SitePerProcessBrowserTest.SubframeKeyboardEventRouting flakily segfaults / asserts |
|||||
Issue descriptionVersion: OS: Mac See for instance https://build.chromium.org/p/chromium.mac/builders/Mac10.9%20Tests%20%28dbg%29/builds/21343 What is the expected output? What do you see instead? [57322:1287:0321/083839:29312363708713:WARNING:histogram_base.cc(131)] 2 histograms were created before reporting was enabled. BrowserTestBase signal handler received SIGTERM. Backtrace: 0 libbase.dylib 0x00000001078d35ef _ZN4base5debug10StackTraceC2Ev + 47 1 libbase.dylib 0x00000001078d3793 _ZN4base5debug10StackTraceC1Ev + 35 2 content_browsertests 0x00000001054779ca _ZN7content12_GLOBAL__N_1L27DumpStackTraceSignalHandlerEi + 74 3 libsystem_platform.dylib 0x00007fff886165aa _sigtramp + 26 4 ??? 0x00007ff8449085d0 0x0 + 140704278939088 5 CoreFoundation 0x00007fff8f092f15 __CFRunLoopServiceMachPort + 181 6 CoreFoundation 0x00007fff8f092539 __CFRunLoopRun + 1161 7 CoreFoundation 0x00007fff8f091e75 CFRunLoopRunSpecific + 309 8 HIToolbox 0x00007fff8b4caa0d RunCurrentEventLoopInMode + 226 9 HIToolbox 0x00007fff8b4ca7b7 ReceiveNextEventCommon + 479 10 HIToolbox 0x00007fff8b4ca5bc _BlockUntilNextEventMatchingListInModeWithFilter + 65 11 AppKit 0x00007fff8f25a24e _DPSNextEvent + 1434 12 AppKit 0x00007fff8f25989b -[NSApplication nextEventMatchingMask:untilDate:inMode:dequeue:] + 122 13 AppKit 0x00007fff8f24d99c -[NSApplication run] + 553 14 libbase.dylib 0x000000010787fd62 _ZN4base24MessagePumpNSApplication5DoRunEPNS_11MessagePump8DelegateE + 306 15 libbase.dylib 0x000000010787e65d _ZN4base24MessagePumpCFRunLoopBase3RunEPNS_11MessagePump8DelegateE + 125 16 libbase.dylib 0x00000001079ab0ca _ZN4base11MessageLoop10RunHandlerEv + 298 17 libbase.dylib 0x0000000107a4c525 _ZN4base7RunLoop3RunEv + 85 18 content_browsertests 0x0000000105669ae4 _ZN7content14RunThisRunLoopEPN4base7RunLoopE + 100 19 content_browsertests 0x000000010566aca5 _ZN7content17MessageLoopRunner3RunEv + 69 20 content_browsertests 0x00000001051e6352 _ZN7content12_GLOBAL__N_120FrameFocusedObserver4WaitEv + 50 21 content_browsertests 0x00000001051b3653 _ZN7content12_GLOBAL__N_110FocusFrameEPNS_13FrameTreeNodeE + 83 22 content_browsertests 0x00000001051bb96d _ZN7content59SitePerProcessBrowserTest_SubframeKeyboardEventRouting_Test19RunTestOnMainThreadEv + 2125 23 content_browsertests 0x000000010549c0a2 _ZN7content18ContentBrowserTest23RunTestOnMainThreadLoopEv + 402 24 content_browsertests 0x000000010547775b _ZN7content15BrowserTestBase28ProxyRunTestOnMainThreadLoopEv + 331 25 content_browsertests 0x0000000105479a07 _ZN4base8internal15RunnableAdapterIMN7content15BrowserTestBaseEFvvEE3RunIJEEEvPS3_DpOT_ + 119 26 content_browsertests 0x000000010547991d _ZN4base8internal12InvokeHelperILb0EvNS0_15RunnableAdapterIMN7content15BrowserTestBaseEFvvEEEE8MakeItSoIJPS4_EEEvS7_DpOT_ + 61 27 content_browsertests 0x00000001054798b8 _ZN4base8internal7InvokerINS_13IndexSequenceIJLm0EEEENS0_9BindStateINS0_15RunnableAdapterIMN7content15BrowserTestBaseEFvvEEEFvPS7_EJSB_EEENS0_12InvokeHelperILb0EvSA_EEFvvEE3RunEPNS0_13BindStateBaseE + 104 28 content_browsertests 0x0000000104d2441f _ZNK4base8CallbackIFvvELNS_8internal8CopyModeE1EE3RunEv + 63 29 content_browsertests 0x00000001055948ac _ZN7content21ShellBrowserMainParts21PreMainMessageLoopRunEv + 844 30 libcontent.dylib 0x000000010bba9760 _ZN7content15BrowserMainLoop21PreMainMessageLoopRunEv + 528 31 libcontent.dylib 