From mboxrd@z Thu Jan  1 00:00:00 1970
From: Jesse Barnes <jbarnes@virtuousgeek.org>
Subject: Re: [PATCH 2/2] drm/i915: load a ring frequency scaling
 table
Date: Tue, 28 Jun 2011 10:40:49 -0700
Message-ID: <20110628104049.1a71d315@jbarnes-desktop>
References: <1308939194-3568-1-git-send-email-jbarnes@virtuousgeek.org>
	<1308939194-3568-2-git-send-email-jbarnes@virtuousgeek.org>
	<87iprruplj.fsf@eliezer.anholt.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Return-path: <intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org>
Received: from oproxy4-pub.bluehost.com (oproxy4-pub.bluehost.com
	[69.89.21.11])
	by gabe.freedesktop.org (Postfix) with SMTP id BC7D79E880
	for <intel-gfx@lists.freedesktop.org>;
	Tue, 28 Jun 2011 10:40:59 -0700 (PDT)
In-Reply-To: <87iprruplj.fsf@eliezer.anholt.net>
List-Unsubscribe: <http://lists.freedesktop.org/mailman/options/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=unsubscribe>
List-Archive: <http://lists.freedesktop.org/archives/intel-gfx>
List-Post: <mailto:intel-gfx@lists.freedesktop.org>
List-Help: <mailto:intel-gfx-request@lists.freedesktop.org?subject=help>
List-Subscribe: <http://lists.freedesktop.org/mailman/listinfo/intel-gfx>,
	<mailto:intel-gfx-request@lists.freedesktop.org?subject=subscribe>
Sender: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
Errors-To: intel-gfx-bounces+gcfxdi-intel-gfx=m.gmane.org@lists.freedesktop.org
To: Eric Anholt <eric@anholt.net>
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On Mon, 27 Jun 2011 10:54:48 -0700
Eric Anholt <eric@anholt.net> wrote:

> > +	for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay;
> > +	     gpu_freq--) {
> > +		int diff = dev_priv->max_delay - gpu_freq;
> > +
> > +		/*
> > +		 * For GPU frequencies less than 750MHz, just use the lowest
> > +		 * ring freq.
> > +		 */
> > +		if (gpu_freq < min_freq)
> > +			ia_freq = 800;
> > +		else
> > +			ia_freq = max_ia_freq - ((diff * scaling_factor) / 2);
> > +		ia_freq = DIV_ROUND_CLOSEST(ia_freq, 100);  
> 
> If the GPU has a wide enough clock range (diff large) and the CPU is low
> enough clocked (max_ia_freq low now), could we end up with the ia_freq <
> 800, and would that be a bad thing?  In other words, should
> scaling_factor be non-constant?

scaling_factor probably should be non-constant, but I don't know what
function it should follow.

ia_freq < 800 shouldn't break anything, but would probably result in
sub-optimal GPU performance.  OTOH it would save power...

-- 
Jesse Barnes, Intel Open Source Technology Center