[Intel-gfx] [PATCH v2 1/2] drm/i915/debugfs: Print sink PSR status

[Souza, Jose]

On Fri, 2018-04-06 at 18:40 -0700, Dhinakaran Pandiyan wrote:
> 
> 
> On Sat, 2018-04-07 at 00:49 +0000, Souza, Jose wrote:
> > On Fri, 2018-04-06 at 16:36 -0700, José Roberto de Souza wrote:
> > > On Mon, 2018-04-02 at 15:38 -0700, Pandiyan, Dhinakaran wrote:
> > > > On Mon, 2018-04-02 at 13:51 -0700, José Roberto de Souza wrote:
> > > > > IGT tests could be improved with sink status, knowing for
> > > > > sure
> > > > > that
> > > > > hardware have activate or exit PSR.
> > > > > 
> > > > > Reviewed-by: Dhinakaran Pandiyan <dhinakaran.pandiyan at intel.c
> > > > > om>
> > > > 
> > > > 
> > > > Please don't merge this patch yet. While the patch itself is
> > > > correct,
> > > > testing it brings up an interesting problem.
> > > > 
> > > > Printing the sink_status() leads to power_get(aux_domain) which
> > > > wakes
> > > > up
> > > > the display engine from DC5/6. This results in a HW triggered
> > > > PSR
> > > > exit,
> > > > effectively altering the state that we are trying to read. I
> > > > would
> > > > like
> > > > to understand the problem fully before merging the patch.
> > > 
> > > In my tests(PSR1 only) with DMC loaded, I can see HW trigerring
> > > PSR
> > > exit when reading i915_edp_psr_status with or without this patch,
> > > the
> > > event that caused is 'SRD Mask Reg Write'.
> > > After masking EDP_PSR_DEBUG_MASK_DISP_REG_WRITE in EDP_PSR_DEBUG
> > > it
> > > do
> > > not happens with or without this patch.
> > > 
> 
> Reads causing PSR exit is a concern, but I have seen edp_psr_status
> showing PSR live status as active. If reads triggered exit, then it
> should always be inactive. Unless the exit happens after the status
> is
> read.

Same thing for sink status, I can see that PSR is exiting but the
status is still active on sink.

> 
> > > Masking this event did not bring more IGT errors than we have
> > > now,
> > > what
> > > do you think Dhinakaran? Should we mask it?
> 
> We'll need PSR to exit on some register writes, example flips or
> cursor
> moves. 

PIPE related writes are still causing PSR to exit, PSR_EVENT have a bit
for Register Update and other for Pipe Registers Update but it don't
have a counter part in PSR_MASK/SRD_DEBUG to make the pipe register
updated not trigger a PSR exit.

> 
> > Another information missed:
> > 
> > When doing a dpcd(I did not look at what exacly is causing it)
> > transaction it is causing the HW to be awaken from DC6 with or
> > without
> > masking EDP_PSR_DEBUG_MASK_DISP_REG_WRITE 
> 
> This is expected, AUX_A is in the DC_OFF power domain, so the driver
> forces a DC6 exit.
> 
> > but when masking it do not
> > cause a PSR exit.
> > Reading i915_sr_status also cause HW to be awaken from DC6.
> > 
> 
> What hardware are you testing on? And is this a PSR2 panel?

KBL it is a PSR2 pannels but I did some changes to only enable PSR1.

> 
> 
> 
> > > 
> > > > 
> > > > 
> > > > > Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
> > > > > Signed-off-by: José Roberto de Souza <jose.souza at intel.com>
> > > > > ---
> > > > >  drivers/gpu/drm/i915/i915_debugfs.c | 29
> > > > > +++++++++++++++++++++++++++++
> > > > >  1 file changed, 29 insertions(+)
> > > > > 
> > > > > diff --git a/drivers/gpu/drm/i915/i915_debugfs.c
> > > > > b/drivers/gpu/drm/i915/i915_debugfs.c
> > > > > index 1dba2c451255..c9ac946b62c9 100644
> > > > > --- a/drivers/gpu/drm/i915/i915_debugfs.c
> > > > > +++ b/drivers/gpu/drm/i915/i915_debugfs.c
> > > > > @@ -2603,6 +2603,26 @@ static const char
> > > > > *psr2_live_status(u32
> > > > > val)
> > > > >  	return "unknown";
> > > > >  }
> > > > >  
> > > > > +static const char *psr_sink_status(u8 val)
> > > > > +{
> > > > > +	static const char * const sink_status[] = {
> > > > > +		"inactive",
> > > > > +		"transition to active, capture and display",
> > > > > +		"active, display from RFB",
> > > > > +		"active, capture and display on sink device
> > > > > timings",
> > > > > +		"transition to inactive, capture and
> > > > > display,
> > > > > timing re-sync",
> > > > > +		"reserved",
> > > > > +		"reserved",
> > > > > +		"sink internal error"
> > > > > +	};
> > > > > +
> > > > > +	val &= DP_PSR_SINK_STATE_MASK;
> > > > > +	if (val < ARRAY_SIZE(sink_status))
> > > > > +		return sink_status[val];
> > > > > +
> > > > > +	return "unknown";
> > > > > +}
> > > > > +
> > > > >  static int i915_edp_psr_status(struct seq_file *m, void
> > > > > *data)
> > > > >  {
> > > > >  	struct drm_i915_private *dev_priv = node_to_i915(m-
> > > > > > private);
> > > > > 
> > > > > @@ -2684,6 +2704,15 @@ static int i915_edp_psr_status(struct
> > > > > seq_file *m, void *data)
> > > > >  		seq_printf(m, "EDP_PSR2_STATUS: %x [%s]\n",
> > > > >  			   psr2, psr2_live_status(psr2));
> > > > >  	}
> > > > > +
> > > > > +	if (dev_priv->psr.enabled) {
> > > > > +		struct drm_dp_aux *aux = &dev_priv-
> > > > > >psr.enabled-
> > > > > > aux;
> > > > > 
> > > > > +		u8 val;
> > > > > +
> > > > > +		if (drm_dp_dpcd_readb(aux, DP_PSR_STATUS,
> > > > > &val)
> > > > > ==
> > > > > 1)
> > > > > +			seq_printf(m, "Sink PSR status: 0x%x
> > > > > [%s]\n", val,
> > > > > +				   psr_sink_status(val));
> > > > > +	}
> > > > >  	mutex_unlock(&dev_priv->psr.lock);
> > > > >  
> > > > >  	intel_runtime_pm_put(dev_priv);
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 
>