[Intel-gfx] [PATCH 16/16] drm/i915: Allow RPS waitboosting to use max GPU frequency

[Chris Wilson]

On Thu, May 21, 2015 at 02:55:16PM +0200, Daniel Vetter wrote:
> On Mon, Apr 27, 2015 at 01:41:27PM +0100, Chris Wilson wrote:
> > Ignore the restriction imposed by the user for when the GPU is stalling
> > the clients and dropping frames. We will return back to the user limits
> > immediately once the stall is over.
> > 
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > ---
> >  drivers/gpu/drm/i915/intel_pm.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> > index 0cc9e95f70d3..9205b5fe2186 100644
> > --- a/drivers/gpu/drm/i915/intel_pm.c
> > +++ b/drivers/gpu/drm/i915/intel_pm.c
> > @@ -4161,7 +4161,7 @@ void intel_rps_boost(struct drm_i915_private *dev_priv,
> >  		rps = NULL;
> >  
> >  	mutex_lock(&dev_priv->rps.hw_lock);
> > -	val = dev_priv->rps.max_freq_softlimit;
> > +	val = dev_priv->rps.max_freq;
> 
> On second thought this has the risk of going into the overclocking range,
> which we probably don't want to do behind the user's back. I've left this
> one out for now.

Hmm, I thought I resent this using rps.boost_freq... Anyway I have
already respun this using boost_freq which fortunately avoids the
overclock range (since it is initialised before we define the
overclocking) and then exposes it to the user through sysfs (which
should also allow the naysayers to disable RPS boosting).
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre