[Intel-gfx] [PATCH 2/5] drm/i915/psr: Kill scheduled work for Core platforms.

Pandiyan, Dhinakaran dhinakaran.pandiyan at intel.com
Mon Feb 26 23:22:35 UTC 2018




On Mon, 2018-02-26 at 15:12 -0800, Rodrigo Vivi wrote:
> On Fri, Feb 23, 2018 at 03:46:09PM -0800, Pandiyan, Dhinakaran wrote:
> > On Tue, 2018-02-13 at 15:26 -0800, Rodrigo Vivi wrote:
> > > It is a fact that scheduled work is now improved.
> > > 
> > > But it is also a fact that on core platforms that shouldn't
> > > be needed. We only need to actually wait on VLV/CHV.
> > > 
> > > The immediate enabling is actually not an issue for the
> > > HW perspective for core platforms that have HW tracking.
> > > HW will wait few identical idle frames before transitioning
> > > to actual psr active anyways.
> > > 
> > > Note that this patch also remove the delayed activation
> > > on HSW and BDW introduced by commit 'd0ac896a477d
> > > ("drm/i915: Delay first PSR activation.")'. This was
> > > introduced to fix a blank screen on VLV/CHV and also
> > > masked some frozen screens on other core platforms.
> > > Probably the same that we are now properly hunting and fixing.
> > > 
> > > Furthermore, if we stop using delayed activation on core
> > > platforms we will be able, on following up patches,
> > > to use available workarounds to make HW tracking properly
> > > exit PSR instead of the big nuke of disabling psr and
> > > re-enable on exit and activate respectively.
> > > At least on few reliable cases.
> > > 
> > > Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan at intel.com>
> > > Signed-off-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
> > > ---
> > >  drivers/gpu/drm/i915/i915_debugfs.c | 14 +++++++-------
> > >  drivers/gpu/drm/i915/intel_psr.c    | 27 +++++++++++++++++++--------
> > >  2 files changed, 26 insertions(+), 15 deletions(-)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
> > > index da80ee16a3cf..541290c307c7 100644
> > > --- a/drivers/gpu/drm/i915/i915_debugfs.c
> > > +++ b/drivers/gpu/drm/i915/i915_debugfs.c
> > > @@ -2522,18 +2522,18 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
> > >  	seq_printf(m, "Busy frontbuffer bits: 0x%03x\n",
> > >  		   dev_priv->psr.busy_frontbuffer_bits);
> > >  
> > > -	if (timer_pending(&dev_priv->psr.activate_timer))
> > > -		seq_printf(m, "Activate scheduled: yes, in %dms\n",
> > > -			   jiffies_to_msecs(dev_priv->psr.activate_timer.expires - jiffies));
> > > -	else
> > > -		seq_printf(m, "Activate scheduled: no\n");
> > > -
> > > -	if (HAS_DDI(dev_priv)) {
> > > +	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) {
> > 
> > I don't get this change, it is better to retain HAS_DDI().
> > 
> > 
> > >  		if (dev_priv->psr.psr2_support)
> > >  			enabled = I915_READ(EDP_PSR2_CTL) & EDP_PSR2_ENABLE;
> > >  		else
> > >  			enabled = I915_READ(EDP_PSR_CTL) & EDP_PSR_ENABLE;
> > >  	} else {
> > > +		if (timer_pending(&dev_priv->psr.activate_timer))
> > > +			seq_printf(m, "Activate scheduled: yes, in %dms\n",
> > > +				   jiffies_to_msecs(dev_priv->psr.activate_timer.expires - jiffies));
> > > +		else
> > > +			seq_printf(m, "Activate scheduled: no\n");
> > > +
> > >  		for_each_pipe(dev_priv, pipe) {
> > >  			enum transcoder cpu_transcoder =
> > >  				intel_pipe_to_cpu_transcoder(dev_priv, pipe);
> > > diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
> > > index 826b480841ac..13409c6301e8 100644
> > > --- a/drivers/gpu/drm/i915/intel_psr.c
> > > +++ b/drivers/gpu/drm/i915/intel_psr.c
> > > @@ -455,6 +455,8 @@ static void intel_psr_schedule(struct drm_i915_private *i915,
> > >  {
> > >  	unsigned long next;
> > >  
> > > +	WARN_ON(!IS_VALLEYVIEW(i915) && !IS_CHERRYVIEW(i915));
> > > +
> > How about using only !(IS_VLV() || IS_CHV) in this file.
> > 
> > I think this is a reasonable check to have, please add a return too.
> > 	WARN_ON(!(IS_VLV() || IS_CHV())
> > 		return;	
> > 
> > >  	lockdep_assert_held(&i915->psr.lock);
> > >  
> > >  	/*
> > > @@ -543,7 +545,7 @@ void intel_psr_enable(struct intel_dp *intel_dp,
> > >  	dev_priv->psr.enable_source(intel_dp, crtc_state);
> > >  	dev_priv->psr.enabled = intel_dp;
> > >  
> > > -	if (INTEL_GEN(dev_priv) >= 9) {
> > > +	if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) {
> > 
> > How about inverting this? 
> > 
> > if (IS_VLV() || IS_CHV())
> > 	intel_psr_schedule()
> > else 
> > 	intel_psr_activate()
> > 
> > is easier to follow IMO.
> > 
> >
> > What is the reason to not use HAS_DDI() ?
> 
> I believe HAS_DDI is not meaningful here. It is just a coincidence.
> 
> maybe we could simplify everything with has_hw_tracking.... but also
> a coincidence in other cases...
> 
> maybe create something meaninfull like VLV_PSR... :/
> 
> no strong feelings actually...
> 

