From mboxrd@z Thu Jan  1 00:00:00 1970
From: Steven Newbury <steve@snewbury.org.uk>
Subject: Re: PCI resources above 4GB
Date: Thu, 12 Apr 2012 17:29:43 +0100
Message-ID: <1334248183.2910.3.camel@Nokia-N900>
References: <1333968563.5678.19.camel@infinity>
	 <CAErSpo7hGvJSmq4U37XN9T7ZaYn55fyy9t3ER5BBCDdQ3XZxqw@mail.gmail.com>
	 <4F84110E.3000400@snewbury.org.uk>
	 <CAE9FiQVHf_pcXUspOtLXx4XminmzdWja3Ra2Wh5jQRbfNOQ1eQ@mail.gmail.com>
	 <4F8467AA.90305@snewbury.org.uk>
	 <CAE9FiQWaop0NE7ZJu6jVW3SnxLkOJ9hRKh4xMNk64XjCAMRMMw@mail.gmail.com>
	 <4F848357.3060007@snewbury.org.uk>
	 <CAE9FiQXTgh3X4x+xD1T0ipsEbhbPtctYucfGeuWK=cq7kHeZ1Q@mail.gmail.com>
	 <4F848E10.1090703@snewbury.org.uk>
	 <CAE9FiQXvwYrJhLi91VDVtD4+OOfbGQ1U20-Cap+8tBZUt7QCkA@mail.gmail.com>
	 <1334089568.4083.2.camel@Nokia-N900>
	 <CAE9FiQWnvB2VbBQQaEnDE0nf+Zwy6nyoGrEuVrUesQEzgYfbvA@mail.gmail.com>
	 <4F84A3EC.6030903@snewbury.org.uk>
	 <CAE9FiQU+EEtBWrc66Sw4GFQjSy9Vh9fyFvhgQUheev_aVpAzrw@mail.gmail.com>
	  <1334229754.30606.7.camel@Nokia-N900>
Reply-To: Steven Newbury <steve@snewbury.org.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Return-path: <linux-pci-owner@vger.kernel.org>
In-Reply-To: <1334229754.30606.7.camel@Nokia-N900>
Content-ID: <1334248182.2910.2.camel@Nokia-N900>
Sender: linux-pci-owner@vger.kernel.org
To: Yinghai Lu <yinghai@kernel.org>, "Barnes, Jesse" <jesse.barnes@intel.com>, Dave Airlie <airlied@linux.ie>
Cc: Bjorn Helgaas <bhelgaas@google.com>, linux-pci@vger.kernel.org, DRI mailing list <dri-devel@lists.freedesktop.org>
List-Id: dri-devel@lists.freedesktop.org

On Thu, 12 Apr 2012, 12:22:34 BST, Steven Newbury <steve@snewbury.org.u=
k> wrote:
> I've attempted to modify probe.c to disable 64-bit BARs not allocated
> above 4G so they get reallocated above when possible later.=C2=A0  It=
 seemed
> to work, but again broke GMA despite the BAR originally containing an
> invalid address as mentioned above, it seems for some reason somethin=
g
> is different when the conflict is detected and rellocated, compared t=
o
> disabling it early then allocating a valid value..?
I understand now why it didn't work.  Memory decoding was enabled in th=
e GMA device.  When it's detected as in conflict with system RAM it get=
s reallocated cleanly, but setting the BAR to 0 prevents that from happ=
ening.  Somehow I need to get the resources I want moved onto a realloc=
 list, but I can't work out where or how...