Re: [RFC 3/5] drm/i915: Add support for CPU mapping to DRM_IOCTL_I915_GEM_MMAP_GTT

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Jan 27, 2016 at 03:24:43PM +0000, Tvrtko Ursulin wrote:
> 
> On 26/01/16 16:59, Chris Wilson wrote:
> >On Tue, Jan 26, 2016 at 04:23:28PM +0000, Tvrtko Ursulin wrote:
> >>Oh yeah, need to pin the pages..
> >
> >But only whilst inserting. Once inserted they need to be evicted, and I
> >was wondering whether we should do the zap on put_pages(). If we don't,
> >it means that the shrinker is neutered.
> 
> Ah I forgot this in v2. So we would need something like
> obj->fault_mappable so that i915_gem_release_mmap runs, yes?

Something like that. I don't have a good idea, but I was wondering if
plonking a callback into the offset-node would be useful, and walking a
list of nodes on the object.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx




[Index of Archives]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux