[Webkit-unassigned] [Bug 67324] filter test_expectations properly for chromium-mac vs chromium-cg-mac

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=67324





--- Comment #11 from epoger at google.com  2011-09-02 11:57:07 PST ---
(In reply to comment #10)
> 
> Also, please add tests. I'm pretty sure most of this code is already unittested. You can run the unittests by running test-webkitpy.

(Ignoring the lower-level stuff until we get this part sorted out...)

> 
> At a higher level, I'm not sure we should tie CG/Skia-ness to GPU/CPU. Should it instead just be a separate modifier that's just ignored on the non-mac ports? e.g.
> BUG_FOO CPU CG : ...
> BUG_FOO SKIA : ....

I agree that it would be simpler to allow separate SKIA and/or CG modifiers, rather than combining it with the CPU/GPU modifiers.

However, the decision had been made previously (in https://bugs.webkit.org/show_bug.cgi?id=66705 ) to organize it this way.  Please see the discussion there, including some clarifying questions I had asked that were never answered. 

Should we undo the decision that was made on that other bug?  Or do I misunderstand something?

> 
> > LayoutTests/platform/chromium/test_expectations.txt:1878
> > +BUGWK45991 CPU GPU : canvas/philip/tests/2d.shadow.canvas.basic.html = TEXT
> > +BUGWK45991 CPU GPU : canvas/philip/tests/2d.shadow.canvas.transparent.2.html = TEXT
> 
> Here and elsewhere, there's no point in writing "CPU GPU" since that's all the graphics types, right? Can you just delete those modifiers?

In the current arrangement, "CPU GPU" means "Skia, not CG".  ("CPU-CG GPU-CG" would mean "CG, not Skia")

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.