userfaultfd: shmem/hugetlbfs: only allow to register VM_MAYWRITE vmas

commit 29ec90660d68bbdd69507c1c8b4e33aa299278b1 upstream.

After the VMA to register the uffd onto is found, check that it has
VM_MAYWRITE set before allowing registration.  This way we inherit all
common code checks before allowing to fill file holes in shmem and
hugetlbfs with UFFDIO_COPY.

The userfaultfd memory model is not applicable for readonly files unless
it's a MAP_PRIVATE.

Link: http://lkml.kernel.org/r/20181126173452.26955-4-aarcange@redhat.com
Fixes: ff62a3421044 ("hugetlb: implement memfd sealing")
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: Mike Rapoport <rppt@linux.ibm.com>
Reviewed-by: Hugh Dickins <hughd@google.com>
Reported-by: Jann Horn <jannh@google.com>
Fixes: 4c27fe4c4c84 ("userfaultfd: shmem: add shmem_mcopy_atomic_pte for userfaultfd support")
Cc: <stable@vger.kernel.org>
Cc: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: stable@vger.kernel.org
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index 3eda623..f92e1f2 100644 (file)
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -1362,6 +1362,19 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
                ret = -EINVAL;
                if (!vma_can_userfault(cur))
                        goto out_unlock;
+
+               /*
+                * UFFDIO_COPY will fill file holes even without
+                * PROT_WRITE. This check enforces that if this is a
+                * MAP_SHARED, the process has write permission to the backing
+                * file. If VM_MAYWRITE is set it also enforces that on a
+                * MAP_SHARED vma: there is no F_WRITE_SEAL and no further
+                * F_WRITE_SEAL can be taken until the vma is destroyed.
+                */
+               ret = -EPERM;
+               if (unlikely(!(cur->vm_flags & VM_MAYWRITE)))
+                       goto out_unlock;
+
                /*
                 * If this vma contains ending address, and huge pages
                 * check alignment.
@@ -1407,6 +1420,7 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
                BUG_ON(!vma_can_userfault(vma));
                BUG_ON(vma->vm_userfaultfd_ctx.ctx &&
                       vma->vm_userfaultfd_ctx.ctx != ctx);
+               WARN_ON(!(vma->vm_flags & VM_MAYWRITE));
 
                /*
                 * Nothing to do: this vma is already registered into this
@@ -1553,6 +1567,7 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx,
                cond_resched();
 
                BUG_ON(!vma_can_userfault(vma));
+               WARN_ON(!(vma->vm_flags & VM_MAYWRITE));
 
                /*
                 * Nothing to do: this vma is already registered into this

diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
index 8da1cd8..5d70fdb 100644 (file)
--- a/mm/userfaultfd.c
+++ b/mm/userfaultfd.c
@@ -228,8 +228,9 @@ retry:
                if (!dst_vma || !is_vm_hugetlb_page(dst_vma))
                        goto out_unlock;
                /*
-                * Only allow __mcopy_atomic_hugetlb on userfaultfd
-                * registered ranges.
+                * Check the vma is registered in uffd, this is
+                * required to enforce the VM_MAYWRITE check done at
+                * uffd registration time.
                 */
                if (!dst_vma->vm_userfaultfd_ctx.ctx)
                        goto out_unlock;
@@ -472,13 +473,9 @@ retry:
        if (!dst_vma)
                goto out_unlock;
        /*
-        * Be strict and only allow __mcopy_atomic on userfaultfd
-        * registered ranges to prevent userland errors going
-        * unnoticed. As far as the VM consistency is concerned, it
-        * would be perfectly safe to remove this check, but there's
-        * no useful usage for __mcopy_atomic ouside of userfaultfd
-        * registered ranges. This is after all why these are ioctls
-        * belonging to the userfaultfd and not syscalls.
+        * Check the vma is registered in uffd, this is required to
+        * enforce the VM_MAYWRITE check done at uffd registration
+        * time.
         */
        if (!dst_vma->vm_userfaultfd_ctx.ctx)
                goto out_unlock;