Re: [Intel-gfx] [PATCH v2 06/20] drm/i915: Polish some dbuf debugs

["Lisovskiy, Stanislav" <stanislav.lisovskiy@intel.com>]

On Thu, 2020-03-05 at 15:46 +0200, Ville Syrjälä wrote:
> On Thu, Mar 05, 2020 at 09:53:34AM +0000, Lisovskiy, Stanislav wrote:
> > On Wed, 2020-03-04 at 20:26 +0200, Ville Syrjälä wrote:
> > > On Wed, Mar 04, 2020 at 04:29:47PM +0000, Lisovskiy, Stanislav
> > > wrote:
> > > > On Tue, 2020-02-25 at 19:11 +0200, Ville Syrjala wrote:
> > > > > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > > > > 
> > > > > Polish some of the dbuf code to give more meaningful debug
> > > > > messages and whatnot. Also we can switch over to the per-
> > > > > device
> > > > > debugs/warns at the same time.
> > > > > 
> > > > > Cc: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
> > > > > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > > > > ---
> > > > >  .../drm/i915/display/intel_display_power.c    | 40
> > > > > +++++++++--
> > > > > ----
> > > > > ----
> > > > >  1 file changed, 19 insertions(+), 21 deletions(-)
> > > > > 
> > > > > diff --git
> > > > > a/drivers/gpu/drm/i915/display/intel_display_power.c
> > > > > b/drivers/gpu/drm/i915/display/intel_display_power.c
> > > > > index 6e25a1317161..e81e561e8ac0 100644
> > > > > --- a/drivers/gpu/drm/i915/display/intel_display_power.c
> > > > > +++ b/drivers/gpu/drm/i915/display/intel_display_power.c
> > > > > @@ -4433,11 +4433,12 @@ static void
> > > > > intel_power_domains_sync_hw(struct drm_i915_private
> > > > > *dev_priv)
> > > > >  	mutex_unlock(&power_domains->lock);
> > > > >  }
> > > > >  
> > > > > -static inline
> > > > > -bool intel_dbuf_slice_set(struct drm_i915_private *dev_priv,
> > > > > -			  i915_reg_t reg, bool enable)
> > > > > +static void intel_dbuf_slice_set(struct drm_i915_private
> > > > > *dev_priv,
> > > > > +				 enum dbuf_slice slice, bool
> > > > > enable)
> > > > >  {
> > > > > -	u32 val, status;
> > > > > +	i915_reg_t reg = DBUF_CTL_S(slice);
> > > > > +	bool state;
> > > > > +	u32 val;
> > > > >  
> > > > >  	val = intel_de_read(dev_priv, reg);
> > > > >  	val = enable ? (val | DBUF_POWER_REQUEST) : (val &
> > > > > ~DBUF_POWER_REQUEST);
> > > > > @@ -4445,13 +4446,10 @@ bool intel_dbuf_slice_set(struct
> > > > > drm_i915_private *dev_priv,
> > > > >  	intel_de_posting_read(dev_priv, reg);
> > > > >  	udelay(10);
> > > > >  
> > > > > -	status = intel_de_read(dev_priv, reg) &
> > > > > DBUF_POWER_STATE;
> > > > > -	if ((enable && !status) || (!enable && status)) {
> > > > > -		drm_err(&dev_priv->drm, "DBus power %s
> > > > > timeout!\n",
> > > > > -			enable ? "enable" : "disable");
> > > > > -		return false;
> > > > > -	}
> > > > > -	return true;
> > > > > +	state = intel_de_read(dev_priv, reg) &
> > > > > DBUF_POWER_STATE;
> > > > > +	drm_WARN(&dev_priv->drm, enable != state,
> > > > > +		 "DBuf slice %d power %s timeout!\n",
> > > > > +		 slice, enable ? "enable" : "disable");
> > > > >  }
> > > > >  
> > > > >  static void gen9_dbuf_enable(struct drm_i915_private
> > > > > *dev_priv)
> > > > > @@ -4467,14 +4465,16 @@ static void gen9_dbuf_disable(struct
> > > > > drm_i915_private *dev_priv)
> > > > >  void icl_dbuf_slices_update(struct drm_i915_private
> > > > > *dev_priv,
> > > > >  			    u8 req_slices)
> > > > >  {
> > > > > -	int i;
> > > > > -	int max_slices = INTEL_INFO(dev_priv)-
> > > > > > num_supported_dbuf_slices;
> > > > > 
> > > > > +	int num_slices = INTEL_INFO(dev_priv)-
