** Description changed:

  Description:
  Invalidation Hint (IH) can be provided for page-selective-within-PASID 
invalidations, the Invalidation Hint specifies if the first-level and nested 
mappings cached in the paging-structure-caches that controls the specified 
address/mask range needs to be invalidated or not. For software usages that 
update only the leaf PTEs, the first-level and nested mappings in the 
paging-structure-caches can be preserved by specifying the Invalidation Hint 
field value of 1.
  
  However, currently our SVM code hardcode IH=0. which is less optimal.
  For vSVM, we have already added a generic flag
  (IOMMU_INV_ADDR_FLAG_LEAF) to indicate passdown cache invalidate API is
  for leaf PTE only. So as long as native SVM is fixed, vSVM could set the
  IH w/o code change.
  
  SW needs to obtain statistics of non-leaf paging structure to provide IH
  bit.
  
  static void intel_flush_svm_range(struct intel_svm *svm, unsigned long 
address,
  unsigned long pages, int ih, int gl)
  
  { struct intel_svm_dev *sdev; rcu_read_lock(); list_for_each_entry_rcu(sdev, 
&svm->devs, list) intel_flush_svm_range_dev(svm, sdev, address, pages, ih, gl); 
rcu_read_unlock(); }
  /* Pages have been freed at this point */
  static void intel_invalidate_range(struct mmu_notifier *mn,
  struct mm_struct *mm,
  unsigned long start, unsigned long end)
  
  { struct intel_svm *svm = container_of(mn, struct intel_svm, notifier);
  intel_flush_svm_range(svm, start, (end - start + PAGE_SIZE - 1) >>
  VTD_PAGE_SHIFT, 0, 0); }
  
- Target Kernel: TBD
- Target Release: 21.04
+ Target Kernel: 5.14
+ Target Release: 22.04

** Tags removed: intel-kernel-21.04
** Tags added: intel-kernel-22.04

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1878523

Title:
  [IOMMU] Provide invalidation hint to IOMMU page selective PASID-based
  IOTLB invalidation

To manage notifications about this bug go to:
https://bugs.launchpad.net/intel/+bug/1878523/+subscriptions

-- 
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to