609784 - [Linux] mochitests-5: test_alerts.html and test_alerts_noobserve.html report "ASSERTION: bad size recorded: 'aInstanceSize == 0 || entry->GetClassSize() == aInstanceSize', file .../nsTraceRefcntImpl.cpp, line 476"

Status: RESOLVED FIXED

(Toolkit :: General, defect)

Description

[Serge Gautherie (:sgautherie)]

Noticed while checking logs for bug 609772.

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1288903673.1288904505.16409.gz&fulltext=1
Rev3 Fedora 12 mozilla-central debug test mochitests-5/5 on 2010/11/04 13:47:53
http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1288903531.1288904243.15514.gz&fulltext=1
Rev3 Fedora 12x64 mozilla-central debug test mochitests-5/5 on 2010/11/04 13:45:31
{
###!!! ASSERTION: bad size recorded: 'aInstanceSize == 0 || entry->GetClassSize() == aInstanceSize', file /builds/slave/mozilla-central-linux64-debug/build/xpcom/base/nsTraceRefcntImpl.cpp, line 476
GetBloatEntry [xpcom/base/nsTraceRefcntImpl.cpp:479]
NS_LogAddRef_P [xpcom/base/nsTraceRefcntImpl.cpp:1001]
NS_LogAddRef [xpcom/stub/nsXPComStub.cpp:279]
nsAlertsService::AddRef [toolkit/system/gnome/nsAlertsService.cpp:42]
...
}

Comment 1

[Serge Gautherie (:sgautherie)]

"blocking2.0=?"
assertion in nsTraceRefcntImpl.cpp seems rather unexpected...

Comment 2

[Joe Drew (not getting mail)]

Bloat log problems don't need to block.

Comment 3

[Ian Neal]

Also get in suite/common/downloads/tests/test_action_keys_respect_focus.xul

###!!! ASSERTION: bad size recorded: 'aInstanceSize == 0 || entry->GetClassSize() == aInstanceSize', file /builds/slave/comm-cen-trunk-lnx-dbg/build/mozilla/xpcom/base/nsTraceRefcntImpl.cpp, line 476
GetBloatEntry [xpcom/base/nsTraceRefcntImpl.cpp:474]
NS_LogAddRef_P [xpcom/base/nsTraceRefcntImpl.cpp:1001]
NS_LogAddRef [xpcom/stub/nsXPComStub.cpp:279]
nsAlertsService::AddRef [toolkit/system/gnome/nsAlertsService.cpp:42]
nsAlertsServiceConstructor [toolkit/system/gnome/nsGnomeModule.cpp:56]
mozilla::GenericFactory::CreateInstance [GenericFactory.cpp:49]
nsComponentManagerImpl::CreateInstanceByContractID [xpcom/components/nsComponentManager.cpp:1309]
nsComponentManagerImpl::GetServiceByContractID [xpcom/components/nsComponentManager.cpp:1670]
CallGetService [nsComponentManagerUtils.cpp:95]
nsGetServiceByContractID::operator() [nsComponentManagerUtils.cpp:278]
nsCOMPtr<nsIAlertsService>::assign_from_gs_contractid [nsCOMPtr.h:1252]
nsCOMPtr<nsIAlertsService>::nsCOMPtr [nsCOMPtr.h:628]
nsAlertsService::ShowAlertNotification [toolkit/components/alerts/nsAlertsService.cpp:110]
nsDownload::SetState [toolkit/components/downloads/nsDownloadManager.cpp:2235]
nsDownload::OnStateChange [toolkit/components/downloads/nsDownloadManager.cpp:2507]
nsWebBrowserPersist::OnStopRequest [embedding/components/webbrowserpersist/src/nsWebBrowserPersist.cpp:810]
nsHttpChannel::OnStopRequest [netwerk/protocol/http/nsHttpChannel.cpp:4057]
nsInputStreamPump::OnStateStop [netwerk/base/src/nsInputStreamPump.cpp:579]
nsInputStreamPump::OnInputStreamReady [netwerk/base/src/nsInputStreamPump.cpp:403]
nsInputStreamReadyEvent::Run [xpcom/io/nsStreamUtils.cpp:113]
nsThread::ProcessNextEvent [xpcom/threads/nsThread.cpp:633]
NS_ProcessNextEvent_P [nsThreadUtils.cpp:250]
mozilla::ipc::MessagePump::Run [ipc/glue/MessagePump.cpp:107]
MessageLoop::RunInternal [ipc/chromium/src/base/message_loop.cc:220]
MessageLoop::RunHandler [ipc/chromium/src/base/message_loop.cc:203]
MessageLoop::Run [ipc/chromium/src/base/message_loop.cc:175]
nsBaseAppShell::Run [widget/src/xpwidgets/nsBaseAppShell.cpp:198]
nsAppStartup::Run [toolkit/components/startup/nsAppStartup.cpp:218]
XRE_main [toolkit/xre/nsAppRunner.cpp:3765]
main [suite/app/nsSuiteApp.cpp:103]
libc.so.6 + 0x15dec

Comment 4

[Karl Tomlinson (:karlt)]

toolkit/components/alerts/nsAlertsService.cpp and toolkit/system/gnome/nsAlertsService.cpp both define nsAlertsService, differently.

We'll also need to avoid the name collision before folding libmozgnome into libxul.

Comment 5

[Karl Tomlinson (:karlt)]

Attachment: rename DBus Notifications nsAlertsService to resolve collision with XP service

Picking a name consistent with the service provided.

Comment 6

[Karl Tomlinson (:karlt)]

https://hg.mozilla.org/integration/mozilla-inbound/rev/0de2844c2c7b

Comment 7

[Matt Brubeck (:mbrubeck)]

Sorry, I backed out this fix and several others on inbound because it looks like one of them broke B2G tests:
https://hg.mozilla.org/integration/mozilla-inbound/rev/e8c8d9373eba

Comment 8

[Phil Ringnalda (:philor)]

Relanded in https://hg.mozilla.org/integration/mozilla-inbound/rev/2e368c01440e

Comment 9

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/2e368c01440e

Comment 10

[Martin Stránský [:stransky] (ni? me)]

Attachment: patch on top of 858919

Comment 11

[Serge Gautherie (:sgautherie)]

(In reply to Martin Stránský from comment #10)
> Created attachment 8512001 [details] [diff] [review]
> patch on top of 858919

Martin, I don't think this patch belongs here (= resolved bug)...