[PATCH] Support PCIe Alternative RID Interpretation (ARI)
Konstantin Belousov
kostikbel at gmail.com
Fri Mar 28 13:35:34 UTC 2014
On Thu, Mar 27, 2014 at 02:48:36PM -0400, Ryan Stone wrote:
> On Tue, Mar 25, 2014 at 5:13 PM, Konstantin Belousov
> <kostikbel at gmail.com> wrote:
> > Well, either the interface I described is provided by pci core, or
> > iommu has to de-facto implement it itself. IMO it is clearer to have
> > it in pci, but I do not want to block your work on this.
>
> Yes, but this amounts to some simple masking and shifting on the RID.
> I don't think that's a very high burden.
Well, as I learned during iommu development, a typo there could lay
unnoticed for long time and than beat. This is due to most modern systems
being PCIe and slot/function often equal to zero, or at least function
equal to zero.
>
> > I mean, that slot and func should be obtained using pci accessors where
> > needed. It is definitely not perf-critical, and I dislike having both
> > bsf and rid in the context structure more, then using accessors.
>
> Ok, I've updated the DMAR patch to use pci accessors instead. This
> required moving the initialization of ctx_tag.owner earlier in the
> initialization of the DMAR ctx, but beyond that the change was
> trivial.
What is the URL ?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/freebsd-hackers/attachments/20140328/b87a0389/attachment.sig>
More information about the freebsd-hackers
mailing list