[v4,21/33] mm/mmap: prevent pagefault handler from racing with mmu_notifier registration

Commit Message

Suren Baghdasaryan Feb. 27, 2023, 5:36 p.m. UTC

Page fault handlers might need to fire MMU notifications while a new
notifier is being registered. Modify mm_take_all_locks to write-lock all
VMAs and prevent this race with page fault handlers that would hold VMA
locks. VMAs are locked before i_mmap_rwsem and anon_vma to keep the same
locking order as in page fault handlers.

Signed-off-by: Suren Baghdasaryan <[e-mail]>
---
 mm/mmap.c | 9 +++++++++
 1 file changed, 9 insertions(+)

Patch

diff --git a/mm/mmap.c b/mm/mmap.c
index ec745586785c..b947d82e8522 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -3486,6 +3486,7 @@  static void vm_lock_mapping(struct mm_struct *mm, struct address_space *mapping)
  * of mm/rmap.c:
  *   - all hugetlbfs_i_mmap_rwsem_key locks (aka mapping->i_mmap_rwsem for
  *     hugetlb mapping);
+ *   - all vmas marked locked
  *   - all i_mmap_rwsem locks;
  *   - all anon_vma->rwseml
  *
@@ -3508,6 +3509,13 @@  int mm_take_all_locks(struct mm_struct *mm)
 
 	mutex_lock(&mm_all_locks_mutex);
 
+	mas_for_each(&mas, vma, ULONG_MAX) {
+		if (signal_pending(current))
+			goto out_unlock;
+		vma_start_write(vma);
+	}
+
+	mas_set(&mas, 0);
 	mas_for_each(&mas, vma, ULONG_MAX) {
 		if (signal_pending(current))
 			goto out_unlock;
@@ -3597,6 +3605,7 @@  void mm_drop_all_locks(struct mm_struct *mm)
 		if (vma->vm_file && vma->vm_file->f_mapping)
 			vm_unlock_mapping(vma->vm_file->f_mapping);
 	}
+	vma_end_write_all(mm);
 
 	mutex_unlock(&mm_all_locks_mutex);
 }