Re: [PATCH v7 6/6] vtd: add lookup_page method to iommu_ops

[Paul Durrant <Paul.Durrant@citrix.com>]

> -----Original Message-----
> From: Tian, Kevin [mailto:kevin.tian@intel.com]
> Sent: 13 September 2018 07:53
> To: Paul Durrant <Paul.Durrant@citrix.com>; xen-devel@lists.xenproject.org
> Cc: Wei Liu <wei.liu2@citrix.com>; Jan Beulich <jbeulich@suse.com>; George
> Dunlap <George.Dunlap@citrix.com>
> Subject: RE: [PATCH v7 6/6] vtd: add lookup_page method to iommu_ops
> 
> > From: Paul Durrant [mailto:paul.durrant@citrix.com]
> > Sent: Wednesday, September 12, 2018 7:30 PM
> >
> > This patch adds a new method to the VT-d IOMMU implementation to find
> > the
> > MFN currently mapped by the specified DFN along with a wrapper function
> > in generic IOMMU code to call the implementation if it exists.
> >
> > This patch also cleans up the initializers in intel_iommu_map_page() and
> > uses array-style dereference there, for consistency. A missing check for
> > shared EPT is also added to intel_iommu_unmap_page().
> 
> then please split into two patches.
> 

Ok.

> >
> > NOTE: This patch only adds a Xen-internal interface. This will be used by
> >       a subsequent patch.
> >       Another subsequent patch will add similar functionality for AMD
> >       IOMMUs.
> >
> > Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
> > ---
> > Cc: Wei Liu <wei.liu2@citrix.com>
> > Cc: Kevin Tian <kevin.tian@intel.com>
> > Cc: Jan Beulich <jbeulich@suse.com>
> > Cc: George Dunlap <george.dunlap@citrix.com>
> >
> > v7:
> >  - Re-base and re-name BFN -> DFN.
> >  - Add missing checks for shared EPT and iommu_passthrough.
> >  - Remove unnecessary initializers and use array-style dereference.
> >  - Drop Wei's R-b because of code churn.
> >
> > v3:
> >  - Addressed comments from George.
> >
> > v2:
> >  - Addressed some comments from Jan.
> > ---
> >  xen/drivers/passthrough/iommu.c     | 11 ++++++++
> >  xen/drivers/passthrough/vtd/iommu.c | 52
> > +++++++++++++++++++++++++++++++++++--
> >  xen/drivers/passthrough/vtd/iommu.h |  3 +++
> >  xen/include/xen/iommu.h             |  4 +++
> >  4 files changed, 68 insertions(+), 2 deletions(-)
> >
> > diff --git a/xen/drivers/passthrough/iommu.c
> > b/xen/drivers/passthrough/iommu.c
> > index a16f1a0c66..52e3f500c7 100644
> > --- a/xen/drivers/passthrough/iommu.c
> > +++ b/xen/drivers/passthrough/iommu.c
> > @@ -296,6 +296,17 @@ int iommu_unmap_page(struct domain *d, dfn_t
> > dfn)
> >      return rc;
> >  }
> >
> > +int iommu_lookup_page(struct domain *d, dfn_t dfn, mfn_t *mfn,
> > +                      unsigned int *flags)
> > +{
> > +    const struct domain_iommu *hd = dom_iommu(d);
> > +
> > +    if ( !iommu_enabled || !hd->platform_ops )
> > +        return -EOPNOTSUPP;
> > +
> > +    return hd->platform_ops->lookup_page(d, dfn, mfn, flags);
> > +}
> > +
> >  static void iommu_free_pagetables(unsigned long unused)
> >  {
> >      do {
> > diff --git a/xen/drivers/passthrough/vtd/iommu.c
> > b/xen/drivers/passthrough/vtd/iommu.c
> > index 0163bb949b..6622c2dd4c 100644
> > --- a/xen/drivers/passthrough/vtd/iommu.c
> > +++ b/xen/drivers/passthrough/vtd/iommu.c
> > @@ -1770,7 +1770,7 @@ static int __must_check
> > intel_iommu_map_page(struct domain *d,
> >                                               unsigned int flags)
> >  {
> >      struct domain_iommu *hd = dom_iommu(d);
> > -    struct dma_pte *page = NULL, *pte = NULL, old, new = { 0 };
> > +    struct dma_pte *page, *pte, old, new = {};
> >      u64 pg_maddr;
> >      int rc = 0;
> >
> > @@ -1790,9 +1790,11 @@ static int __must_check
> > intel_iommu_map_page(struct domain *d,
> >          spin_unlock(&hd->arch.mapping_lock);
> >          return -ENOMEM;
> >      }
> > +
> >      page = (struct dma_pte *)map_vtd_domain_page(pg_maddr);
> > -    pte = page + (dfn_x(dfn) & LEVEL_MASK);
> > +    pte = &page[dfn_x(dfn) & LEVEL_MASK];
> >      old = *pte;
> > +
> >      dma_set_pte_addr(new, mfn_to_maddr(mfn));
> >      dma_set_pte_prot(new,
> >                       ((flags & IOMMUF_readable) ? DMA_PTE_READ  : 0) |
> > @@ -1808,6 +1810,7 @@ static int __must_check
> > intel_iommu_map_page(struct domain *d,
> >          unmap_vtd_domain_page(page);
> >          return 0;
> >      }
> > +
> >      *pte = new;
> >
> >      iommu_flush_cache_entry(pte, sizeof(struct dma_pte));
> > @@ -1823,6 +1826,10 @@ static int __must_check
> > intel_iommu_map_page(struct domain *d,
> >  static int __must_check intel_iommu_unmap_page(struct domain *d,
> >                                                 dfn_t dfn)
> >  {
> > +    /* Do nothing if VT-d shares EPT page table */
> > +    if ( iommu_use_hap_pt(d) )
> > +        return 0;
> > +
> >      /* Do nothing if hardware domain and iommu supports pass thru. */
> >      if ( iommu_passthrough && is_hardware_domain(d) )
> >          return 0;
> > @@ -1830,6 +1837,46 @@ static int __must_check
> > intel_iommu_unmap_page(struct domain *d,
> >      return dma_pte_clear_one(d, dfn_to_daddr(dfn));
> >  }
> >
> > +static int intel_iommu_lookup_page(struct domain *d, dfn_t dfn, mfn_t
> > *mfn,
> > +                                   unsigned int *flags)
> > +{
> > +    struct domain_iommu *hd = dom_iommu(d);
> > +    struct dma_pte *page, val;
> > +    u64 pg_maddr;
> > +
> > +    /* Fail if VT-d shares EPT page table */
> > +    if ( iommu_use_hap_pt(d) )
> > +        return -ENOENT;
> > +
> > +    /* Fail if hardware domain and iommu supports pass thru. */
> > +    if ( iommu_passthrough && is_hardware_domain(d) )
> > +        return -ENOENT;
> 
> why fail instead of returning dfn as mfn? passthrough is just one
> special translation mode in IOMMU, which doesn't mean lookup
> is not possible.
> 

