Re: [PATCH 1/1] iommu/vt-d: PRS isn't usable if PDS isn't supported

[Baolu Lu]

On 9/15/25 19:30, Joel Granados wrote:
> On Mon, Sep 15, 2025 at 02:29:46PM +0800, Lu Baolu wrote:
> > The specification, Section 7.10, "Software Steps to Drain Page Requests &
> > Responses," requires software to submit an Invalidation Wait Descriptor
> > (inv_wait_dsc) with the Page-request Drain (PD=1) flag set, along with
> > the Invalidation Wait Completion Status Write flag (SW=1). It then waits
> > for the Invalidation Wait Descriptor's completion.
> >
> > However, the PD field in the Invalidation Wait Descriptor is optional, as
> > stated in Section 6.5.2.9, "Invalidation Wait Descriptor":
> >
> > "Page-request Drain (PD): Remapping hardware implementations reporting
> >   Page-request draining as not supported (PDS = 0 in ECAP_REG) treat this
> >   field as reserved."
> >
> > This implies that if the IOMMU doesn't support the PDS capability, software
> > can't drain page requests and group responses as expected.
> >
> > Do not enable PCI/PRI if the IOMMU doesn't support PDS.
>
> After giving the spec another look, this is probably the way to go.
> However the PDS also mentions that DT must be set. Should we check
> ecap_dev_iotlb_support(iommu->ecap)  as well?

It has already been checked.

> > Reported-by: Joel Granados <joel.granados@xxxxxxxxxx>
> > Closes: https://lore.kernel.org/r/20250909-jag-pds-v1-1-ad8cba0e494e@xxxxxxxxxx
> > Fixes: 66ac4db36f4c ("iommu/vt-d: Add page request draining support")
> > Cc: stable@xxxxxxxxxxxxxxx
> > Signed-off-by: Lu Baolu <baolu.lu@xxxxxxxxxxxxxxx>
> > ---
> >   drivers/iommu/intel/iommu.c | 2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c
> > index 9c3ab9d9f69a..92759a8f8330 100644
> > --- a/drivers/iommu/intel/iommu.c
> > +++ b/drivers/iommu/intel/iommu.c
> > @@ -3812,7 +3812,7 @@ static struct iommu_device *intel_iommu_probe_device(struct device *dev)
> >   			}
> >   
> >   			if (info->ats_supported && ecap_prs(iommu->ecap) &&
> > -			    pci_pri_supported(pdev))
> > +			    ecap_pds(iommu->ecap) && pci_pri_supported(pdev))
> Should this be
>   +			    ecap_dev_iotlb_support(iommu->ecap) && ecap_pds(iommu->ecap) && pci_pri_supported(pdev))
>
> ???

ecap_dev_iotlb_support() has already been checked as it's part of the
info->ats_supported check. The dependency chain is:
ecap_dev_iotlb_support() -> ats_support -> pri_support.

Thanks,
baolu