[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <0b7d3e6c770a2bdc2c307ddde268561135ef3395.1679137163.git.lstoakes@gmail.com>
Date: Sat, 18 Mar 2023 11:13:20 +0000
From: Lorenzo Stoakes <lstoakes@...il.com>
To: linux-mm@...ck.org, linux-kernel@...r.kernel.org,
Andrew Morton <akpm@...ux-foundation.org>
Cc: David Hildenbrand <david@...hat.com>,
Matthew Wilcox <willy@...radead.org>,
Vlastimil Babka <vbabka@...e.cz>,
"Liam R . Howlett" <Liam.Howlett@...cle.com>,
maple-tree@...ts.infradead.org,
Lorenzo Stoakes <lstoakes@...il.com>
Subject: [PATCH 3/4] mm/mmap/vma_merge: extend invariants, avoid invalid res, vma
Previously, vma was an uninitialised variable which was only definitely
assigned as a result of the logic covering all possible input cases - for
it to have remained uninitialised, prev would have to be NULL, and next
would _have_ to be mergeable.
We now reuse vma to assign curr and next, so to be absolutely explicit,
ensure this variable is _always_ assigned, and while we're at it remove the
redundant assignment of both res and vma (if prev is NULL then we simply
assign to NULL).
In addition, we absolutely do rely on addr == curr->vm_start should curr
exist, so assert as much.
Signed-off-by: Lorenzo Stoakes <lstoakes@...il.com>
---
mm/mmap.c | 19 ++++++++++++++-----
1 file changed, 14 insertions(+), 5 deletions(-)
diff --git a/mm/mmap.c b/mm/mmap.c
index 66893fc72e03..e60c637f4e49 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -911,7 +911,7 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
{
pgoff_t pglen = (end - addr) >> PAGE_SHIFT;
pgoff_t vma_pgoff;
- struct vm_area_struct *curr, *next, *res = NULL;
+ struct vm_area_struct *curr, *next, *res;
struct vm_area_struct *vma, *adjust, *remove, *remove2;
int err = -1;
bool merge_prev = false;
@@ -978,14 +978,18 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
else
next = NULL;
- /* verify some invariant that must be enforced by the caller */
+ /*
+ * By default, we return prev. Cases 3, 4, 8 will instead return next
+ * and cases 3, 8 will also update vma to point at next.
+ */
+ res = vma = prev;
+
+ /* Verify some invariant that must be enforced by the caller. */
VM_WARN_ON(prev && addr <= prev->vm_start);
- VM_WARN_ON(curr && end > curr->vm_end);
+ VM_WARN_ON(curr && (addr != curr->vm_start || end > curr->vm_end));
VM_WARN_ON(addr >= end);
if (prev) {
- res = prev;
- vma = prev;
vma_start = prev->vm_start;
vma_pgoff = prev->vm_pgoff;
/* Can we merge the predecessor? */
@@ -996,6 +1000,7 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
vma_prev(vmi);
}
}
+
/* Can we merge the successor? */
if (next && mpol_equal(policy, vma_policy(next)) &&
can_vma_merge_before(next, vm_flags,
@@ -1036,6 +1041,10 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
adj_start = -(prev->vm_end - addr);
err = dup_anon_vma(next, prev);
} else {
+ /*
+ * Note that cases 3 and 8 are the ONLY ones where prev
+ * is permitted to be (but is not necessarily) NULL.
+ */
vma = next; /* case 3 */
vma_start = addr;
vma_end = next->vm_end;
--
2.39.2
Powered by blists - more mailing lists