[Intel-gfx] i915 resume-from-hibernation problems on resume with current Linus' tree

[Jiri Kosina]

On Fri, 21 Mar 2014, Daniel Vetter wrote:

> > So I am still seeing this with current Linus' tree rather regularly (but
> > it's not deterministic enough for a reliable bisect).
> >
> > Unfortunately I haven't received any patches to test; what do you propose?
> > Would reporting this on bugs.freedesktop.org instead of ranting here
> > change anything?
> 
> Yeah, filing a bug on fdo to keep track of this should help. Please
> add a [gm45 regression] tag to the summary so that it shows up in our
> queries at the right spot ;-)
> 
> One thing to test would be to give -nightly (or drm-intel-testing as
> of today) a shot. We've merged a few patches recently to slightly
> improve the ring init sequence in the generic code, they might
> actually help here. Sorry for not correlating them with your patch
> when I've merged them, only just realized this now.

I just tested current Linus' tree + drm-intel-testing (HEAD == 14347c2 
("drm-intel-nightly: 2014y-03m-21d-16h-13m-30s integration manifest")), 
and on 3rd suspend-resume cycle I got the ring initialization failure 
again, with Xorg going crazy afterwards. So unfortunately it's still 
unfixed.

Now submitted as

	https://bugs.freedesktop.org/show_bug.cgi?id=76554

as well.

Thanks,

-- 
Jiri Kosina
SUSE Labs