intel-iommu: Avoid global flushes with caching mode.

While it may be efficient on real hardware, emulation of global
invalidations is very expensive as all shadow entries must be examined.
This patch changes the behaviour when caching mode is enabled (which is
the case when IOMMU emulation takes place). In this case, page specific
invalidation is used instead.

Signed-off-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>

diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 1880ee0..9ce79b1 100644 (file)
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -2647,15 +2647,24 @@ static void flush_unmaps(void)
                if (!deferred_flush[i].next)
                        continue;
 
                if (!deferred_flush[i].next)
                        continue;
 

-               iommu->flush.flush_iotlb(iommu, 0, 0, 0,
+               /* In caching mode, global flushes turn emulation expensive */
+               if (!cap_caching_mode(iommu->cap))
+                       iommu->flush.flush_iotlb(iommu, 0, 0, 0,

                                         DMA_TLB_GLOBAL_FLUSH);
                for (j = 0; j < deferred_flush[i].next; j++) {
                        unsigned long mask;
                        struct iova *iova = deferred_flush[i].iova[j];
                                         DMA_TLB_GLOBAL_FLUSH);
                for (j = 0; j < deferred_flush[i].next; j++) {
                        unsigned long mask;
                        struct iova *iova = deferred_flush[i].iova[j];

-
-                       mask = ilog2(mm_to_dma_pfn(iova->pfn_hi - iova->pfn_lo + 1));
-                       iommu_flush_dev_iotlb(deferred_flush[i].domain[j],
-                                       (uint64_t)iova->pfn_lo << PAGE_SHIFT, mask);
+                       struct dmar_domain *domain = deferred_flush[i].domain[j];
+
+                       /* On real hardware multiple invalidations are expensive */
+                       if (cap_caching_mode(iommu->cap))
+                               iommu_flush_iotlb_psi(iommu, domain->id,
+                               iova->pfn_lo, iova->pfn_hi - iova->pfn_lo + 1, 0);
+                       else {
+                               mask = ilog2(mm_to_dma_pfn(iova->pfn_hi - iova->pfn_lo + 1));
+                               iommu_flush_dev_iotlb(deferred_flush[i].domain[j],
+                                               (uint64_t)iova->pfn_lo << PAGE_SHIFT, mask);
+                       }

                        __free_iova(&deferred_flush[i].domain[j]->iovad, iova);
                }
                deferred_flush[i].next = 0;
                        __free_iova(&deferred_flush[i].domain[j]->iovad, iova);
                }
                deferred_flush[i].next = 0;