| Message ID | 20180802100631.31305-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/4] drm/i915: Drop stray clearing of rps->last_adj | expand |
Chris Wilson <chris@chris-wilson.co.uk> writes: > We used to reset last_adj to 0 on crossing a power domain boundary, to > slow down our rate of change. However, commit 60548c554be2 ("drm/i915: > Interactive RPS mode") accidentally caused it to be reset on every > frequency update, nerfing the fast response granted by the slow start > algorithm. > > Fixes: 60548c554be2 ("drm/i915: Interactive RPS mode") > Testcase: igt/pm_rps/mix-max-config-loaded > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> > --- > drivers/gpu/drm/i915/intel_pm.c | 1 - > 1 file changed, 1 deletion(-) > > diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c > index 2531eb75bdce..f90a3c7f1c40 100644 > --- a/drivers/gpu/drm/i915/intel_pm.c > +++ b/drivers/gpu/drm/i915/intel_pm.c > @@ -6371,7 +6371,6 @@ static void gen6_set_rps_thresholds(struct drm_i915_private *dev_priv, u8 val) > new_power = HIGH_POWER; > rps_set_power(dev_priv, new_power); > mutex_unlock(&rps->power.mutex); > - rps->last_adj = 0; > } > > void intel_rps_mark_interactive(struct drm_i915_private *i915, bool interactive) > -- > 2.18.0 > > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@lists.freedesktop.org > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoting Mika Kuoppala (2018-08-02 16:30:24) > Chris Wilson <chris@chris-wilson.co.uk> writes: > > > We used to reset last_adj to 0 on crossing a power domain boundary, to > > slow down our rate of change. However, commit 60548c554be2 ("drm/i915: > > Interactive RPS mode") accidentally caused it to be reset on every > > frequency update, nerfing the fast response granted by the slow start > > algorithm. > > > > Fixes: 60548c554be2 ("drm/i915: Interactive RPS mode") > > Testcase: igt/pm_rps/mix-max-config-loaded > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> > > Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Plonked this one as fixes the regression. 2&3 are nice cleanups, 4 seems like an unlikely boundary case, but I suspect kodi might be a good example. -Chris
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 2531eb75bdce..f90a3c7f1c40 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -6371,7 +6371,6 @@ static void gen6_set_rps_thresholds(struct drm_i915_private *dev_priv, u8 val) new_power = HIGH_POWER; rps_set_power(dev_priv, new_power); mutex_unlock(&rps->power.mutex); - rps->last_adj = 0; } void intel_rps_mark_interactive(struct drm_i915_private *i915, bool interactive)
We used to reset last_adj to 0 on crossing a power domain boundary, to slow down our rate of change. However, commit 60548c554be2 ("drm/i915: Interactive RPS mode") accidentally caused it to be reset on every frequency update, nerfing the fast response granted by the slow start algorithm. Fixes: 60548c554be2 ("drm/i915: Interactive RPS mode") Testcase: igt/pm_rps/mix-max-config-loaded Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> --- drivers/gpu/drm/i915/intel_pm.c | 1 - 1 file changed, 1 deletion(-)