On 01/08/2019 11:29, Chris Wilson wrote:
Quoting Tvrtko Ursulin (2019-08-01 11:18:24)static void -frequency_sample(struct drm_i915_private *dev_priv, unsigned int period_ns) +frequency_sample(struct intel_gt *gt, unsigned int period_ns) { - if (dev_priv->pmu.enable & - config_enabled_mask(I915_PMU_ACTUAL_FREQUENCY)) { + struct drm_i915_private *i915 = gt->i915; + struct intel_uncore *uncore = gt->uncore; + struct i915_pmu *pmu = &i915->pmu; + + if (pmu->enable & config_enabled_mask(I915_PMU_ACTUAL_FREQUENCY)) { u32 val;- val = dev_priv->gt_pm.rps.cur_freq;- if (dev_priv->gt.awake) { + val = i915->gt_pm.rps.cur_freq; + if (gt->awake) { intel_wakeref_t wakeref;- with_intel_runtime_pm_if_in_use(&dev_priv->runtime_pm,+ with_intel_runtime_pm_if_in_use(&i915->runtime_pm, wakeref) {We can replace this wakeref with if (intel_gt_pm_get_if_awake()) { instead of if (gt->awake);
Don't see it in my tree?
Which reminds me I had some patches to change the rc6 sampling over runtime suspend -- from your last vacation! My timing is perfect.
Okay, maybe doable. Can put on top of this series if I can merge the trivial refactors quickly today?
Regards, Tvrtko _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
