[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [PATCH] PCI/MSI: don't disable AMD IOMMU MSI on Xen dom0
- To: Jan Beulich <JBeulich@xxxxxxxx>
- From: Wei Wang <wei.wang2@xxxxxxx>
- Date: Thu, 21 Jun 2012 15:27:12 +0200
- Cc: Jeremy Fitzhardinge <jeremy@xxxxxxxx>, "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>, KonradRzeszutek Wilk <konrad.wilk@xxxxxxxxxx>, Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, linux-pci@xxxxxxxxxxxxxxx, Jesse Barnes <jbarnes@xxxxxxxxxxxxxxxx>, ebiederm@xxxxxxxxxxxx, SherryHurwitz <sherry.hurwitz@xxxxxxx>, stable@xxxxxxxxxx
- Delivery-date: Thu, 21 Jun 2012 13:29:34 +0000
- List-id: Xen developer discussion <xen-devel.lists.xen.org>
On 06/21/2012 03:24 PM, Jan Beulich wrote:
On 21.06.12 at 15:10, Wei Wang<wei.wang2@xxxxxxx> wrote:
On 06/21/2012 02:45 PM, Jan Beulich wrote:
On 21.06.12 at 14:28, Wei Wang<wei.wang2@xxxxxxx> wrote:
AMD IOMMU is an independent pci-e endpoint, and this function will not
be used for other purposes other than containing an iommu. So I don't
see that iommu will share bdf value with other devices.
The question is not regarding bdf, but regarding whether under
the same seg:bus:dev there might be multiple functions, one of
which is the IOMMU, and if so, whether the IOMMU would be
guaranteed to have a non-zero function number.
In a real system (single or multiple iommu), amd iommu shares the same
device number with north bridge but has function number 2.. (e.g
bus:00.2) Howerver according to spec, it does not guaranteed to have
non-zero function number. So what is the problem you see if iommu uses
fun0 on a multi-func device?
If it's on func 0 and gets hidden completely (as done by your
partial patch), other functions won't be found when scanning
for them (because secondary functions get looked at only
when func 0 actually exists, as otherwise evaluating the header
type register is invalid).
OK, understood. Then I think we do need to allow pci cfg read for iommu
Xen-devel mailing list