0x000000010bbb0017 _ZN4base8internal15RunnableAdapterIMN7content15BrowserMainLoopEFivEE3RunIJEEEiPS3_DpOT_ + 119 32 libcontent.dylib 0x000000010bbaff1d _ZN4base8internal12InvokeHelperILb0EiNS0_15RunnableAdapterIMN7content15BrowserMainLoopEFivEEEE8MakeItSoIJPS4_EEEiS7_DpOT_ + 61 33 libcontent.dylib 0x000000010bbafeb3 _ZN4base8internal7InvokerINS_13IndexSequenceIJLm0EEEENS0_9BindStateINS0_15RunnableAdapterIMN7content15BrowserMainLoopEFivEEEFiPS7_EJNS0_17UnretainedWrapperIS7_EEEEENS0_12InvokeHelperILb0EiSA_EEFivEE3RunEPNS0_13BindStateBaseE + 115 34 libcontent.dylib 0x000000010c5385bf _ZNK4base8CallbackIFivELNS_8internal8CopyModeE1EE3RunEv + 63 35 libcontent.dylib 0x000000010cbcdfb3 _ZN7content17StartupTaskRunner14RunAllTasksNowEv + 531 36 libcontent.dylib 0x000000010bba64af _ZN7content15BrowserMainLoop18CreateStartupTasksEv + 1983 37 libcontent.dylib 0x000000010bbb3305 _ZN7content21BrowserMainRunnerImpl10InitializeERKNS_18MainFunctionParamsE + 2021 38 content_browsertests 0x000000010559336d _Z16ShellBrowserMainRKN7content18MainFunctionParamsERKNSt3__110unique_ptrINS_17BrowserMainRunnerENS3_14default_deleteIS5_EEEE + 141 39 content_browsertests 0x00000001054c0437 _ZN7content17ShellMainDelegate10RunProcessERKNSt3__112basic_stringIcNS1_11char_traitsIcEENS1_9allocatorIcEEEERKNS_18MainFunctionParamsE + 983 40 libcontent.dylib 0x000000010b884c2c _ZN7content23RunNamedProcessTypeMainERKNSt3__112basic_stringIcNS0_11char_traitsIcEENS0_9allocatorIcEEEERKNS_18MainFunctionParamsEPNS_19ContentMainDelegateE + 524 41 libcontent.dylib 0x000000010b886c42 _ZN7content21ContentMainRunnerImpl3RunEv + 626 42 libcontent.dylib 0x000000010b8842cd _ZN7content11ContentMainERKNS_17ContentMainParamsE + 349 43 content_browsertests 0x00000001054773b2 _ZN7content15BrowserTestBase5SetUpEv + 2418 44 content_browsertests 0x000000010549bce1 _ZN7content18ContentBrowserTest5SetUpEv + 1009 45 content_browsertests 0x000000010587ba78 _ZN7testing8internal38HandleSehExceptionsInMethodIfSupportedINS_4TestEvEET0_PT_MS4_FS3_vEPKc + 136 46 content_browsertests 0x000000010585041c _ZN7testing8internal35HandleExceptionsInMethodIfSupportedINS_4TestEvEET0_PT_MS4_FS3_vEPKc + 124 47 content_browsertests 0x00000001058502e1 _ZN7testing4Test3RunEv + 129 48 content_browsertests 0x00000001058516a6 _ZN7testing8TestInfo3RunEv + 230 49 content_browsertests 0x0000000105852d85 _ZN7testing8TestCase3RunEv + 245 50 content_browsertests 0x0000000105861c8a _ZN7testing8internal12UnitTestImpl11RunAllTestsEv + 922 51 content_browsertests 0x000000010587f138 _ZN7testing8internal38HandleSehExceptionsInMethodIfSupportedINS0_12UnitTestImplEbEET0_PT_MS4_FS3_vEPKc + 136 52 content_browsertests 0x000000010586184f _ZN7testing8internal35HandleExceptionsInMethodIfSupportedINS0_12UnitTestImplEbEET0_PT_MS4_FS3_vEPKc + 127 53 content_browsertests 0x00000001058616e9 _ZN7testing8UnitTest3RunEv + 393 54 content_browsertests 0x00000001056e8933 _Z13RUN_ALL_TESTSv + 35 55 content_browsertests 0x00000001056e7df9 _ZN4base9TestSuite3RunEv + 713 56 content_browsertests 0x00000001054bae00 _ZN7content27ContentTestLauncherDelegate12RunTestSuiteEiPPc + 64 57 content_browsertests 0x0000000105654e6d _ZN7content11LaunchTestsEPNS_20TestLauncherDelegateEiiPPc + 829 58 content_browsertests 0x00000001054bac26 main + 214 59 content_browsertests 0x0000000104c38874 start + 52 60 ??? 0x0000000000000008 0x0 + 8 Please use labels and text to provide additional information.
