132022 2014-04-22 13:25:01 -0700 REGRESSION: JSRegress's js/slow-stress/new-spread.html fails sometimes 2014-04-25 11:21:50 -0700 1 1 1 Unclassified WebKit JavaScriptCore 528+ (Nightly build) Unspecified Unspecified RESOLVED FIXED InRadar, LayoutTestFailure, Regression P2 Normal --- 1 thorton oliver ap fpizlo ggaren mark.lam mhahnenberg msaboff webkit-bug-importer oldest_to_newest 1003045 0 thorton 2014-04-22 13:25:01 -0700 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) 1003046 1 thorton 2014-04-22 13:28:22 -0700 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 1003048 2 webkit-bug-importer 2014-04-22 13:29:20 -0700 <rdar://problem/16690671> 1003052 3 thorton 2014-04-22 13:41:29 -0700 Marked as flaky in http://trac.webkit.org/changeset/167680 1003056 4 msaboff 2014-04-22 13:54:18 -0700 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. 1003813 5 thorton 2014-04-24 15:39:07 -0700 Is this fixed by bug 132079? Can we un-flaky the test? 1004136 6 ap 2014-04-25 10:43:17 -0700 Ping? 1004138 7 oliver 2014-04-25 10:45:44 -0700 (In reply to comment #5) > Is this fixed by bug 132079? Can we un-flaky the test? yes1 1004150 8 thorton 2014-04-25 11:20:55 -0700 (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