* [PATCH] mm/damon/vaddr: check address space stability before VMA traversal
@ 2026-06-30 15:25 Igor Putko
2026-07-01 1:08 ` SJ Park
0 siblings, 1 reply; 2+ messages in thread
From: Igor Putko @ 2026-06-30 15:25 UTC (permalink / raw)
To: sj; +Cc: akpm, damon, linux-mm, linux-kernel, Igor Putko
In __damon_va_three_regions(), the VMA tree is traversed without
checking if the address space is stable. If the OOM reaper runs
concurrently, it may mark the mm as MMF_UNSTABLE and leave empty
markers in the maple tree.
Add a check_stable_address_space(mm) call before iterating over
the VMAs to avoid potential NULL pointer dereferences.
Signed-off-by: Igor Putko <igorpetindev@gmail.com>
---
mm/damon/vaddr.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/mm/damon/vaddr.c b/mm/damon/vaddr.c
index 345b89b5d..e756ac147 100644
--- a/mm/damon/vaddr.c
+++ b/mm/damon/vaddr.c
@@ -11,6 +11,7 @@
#include <linux/hugetlb.h>
#include <linux/mman.h>
#include <linux/mmu_notifier.h>
+#include <linux/oom.h>
#include <linux/page_idle.h>
#include <linux/pagewalk.h>
#include <linux/sched/mm.h>
@@ -84,6 +85,9 @@ static int __damon_va_three_regions(struct mm_struct *mm,
* If this is too slow, it can be optimised to examine the maple
* tree gaps.
*/
+ if (!check_stable_address_space(mm))
+ return -EINVAL;
+
for_each_vma(vmi, vma) {
unsigned long gap;
--
2.47.3
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [PATCH] mm/damon/vaddr: check address space stability before VMA traversal
2026-06-30 15:25 [PATCH] mm/damon/vaddr: check address space stability before VMA traversal Igor Putko
@ 2026-07-01 1:08 ` SJ Park
0 siblings, 0 replies; 2+ messages in thread
From: SJ Park @ 2026-07-01 1:08 UTC (permalink / raw)
To: Igor Putko; +Cc: SJ Park, akpm, damon, linux-mm, linux-kernel
On Tue, 30 Jun 2026 18:25:58 +0300 Igor Putko <igorpetindev@gmail.com> wrote:
> In __damon_va_three_regions(), the VMA tree is traversed without
> checking if the address space is stable. If the OOM reaper runs
> concurrently, it may mark the mm as MMF_UNSTABLE and leave empty
> markers in the maple tree.
I initially thought this is correct, but... Hasn't the series starting from
commit bed76bec3111 ("mm: relocate the page table ceiling and floor
definitions") fixed the issue?
Thanks,
SJ
[...]
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-07-01 1:08 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-06-30 15:25 [PATCH] mm/damon/vaddr: check address space stability before VMA traversal Igor Putko
2026-07-01 1:08 ` SJ Park
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox