880037 - Lots of WARNING: NS_ENSURE_TRUE(IsChromeProcess()) failed: file /content/base/src/nsFrameMessageManager.cpp, line 728

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Reuben Morais [:reuben]]

I think these come from http://mxr.mozilla.org/mozilla-central/source/content/base/src/nsCCUncollectableMarker.cpp?mark=97-98#94

Comment 1

[David Zbarsky (:dzbarsky)]

Attachment: PAtch

Comment 2

[Justin Lebar (not reading bugmail)]

Comment on attachment 759448 [details] [diff] [review]
PAtch

The idea here is fine, but the one-liner to check for is-main-process is

   XRE_GetProcessType() == GeckoProcessType_Default

(I'd love to improve this, but we should probably do so all at once, and not in nsFrameMessageManager.h.  But that's probably out of scope for this bug.)

Comment 3

[Justin Lebar (not reading bugmail)]

I'd probably also leave the NS_NewGlobalMessageManager code unchanged.  We have very poor testing of debug builds in B2G, so there's no guarantee that we'd notice if we were hitting this assertion.  And anyway, XPCOM semantics are to return an error, rather than to assert, and this is definitely XPCOM code.

Comment 4

[David Zbarsky (:dzbarsky)]

Attachment: Patch

Comment 5

[Justin Lebar (not reading bugmail)]

Comment on attachment 759858 [details] [diff] [review]
Patch

r=me, but could you add a comment saying that the globalmm only exists in the root process?

Comment 6

[David Zbarsky (:dzbarsky)]

Sure.
https://hg.mozilla.org/integration/mozilla-inbound/rev/b9b9d53a5c2d

Comment 7

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/b9b9d53a5c2d