[PATCH 5.4 01/38] mm: pagewalk: Fix race between unmap and page walker

From: Greg Kroah-Hartman
Date: Thu Oct 13 2022 - 13:52:47 EST


From: Steven Price <steven.price@xxxxxxx>

commit 8782fb61cc848364e1e1599d76d3c9dd58a1cc06 upstream.

The mmap lock protects the page walker from changes to the page tables
during the walk. However a read lock is insufficient to protect those
areas which don't have a VMA as munmap() detaches the VMAs before
downgrading to a read lock and actually tearing down PTEs/page tables.

For users of walk_page_range() the solution is to simply call pte_hole()
immediately without checking the actual page tables when a VMA is not
present. We now never call __walk_page_range() without a valid vma.

For walk_page_range_novma() the locking requirements are tightened to
require the mmap write lock to be taken, and then walking the pgd
directly with 'no_vma' set.

This in turn means that all page walkers either have a valid vma, or
it's that special 'novma' case for page table debugging. As a result,
all the odd '(!walk->vma && !walk->no_vma)' tests can be removed.

Fixes: dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap")
Reported-by: Jann Horn <jannh@xxxxxxxxxx>
Signed-off-by: Steven Price <steven.price@xxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Thomas Hellström <thomas.hellstrom@xxxxxxxxxxxxxxx>
Cc: Konstantin Khlebnikov <koct9i@xxxxxxxxx>
Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
[manually backported. backport note: walk_page_range_novma() does not exist in
5.4, so I'm omitting it from the backport]
Signed-off-by: Jann Horn <jannh@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
mm/pagewalk.c | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)

--- a/mm/pagewalk.c
+++ b/mm/pagewalk.c
@@ -38,7 +38,7 @@ static int walk_pmd_range(pud_t *pud, un
do {
again:
next = pmd_addr_end(addr, end);
- if (pmd_none(*pmd) || !walk->vma) {
+ if (pmd_none(*pmd)) {
if (ops->pte_hole)
err = ops->pte_hole(addr, next, walk);
if (err)
@@ -84,7 +84,7 @@ static int walk_pud_range(p4d_t *p4d, un
do {
again:
next = pud_addr_end(addr, end);
- if (pud_none(*pud) || !walk->vma) {
+ if (pud_none(*pud)) {
if (ops->pte_hole)
err = ops->pte_hole(addr, next, walk);
if (err)
@@ -254,7 +254,7 @@ static int __walk_page_range(unsigned lo
int err = 0;
struct vm_area_struct *vma = walk->vma;

- if (vma && is_vm_hugetlb_page(vma)) {
+ if (is_vm_hugetlb_page(vma)) {
if (walk->ops->hugetlb_entry)
err = walk_hugetlb_range(start, end, walk);
} else
@@ -324,9 +324,13 @@ int walk_page_range(struct mm_struct *mm
if (!vma) { /* after the last vma */
walk.vma = NULL;
next = end;
+ if (ops->pte_hole)
+ err = ops->pte_hole(start, next, &walk);
} else if (start < vma->vm_start) { /* outside vma */
walk.vma = NULL;
next = min(end, vma->vm_start);
+ if (ops->pte_hole)
+ err = ops->pte_hole(start, next, &walk);
} else { /* inside vma */
walk.vma = vma;
next = min(end, vma->vm_end);
@@ -344,9 +348,8 @@ int walk_page_range(struct mm_struct *mm
}
if (err < 0)
break;
- }
- if (walk.vma || walk.ops->pte_hole)
err = __walk_page_range(start, next, &walk);
+ }
if (err)
break;
} while (start = next, start < end);