mm: write-lock VMAs before removing them from VMA tree

Write-locking VMAs before isolating them ensures that page fault handlers
don't operate on isolated VMAs.

[surenb@google.com: mm/nommu: remove unnecessary VMA locking]
Link: https://lkml.kernel.org/r/20230301190457.1498985-1-surenb@google.com
Link: https://lore.kernel.org/all/Y%2F8CJQGNuMUTdLwP@localhost/
Link: https://lkml.kernel.org/r/20230227173632.3292573-19-surenb@google.com
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/mmap.c b/mm/mmap.c
index 8ba51e73f6265b9febf97ea3bc4f060332d6e673..83adf86fd62b1f4bda7f3c3ab8927380626d4b7c 100644 (file)
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2270,6 +2270,7 @@ int split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 static inline int munmap_sidetree(struct vm_area_struct *vma,
                                   struct ma_state *mas_detach)
 {
+       vma_start_write(vma);
        mas_set_range(mas_detach, vma->vm_start, vma->vm_end - 1);
        if (mas_store_gfp(mas_detach, vma, GFP_KERNEL))
                return -ENOMEM;