Bug 132022

Summary: REGRESSION: JSRegress's js/slow-stress/new-spread.html fails sometimes
Product: WebKit Reporter: Tim Horton <thorton>
Component: JavaScriptCoreAssignee: Oliver Hunt <oliver>
Status: RESOLVED FIXED    
Severity: Normal CC: ap, fpizlo, ggaren, mark.lam, mhahnenberg, msaboff, webkit-bug-importer
Priority: P2 Keywords: InRadar, LayoutTestFailure, Regression
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   

Description Tim Horton 2014-04-22 13:25:01 PDT
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)
Comment 1 Tim Horton 2014-04-22 13:28:22 PDT
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
Comment 2 Radar WebKit Bug Importer 2014-04-22 13:29:20 PDT
<rdar://problem/16690671>
Comment 3 Tim Horton 2014-04-22 13:41:29 PDT
Marked as flaky in http://trac.webkit.org/changeset/167680
Comment 4 Michael Saboff 2014-04-22 13:54:18 PDT
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.
Comment 5 Tim Horton 2014-04-24 15:39:07 PDT
Is this fixed by bug 132079? Can we un-flaky the test?
Comment 6 Alexey Proskuryakov 2014-04-25 10:43:17 PDT
Ping?
Comment 7 Oliver Hunt 2014-04-25 10:45:44 PDT
(In reply to comment #5)
> Is this fixed by bug 132079? Can we un-flaky the test?

yes1
Comment 8 Tim Horton 2014-04-25 11:20:55 PDT
(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