Hi Bastian,
On Sat, Mar 31, 2007 at 04:05:33PM +0200, Bastian Blank wrote:
> On Thu, Mar 22, 2007 at 01:15:31AM -0700, Steve Langasek wrote:
> > In the meantime, I don't see any reason why we shouldn't patch the kernel to
> > disable hw iommu on nvidia systems only. I believe the attached patch
> > should do this. Are you in a position to confirm that this does disable hw
> > iommu for you?
> The kernel already includes similar code for VIA chipsets. Why is this
> snippet slightly different to the original one?
> Upstream head uses
> | static void __init via_bugs(void)
> | {
> | #ifdef CONFIG_IOMMU
> | if ((end_pfn > MAX_DMA32_PFN || force_iommu) &&
> | !iommu_aperture_allowed) {
> | printk(KERN_INFO
> | "Looks like a VIA chipset. Disabling IOMMU. Override with
> iommu=allowed\n");
> | iommu_aperture_disabled = 1;
> | }
> | #endif
> | }
It's my understanding that this code for VIA is disabling *all* use of an
IOMMU aperture for DMA, whereas what we want for NVidia chips is only to
disable use of *hardware* IOMMU, hence the difference.
If I've misunderstood, I'll be happy to correct.
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
[EMAIL PROTECTED] http://www.debian.org/
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
