[Mesa-dev] [PATCH 1/4] i965: Also examine _BaseFormat when deciding to perform xRGB_alpha fixups

Michel Dänzer michel at daenzer.net
Wed Jan 23 08:23:34 PST 2013


On Mit, 2013-01-23 at 08:17 -0800, Carl Worth wrote: 
> Ian Romanick <idr at freedesktop.org> writes:
> > These tests intermittently pass / fail for me for quite some time (since 
> > always?).
> 
> Oh, well in that case I feel better about my patch series.
> 
> We should come up with a better plan for unreliable tests in piglit. It
> seems that the current approach is developing human filters to ignore
> unreliable test results. That's obviously fragile in lots of ways, (and
> especially painful for new users with no filters in place).
> 
> How about moving tests that like these that are known to misbehave out
> of all.tests and into something like unreliable.tests? Or just removing
> them, leaving them just in the git history?

Are these tests unreliable themselves, or rather the driver paths
they're hitting? If it's the latter (a phenomenon which is well-known to
me with radeonsi :), neither approach seems appropriate.


-- 
Earthling Michel Dänzer           |                   http://www.amd.com
Libre software enthusiast         |          Debian, X and DRI developer


More information about the mesa-dev mailing list