Filed by: wkocher [at] mozilla.com https://treeherder.mozilla.org/logviewer.html#?job_id=79745050&repo=autoland https://queue.taskcluster.net/v1/task/X7tk7Zq1RxqxYuXN5IJnUg/runs/0/artifacts/public/logs/live_backing.log https://hg.mozilla.org/mozilla-central/raw-file/tip/layout/tools/reftest/reftest-analyzer.xhtml#logurl=https://queue.taskcluster.net/v1/task/X7tk7Zq1RxqxYuXN5IJnUg/runs/0/artifacts/public/logs/live_backing.log&only_show_unexpected=1
38 failures in 146 pushes (0.26 failures/push) were associated with this bug yesterday. Repository breakdown: * autoland: 34 * mozilla-central: 4 Platform breakdown: * linux64-stylo: 37 * osx-10-10: 1 For more details, see: https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1342228&startday=2017-02-23&endday=2017-02-23&tree=all
The first assertion is ASSERTION: Disagreement about whether it's a block or not: 'fromLine->IsBlock() == fromLine->mFirstChild->IsBlockOutside()', file /home/worker/workspace/build/src/layout/generic/nsBlockFrame.cpp, line 2877 the rest are ASSERTION: How'd we get a floated inline frame? The frame ctor should've dealt with this.: '!aReflowInput.IsFloating()', file /home/worker/workspace/build/src/layout/generic/nsLineLayout.cpp, line 1173
:tylin - This failure started shortly after your landing in bug 775624. I don't see a strong association, but wonder if you think your changes might be related?
I see now that object-fit-scale-down-png-001o.html was skipped on stylo, until it was re-enabled in https://hg.mozilla.org/integration/autoland/rev/75228b956df12f72c324e25be2602202eb4ab6db, bug 1341988. With 38 failures on its first day, we should fix or re-skip this right away.
Flags: needinfo?(tlin) → needinfo?(bzbarsky)
If the fix for bug 1322317 (which per bug 1342188 should fix these assertions altogether) didn't make this stop happening, then we should just re-disable. If it _did_ make this stop happening, we should probably reeable a bunch more of the object-fit and image-fit tests instead.... ;)
Flags: needinfo?(bzbarsky) → needinfo?(emilio+bugs)
48 failures in 812 pushes (0.059 failures/push) were associated with this bug in the last 7 days. This is the #30 most frequent failure this week. ** This failure happened more than 30 times this week! Resolving this bug is a high priority. ** ** Try to resolve this bug as soon as possible. If unresolved for 2 weeks, the affected test(s) may be disabled. ** Repository breakdown: * autoland: 39 * mozilla-central: 5 * mozilla-inbound: 4 Platform breakdown: * linux64-stylo: 47 * osx-10-10: 1 For more details, see: https://brasstacks.mozilla.com/orangefactor/?display=Bug&bugid=1342228&startday=2017-02-20&endday=2017-02-26&tree=all
We should re-disable this test; it's still happening.
Actually, I see no new failures since Feb 24...maybe all's well?
Bug 1342342 is certainly happening, and I expect this is the same problem fundamentally... Maybe the frequency is lower, though.
This one should be disabled too right? http://searchfox.org/mozilla-central/source/layout/reftests/w3c-css/submitted/images3/reftest-stylo.list#47.
Hmm, yes. So it was only enabled for a few hours on the 23rd...
Status: NEW → RESOLVED
Last Resolved: a year ago
Resolution: --- → FIXED
status-firefox52: --- → unaffected
status-firefox53: --- → unaffected
status-firefox54: --- → fixed
You need to log in before you can comment on or make changes to this bug.