Open Bug 742682 Opened 13 years ago Updated 2 years ago

crashtest: 329884-1.xul leaks to next test (330010-1.xul), and 4 other leaking tests too

Categories

(Core :: XUL, defect)

defect

Tracking

()

mozilla14
Tracking Status
firefox11 --- wontfix
firefox12 --- affected
firefox13 --- affected
firefox-esr10 --- wontfix

People

(Reporter: sgautherie, Unassigned)

References

(Depends on 1 open bug, )

Details

https://tbpl.mozilla.org/php/getParsedLog.php?id=10662095&tree=Firefox&full=1 Rev3 WINNT 5.1 mozilla-central debug test crashtest on 2012-04-05 03:23:37 PDT for push ac5721fa4acb + http://tinderbox.mozilla.org/showlog.cgi?log=SeaMonkey/1333608083.1333608571.30263.gz WINNT 5.2 comm-central-trunk debug test crashtest on 2012/04/04 23:41:23 { REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/content/xul/templates/src/crashtests/329884-1.xul | (LOAD ONLY) REFTEST TEST-START | http://localhost:4444/1333621541986/1/330010-1.xul | 391 / 2040 (19%) Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/content/xul/templates/src/crashtests/329884-1.xul may not load data from file:///c:/talos-slave/test/build/reftest/tests/content/xul/templates/src/crashtests/1.rdf. JavaScript error: http://localhost:4444/1333621541986/1/330010-1.xul, line 14: generatedShape is undefined REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/content/xul/templates/src/crashtests/330010-1.xul | (LOAD ONLY) } *** Test added by http://hg.mozilla.org/mozilla-central/rev/1c28920d2a30 "Power can corrupt. So can freeing objects twice. Adding a crashtest." 329884-1.xul leaks to next test (330010-1.xul): it misses http://mxr.mozilla.org/comm-central/source/mozilla/layout/tools/reftest/README.txt#373
Just adding reftest-wait wouldn't solve this, because the dynamic portion runs in a load event (there are no timeouts). How do you wait for RDF to load/error out?
(In reply to Jesse Ruderman from comment #1) Then I have no answers :-/ Fwiw, http://mxr.mozilla.org/mozilla-central/search?string=datasources%3D&case=1 wrt 'rdf:null'. *** Ftr, http://mxr.mozilla.org/mozilla-central/search?string=datasources%3D&case=1&find=crashtest There are a few other "similar" crashtests: some with non-existent rdf datasources too, some that have code that runs off the load event too, none have 'reftest-wait'. They should be checked too. If there is no solution, then tests which misbehave should be disabled. ***** 4 other tests which leak, from that FF log: { REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/561981-1.html | (LOAD ONLY) Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/561981-1.html may not load data from data:text/xml,%3Cbindings%20xmlns%3D%22http%3A%2F%2Fwww.mozilla.org%2Fxbl%22%3E%3Cbinding%20id%3D%22foo%22%20g%3D%220.9988319823444692%22%3E%3Ccontent%3E%0A%3C%2Fcontent%3E%3C%2Fbinding%3E%3C%2Fbindings%3E%0A. [...] REFTEST TEST-START | file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/561981-2.html | 75 / 2040 (3%) REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/561981-2.html | (LOAD ONLY) Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/561981-2.html may not load data from data:text/xml,%3Cbindings%20xmlns%3D%22http%3A%2F%2Fwww.mozilla.org%2Fxbl%22%3E%3Cbinding%20id%3D%22foo%22%20g%3D%220.11745441848751925%22%3E%3Ccontent%3E%0A%0A%3C%2Fcontent%3E%3C%2Fbinding%3E%3C%2Fbindings%3E%0A. [...] REFTEST TEST-START | file:///c:/talos-slave/test/build/reftest/tests/content/base/crashtests/564079-1.html | 76 / 2040 (3%) REFTEST TEST-START | file:///c:/talos-slave/test/build/reftest/tests/layout/style/crashtests/665209-1.html | 1479 / 2040 (72%) Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/layout/style/crashtests/645951-1.html may not load or link to chrome:///browser/skin/. REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/layout/style/crashtests/665209-1.html | (LOAD ONLY) REFTEST TEST-START | file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/614367-1.svg | 1594 / 2040 (78%) Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/612662-1.svg may not load data from file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/612662-1.svg#a. REFTEST TEST-PASS | file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/614367-1.svg | (LOAD ONLY) }
Summary: crashtest: 329884-1.xul leaks to next test (330010-1.xul) → crashtest: 329884-1.xul leaks to next test (330010-1.xul), and 4 other leaking tests too
Why do you think this is a problem? I don't see any test failures in the log you linked to.
Depends on: 742830
(In reply to Jesse Ruderman from comment #3) > Why do you think this is a problem? Why do you think it can't be (= cause or hide) a problem? > I don't see any test failures in the log you linked to. For example, { JavaScript error: http://localhost:4444/1333621541986/1/330010-1.xul, line 14: generatedShape is undefined } and { Security Error: Content at file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/612662-1.svg may not load data from file:///c:/talos-slave/test/build/reftest/tests/layout/svg/crashtests/612662-1.svg#a. } seem doubtful enough to me...
Meh, it's not causing Tinderbox oranges now. I don't think it makes sense to disable the tests over this.
Priority: P2 → --
Whiteboard: [good first bug][mentor=sgautherie][lang=js]
Severity: normal → S3
You need to log in before you can comment on or make changes to this bug.