Closed Bug 792300 Opened 12 years ago Closed 12 years ago

armv6 builds of fennec with gcc-4.4.3 fail test_bug560780.html in mochitest-1 which passes in armv7 builds

Categories

(Firefox for Android Graveyard :: General, defect)

Product:
Firefox for Android Graveyard
Component:
General
Platform:
ARM
Android
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(firefox16 affected, firefox17+ fixed)

Status:
RESOLVED FIXED
Tracking Flags:
Tracking Status
firefox16 --- affected
firefox17 + fixed

People

(Reporter: philor, Assigned: bugs)

References

Details

(Whiteboard: [armv6]) 

Presents a pretty little problem, since we promised that the armv6 tests would replace the noion tests, but we forgot to mention that they wouldn't replace mochitest-1 because there's permaorange in it, and we don't have any convenient way to turn that test off for armv6 without turning it off for armv7. So, armv6 mochitest-1 is hidden on every trunk tree, and on aurora and on beta.

https://tbpl.mozilla.org/php/getParsedLog.php?id=15326697&tree=Mozilla-Inbound
Android Armv6 Tegra 250 mozilla-inbound opt test mochitest-1 on 2012-09-18 19:01:02 PDT for push a04dc1a6fe29
slave: tegra-096

25883 INFO TEST-START | /tests/content/base/test/test_bug560780.html
25884 INFO TEST-INFO | /tests/content/base/test/test_bug560780.html | must wait for load
25885 ERROR TEST-UNEXPECTED-FAIL | /tests/content/base/test/test_bug560780.html | got the mousedown events - got 1, expected 12
25886 INFO TEST-END | /tests/content/base/test/test_bug560780.html | finished in 875ms
Blocks: 784681
We're going to be getting mats access to an ARMv6 phone to try to reproduce and fix for FF17.
Assignee: nobody → matspal
tracking-firefox17: --- → +
This is likely the same underlying tool chain bug as explained in bug
790624 comment 17.
Assignee: matspal → nobody
Whiteboard: [permaorange] → [permaorange][armv6]
As with bug 790624 over to Jet to help decide next steps here, we're still tracking this for 17 but if there's a case to be made that these tests are not key to our being able to ship a stable ARMv6 build when we release Firefox 17, please let us know.
Assignee: nobody → bugs
The latest failure is for Fx-team rev 8116fd47e9d8, which doesn't have
the fix in bug 804641.  The next rev after it does.

I'm not sure if this failure is related to bug 804641 though. Let's wait
a bit and see if there's a new failure with that fix...
It is related - it was permaorange before, and is permagreen after.
should be addressed by bug 804641 - adding qawanted to confirm.
Keywords: qawanted, verifyme
Looks good to me. Android Armv6 opt M(1) tests don't seem to be failing in this test.
Keywords: qawanted
Cleaning up/fixing whiteboards containing 'orange' but not '[orange]' in advance of bug 791758.
Whiteboard: [permaorange][armv6] → [armv6]
No longer needs to block bug 784681, since Armv6 tests are not hidden, since we worked around it for now.
No longer blocks: 784681
The toolchain upgrades in bug 825453 mean that this bug is no longer a problem. Resolving as fixed.
Status: NEW → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
Summary: armv6 builds of fennec fail test_bug560780.html in mochitest-1 which passes in armv7 builds → armv6 builds of fennec with gcc-4.4.3 fail test_bug560780.html in mochitest-1 which passes in armv7 builds
Keywords: verifyme
Product: Firefox for Android → Firefox for Android Graveyard
You need to log in before you can comment on or make changes to this bug.