On Fri, Feb 17, 2023 at 8:43 PM Jason Gunthorpe [off-list ref] wrote:

On Fri, Feb 17, 2023 at 05:12:29AM -0500, Michael S. Tsirkin wrote:

quoted

On Thu, Feb 16, 2023 at 08:14:50PM -0400, Jason Gunthorpe wrote:

quoted

On Tue, Feb 07, 2023 at 08:08:43PM +0800, Nanyong Sun wrote:

quoted

From: Rong Wang <redacted>

Once enable iommu domain for one device, the MSI
translation tables have to be there for software-managed MSI.
Otherwise, platform with software-managed MSI without an
irq bypass function, can not get a correct memory write event
from pcie, will not get irqs.
The solution is to obtain the MSI phy base address from
iommu reserved region, and set it to iommu MSI cookie,
then translation tables will be created while request irq.

Probably not what anyone wants to hear, but I would prefer we not add
more uses of this stuff. It looks like we have to get rid of
iommu_get_msi_cookie() :\

I'd like it if vdpa could move to iommufd not keep copying stuff from
it..

Absolutely but when is that happening?

Don't know, I think it has to come from the VDPA maintainers, Nicolin
made some drafts but wasn't able to get it beyond that.

Cindy (cced) will carry on the work.

Thanks

Please have people who need more iommu platform enablement to pick it
up instead of merging hacks like this..

We are very close to having nested translation on ARM so anyone who is
serious about VDPA on ARM is going to need iommufd anyhow.

Jason