Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-23 Thread Chris Wilson
Quoting Joonas Lahtinen (2018-03-23 13:20:40) > So far nobody has been succesful in selling this to the userspace > compositors (the most likely user) or has somebody? I hadn't even contemplated selling it. However, it does seem applicable to the RealTime priorities for Vk and https://www.khronos

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-23 Thread Joonas Lahtinen
s.freedesktop.org; b...@bwidawsk.net > > Subject: Re: [Intel-gfx] [RFC 0/8] Force preemption > > > > On Thu, Mar 22, 2018 at 05:41:57PM +, Tvrtko Ursulin wrote: > > > > > > On 22/03/2018 16:01, Jeff McGee wrote: > > > >On Thu, Mar 22, 2018 at 03:5

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Bloomfield, Jon
> From: Intel-gfx [mailto:intel-gfx-boun...@lists.freedesktop.org] On Behalf > Of Jeff McGee > Sent: Thursday, March 22, 2018 12:09 PM > To: Tvrtko Ursulin > Cc: Kondapally, Kalyan ; intel- > g...@lists.freedesktop.org; b...@bwidawsk.net > Subject: Re: [Intel-gfx] [RFC

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Jeff McGee
On Thu, Mar 22, 2018 at 05:41:57PM +, Tvrtko Ursulin wrote: > > On 22/03/2018 16:01, Jeff McGee wrote: > >On Thu, Mar 22, 2018 at 03:57:49PM +, Tvrtko Ursulin wrote: > >> > >>On 22/03/2018 14:34, Jeff McGee wrote: > >>>On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: > Quo

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Tvrtko Ursulin
On 22/03/2018 16:01, Jeff McGee wrote: On Thu, Mar 22, 2018 at 03:57:49PM +, Tvrtko Ursulin wrote: On 22/03/2018 14:34, Jeff McGee wrote: On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-03-22 09:22:55) On 21/03/2018 17:26, jeff.mc...@intel.com

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Jeff McGee
On Thu, Mar 22, 2018 at 03:57:49PM +, Tvrtko Ursulin wrote: > > On 22/03/2018 14:34, Jeff McGee wrote: > >On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: > >>Quoting Tvrtko Ursulin (2018-03-22 09:22:55) > >>> > >>>On 21/03/2018 17:26, jeff.mc...@intel.com wrote: > From: Jeff

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Jeff McGee
On Thu, Mar 22, 2018 at 03:35:19PM +, Chris Wilson wrote: > Quoting Jeff McGee (2018-03-22 14:34:58) > > On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: > > > Quoting Tvrtko Ursulin (2018-03-22 09:22:55) > > > > > > > > On 21/03/2018 17:26, jeff.mc...@intel.com wrote: > > > > > F

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Tvrtko Ursulin
On 22/03/2018 14:34, Jeff McGee wrote: On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-03-22 09:22:55) On 21/03/2018 17:26, jeff.mc...@intel.com wrote: From: Jeff McGee Force preemption uses engine reset to enforce a limit on the time that a reque

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Chris Wilson
Quoting Jeff McGee (2018-03-22 14:34:58) > On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: > > Quoting Tvrtko Ursulin (2018-03-22 09:22:55) > > > > > > On 21/03/2018 17:26, jeff.mc...@intel.com wrote: > > > > From: Jeff McGee > > > > > > > > Force preemption uses engine reset to en

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Jeff McGee
On Thu, Mar 22, 2018 at 09:28:00AM +, Chris Wilson wrote: > Quoting Tvrtko Ursulin (2018-03-22 09:22:55) > > > > On 21/03/2018 17:26, jeff.mc...@intel.com wrote: > > > From: Jeff McGee > > > > > > Force preemption uses engine reset to enforce a limit on the time > > > that a request targeted

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Chris Wilson
Quoting Tvrtko Ursulin (2018-03-22 09:22:55) > > On 21/03/2018 17:26, jeff.mc...@intel.com wrote: > > From: Jeff McGee > > > > Force preemption uses engine reset to enforce a limit on the time > > that a request targeted for preemption can block. This feature is > > a requirement in automotive s

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-22 Thread Tvrtko Ursulin
On 21/03/2018 17:26, jeff.mc...@intel.com wrote: From: Jeff McGee Force preemption uses engine reset to enforce a limit on the time that a request targeted for preemption can block. This feature is a requirement in automotive systems where the GPU may be shared by clients of critically high pr

[Intel-gfx] [RFC 0/8] Force preemption

2018-03-21 Thread jeff . mcgee
From: Jeff McGee Force preemption uses engine reset to enforce a limit on the time that a request targeted for preemption can block. This feature is a requirement in automotive systems where the GPU may be shared by clients of critically high priority and clients of low priority that may not have

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-16 Thread Chris Wilson
Quoting jeff.mc...@intel.com (2018-03-16 18:30:57) > From: Jeff McGee > > Force preemption uses engine reset to enforce a limit on the time > that a request targeted for preemption can block. This feature is > a requirement in automotive systems where the GPU may be shared by > clients of critica

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-16 Thread Chris Wilson
Quoting Chris Wilson (2018-03-16 20:53:01) > +static void preempt_reset(struct work_struct *work) > +{ > + struct intel_engine_cs *engine = > + container_of(work, typeof(*engine), execlists.preempt_reset); > + So thinking about the races you had in the reset, you need tasklet_

Re: [Intel-gfx] [RFC 0/8] Force preemption

2018-03-16 Thread Chris Wilson
Quoting jeff.mc...@intel.com (2018-03-16 18:30:57) > From: Jeff McGee > > Force preemption uses engine reset to enforce a limit on the time > that a request targeted for preemption can block. This feature is > a requirement in automotive systems where the GPU may be shared by > clients of critica

[Intel-gfx] [RFC 0/8] Force preemption

2018-03-16 Thread jeff . mcgee
From: Jeff McGee Force preemption uses engine reset to enforce a limit on the time that a request targeted for preemption can block. This feature is a requirement in automotive systems where the GPU may be shared by clients of critically high priority and clients of low priority that may not have