mm/mmu_notifier: use MMU_NOTIFY_CLEAR in remove_device_exclusive_entry()

Let's limit the use of MMU_NOTIFY_EXCLUSIVE to the case where we convert a
present PTE to device-exclusive.  For the other case, we can simply use
MMU_NOTIFY_CLEAR, because it really is clearing the device-exclusive entry
first, to then install the present entry.

Update the documentation of MMU_NOTIFY_EXCLUSIVE, to document the single
use case more thoroughly.

If ever required, we could add a separate MMU_NOTIFY_CLEAR_EXCLUSIVE; for
now using MMU_NOTIFY_CLEAR seems to be sufficient.

Link: https://lkml.kernel.org/r/20250226132257.2826043-6-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Jérôme Glisse <jglisse@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index d4e7146618262f08d45707ad83afbb7cf5c4c462..bc2402a45741d82c2b912e94d63691a77801acc6 100644 (file)
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -43,10 +43,10 @@ struct mmu_interval_notifier;
  * a device driver to possibly ignore the invalidation if the
  * owner field matches the driver's device private pgmap owner.
  *
- * @MMU_NOTIFY_EXCLUSIVE: to signal a device driver that the device will no
- * longer have exclusive access to the page. When sent during creation of an
- * exclusive range the owner will be initialised to the value provided by the
- * caller of make_device_exclusive(), otherwise the owner will be NULL.
+ * @MMU_NOTIFY_EXCLUSIVE: conversion of a page table entry to device-exclusive.
+ * The owner is initialized to the value provided by the caller of
+ * make_device_exclusive(), such that this caller can filter out these
+ * events.
  */
 enum mmu_notifier_event {
        MMU_NOTIFY_UNMAP = 0,

diff --git a/mm/memory.c b/mm/memory.c
index cb9549477556044292f88f8b1b16598321dc99cf..1efc393e32b6d0ba78e05ad960f322d3d9a10ce3 100644 (file)
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -4048,7 +4048,7 @@ static vm_fault_t remove_device_exclusive_entry(struct vm_fault *vmf)
                folio_put(folio);
                return ret;
        }
-       mmu_notifier_range_init_owner(&range, MMU_NOTIFY_EXCLUSIVE, 0,
+       mmu_notifier_range_init_owner(&range, MMU_NOTIFY_CLEAR, 0,
                                vma->vm_mm, vmf->address & PAGE_MASK,
                                (vmf->address & PAGE_MASK) + PAGE_SIZE, NULL);
        mmu_notifier_invalidate_range_start(&range);