Starting some time yesterday, js/slow-stress/new-spread started failing like so: -PASS no exception thrown -PASS successfullyParsed is true +FAIL caught exception: Result was -1743984020 expected -2371153088 +FAIL successfullyParsed should be true (of type boolean). Was undefined (of type undefined). (see http://build.webkit.org/results/Apple%20Mavericks%20Release%20WK1%20%28Tests%29/r167675%20%285412%29/results.html for example)
This is a sometimes fail, as you can see here, but just started: http://webkit-test-results.appspot.com/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=js%2Fslow-stress%2Fnew-spread.html
<rdar://problem/16690671>
Marked as flaky in http://trac.webkit.org/changeset/167680
Running repeatedly, I can get this to fail ~25% of the time: $ for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20; do echo "Try #$i"; DYLD_FRAMEWORK_PATH=WebKitBuild/Release WebKitBuild/Release/jsc LayoutTests/js/slow-stress/script-tests/new-spread.js ; done Try #1 Try #2 Exception: Result was -872245192 expected -2371153088 Try #3 Try #4 Try #5 Exception: Result was 1614215892 expected -2371153088 Try #6 Try #7 Try #8 Try #9 Try #10 Try #11 Try #12 Exception: Result was 1614215892 expected -2371153088 Try #13 Try #14 Try #15 Try #16 Try #17 Exception: Result was -872245192 expected -2371153088 Try #18 Try #19 Try #20 Exception: Result was 1614215892 expected -2371153088 It happens less frequently with a debug build, roughly 10% of the time.
Is this fixed by bug 132079? Can we un-flaky the test?
Ping?
(In reply to comment #5) > Is this fixed by bug 132079? Can we un-flaky the test? yes1
(In reply to comment #7) > (In reply to comment #5) > > Is this fixed by bug 132079? Can we un-flaky the test? > > yes1 un-flaky'd in http://trac.webkit.org/changeset/167812