On Mon, Oct 17, 2016 at 10:56:27AM +0100, Tvrtko Ursulin wrote:
>
> On 14/10/2016 15:07, Chris Wilson wrote:
> >On Fri, Oct 14, 2016 at 02:32:03PM +0100, Tvrtko Ursulin wrote:
> >>On 14/10/2016 13:18, Chris Wilson wrote:
> >>>A while ago we switched from a contiguous array of pages into an sglist,
On 14/10/2016 15:07, Chris Wilson wrote:
On Fri, Oct 14, 2016 at 02:32:03PM +0100, Tvrtko Ursulin wrote:
On 14/10/2016 13:18, Chris Wilson wrote:
A while ago we switched from a contiguous array of pages into an sglist,
for that was both more convenient for mapping to hardware and avoided
the r
On Fri, Oct 14, 2016 at 02:32:03PM +0100, Tvrtko Ursulin wrote:
>
> On 14/10/2016 13:18, Chris Wilson wrote:
> >A while ago we switched from a contiguous array of pages into an sglist,
> >for that was both more convenient for mapping to hardware and avoided
> >the requirement for a vmalloc array o
On 14/10/2016 13:18, Chris Wilson wrote:
A while ago we switched from a contiguous array of pages into an sglist,
for that was both more convenient for mapping to hardware and avoided
the requirement for a vmalloc array of pages on every object. However,
certain GEM API calls (like pwrite, pread