Re: [Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-07 Thread Rafael J. Wysocki
On 2/6/2018 10:54 PM, Imre Deak wrote: Hi Rafael, On Tue, Feb 06, 2018 at 09:11:02PM +, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-02-06 18:33:11) From: Tvrtko Ursulin We are not allowed to call intel_runtime_pm_get from the PMU counter read callback

Re: [Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-07 Thread Chris Wilson
Quoting Tvrtko Ursulin (2018-02-07 09:20:27) > > On 06/02/2018 21:54, Imre Deak wrote: > > Hi Rafael, > > > > On Tue, Feb 06, 2018 at 09:11:02PM +, Chris Wilson wrote: > >> Quoting Tvrtko Ursulin (2018-02-06 18:33:11) > >>> From: Tvrtko Ursulin > >>> > >>> We are

Re: [Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-07 Thread Tvrtko Ursulin
On 06/02/2018 21:54, Imre Deak wrote: Hi Rafael, On Tue, Feb 06, 2018 at 09:11:02PM +, Chris Wilson wrote: Quoting Tvrtko Ursulin (2018-02-06 18:33:11) From: Tvrtko Ursulin We are not allowed to call intel_runtime_pm_get from the PMU counter read callback

Re: [Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-06 Thread Imre Deak
Hi Rafael, On Tue, Feb 06, 2018 at 09:11:02PM +, Chris Wilson wrote: > Quoting Tvrtko Ursulin (2018-02-06 18:33:11) > > From: Tvrtko Ursulin > > > > We are not allowed to call intel_runtime_pm_get from the PMU counter read > > callback since the former can sleep,

Re: [Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-06 Thread Chris Wilson
Quoting Tvrtko Ursulin (2018-02-06 18:33:11) > From: Tvrtko Ursulin > > We are not allowed to call intel_runtime_pm_get from the PMU counter read > callback since the former can sleep, and the latter is running under IRQ > context. > > To workaround this, we record the

[Intel-gfx] [PATCH v2] drm/i915/pmu: Fix sleep under atomic in RC6 readout

2018-02-06 Thread Tvrtko Ursulin
From: Tvrtko Ursulin We are not allowed to call intel_runtime_pm_get from the PMU counter read callback since the former can sleep, and the latter is running under IRQ context. To workaround this, we record the last known RC6 and while runtime suspended estimate its