On Wed, Jul 09, 2025 at 02:09:04PM -0300, Jason Gunthorpe wrote:
> On Fri, Jul 04, 2025 at 06:13:30PM -0700, Nicolin Chen wrote:
> > +static struct iommufd_access *
> > +iommufd_hw_queue_alloc_phys(struct iommu_hw_queue_alloc *cmd,
> > + struct iommufd_viommu *viommu, phys_addr_t *base_pa)
> > +{
> > + struct iommufd_access *access;
> > + struct page **pages;
> > + size_t max_npages;
> > + size_t length;
> > + u64 offset;
> > + size_t i;
> > + int rc;
> > +
> > + offset =
> > + cmd->nesting_parent_iova - PAGE_ALIGN(cmd->nesting_parent_iova);
>
> PAGE_ALIGN is ALIGN UP, that is the wrong direction?
>
> It is just:
>
> offset = cmd->nesting_parent_iova % PAGE_SIZE;
>
> And this is missing:
>
> *base_pa = (page_to_pfn(pages[0]) << PAGE_SHIFT) + offset;
>
> ??
Yes, my bad. And I realized that all my testings are page aligned.
Maybe we could add an IOMMU_TEST_OP_HW_QUEUE_CHECK_ADDR..
Thanks
Nicolin
