userfaultfd: hugetlbfs: prevent UFFDIO_COPY to fill beyond the end of i_size

This oops:

  kernel BUG at fs/hugetlbfs/inode.c:484!
  RIP: remove_inode_hugepages+0x3d0/0x410
  Call Trace:
    hugetlbfs_setattr+0xd9/0x130
    notify_change+0x292/0x410
    do_truncate+0x65/0xa0
    do_sys_ftruncate.constprop.3+0x11a/0x180
    SyS_ftruncate+0xe/0x10
    tracesys+0xd9/0xde

was caused by the lack of i_size check in hugetlb_mcopy_atomic_pte.

mmap() can still succeed beyond the end of the i_size after vmtruncate
zapped vmas in those ranges, but the faults must not succeed, and that
includes UFFDIO_COPY.

We could differentiate the retval to userland to represent a SIGBUS like
a page fault would do (vs SIGSEGV), but it doesn't seem very useful and
we'd need to pick a random retval as there's no meaningful syscall
retval that would differentiate from SIGSEGV and SIGBUS, there's just
-EFAULT.

Link: http://lkml.kernel.org/r/20171016223914.2421-2-aarcange@redhat.com
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: "Dr. David Alan Gilbert" <dgilbert@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>

mm/hugetlb.c

@@ -3984,6 +3984,9 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm,
 			    unsigned long src_addr,
 			    struct page **pagep)
 {
+	struct address_space *mapping;
+	pgoff_t idx;
+	unsigned long size;
 	int vm_shared = dst_vma->vm_flags & VM_SHARED;
 	struct hstate *h = hstate_vma(dst_vma);
 	pte_t _dst_pte;
@@ -4021,13 +4024,24 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm,
 	__SetPageUptodate(page);
 	set_page_huge_active(page);
 
+	mapping = dst_vma->vm_file->f_mapping;
+	idx = vma_hugecache_offset(h, dst_vma, dst_addr);
+
 	/*
 	 * If shared, add to page cache
 	 */
 	if (vm_shared) {
-		struct address_space *mapping = dst_vma->vm_file->f_mapping;
+		size = i_size_read(mapping->host) >> huge_page_shift(h);
-		pgoff_t idx = vma_hugecache_offset(h, dst_vma, dst_addr);
+		ret = -EFAULT;
+		if (idx >= size)
+			goto out_release_nounlock;
 
+		/*
+		 * Serialization between remove_inode_hugepages() and
+		 * huge_add_to_page_cache() below happens through the
+		 * hugetlb_fault_mutex_table that here must be hold by
+		 * the caller.
+		 */
 		ret = huge_add_to_page_cache(page, mapping, idx);
 		if (ret)
 			goto out_release_nounlock;
@@ -4036,6 +4050,20 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm,
 	ptl = huge_pte_lockptr(h, dst_mm, dst_pte);
 	spin_lock(ptl);
 
+	/*
+	 * Recheck the i_size after holding PT lock to make sure not
+	 * to leave any page mapped (as page_mapped()) beyond the end
+	 * of the i_size (remove_inode_hugepages() is strict about
+	 * enforcing that). If we bail out here, we'll also leave a
+	 * page in the radix tree in the vm_shared case beyond the end
+	 * of the i_size, but remove_inode_hugepages() will take care
+	 * of it as soon as we drop the hugetlb_fault_mutex_table.
+	 */
+	size = i_size_read(mapping->host) >> huge_page_shift(h);
+	ret = -EFAULT;
+	if (idx >= size)
+		goto out_release_unlock;
+
 	ret = -EEXIST;
 	if (!huge_pte_none(huge_ptep_get(dst_pte)))
 		goto out_release_unlock;