Re: [Intel-gfx] [RFC 00/44] GPU scheduler for i915 driver

2014-10-20 Thread John Harrison

On 10/10/2014 11:35, Steven Newbury wrote:

On Thu, 2014-06-26 at 18:23 +0100, john.c.harri...@intel.com wrote:

From: John Harrison harri...@intel.com

Implemented a batch buffer submission scheduler for the i915 DRM
driver.

Hi John,

I was just wondering what's happening with this patch series?  Are you
still working on it?  Does it need any testing?


It is still in progress. Although currently stalled because it was 
decided to first replace the driver's seqno usage with request 
structures wherever possible. The theory being that this is safer and 
more sensible that having the scheduler cause out-of-order seqnos and 
the various work-arounds to cope with that.


___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [RFC 00/44] GPU scheduler for i915 driver

2014-10-10 Thread Steven Newbury
On Thu, 2014-06-26 at 18:23 +0100, john.c.harri...@intel.com wrote:
 From: John Harrison harri...@intel.com
 
 Implemented a batch buffer submission scheduler for the i915 DRM 
 driver.

Hi John,

I was just wondering what's happening with this patch series?  Are you 
still working on it?  Does it need any testing?


signature.asc
Description: This is a digitally signed message part
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [RFC 00/44] GPU scheduler for i915 driver

2014-07-07 Thread Daniel Vetter
On Fri, Jun 27, 2014 at 06:44:04AM +1000, Dave Airlie wrote:
 
  Implemented a batch buffer submission scheduler for the i915 DRM driver.
 
 
 While this seems very interesting, you might want to address in the commit msg
 or the cover email
 
 a) why this is needed,
 b) any improvements in speed, power consumption or throughput it generates,
 i.e. benchmarks.
 
 also some notes on what hw supports preemption.

Also tests to both exercise the rescheduling (i.e. let a high-prio batch
compete agains a large pile of low-prio batches) and augmenting our gpu
sync tests (from Damien) with that, too.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx


Re: [Intel-gfx] [RFC 00/44] GPU scheduler for i915 driver

2014-06-26 Thread Dave Airlie

 Implemented a batch buffer submission scheduler for the i915 DRM driver.


While this seems very interesting, you might want to address in the commit msg
or the cover email

a) why this is needed,
b) any improvements in speed, power consumption or throughput it generates,
i.e. benchmarks.

also some notes on what hw supports preemption.

Dave.
___
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx