Closed
Bug 1343666
Opened 7 years ago
Closed 7 years ago
Assertion failure: sVRManagerChildSingleton
Categories
(Core :: Graphics, defect, P3)
Core
Graphics
Tracking
()
RESOLVED
FIXED
mozilla54
Tracking | Status | |
---|---|---|
firefox52 | --- | unaffected |
firefox-esr52 | --- | unaffected |
firefox53 | --- | unaffected |
firefox54 | --- | fixed |
People
(Reporter: truber, Assigned: kip)
Details
(Keywords: assertion, testcase, Whiteboard: [gfx-noted])
Attachments
(3 files, 1 obsolete file)
The attached testcase causes an assertion failure at shutdown in mozilla-central rev 34c6c2f302e7. Assertion failure: sVRManagerChildSingleton, at /home/worker/workspace/build/src/gfx/vr/ipc/VRManagerChild.cpp:75 #01: mozilla::dom::VREventObserver::DisconnectFromOwner at dom/vr/VREventObserver.cpp:49 #02: mozilla::dom::VREventObserver::Release at memory/mozalloc/mozalloc.h:218 #03: nsGlobalWindow::~nsGlobalWindow at xpcom/ds/nsTArray.h:2270 #04: nsGlobalWindow::~nsGlobalWindow at dom/base/nsGlobalWindow.cpp:1707 #05: SnowWhiteKiller::~SnowWhiteKiller at mfbt/SegmentedVector.h:309 #06: nsCycleCollector::FreeSnowWhite at xpcom/base/nsCycleCollector.cpp:2831 #07: nsCycleCollector::BeginCollection at xpcom/base/nsCycleCollector.cpp:3828 #08: nsCycleCollector::Collect at xpcom/base/nsCycleCollector.cpp:3650 #09: nsCycleCollector::ShutdownCollect at xpcom/base/nsCycleCollector.cpp:3591
Updated•7 years ago
|
Flags: needinfo?(kgilbert)
Priority: -- → P3
Whiteboard: [gfx-noted]
Assignee | ||
Comment 1•7 years ago
|
||
I'll take this and investigate, thanks!
Assignee: nobody → kgilbert
Flags: needinfo?(kgilbert)
Assignee | ||
Comment 2•7 years ago
|
||
I see the logic error in VREventObserver::DisconnectFromOwner()... I need to check VRManagerChild::IsCreated() rather than checking if VRManagerChild::Get() returns a nullptr. This bug should affect only debug builds. Patch incoming..
Comment hidden (mozreview-request) |
Assignee | ||
Updated•7 years ago
|
Attachment #8843056 -
Flags: review?(dmu)
Comment 4•7 years ago
|
||
mozreview-review |
Comment on attachment 8843056 [details] Bug 1343666 - Prevent crash on shutdown due to assertion in VRManagerChild::Get() https://reviewboard.mozilla.org/r/116806/#review118582 LGTM
Attachment #8843056 -
Flags: review?(dmu) → review+
Pushed by kgilbert@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/06074f7e8901 Prevent crash on shutdown due to assertion in VRManagerChild::Get() r=daoshengmu
Comment 6•7 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/06074f7e8901
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla54
Comment 7•7 years ago
|
||
Can we land this attached testcase as a crashtest? Also, does this need to be backported or can it ride the 54 train?
Flags: needinfo?(kgilbert)
Flags: in-testsuite?
Assignee | ||
Comment 8•7 years ago
|
||
(In reply to Ryan VanderMeulen [:RyanVM] from comment #7) > Can we land this attached testcase as a crashtest? Also, does this need to > be backported or can it ride the 54 train? WebVR will not be enabled by default in release 54 (it has slipped to 55); however, it is already enabled in Aurora. I expect the incidence to be low of users encountering this. It would be low-risk to back-port this to 54 if we see any reports there. The testcase would be useful as a crashtest. I'd be glad to get this in.
Flags: needinfo?(kgilbert)
Comment 9•7 years ago
|
||
This landed in time for 54 prior to Monday's uplift to Aurora, so it sounds like we're good in that respect. Thanks for taking care of getting the testcase landed as a crashtest too!
status-firefox52:
--- → unaffected
status-firefox53:
--- → unaffected
status-firefox-esr52:
--- → unaffected
Assignee | ||
Comment 10•7 years ago
|
||
Assignee | ||
Comment 11•7 years ago
|
||
Attachment #8846195 -
Attachment is obsolete: true
Assignee | ||
Updated•7 years ago
|
Attachment #8846853 -
Flags: review?(daoshengmu)
Updated•7 years ago
|
Attachment #8846853 -
Flags: review?(daoshengmu) → review?(dmu)
Comment 12•7 years ago
|
||
Comment on attachment 8846853 [details] [diff] [review] Bug 1343666 - Part 2: Add Crashtest Review of attachment 8846853 [details] [diff] [review]: ----------------------------------------------------------------- r=me, although I am interested in why you add "document.documentElement.removeAttribute("class");" at the end of this test file?
Attachment #8846853 -
Flags: review?(dmu) → review+
Comment 13•7 years ago
|
||
Pushed by kgilbert@mozilla.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/b70211869d2e Part 2: Add crashtest; r=daoshengmu
Comment 14•7 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/b70211869d2e
Hi Kip and Daosheng We're fixing data: URI to have its own unique origin in bug 1324406, I noticed you added a test that using data: URI in this bug, please don't do this anymore unless you're pretty sure what you're doing. Thanks
Comment hidden (Intermittent Failures Robot) |
You need to log in
before you can comment on or make changes to this bug.
Description
•