Thanks for the clarification, IS_VLV() || IS_CHV() is good enough in
that case. Since you also have a comment explaining that a blank screen
is seen if we activate PSR immediately on VLV/CHV, let's go ahead with
the inverted 'if'.



> > 
> > >  		intel_psr_activate(intel_dp);
> > >  	} else {
> > >  		/*
> > > @@ -553,8 +555,6 @@ void intel_psr_enable(struct intel_dp *intel_dp,
> > >  		 * However on some platforms we face issues when first
> > >  		 * activation follows a modeset so quickly.
> > >  		 *     - On VLV/CHV we get bank screen on first activation
> > > -		 *     - On HSW/BDW we get a recoverable frozen screen until
> > > -		 *       next exit-activate sequence.
> > >  		 */
> > >  		intel_psr_schedule(dev_priv,
> > >  				   intel_dp->panel_power_cycle_delay * 5);
> > > @@ -687,6 +687,8 @@ static void intel_psr_work(struct work_struct *work)
> > >  	struct drm_crtc *crtc = dp_to_dig_port(intel_dp)->base.base.crtc;
> > >  	enum pipe pipe = to_intel_crtc(crtc)->pipe;
> > >  
> > > +	WARN_ON(!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv));
> > > +
> > 
> > This is not needed, we don't even setup the work function for VLV/CHV.
> > Since the functions are all contained in this one file, I don't see much
> > risk of somehow ending up here.
> > 
> > >  	/* We have to make sure PSR is ready for re-enable
> > >  	 * otherwise it keeps disabled until next full enable/disable cycle.
> > >  	 * PSR might take some time to get fully disabled
> > > @@ -757,6 +759,8 @@ static void intel_psr_timer_fn(struct timer_list *timer)
> > >  	struct drm_i915_private *i915 =
> > >  		from_timer(i915, timer, psr.activate_timer);
> > >  
> > > +	WARN_ON(!IS_VALLEYVIEW(i915) && !IS_CHERRYVIEW(i915));
> > > +
> > 
> > This is not needed too.
> > 
> > >  	/*
> > >  	 * We need a non-atomic context to activate PSR.  Using
> > >  	 * delayed_work wouldn't be an improvement -- delayed_work is
> > > @@ -945,9 +949,12 @@ void intel_psr_flush(struct drm_i915_private *dev_priv,
> > >  	if (frontbuffer_bits)
> > >  		intel_psr_exit(dev_priv);
> > >  
> > > -	if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits)
> > > -		intel_psr_schedule(dev_priv, 100);
> > > -
> > > +	if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits) {
> > > +		if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
> > > +			intel_psr_schedule(dev_priv, 100);
> > > +		else
> > > +			intel_psr_activate(dev_priv->psr.enabled);
> > > +	}
> > >  	mutex_unlock(&dev_priv->psr.lock);
> > >  }
> > >  
> > > @@ -994,8 +1001,12 @@ void intel_psr_init(struct drm_i915_private *dev_priv)
> > >  		dev_priv->psr.link_standby = false;
> > >  	}
> > >  
> > > -	timer_setup(&dev_priv->psr.activate_timer, intel_psr_timer_fn, 0);
> > > -	INIT_WORK(&dev_priv->psr.activate_work, intel_psr_work);
> > > +	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {
> > > +		timer_setup(&dev_priv->psr.activate_timer,
> > > +			    intel_psr_timer_fn, 0);
> > > +		INIT_WORK(&dev_priv->psr.activate_work, intel_psr_work);
> > > +	}
> > > +
> > >  	mutex_init(&dev_priv->psr.lock);
> > >  
> > >  	if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) {


More information about the Intel-gfx mailing list