> > > > > > num_supported_dbuf_slices;
> > > > > 
> > > > >  	struct i915_power_domains *power_domains = &dev_priv-
> > > > > > power_domains;
> > > > > 
> > > > > +	enum dbuf_slice slice;
> > > > >  
> > > > > -	drm_WARN(&dev_priv->drm, hweight8(req_slices) >
> > > > > max_slices,
> > > > > -		 "Invalid number of dbuf slices requested\n");
> > > > > +	drm_WARN(&dev_priv->drm, req_slices & ~(BIT(num_slices)
> > > > > - 1),
> > > > > +		 "Invalid set of dbuf slices (0x%x) requested
> > > > > (num dbuf
> > > > > slices %d)\n",
> > > > > +		 req_slices, num_slices);
> > > > >  
> > > > > -	DRM_DEBUG_KMS("Updating dbuf slices to 0x%x\n",
> > > > > req_slices);
> > > > > +	drm_dbg_kms(&dev_priv->drm,
> > > > > +		    "Updating dbuf slices to 0x%x\n",
> > > > > req_slices);
> > > > >  
> > > > >  	/*
> > > > >  	 * Might be running this in parallel to
> > > > > gen9_dc_off_power_well_enable
> > > > > @@ -4485,11 +4485,9 @@ void icl_dbuf_slices_update(struct
> > > > > drm_i915_private *dev_priv,
> > > > >  	 */
> > > > >  	mutex_lock(&power_domains->lock);
> > > > >  
> > > > > -	for (i = 0; i < max_slices; i++) {
> > > > > -		intel_dbuf_slice_set(dev_priv,
> > > > > -				     DBUF_CTL_S(i),
> > > > > -				     (req_slices & BIT(i)) !=
> > > > > 0);
> > > > > -	}
> > > > > +	for (slice = DBUF_S1; slice < num_slices; slice++)
> > > > > +		intel_dbuf_slice_set(dev_priv, slice,
> > > > > +				     req_slices & BIT(slice));
> > > > 
> > > > Would be cool to completely get rid of any magic numbers or
> > > > definitions, 0 in a sense is more universal here than DBUF_S1.
> > > > 
> > > > If we are counting slices as numbers it seems logical that we 
> > > > iterate [0..num_slices) range. If you want to name the first
> > > > slice
> > > > explicitly then it probably has to be something like iterator
> > > > logic, i.e for (slice = FIRST_SLICE; slice != LAST_SLICE;
> > > > slice++).
> > > > 
> > > > But trying to name it at the same time with comparing to total
> > > > _amount_
> > > > looks a bit confusing.
> > > 
> > > This is the standard pattern used all over the driver.
> > 
> > Well, you can enumerate objects using their qualitative or
> > quantative
> > characteristics, for instance if you take alphabet you would be
> > either enumerating letters like first is A and count until it
> > becomes
> > Z, or
> > you take indexes and say start from index 0 and count until it
> > becomes
> > 26.
> > 
> > What happens here is mixing those: i.e take letter A and count
> > until it
> > becomes 26, i.e mixing a name of an object with it's index, so
> > hopefully DBUF_S1 will always be defined as 0 :D
> 
> The old code assumed DBUF_S1==0 for the purposes of passing it to
> DBUF_CTL(), and for the purposes of the BIT(int).
> 
> The new code assumes DBUF_S1 == 0 for the purposes of terminating
> the iteration.
> 
> Suo siellä, vetelä täällä.
> 
> We may actually need to change the device info to contain a dbuf
> slice
> mask instead of just the number of slices. That's in case some hw has
> some slices fused off (not sure that's actually a thing but maybe,
> need to check the spec at some point for this). At that point we
> probably want to stash the whole thing into a for_each_dbuf_slice().

Yes, for_each_dbuf_slice sounds much better indeed, that way you
neither have to use indexes nor start from some explicitly hardcoded
slice.

> 
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx