On Thu, Mar 17, 2016 at 04:30:47PM +, Tvrtko Ursulin wrote:
>
> On 17/03/16 13:14, Chris Wilson wrote:
> >On Thu, Mar 17, 2016 at 12:59:46PM +, Tvrtko Ursulin wrote:
> >>From: Tvrtko Ursulin
> >>
> >>By reading the CSB (slow MMIO accesses) into a temporary local
> >>buffer we can decrease
From: Tvrtko Ursulin
By reading the CSB (slow MMIO accesses) into a temporary local
buffer we can decrease the duration of holding the execlist
lock.
Main advantage is that during heavy batch buffer submission we
reduce the execlist lock contention, which should decrease the
latency and CPU usag
On 17/03/16 13:14, Chris Wilson wrote:
On Thu, Mar 17, 2016 at 12:59:46PM +, Tvrtko Ursulin wrote:
From: Tvrtko Ursulin
By reading the CSB (slow MMIO accesses) into a temporary local
buffer we can decrease the duration of holding the execlist
lock.
Main advantage is that during heavy bat
On Thu, Mar 17, 2016 at 12:59:46PM +, Tvrtko Ursulin wrote:
> From: Tvrtko Ursulin
>
> By reading the CSB (slow MMIO accesses) into a temporary local
> buffer we can decrease the duration of holding the execlist
> lock.
>
> Main advantage is that during heavy batch buffer submission we
> red