902013 - CPOWs should be preffed off by default

Status: RESOLVED FIXED

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

Description

[Benjamin Smedberg]

In bug 868859 CPOW support landed for Firefox desktop experiments in making addons compatible. But they should not be available to B2G at all, so they should be behind a pref which is off by default and only enabled on the e10s desktop branch currently.

Comment 1

[Tom S [:evilpie]]

Attachment: v1

Something like this?

Comment 2

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

I think CPOWs should be enabled if this pref is set OR if browser.tabs.remote is set. Also, we'll have to change the CPOW tests to set this special pref.

Comment 3

[Tom S [:evilpie]]

Well the way I did this is that the pref is always on when on Desktop. Because if we enable it automatically with browser.tabs.remote, we can't really test with remote tabs and CPOWs disabled. And because it would always be enabled on Desktop we don't have to bother with any special code.

Comment 4

[Tom S [:evilpie]]

Comment on attachment 788302 [details] [diff] [review]
v1

Review of attachment 788302 [details] [diff] [review]:
-----------------------------------------------------------------

Not sure if you can review this?

Comment 5

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

Comment on attachment 788302 [details] [diff] [review]
v1

Review of attachment 788302 [details] [diff] [review]:
-----------------------------------------------------------------

David should probably review this, since he wrote the original code here.

However, the reason the bug was filed was to ensure that CPOWs are disabled by default, even on desktop. I'd rather we do it the way I suggested in comment 2. Is there a reason we would want to test e10s with CPOWs disabled? If we just want to find places where CPOWs are used when we don't expect, I've found it useful to dump the stack whenever we send a CPOW message. That seems easier than disabling them and seeing what breaks.

Comment 6

[Tom S [:evilpie]]

Attachment: v2

So I guess this is more what Bill has in mind. We could even remove the force-disabled based on his feedback.

Comment 7

[David Anderson [:dvander] - inactive, e-mail if emergency]

Comment on attachment 792873 [details] [diff] [review]
v2

Review of attachment 792873 [details] [diff] [review]:
-----------------------------------------------------------------

lgtm

Comment 8

[Tom S [:evilpie]]

http://hg.mozilla.org/integration/mozilla-inbound/rev/bf471c1e49ba

Comment 9

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

https://hg.mozilla.org/mozilla-central/rev/bf471c1e49ba