New issue
Advanced search Search tips

Issue 727994 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: May 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 1
Type: Bug-Regression



Sign in to add a comment

ServiceWorkerResourceStorageDiskTest.* are crashing

Project Member Reported by falken@chromium.org, May 31 2017

Issue description

$ out/Default/content_unittests --gtest_filter="ServiceWorkerResourceStorageDiskTest.*"



[==========] Running 4 tests from 1 test case.
[----------] Global test environment set-up.
[----------] 4 tests from ServiceWorkerResourceStorageDiskTest
[ RUN      ] ServiceWorkerResourceStorageDiskTest.CleanupOnRestart
[       OK ] ServiceWorkerResourceStorageDiskTest.CleanupOnRestart (1822 ms)
[ RUN      ] ServiceWorkerResourceStorageDiskTest.DeleteAndStartOver
[       OK ] ServiceWorkerResourceStorageDiskTest.DeleteAndStartOver (44 ms)
[ RUN      ] ServiceWorkerResourceStorageDiskTest.DeleteAndStartOver_UnrelatedFileExists
Received signal 11 SEGV_MAPERR fffffffc01326266
#0 0x7fd9b3a7abc7 base::debug::StackTrace::StackTrace()
#1 0x7fd9b3a7a73f base::debug::(anonymous namespace)::StackDumpSignalHandler()
#2 0x7fd9b4e63330 <unknown>
#3 0x7fd9b3b0b240 base::internal::(anonymous namespace)::SchedulerParallelTaskRunner::PostDelayedTask()
#4 0x7fd9b3b1c537 base::SequencedWorkerPool::Inner::PostTask()
#5 0x7fd9b3b1ef22 base::(anonymous namespace)::SequencedWorkerPoolTaskRunner::PostDelayedTask()
#6 0x7fd9b3b035e6 base::(anonymous namespace)::PostTaskAndReplyTaskRunner::PostTask()
#7 0x7fd9b3b19628 base::internal::PostTaskAndReplyImpl::PostTaskAndReply()
#8 0x7fd9b3b0352f base::TaskRunner::PostTaskAndReply()
#9 0x7fd9b133d7ee disk_cache::SimpleIndexFile::LoadIndexEntries()
#10 0x7fd9b1338b4c disk_cache::SimpleIndex::Initialize()
#11 0x7fd9b1328392 disk_cache::SimpleBackendImpl::InitializeIndex()
#12 0x7fd9b132b53b _ZN4base8internal7InvokerINS0_9BindStateIMN10disk_cache17SimpleBackendImplEFvRKNS_8CallbackIFviELNS0_8CopyModeE1ELNS0_10RepeatModeE1EEERKNS4_14DiskStatResultEEJNS_7WeakPtrIS4_EES9_EEEFvSE_EE3RunEPNS0_13BindStateBaseESE_
#13 0x7fd9b12a1e32 base::internal::ReplyAdapter<>()
#14 0x7fd9b12a1adb _ZN4base8internal7InvokerINS0_9BindStateIPFvNS_8CallbackIFvN3net10FileStream7Context10OpenResultEELNS0_8CopyModeE0ELNS0_10RepeatModeE0EEEPS7_EJSB_NS0_12OwnedWrapperIS7_EEEEEFvvEE7RunOnceEPNS0_13BindStateBaseE
#15 0x7fd9b3a65ef1 _ZNO4base8CallbackIFvvELNS_8internal8CopyModeE0ELNS2_10RepeatModeE0EE3RunEv
#16 0x7fd9b3b1998b base::(anonymous namespace)::PostTaskAndReplyRelay::RunReplyAndSelfDestruct()
#17 0x7fd9b3a65ef1 _ZNO4base8CallbackIFvvELNS_8internal8CopyModeE0ELNS2_10RepeatModeE0EE3RunEv
#18 0x7fd9b3a7b4e3 base::debug::TaskAnnotator::RunTask()
#19 0x7fd9b3aad96d base::MessageLoop::RunTask()
#20 0x7fd9b3aadc48 base::MessageLoop::DeferOrRunPendingTask()
#21 0x7fd9b3aae036 base::MessageLoop::DoWork()
#22 0x7fd9b3ab1159 base::MessagePumpLibevent::Run()
#23 0x7fd9b3aad5b2 base::MessageLoop::Run()
#24 0x7fd9b3ae34d7 base::RunLoop::Run()
#25 0x000001a648db net::internal::TestCompletionCallbackBaseInternal::WaitForResult()
#26 0x00000127419f content::(anonymous namespace)::WriteBasicResponse()
#27 0x0000012a650c content::ServiceWorkerResourceStorageTest::SetUp()
#28 0x000001918c0e testing::Test::Run()
#29 0x0000019196c0 testing::TestInfo::Run()
#30 0x000001919be7 testing::TestCase::Run()
#31 0x000001920d87 testing::internal::UnitTestImpl::RunAllTests()
#32 0x000001920a07 testing::UnitTest::Run()
#33 0x0000019fd991 base::TestSuite::Run()
#34 0x000001a04b00 base::(anonymous namespace)::LaunchUnitTestsInternal()
#35 0x000001a04974 base::LaunchUnitTests()
#36 0x0000016676c6 main

