From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ben Widawsky <ben@bwidawsk.net>
Subject: Re: [PATCH 04/10] drm/i915: add RPS configuration for
 Haswell
Date: Mon, 2 Jul 2012 14:19:11 -0700
Message-ID: <20120702141911.1aa63642@bwidawsk.net>
References: <1341240671-5843-1-git-send-email-eugeni.dodonov@intel.com>
	<1341240671-5843-5-git-send-email-eugeni.dodonov@intel.com>
	<20120702104930.5e992630@bwidawsk.net>
	<4FF1FE73.8090408@linux.intel.com>
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 cloud01.chad-versace.us (184-106-247-128.static.cloud-ips.com
	[184.106.247.128])
	by gabe.freedesktop.org (Postfix) with ESMTP id 8D63F9EAFA
	for <intel-gfx@lists.freedesktop.org>;
	Mon,  2 Jul 2012 14:19:18 -0700 (PDT)
In-Reply-To: <4FF1FE73.8090408@linux.intel.com>
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: eugeni.dodonov@intel.com
Cc: intel-gfx@lists.freedesktop.org
List-Id: intel-gfx@lists.freedesktop.org

On Mon, 02 Jul 2012 17:02:59 -0300
Eugeni Dodonov <eugeni.dodonov@linux.intel.com> wrote:

> On 07/02/2012 02:49 PM, Ben Widawsky wrote:
> > On Mon,  2 Jul 2012 11:51:05 -0300
> > Eugeni Dodonov <eugeni.dodonov@intel.com> wrote:
> > 
> >> Most of the RPS and RC6 enabling functionality is similar to what we had
> >> on Gen6/Gen7, so we preserve most of the registers.
> >>
> >> Note that Haswell only has RC6, so account for that as well. As suggested
> >> by Daniel Vetter, to reduce the amount of changes in the patch, we still
> >> write the RC6p/RC6pp thresholds, but those are ignored on Haswell.
> >>
> >> Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
> > Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
> 
> Daniel, to please Ben Widawsky, you could add an extra paragraph in my
> message, saying:
> 
> 'On Haswell, the suggested values for the RP UP and RP DOWN interrupts
> have changed in a way to have lower thresholds for all the operations.
> However, the suggested values for pre-Haswell machines stay the same. In
> order to follow what the hardware designers suggest, we program those as
> suggested for each generation now.
> 
> Also, Haswell introduces a new scheduling bit for the RP DOWN
> scheduling, which we now use by default'.
> 
> Eugeni

The code itself states the above perfectly well. I was really trying
to understand the, "why."


-- 
Ben Widawsky, Intel Open Source Technology Center