Hmm. Given that map and unmap don't return errors then maybe that is best. Will do.

  Paul

> > +
> > +    spin_lock(&hd->arch.mapping_lock);
> > +
> > +    pg_maddr = addr_to_dma_page_maddr(d, dfn_to_daddr(dfn), 0);
> > +    if ( pg_maddr == 0 )
> > +    {
> > +        spin_unlock(&hd->arch.mapping_lock);
> > +        return -ENOMEM;
> > +    }
> > +
> > +    page = map_vtd_domain_page(pg_maddr);
> > +    val = page[dfn_x(dfn) & LEVEL_MASK];
> > +
> > +    unmap_vtd_domain_page(page);
> > +    spin_unlock(&hd->arch.mapping_lock);
> > +
> > +    if ( !dma_pte_present(val) )
> > +        return -ENOENT;
> > +
> > +    *mfn = maddr_to_mfn(dma_pte_addr(val));
> > +    *flags = dma_pte_read(val) ? IOMMUF_readable : 0;
> > +    *flags |= dma_pte_write(val) ? IOMMUF_writable : 0;
> > +
> > +    return 0;
> > +}
> > +
> >  int iommu_pte_flush(struct domain *d, uint64_t dfn, uint64_t *pte,
> >                      int order, int present)
> >  {
> > @@ -2655,6 +2702,7 @@ const struct iommu_ops intel_iommu_ops = {
> >      .teardown = iommu_domain_teardown,
> >      .map_page = intel_iommu_map_page,
> >      .unmap_page = intel_iommu_unmap_page,
> > +    .lookup_page = intel_iommu_lookup_page,
> >      .free_page_table = iommu_free_page_table,
> >      .reassign_device = reassign_device_ownership,
> >      .get_device_group_id = intel_iommu_group_id,
> > diff --git a/xen/drivers/passthrough/vtd/iommu.h
> > b/xen/drivers/passthrough/vtd/iommu.h
> > index 72c1a2e3cd..47bdfcb5ea 100644
> > --- a/xen/drivers/passthrough/vtd/iommu.h
> > +++ b/xen/drivers/passthrough/vtd/iommu.h
> > @@ -272,6 +272,9 @@ struct dma_pte {
> >  #define dma_set_pte_prot(p, prot) do { \
> >          (p).val = ((p).val & ~DMA_PTE_PROT) | ((prot) & DMA_PTE_PROT); \
> >      } while (0)
> > +#define dma_pte_prot(p) ((p).val & DMA_PTE_PROT)
> > +#define dma_pte_read(p) (dma_pte_prot(p) & DMA_PTE_READ)
> > +#define dma_pte_write(p) (dma_pte_prot(p) & DMA_PTE_WRITE)
> >  #define dma_pte_addr(p) ((p).val & PADDR_MASK & PAGE_MASK_4K)
> >  #define dma_set_pte_addr(p, addr) do {\
> >              (p).val |= ((addr) & PAGE_MASK_4K); } while (0)
> > diff --git a/xen/include/xen/iommu.h b/xen/include/xen/iommu.h
> > index 9e0b4e8638..bebddc2db4 100644
> > --- a/xen/include/xen/iommu.h
> > +++ b/xen/include/xen/iommu.h
> > @@ -100,6 +100,8 @@ void iommu_teardown(struct domain *d);
> >  int __must_check iommu_map_page(struct domain *d, dfn_t dfn,
> >                                  mfn_t mfn, unsigned int flags);
> >  int __must_check iommu_unmap_page(struct domain *d, dfn_t dfn);
> > +int __must_check iommu_lookup_page(struct domain *d, dfn_t dfn,
> > mfn_t *mfn,
> > +                                   unsigned int *flags);
> >
> >  enum iommu_feature
> >  {
> > @@ -190,6 +192,8 @@ struct iommu_ops {
> >      int __must_check (*map_page)(struct domain *d, dfn_t dfn, mfn_t
> mfn,
> >                                   unsigned int flags);
> >      int __must_check (*unmap_page)(struct domain *d, dfn_t dfn);
> > +    int __must_check (*lookup_page)(struct domain *d, dfn_t dfn, mfn_t
> > *mfn,
> > +                                    unsigned int *flags);
> >      void (*free_page_table)(struct page_info *);
> >  #ifdef CONFIG_X86
> >      void (*update_ire_from_apic)(unsigned int apic, unsigned int reg,
> > unsigned int value);
> > --
> > 2.11.0


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel