mm: Fix TLB flush for not-first PFNMAP mappings in unmap_region()

This is a stable-specific patch.
I botched the stable-specific rewrite of
commit b67fbeb ("mmu_gather: Force tlb-flush VM_PFNMAP vmas"):
As Hugh pointed out, unmap_region() actually operates on a list of VMAs,
and the variable "vma" merely points to the first VMA in that list.
So if we want to check whether any of the VMAs we're operating on is
PFNMAP or MIXEDMAP, we have to iterate through the list and check each VMA.

Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mm/mmap.c

@@ -2638,6 +2638,7 @@ static void unmap_region(struct mm_struct *mm,
 {
 	struct vm_area_struct *next = vma_next(mm, prev);
 	struct mmu_gather tlb;
+	struct vm_area_struct *cur_vma;
 
 	lru_add_drain();
 	tlb_gather_mmu(&tlb, mm);
@@ -2652,8 +2653,12 @@ static void unmap_region(struct mm_struct *mm,
 	 * concurrent flush in this region has to be coming through the rmap,
 	 * and we synchronize against that using the rmap lock.
 	 */
-	if ((vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) != 0)
-		tlb_flush_mmu(&tlb);
+	for (cur_vma = vma; cur_vma; cur_vma = cur_vma->vm_next) {
+		if ((cur_vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)) != 0) {
+			tlb_flush_mmu(&tlb);
+			break;
+		}
+	}
 
 	free_pgtables(&tlb, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS,
 				 next ? next->vm_start : USER_PGTABLES_CEILING);