[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20070411024021.GA5897@zhen-devel.sh.intel.com>
Date: Wed, 11 Apr 2007 10:40:21 +0800
From: Wang Zhenyu <zhenyu.z.wang@...el.com>
To: Andi Kleen <ak@...e.de>, Dave Jones <davej@...hat.com>
Cc: David Miller <davem@...emloft.net>, hch@...radead.org,
ashok.raj@...el.com, linux-kernel@...r.kernel.org, akpm@...l.org,
gregkh@...e.de, muli@...ibm.com, asit.k.mallick@...el.com,
suresh.b.siddha@...el.com, anil.s.keshavamurthy@...el.com,
arjan@...ux.intel.com, shaohua.li@...el.com
Subject: Re: [patch 5/8] [Intel IOMMU] Graphics driver workarounds to provide unity map
On 2007.04.10 11:12:17 +0000, Andi Kleen wrote:
> > > On Mon, Apr 09, 2007 at 02:55:57PM -0700, Ashok Raj wrote:
> > > > Most GFX drivers don't call standard PCI DMA APIs to allocate DMA buffer,
> > > > Such drivers will be broken with IOMMU enabled. To workaround this issue,
> > > > we added two options.
> > >
> > > All drm drivers do it. If the usual out of tree crap vendors are too
> > > stupid for their own sake it's their fault.
> > >
> > > So NACK to this patch.
> >
> > That's my feeling as well, everything we care about should be using
> > the proper APIs or else what is the point of them...
>
> They can't. There is no proper API to do IOMMU mappings from user space.
> And that is how Xorg on x86 works.
>
> I had some hackish patches to enable mapping on /sys/bus/pci/.../coherent_mem, but
> you need ioctls to pass out the translated address and it wasn't
> exactly pretty. Still also not sure that's the right way.
>
For agpgart based gfx driver, we need to enable dma mapping on agpgart module,
which would work in the IOMMU case. I attach my current patches to do this.
Dave, how do you think about it?
View attachment "agpgart_dma_mapping.patch" of type "text/plain" (5324 bytes)
Powered by blists - more mailing lists