 PM, Andrew Halberstadt wrote:
[copy from bug 841467]
On cedar and b2g18 we are running debug b2g builds on emulators.
Sometime on Feb. 9th a check-in caused every job (on cedar) to turn
perma-red with a logcat like:
https://tbpl.mozilla.org/php/getParsedLog.php?id=19606312&tree=Cedar&full=1
It looks like the b2g process is constantly seg faulting and re-starting
until the job gets killed. Of note might be the line:
"F/MOZ_Assert( 871): Assertion failure: false (The number of items in
sClassInfoData doesn't match the number of nsIDOMClassInfo ID's, this is
bad! Fix it!), at
/builds/slave/ced-ics-armv7a-g-dbg/build/dom/base/nsDOMClassInfo.cpp:3648"
Unfortunately we aren't running these on inbound or central so the best
I can do is say that it was somewhere in this push:
https://hg.mozilla.org/projects/cedar/pushloghtml?startID=51&endID=52 :(
Does anyone know what might be causing this? Thanks for your help!
Andrew
_______________________________________________
dev-b2g mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-b2g
