Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Imre Deak
On Wed, Nov 29, 2017 at 10:59:27AM +, Tvrtko Ursulin wrote: > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states a lot when > there are no connected displays (no active power domains) during command > submission. > > This activity on DC states has a nega

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Imre Deak
On Wed, Nov 29, 2017 at 11:59:04AM +, Chris Wilson wrote: > Quoting Tvrtko Ursulin (2017-11-29 11:53:41) > > > > On 29/11/2017 11:40, Chris Wilson wrote: > > > Quoting Tvrtko Ursulin (2017-11-29 11:34:27) > > >> > > >> On 29/11/2017 11:12, Daniel Vetter wrote: > > >>> I think given that DMC is

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Daniel Vetter
On Wed, Nov 29, 2017 at 11:34:27AM +, Tvrtko Ursulin wrote: > > On 29/11/2017 11:12, Daniel Vetter wrote: > > On Wed, Nov 29, 2017 at 10:59:27AM +, Tvrtko Ursulin wrote: > > > From: Tvrtko Ursulin > > > > > > It seems that the DMC likes to transition between the DC states a lot when > >

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Imre Deak
On Wed, Nov 29, 2017 at 11:10:58AM +, Chris Wilson wrote: > Quoting Tvrtko Ursulin (2017-11-29 10:59:27) > > From: Tvrtko Ursulin > > > > It seems that the DMC likes to transition between the DC states a lot when > > there are no connected displays (no active power domains) during command > >

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Chris Wilson
Quoting Tvrtko Ursulin (2017-11-29 11:53:41) > > On 29/11/2017 11:40, Chris Wilson wrote: > > Quoting Tvrtko Ursulin (2017-11-29 11:34:27) > >> > >> On 29/11/2017 11:12, Daniel Vetter wrote: > >>> I think given that DMC is strongly recommended there shouldn't be a real > >>> cost with making this

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Tvrtko Ursulin
On 29/11/2017 11:40, Chris Wilson wrote: Quoting Tvrtko Ursulin (2017-11-29 11:34:27) On 29/11/2017 11:12, Daniel Vetter wrote: I think given that DMC is strongly recommended there shouldn't be a real cost with making this unconditional. I don't know, not liking it on the first go. But then

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Chris Wilson
Quoting Tvrtko Ursulin (2017-11-29 11:34:27) > > On 29/11/2017 11:12, Daniel Vetter wrote: > > I think given that DMC is strongly recommended there shouldn't be a real > > cost with making this unconditional. > > I don't know, not liking it on the first go. But then again I have no > idea how mu

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Tvrtko Ursulin
On 29/11/2017 11:12, Daniel Vetter wrote: On Wed, Nov 29, 2017 at 10:59:27AM +, Tvrtko Ursulin wrote: From: Tvrtko Ursulin It seems that the DMC likes to transition between the DC states a lot when there are no connected displays (no active power domains) during command submission. This

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Daniel Vetter
On Wed, Nov 29, 2017 at 10:59:27AM +, Tvrtko Ursulin wrote: > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states a lot when > there are no connected displays (no active power domains) during command > submission. > > This activity on DC states has a nega

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Chris Wilson
Quoting Tvrtko Ursulin (2017-11-29 10:59:27) > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states a lot when > there are no connected displays (no active power domains) during command > submission. > > This activity on DC states has a negative impact on the

[Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-11-29 Thread Tvrtko Ursulin
From: Tvrtko Ursulin It seems that the DMC likes to transition between the DC states a lot when there are no connected displays (no active power domains) during command submission. This activity on DC states has a negative impact on the performance of the chip with huge latencies observed in the

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-08 Thread Jani Nikula
On Fri, 05 May 2017, Tvrtko Ursulin wrote: > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states > a lot when there are no connected displays (no active power > domains) during simple command submission. > > This frantic activity on DC states has a terrible im

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-05 Thread Ville Syrjälä
On Fri, May 05, 2017 at 05:13:58PM +0100, Tvrtko Ursulin wrote: > > On 05/05/2017 15:49, Ville Syrjälä wrote: > > On Fri, May 05, 2017 at 12:43:21PM +0100, Tvrtko Ursulin wrote: > >> From: Tvrtko Ursulin > >> > >> It seems that the DMC likes to transition between the DC states > >> a lot when the

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-05 Thread Tvrtko Ursulin
On 05/05/2017 15:49, Ville Syrjälä wrote: On Fri, May 05, 2017 at 12:43:21PM +0100, Tvrtko Ursulin wrote: From: Tvrtko Ursulin It seems that the DMC likes to transition between the DC states a lot when there are no connected displays (no active power domains) during simple command submission.

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-05 Thread Ville Syrjälä
On Fri, May 05, 2017 at 12:43:21PM +0100, Tvrtko Ursulin wrote: > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states > a lot when there are no connected displays (no active power > domains) during simple command submission. Is it trapping on some interrupt r

Re: [Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-05 Thread Chris Wilson
On Fri, May 05, 2017 at 12:43:21PM +0100, Tvrtko Ursulin wrote: > From: Tvrtko Ursulin > > It seems that the DMC likes to transition between the DC states > a lot when there are no connected displays (no active power > domains) during simple command submission. > > This frantic activity on DC st

[Intel-gfx] [PATCH] drm/i915: Restore GT performance in headless mode with DMC loaded

2017-05-05 Thread Tvrtko Ursulin
From: Tvrtko Ursulin It seems that the DMC likes to transition between the DC states a lot when there are no connected displays (no active power domains) during simple command submission. This frantic activity on DC states has a terrible impact on the performance of the overall chip with huge la