Closed Bug 611164 Opened 15 years ago Closed 15 years ago

Intermittent failure in 580160-1.html | reftest unexpected-fail | image comparison (==)

Categories

(Core :: Layout, defect)

x86
Windows 7
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: Felipe, Assigned: roc)

References

Details

(Keywords: intermittent-failure)

Attachments

(4 files)

WINNT 5.2 mozilla-central debug test reftest http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1289428702.1289431399.9503.gz s: win32-slave36 REFTEST TEST-UNEXPECTED-FAIL | file:///e:/builds/moz2_slave/mozilla-central-win32-debug-unittest-reftest/build/reftest/tests/layout/reftests/bugs/580160-1.html | PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found)
WINNT 5.2 mozilla-central debug test reftest http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1289434160.1289437757.10210.gz s: win32-slave17 REFTEST TEST-UNEXPECTED-FAIL | file:///e:/builds/moz2_slave/mozilla-central-win32-debug-unittest-reftest/build/reftest/tests/layout/reftests/bugs/580160-1.html | PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found) PROCESS-CRASH | Main app process exited normally | application crashed (minidump found)
Even though this didn't show up on Maple (at least not while I was watching), it seems likely to be async plugins fallout.
The images in the log files indicate the same error for comment 0 and 1.
Attached image Test and reference —
http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1289456795.1289460240.21740.gz WINNT 5.2 mozilla-central debug test reftest on 2010/11/10 22:26:35 s: win32-slave12
Blocks: 438871
Whiteboard: [orange]
Blocks: 580160
I don't think we need a layout flush, the harness will do that already. This is a problem with async plugin painting. We simply haven't received the new plugin surface yet. I have a patch we can try.
Attached patch potential fix — — Splinter Review
This should work.
Assignee: nobody → roc
Attachment #491634 - Flags: review?
Attachment #491634 - Flags: review? → review?(romaxa)
Comment on attachment 491634 [details] [diff] [review] potential fix is it for windows only? Fix seems make sense, but I wondering why it was working fine on linux before... did we change logic for Windows plugin layers?
Attachment #491634 - Flags: review?(romaxa) → review+
Maybe we were just lucky before.
Whiteboard: [orange] → [orange][needs landing]
Summary: Intermittent failure in 580160-1.html | reftest unexpected-fail → Intermittent failure in 580160-1.html | reftest unexpected-fail | image comparison (==)
Blocks: 615235
Pushed http://hg.mozilla.org/mozilla-central/rev/12c31d0065ca I took the [needs landing] as a=roc.
Whiteboard: [orange][needs landing] → [orange]
That probably wasn't a safe assumption, but this was test-only (well, mostly) so I think this was OK. This bug still needs bug 615417 though.
(In reply to comment #85) > That probably wasn't a safe assumption Noted. It occurred to me after I'd pushed, so I thought I'd better at least note what I'd done.
This should be fixed now that this patch and the patch in bug 615417 have landed.
Status: NEW → RESOLVED
Closed: 15 years ago
Resolution: --- → FIXED
Blocks: 803473
Whiteboard: [orange]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: