776328 - "ASSERTION: Forgot to check if this is a wrapper?" with Components.lookupMethod

Status: RESOLVED FIXED

(Core :: XPConnect, defect)

Description

[Jesse Ruderman]

Attachment: testcase (asserts fatally when loaded)

###!!! ASSERTION: Forgot to check if this is a wrapper?: 'IS_WRAPPER_CLASS(js::GetObjectClass(obj))', file ../../../dist/include/xpcpublic.h, line 70

Assertion failure: (DebugCheckWrapperClass(obj) && !js::GetReservedSlot(obj, 0).isDouble()), at /Users/jruderman/trees/mozilla-central/js/xpconnect/wrappers/XrayWrapper.cpp:440

Might be a regression from bug 774245.

Comment 1

[Jesse Ruderman]

Attachment: stack traces

Comment 2

[Bobby Holley (:bholley)]

Same as bug 776333.

Comment 3

[Jesse Ruderman]

This testcase still asserts for me.

Comment 4

[Lukas Blakk [:lsblakk] use ?needinfo]

Testcase still asserts even though bug 776333 is now resolved, so what's the impact here of this assert being tripped?  How does this manifest?  What's the ramification for the end user?  Trying to get a sense of the importance for tracking here.

Comment 5

[Bobby Holley (:bholley)]

Attachment: Only create holders for WNs. v1

Fixed a stupid mistake. Flagging Blake for review.

Comment 6

[Bobby Holley (:bholley)]

(In reply to Lukas Blakk [:lsblakk] from comment #4)
> Testcase still asserts even though bug 776333 is now resolved, so what's the
> impact here of this assert being tripped?  How does this manifest?  What's
> the ramification for the end user?  Trying to get a sense of the importance
> for tracking here.

We should track. Simple fix, and potential crashes etc if left unfixed.

Comment 7

[Bobby Holley (:bholley)]

Pushed to m-i:

https://hg.mozilla.org/integration/mozilla-inbound/rev/4e0f55796e6f

(this is very low risk for bustage, so I don't think it's worth the try resources to do a push there)

Comment 8

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/4e0f55796e6f

Comment 9

[Lukas Blakk [:lsblakk] use ?needinfo]

Already fixed, setting status & tracking for completeness in case there's fallout later for any reason.

Comment 10

[Mihaela Velimiroviciu (:mihaelav)]

Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:17.0) Gecko/17.0 Firefox/17.0

Verified the fix with the latest beta-debug build (id 20121017141413) on Mac OS X 10.7.5 and doesn't assert anymore.
Marking verified for Firefox 17.

Comment 11

[Tracy Walker [:tracy]]

mass remove verifyme requests greater than 4 months old