,
Mar 21 2016
Actually, it appears to be flaky on all platforms since a while back (can't find a clear culprit to revert), but on the other platforms it's an assert failing instead. I'll disable the test everywhere for now.
,
Mar 21 2016
,
Mar 21 2016
+kenrb, alexmos
,
Mar 21 2016
I'll take a look and see if I can repro locally. Looking at the flakiness dashboard, I actually don't see any failures on Linux and very low rate of flakiness on Windows. Do we have to disable it on all platforms rather than just Mac (or Mac+Windows), or did I miss something on the dashboard?
,
Mar 21 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/89f536fa94877c313a302c00b1dbc06e48be8384 commit 89f536fa94877c313a302c00b1dbc06e48be8384 Author: phoglund <phoglund@chromium.org> Date: Mon Mar 21 16:47:09 2016 Disabling flaky SitePerProcessBrowserTest.SubframeKeyboardEventRouting TBR=kinuko@chromium.org BUG= 596508 Review URL: https://codereview.chromium.org/1820953002 Cr-Commit-Position: refs/heads/master@{#382307} [modify] https://crrev.com/89f536fa94877c313a302c00b1dbc06e48be8384/content/browser/site_per_process_browsertest.cc
,
Mar 21 2016
http://chromium-build-logs.appspot.com/gtest_query?gtest_query=SitePerProcessBrowserTest.SubframeKeyboardEventRouting says there were a few Linux flakes, for instance Feb. 18, 2016, 2:34 p.m. http://build.chromium.org/p/chromium.linux/json Linux Tests I don't know which source is the most trustworthy.
,
Mar 21 2016
I haven't been able to repro this locally on 10.11. The failing bot is 10.9, so perhaps that makes a difference. I did manage to hit the failing assert a couple of times, which is: [ RUN ] SitePerProcessBrowserTest.SubframeKeyboardEventRouting e:\b\build\slave\win_x64_gn\build\src\content\browser\site_per_process_browsertest.cc(4926): error: Value of: result Actual: "input-blur" Expected: "FOO" That happens if I enable pixel output and randomly click around on the frames in the test window to screw up focus. So my suspicion is that something else on the bot is tinkering with focus, and my plan is to move this test to interactive_ui_tests, similarly to what I'm also planning to do for issue 559273 . At that point, we can reevaluate whether that helps with the Mac problem as well.
,
Mar 21 2016
In the meantime, I'm wondering if it's ok to reenable this for Linux, as the last Linux flakes were almost a month ago, and this is an important area where we want to maintain at least some coverage.
,
Mar 22 2016
Re #9: sounds reasonable to me (but watch out for more flakes).
,
Apr 12 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/e16154fa1a98b487a9e04feaad87fd31a3ae3fa4 commit e16154fa1a98b487a9e04feaad87fd31a3ae3fa4 Author: alexmos <alexmos@chromium.org> Date: Tue Apr 12 22:20:26 2016 Move the SubframeKeyboardEventRouting test to interactive_ui_tests. Since this test heavily relies on correctly focused frames, it should be more reliable as an interactive test. Hopefully, this will help with flakiness that has been recently observed in https://crbug.com/596508 . BUG= 596508 Review URL: https://codereview.chromium.org/1869853002 Cr-Commit-Position: refs/heads/master@{#386833} [modify] https://crrev.com/e16154fa1a98b487a9e04feaad87fd31a3ae3fa4/chrome/browser/site_per_process_interactive_browsertest.cc [modify] https://crrev.com/e16154fa1a98b487a9e04feaad87fd31a3ae3fa4/content/browser/site_per_process_browsertest.cc [modify] https://crrev.com/e16154fa1a98b487a9e04feaad87fd31a3ae3fa4/content/test/data/frame_tree/page_with_one_frame.html
,
Apr 13 2016
I hope that r386833 fixed this. So far, there are no failures on the flakiness dashboard: http://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=interactive_ui_tests&tests=SubframeKeyboardEventRouting |
|||||
►
Sign in to add a comment |
|||||
Comment 1 by phoglund@chromium.org
, Mar 21 2016Labels: OS-Mac