[Intel-gfx] [PATCH] drm/i915/icl: Prevent incorrect DBuf enabling

Imre Deak imre.deak at intel.com
Thu Mar 7 19:45:29 UTC 2019


On Thu, Mar 07, 2019 at 09:59:39AM -0800, Rodrigo Vivi wrote:
> On Thu, Mar 07, 2019 at 12:32:35PM +0200, Imre Deak wrote:
> > Pretend that we have only 1 DBuf slice and that 1 slice is always
> > enabled, until we have a proper way for on-demand toggling of the second
> > slice.  Currently we'll try to incorrectly enable DBuf even when all
> > pipes are disabled and we are already runtime suspended (as the computed
> > number of DBuf slices will be 1 in that case).
> > 
> > This also means we'll leave the second slice enabled redundantly (except
> > when suspended), but that's an acceptable tradeoff until we have a
> > proper solution.
> 
> redundant is the word of the week :)
> 
> > 
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=108756
> > Cc: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > Signed-off-by: Imre Deak <imre.deak at intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_pm.c         |  7 ++++++-
> >  drivers/gpu/drm/i915/intel_runtime_pm.c | 12 ++++++++++--
> >  2 files changed, 16 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> > index 9c97a95c1816..bece16ae6d15 100644
> > --- a/drivers/gpu/drm/i915/intel_pm.c
> > +++ b/drivers/gpu/drm/i915/intel_pm.c
> > @@ -3624,7 +3624,12 @@ static u8 intel_enabled_dbuf_slices_num(struct drm_i915_private *dev_priv)
> >  	if (INTEL_GEN(dev_priv) < 11)
> >  		return enabled_slices;
> >  
> > -	if (I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE)
> > +	/*
> > +	 * FIXME: for now we'll only ever use 1 slice; pretend that we have
> > +	 * only that 1 slice enabled until we have a proper way for on-demand
> > +	 * toggling of the second slice.
> > +	 */
> > +	if (0 && I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE)
> 
> what if just add a check for enabled_slices >= 2 and avoid the FIXME here?

You mean >= 1 during state verification? That would also need its own
FIXME:) And this one here matches better what we do elsewhere to keep
things working.

> 
> but up to you:
> 
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
> 
> >  		enabled_slices++;
> >  
> >  	return enabled_slices;
> > diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
> > index aa974b11928a..676a89bb8194 100644
> > --- a/drivers/gpu/drm/i915/intel_runtime_pm.c
> > +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
> > @@ -3576,7 +3576,11 @@ static void icl_dbuf_enable(struct drm_i915_private *dev_priv)
> >  	    !(I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE))
> >  		DRM_ERROR("DBuf power enable timeout\n");
> >  	else
> > -		dev_priv->wm.skl_hw.ddb.enabled_slices = 2;
> > +		/*
> > +		 * FIXME: for now pretend that we only have 1 slice, see
> > +		 * intel_enabled_dbuf_slices_num().
> > +		 */
> > +		dev_priv->wm.skl_hw.ddb.enabled_slices = 1;
> >  }
> >  
> >  static void icl_dbuf_disable(struct drm_i915_private *dev_priv)
> > @@ -3591,7 +3595,11 @@ static void icl_dbuf_disable(struct drm_i915_private *dev_priv)
> >  	    (I915_READ(DBUF_CTL_S2) & DBUF_POWER_STATE))
> >  		DRM_ERROR("DBuf power disable timeout!\n");
> >  	else
> > -		dev_priv->wm.skl_hw.ddb.enabled_slices = 0;
> > +		/*
> > +		 * FIXME: for now pretend that the first slice is always
> > +		 * enabled, see intel_enabled_dbuf_slices_num().
> > +		 */
> > +		dev_priv->wm.skl_hw.ddb.enabled_slices = 1;
> >  }
> >  
> >  static void icl_mbus_init(struct drm_i915_private *dev_priv)
> > -- 
> > 2.13.2
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx at lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/intel-gfx


More information about the Intel-gfx mailing list