The goal is to land new qcms ICC v4 features such as CLUT support on m-c with a preference to enable them.

Feel free to nit pick about the name. Waiting for tryserver results before flagging review.

Try run: http://tbpl.mozilla.org/?tree=Try&pusher=b56girard@gmail.com&rev=661b4dd715a4

Comment on attachment 548770 [details] [diff] [review] Part 2: Add gfx.color_management.force_v4 I don't really see the point of CMVersionFourPrefName. The only thing I can think of is to keep all of the different prefs close together. However, I think documentation would serve this just as well. I suggest just using the name directly. I'd also take a patch to inline all of the other names.

Carrying forward r+

Also: http://hg.mozilla.org/mozilla-central/rev/86ff239e3948

Also: http://hg.mozilla.org/mozilla-central/rev/8d309b30299a

I just tested with the following profile: http://www.colorkeeper.com/?page_id=235 We get the un managed result without v4 and the perceptual with the v4 pref. I would of expected to get the TRC result however we are less strict with the profiles we reject now. The profile in the package is very interesting. It shows very obviously what is being managed and what is not (i.e. the browser chrome).

I just stumbled across this feature when visiting http://cameratico.com/tools/web-browser-color-management-test/ and noticing the "v4" brokenness there in Firefox. (though it works if I toggle this bug's pref & restart Firefox) Does anyone recall why this feature was pref-controlled & preffed-off-by-default? i.e. was there some remaining work to be done before it could be enabled? Or was the feature known-to-be-permanently-problematic so we wanted to make it opt in? Or something else?

Performance isn't great and there might be security problems.