From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joerg Roedel Subject: Re: Linux 2.6.39-rc3 Date: Fri, 15 Apr 2011 10:55:26 +0200 Message-ID: <20110415085526.GI18463@8bytes.org> References: <20110412184433.GF19819@8bytes.org> <20110413064609.GA18777@elte.hu> <20110413172147.GI19819@8bytes.org> <4DA5F62F.3030504@kernel.org> <4DA654D6.2060103@zytor.com> <20110414085624.GC18463@8bytes.org> <20110414210932.GG18463@8bytes.org> <1302855994.15520.158.camel@thor.local> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <1302855994.15520.158.camel@thor.local> Sender: linux-kernel-owner@vger.kernel.org To: Michel =?iso-8859-1?Q?D=E4nzer?= Cc: Alex Deucher , Yinghai Lu , Linux Kernel Mailing List , dri-devel@lists.freedesktop.org, "H. Peter Anvin" , Tejun Heo , Linus Torvalds , Thomas Gleixner List-Id: dri-devel@lists.freedesktop.org On Fri, Apr 15, 2011 at 10:26:34AM +0200, Michel D=E4nzer wrote: > On Don, 2011-04-14 at 23:09 +0200, Joerg Roedel wrote:=20 > > On Thu, Apr 14, 2011 at 10:28:43AM -0400, Alex Deucher wrote: > > > On Thu, Apr 14, 2011 at 4:56 AM, Joerg Roedel w= rote: > > > > And this makes a difference, with this change on-top of -rc3 th= e box boots > > > > fine. So there seems to be some dependency between the GART bas= e and the GTT > > > > base even when they are in different address spaces. > > > > > > > > Alex, can you comment on this? > > >=20 > > > As Dave said, they are completely different addresses spaces. Yo= u > > > could put the GPU aperture at 0 if you wanted (in fact we do on s= ome > > > chips). Perhaps there's some strange interaction with the nb gar= t > > > since the nb gart on that chipset was designed to be used for gra= phics > > > and the rs780/880 can be configured to use an agp aperture. > > > Unfortunately, I'm not that familiar with the nb gart. > >=20 > > Actually, the nb gart is part of the cpu. It is part of the cpu nor= th > > bridge and can translate io and cpu accesses. In fact, it is a rema= pper > > of physical memory addresses. > >=20 > > The problem seems to be related to specific gpu chips. On another > > notebook with an hd3000 card gtt and the nb gart aperture are both = on > > 0xa0000000 too but the box works fine. >=20 > Wasn't the working theory that the problem occurs if those two values > aren't the same? Yes it is, but this doesn't seem to be problematic on all readeon GPU chips. Joerg