Rebasing then bisecting...
 

Comment 2 by falken@chromium.org, May 31 2017

git bisect start
# bad: [4eba08ea7191979e16f482ff78ce72597cbab6fa] Record histogram when succeeding in creating a DC layer overlay.
git bisect bad 4eba08ea7191979e16f482ff78ce72597cbab6fa
# good: [802480762c4dcb5e61bcafc21edf49002ca003f8] rebaseline-cl: If --fill-missing not given, prompt user to fill in missing results.
git bisect good 802480762c4dcb5e61bcafc21edf49002ca003f8
# bad: [10ae3c4b866897e7097ee9606d35b71f789e8532] Convert Spellcheck host MessageFilter IPC to mojo
git bisect bad 10ae3c4b866897e7097ee9606d35b71f789e8532
# bad: [dac60ab4c5c107622c519dfd58aaf0ff417d142d] Roll src/third_party/skia/ 6533159f2..f7beee7cd (45 commits)
git bisect bad dac60ab4c5c107622c519dfd58aaf0ff417d142d

Comment 3 by falken@chromium.org, May 31 2017

Cc: fdoray@chromium.org
Bisect result:
# bad: [e1aa28915a28da266d406202a07112d44d6cc018] Enable redirection of SequencedWorkerPool to TaskScheduler in TestBrowserThreadBundle.
https://codereview.chromium.org/2769693002

fdoray: does this look like a bug in your CL or did the CL uncover a bug in the tests?

Project Member

Comment 4 by bugdroid1@chromium.org, May 31 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/2a1abaac62e57773e4abe594f1de2b19dbb83e58

commit 2a1abaac62e57773e4abe594f1de2b19dbb83e58
Author: falken <falken@chromium.org>
Date: Wed May 31 05:48:24 2017

Revert of Enable redirection of SequencedWorkerPool to TaskScheduler in TestBrowserThreadBundle. (patchset #3 id:40001 of https://codereview.chromium.org/2769693002/ )

Reason for revert:
Caused ServiceWorkerResourceStorageDiskTest.* tests in content_unittests to crash.

See https://bugs.chromium.org/p/chromium/issues/detail?id=727994

Reverting locally fixed the crashes, so I'm going to revert this CL.

Original issue's description:
> Enable redirection of SequencedWorkerPool to TaskScheduler in TestBrowserThreadBundle.
>
> Motivation:
> - Redirection of SequencedWorkerPool to TaskScheduler is enabled in
>   browser tests and on Canary and Dev. It makes sense to enable it
>   in unit tests too.
> - It is not possible to wait until there are no more blocking pool
>   *and* TaskScheduler tasks (new tasks in the blocking pool during
>   a FlushForTesting() call on TaskScheduler and vice versa). If
>   the blocking pool is redirected to TaskScheduler, calling
>   FlushForTesting() on TaskScheduler guarantees that there are no
>   more blocking pool *and* TaskScheduler tasks.
>
> BUG= 667892 
>
> Review-Url: https://codereview.chromium.org/2769693002
> Cr-Commit-Position: refs/heads/master@{#475144}
> Committed: https://chromium.googlesource.com/chromium/src/+/e1aa28915a28da266d406202a07112d44d6cc018

TBR=gab@chromium.org,robliao@chromium.org,sky@chromium.org,fdoray@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG= 667892 , 727994 

Review-Url: https://codereview.chromium.org/2909393003
Cr-Commit-Position: refs/heads/master@{#475816}

[modify] https://crrev.com/2a1abaac62e57773e4abe594f1de2b19dbb83e58/content/public/test/test_browser_thread_bundle.cc

Comment 5 by falken@chromium.org, May 31 2017

Status: Fixed (was: Started)

Comment 6 by falken@chromium.org, May 31 2017

Labels: M-61

Sign in to add a comment