Re: [PATCH] drm/i915: annote drop_caches debugfs interface with lockdep

[Peter Zijlstra]

On Mon, Mar 13, 2017 at 09:15:17AM +0100, Daniel Vetter wrote:
> On Mon, Mar 13, 2017 at 09:01:57AM +0100, Peter Zijlstra wrote:
> > On Sun, Mar 12, 2017 at 09:53:40PM +0100, Daniel Vetter wrote:
> > 
> > > Peter/Ingo,
> > > 
> > > We want this to validate the i915 shrinker locking in our fast tests
> > > without thrashing badly (that takes too long, we can only thrash in
> > > the extended runs). Can you pls take a look and if it's ok ack for
> > > merging through drm-intel.git?
> > 
> > Hurm, I was going to rework all that soonish; have a look here:
> > 
> >  https://lkml.kernel.org/r/20170302134031.GG6536@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> > 
> > The immediate problem is that I made the annotation private to mm/
> > there, I suppose I could fix that.
> 
> Yeah, we'd really like to have that, and even when switched to a
> lockdep_map instead of reusing the context stuff the semantic interface
> would be the same (and I think we should keep the gfp_flags stuff, in case
> someone adds a nesting lockdep map for GFP_IO).
> 
> Do you want a topic branch with just this patch (the shrink_all is new so
> there will be a conflict and we can't mege it through one tree alone) so
> that you can refactor things with i915 included?

Just take your patch; I'll sort it out when I get time for things and
take i915 along for the ride.

Acked-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>

Thanks!