Intermittent TEST-UNEXPECTED-FAIL | valgrind-test | non-zero exit code

RESOLVED WORKSFORME

Status

()

Core
General
RESOLVED WORKSFORME
4 years ago
4 years ago

People

(Reporter: Tomcat, Unassigned)

Tracking

({intermittent-failure})

Trunk
x86
Linux
intermittent-failure
Points:
---
Dependency tree / graph

Firefox Tracking Flags

(Not tracked)

Details

(URL)

(Reporter)

Description

4 years ago
Linux x86-64 mozilla-inbound valgrind on 2014-01-18 04:45:34 PST for push 103c0b6c135d

slave: bld-linux64-ec2-096

https://tbpl.mozilla.org/php/getParsedLog.php?id=33224002&tree=Mozilla-Inbound



TEST-UNEXPECTED-FAIL | valgrind-test | non-zero exit code
The nice thing about having only one error message is that you'll get to see every single failure ;)
Comment hidden (Treeherder Robot)
> The nice thing about having only one error message is that you'll get to see
> every single failure ;)

Bug 957410 will open the door for better parsing that will avoid this problem.

In the meantime:

> ==20525== Invalid read of size 8
> ==20525==    at 0x80FC00F: mozilla::net::Seer::Observe(nsISupports*, char const*, char16_t const*) (Seer.cpp:355)
> ==20525==    by 0x80AC7EF: nsTimerImpl::Fire() (nsTimerImpl.cpp:559)
> ==20525==    by 0x80AC8C9: nsTimerEvent::Run() (nsTimerImpl.cpp:635)
> ==20525==    by 0x80A880C: nsThread::ProcessNextEvent(bool, bool*) (nsThread.cpp:637)
> ==20525==    by 0x80580F4: NS_ProcessNextEvent(nsIThread*, bool) (nsThreadUtils.cpp:263)
> ==20525==    by 0x80AD128: nsThread::Shutdown() (nsThread.cpp:488)
> ==20525==    by 0x8E79DE3: mozilla::dom::workers::RuntimeService::NoteIdleThread(mozilla::dom::workers::RuntimeService::WorkerThread*) (RuntimeService.cpp:2198)
> ==20525==    by 0x8E79F21: (anonymous namespace)::WorkerThreadPrimaryRunnable::FinishedRunnable::Run() (RuntimeService.cpp:2597)
> ==20525==    by 0x80A880C: nsThread::ProcessNextEvent(bool, bool*) (nsThread.cpp:637)
> ==20525==    by 0x80580F4: NS_ProcessNextEvent(nsIThread*, bool) (nsThreadUtils.cpp:263)
> ==20525==    by 0x80AD128: nsThread::Shutdown() (nsThread.cpp:488)
> ==20525==    by 0x8CC2E86: nsHtml5ParserThreadTerminator::Observe(nsISupports*, char const*, char16_t const*) (nsHtml5Module.cpp:100)
> ==20525==    by 0x808726B: nsObserverList::NotifyObservers(nsISupports*, char const*, char16_t const*) (nsObserverList.cpp:96)
> ==20525==    by 0x8087327: nsObserverService::NotifyObservers(nsISupports*, char const*, char16_t const*) (nsObserverService.cpp:302)
> ==20525==    by 0x805B69F: mozilla::ShutdownXPCOM(nsIServiceManager*) (nsXPComInit.cpp:708)
> ==20525==    by 0x805BB87: NS_ShutdownXPCOM (nsXPComInit.cpp:656)
> ==20525==    by 0x97B4512: ScopedXPCOMStartup::~ScopedXPCOMStartup() (nsAppRunner.cpp:1163)
> ==20525==    by 0x97BAB55: XREMain::XRE_main(int, char**, nsXREAppData const*) (nsAppRunner.cpp:4116)
> ==20525==    by 0x97BAD9E: XRE_main (nsAppRunner.cpp:4331)
> ==20525==    by 0x404933: do_main(int, char**, nsIFile*) (nsBrowserApp.cpp:280)
> ==20525==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
> ==20525==

This is a simple NULL-deref:

    gSeer->mIOThread->Dispatch(event, NS_DISPATCH_NORMAL);

Either |gSeer| or |gSeer->mIOThread| is NULL. This causes a segfault, and we enter Breakpad, and Breakpad does a clone() call that Valgrind can't handle.

philor, should I file a new bug for that, and leave this bug as the catch-all for other such problems?
> > ==20525== Invalid read of size 8
> > ==20525==    at 0x80FC00F: mozilla::net::Seer::Observe(nsISupports*, char const*, char16_t const*) (Seer.cpp:355)

I field bug 961860 for that.
Depends on: 961860
Comment hidden (Treeherder Robot)
(In reply to TBPL Robot from comment #5)

That's bug 961860 again.
Comment hidden (Treeherder Robot)
Comment hidden (Treeherder Robot)
(In reply to TBPL Robot from comment #8)
> TEST-UNEXPECTED-FAIL | valgrind-test | non-zero exit code

That is bug 961880 again.
(In reply to TBPL Robot from comment #7)
> 
> ==27371== ERROR SUMMARY: 101 errors from 101 contexts (suppressed: 71 from 65)
> TEST-UNEXPECTED-FAIL | valgrind-test | non-zero exit code

That's a new one. I filed bug 962921 for it.
Comment hidden (Treeherder Robot)
(In reply to TBPL Robot from comment #11)
> RyanVM
> https://tbpl.mozilla.org/php/getParsedLog.php?id=33537270&tree=Mozilla-
> Inbound
> Linux x86-64 mozilla-inbound valgrind on 2014-01-24 10:29:58
> revision: fc6dc7e3ae7c
> slave: bld-linux64-ec2-351
> 
> TEST-UNEXPECTED-FAIL | valgrind-test | non-zero exit code

Maybe relevant - NSPR_4_10_3_BETA3 landed a few pushes before this. Bug 936320.
Actually, this may be real bustage from bug 963077.
(In reply to Ryan VanderMeulen [:RyanVM UTC-5] from comment #13)
> Actually, this may be real bustage from bug 963077.

It was.
Now that the Valgrind errors are better distinguished on TBPL, we shouldn't need this bug any more.
Status: NEW → RESOLVED
Last Resolved: 4 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.