Bug 162567

Summary: Intermittent failures in ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js
Product: WebKit Reporter: Alexey Proskuryakov <ap>
Component: JavaScriptCoreAssignee: Mark Lam <mark.lam>
Status: RESOLVED DUPLICATE    
Severity: Normal CC: commit-queue, fpizlo, ggaren, mark.lam, mcatanzaro, msaboff, ryanhaddad, saam, webkit-bug-importer
Priority: P2 Keywords: InRadar
Version: Safari 9   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Skip the test none

Alexey Proskuryakov
Reported 2016-09-26 10:31:01 PDT
I've seen ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default fail more than once. ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: DIFF FAILURE! ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: --- ../.tests/ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.baseline 2016-09-24 18:04:02.000000000 -0700 ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +++ ../ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default.out 2016-09-24 20:37:46.000000000 -0700 ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: @@ -18,18 +18,18 @@ ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa2[5]: { a: 101, b: 1, c: 2, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa2[6]: { a: 101, b: 1, c: 2, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa2[7]: { a: 101, b: 1, c: 2, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa2[8]: { a: 101, b: 1, c: 2, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa2[8]: { a: 0, b: 1, c: 2, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: Test 3: ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa3[0]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa3[1]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa3[2]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa3[3]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa3[4]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa3[5]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa3[6]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa3[7]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: -oa3[8]: { a: 0, b: 1, p: 100, x: 102, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa3[4]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa3[5]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa3[6]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa3[7]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: +oa3[8]: { a: 0, b: 1, p: 100, x: 10, y: 11 } ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: Test 4: ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default: oa4[0]: { a: 100, b: 101, c: 102, i: 108, j: 109, p: 200, x: 110, y: 111 } FAIL: ChakraCore.yaml/ChakraCore/test/fieldopts/objtypespec-newobj-invalidation.1.js.default
Attachments
Skip the test (1.51 KB, patch)
2016-12-01 15:55 PST, Ryan Haddad
no flags
Radar WebKit Bug Importer
Comment 1 2016-09-26 10:32:57 PDT
Mark Lam
Comment 2 2016-11-28 09:12:20 PST
*** Bug 165064 has been marked as a duplicate of this bug. ***
Mark Lam
Comment 3 2016-11-28 09:13:21 PST
This test is not flaky. There appears to be an intermittent bug that is causing the test failure.
Alexey Proskuryakov
Comment 4 2016-11-28 09:15:05 PST
When we say "flaky test", that doesn't imply that the test is broken. A much more common cause of flakyness is bugs in WebKit.
Alexey Proskuryakov
Comment 5 2016-12-01 13:22:36 PST
*** Bug 163692 has been marked as a duplicate of this bug. ***
Alexey Proskuryakov
Comment 6 2016-12-01 13:22:48 PST
*** Bug 163847 has been marked as a duplicate of this bug. ***
Alexey Proskuryakov
Comment 7 2016-12-01 13:24:21 PST
The test fails fairly frequently (right now, both debug and release bots are red because of it). A fix would be highly appreciated.
Ryan Haddad
Comment 8 2016-12-01 15:55:22 PST
Created attachment 295908 [details] Skip the test Patch to skip the test.
Saam Barati
Comment 9 2016-12-01 15:56:38 PST
Comment on attachment 295908 [details] Skip the test r=me
Ryan Haddad
Comment 10 2016-12-01 16:04:34 PST
Comment on attachment 295908 [details] Skip the test Clearing flags on attachment: 295908 Committed r209219: <http://trac.webkit.org/changeset/209219>
Saam Barati
Comment 11 2016-12-01 20:47:43 PST
My hypothesis for what the bug is is that we're failing to register for some watchpoint inside the DFG inside the bytecode parser when we optimize create_this to lower to NewObject instead of CreateThis
Michael Saboff
Comment 12 2017-06-13 15:42:36 PDT
Opened a new bug for the underlying problem <https://bugs.webkit.org/show_bug.cgi?id=173321> - "DFG doesn't properly handle a property that is change to read only in a prototype" Fixed the issue in change set r218203: <http://trac.webkit.org/changeset/218203>
Michael Saboff
Comment 13 2017-06-13 15:42:42 PDT
*** This bug has been marked as a duplicate of bug 173321 ***
Note You need to log in before you can comment on or make changes to this bug.