[Intel-gfx] [PATCH] drm/i915: Handle concurrent GTT faults gracefully

Chris Wilson chris at chris-wilson.co.uk
Thu Jun 12 19:36:04 CEST 2014


On Thu, Jun 12, 2014 at 10:10:58AM -0700, Volkin, Bradley D wrote:
> On Thu, Jun 12, 2014 at 03:13:14PM +0100, Chris Wilson wrote:
> > +	/* Check if a second thread completed the prefaulting for us */
> > +	if (obj->fault_mappable && pte_exists(vma, vma->vm_start))
> > +		goto unlock;
> > +
> 
> We only set fault_mappable if we successfully remapped the whole range and clear
> it on unmapping the whole range, so do we need the additional pte_exists() check?

Yes. You have you have to consider that we can have multiple vma
pointing to the same object. See i-g-t/gem_mmap_gtt/read-write-distinct.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre



More information about the Intel-gfx mailing list