Re: [PATCH v3 1/7] iommu/amd: Warn when found inconsistency EFR mask

[Suthikulpanit, Suravee]

On 6/23/2022 3:22 PM, Joerg Roedel wrote:
> On Wed, Jun 22, 2022 at 12:11:25PM -0500, Suravee Suthikulpanit wrote:
> >   #ifdef CONFIG_IRQ_REMAP
> > +/*
> > + * Iterate through all the IOMMUs to verify if the specified
> > + * EFR bitmask of IOMMU feature are set.
> > + * Warn and return false if found inconsistency.
> > + */
> >   static bool check_feature_on_all_iommus(u64 mask)
> >   {
> >   	bool ret = false;
> >   	struct amd_iommu *iommu;
> >   
> >   	for_each_iommu(iommu) {
> > -		ret = iommu_feature(iommu, mask);
> > -		if (!ret)
> > +		bool tmp = iommu_feature(iommu, mask);
> > +
> > +		if ((ret != tmp) &&
> > +		    !list_is_first(&iommu->list, &amd_iommu_list)) {
> > +			pr_err(FW_BUG "Found inconsistent EFR mask (%#llx) on iommu%d (%04x:%02x:%02x.%01x).\n",
> > +			       mask, iommu->index, iommu->pci_seg->id, PCI_BUS_NUM(iommu->devid),
> > +			       PCI_SLOT(iommu->devid), PCI_FUNC(iommu->devid));
> >   			return false;
> > +		}
> > +		ret = tmp;
>
> It is better to implement this by introducing a global feature mask,
> which represents the minial set of features supported by any IOMMU in
> the system.
>
> The warning is then something like:
>
> 	if ((global_feature_mask & iommu_features) != global_feature_mask)
> 		pr_warn(...);
>
> This also makes the global variable to track SNP support obsolete.
>
> Regards,
>
> 	Joerg

That's actually better. I'll send out v4 w/ global EFR variable.

Thanks,
Suravee