[Intel-gfx] [RFC i-g-t 0/6] Drop tests for object creation from stolen

[Petri Latvala]

On Tue, Oct 03, 2017 at 10:31:10AM +0200, Daniel Vetter wrote:
> On Mon, Oct 02, 2017 at 04:00:12PM -0700, Daniele Ceraolo Spurio wrote:
> > The tests were merged before the feature, which never made it to the
> > driver; the last update I could find ([1]) is more than a year old.
> > The tests are using a wrong getparam number and planned extensions
> > to the gem_create and get_aperture ioctl structures that don't match
> > what's currently in i915. Due to this all stolen-related subtests are
> > skipping.
> > Given the fact that there has been no traction behind the feature for
> > a long time it doesn't look like we'll need those tests anytime soon
> > and thus it makes sense to remove them.
> > 
> > The patches didn't revert cleanly and also contained some improvements
> > that are worth keeping, so instead of a revert I've organized the removal
> > in a single patch per test file to drop only the unneeded code.
> > 
> > [1] https://patchwork.freedesktop.org/series/659/
> > 
> > Cc: Ankitprasad Sharma <ankitprasad.r.sharma at intel.com>
> > Cc: Petri Latvala <petri.latvala at intel.com>
> > Cc: Arkadiusz Hiler <arkadiusz.hiler at intel.com>
> > Cc: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > Cc: Chris Wilson <chris at chris-wilson.co.uk> 
> 
> Ack from my side on the entire series. Merging stuff before the kernel
> side has landed is a good way to piss of Dave Airlie. Yes CI right now
> doesn't support testing both a kernel and an igt branch at the same time,
> so not that useful for testing new features. It's on the list, but then CI
> is primarily about preventing regressions, not so much about making sure
> new stuff works perfectly out of the box (would be good too ofc).

Series is
Acked-by: Petri Latvala <petri.latvala at intel.com>
as well, with an addition posted separately to patch 4. GEM people
could give their blessing as well before landing this, Joonas?


-- 
Petri Latvala