Re: [RFC 2/7] iommu/vt-d: Remove iova handling code from non-dma ops path

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi,

On 5/5/19 9:19 AM, Lu Baolu wrote:
Hi,

On 5/4/19 9:23 PM, Tom Murphy via iommu wrote:
@@ -4181,58 +4168,37 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb,
                         unsigned long val, void *v)
  {
      struct memory_notify *mhp = v;
-    unsigned long long start, end;
-    unsigned long start_vpfn, last_vpfn;
+    unsigned long start_vpfn = mm_to_dma_pfn(mhp->start_pfn);
+    unsigned long last_vpfn = mm_to_dma_pfn(mhp->start_pfn +
+            mhp->nr_pages - 1);
      switch (val) {
      case MEM_GOING_ONLINE:
-        start = mhp->start_pfn << PAGE_SHIFT;
-        end = ((mhp->start_pfn + mhp->nr_pages) << PAGE_SHIFT) - 1;
-        if (iommu_domain_identity_map(si_domain, start, end)) {
-            pr_warn("Failed to build identity map for [%llx-%llx]\n",
-                start, end);
+        if (iommu_domain_identity_map(si_domain, start_vpfn,
+                    last_vpfn)) {
+            pr_warn("Failed to build identity map for [%lx-%lx]\n",
+                start_vpfn, last_vpfn);
              return NOTIFY_BAD;
          }
          break;

Actually we don't need to update the si_domain if iommu hardware
supports pass-through mode. This should be made in a separated patch
anyway.

Oh! please ignore it.

This callback is only registered when hardware doesn't support pass
through mode.

        if (si_domain && !hw_pass_through)
                register_memory_notifier(&intel_iommu_memory_nb);

Best regards,
Lu Baolu




[Index of Archives]     [Linux SoC Development]     [Linux Rockchip Development]     [Linux for Synopsys ARC Processors]    
  • [Linux on Unisoc (RDA Micro) SoCs]     [Linux Actions SoC]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Linux SCSI]     [Yosemite News]

  •   Powered by Linux