From: sashiko-bot@kernel.org
To: "Gregory Price (Meta)" <gourry@gourry.net>, linux-mm@kvack.org
Cc: kernel-team@meta.com, david@kernel.org, ziy@nvidia.com,
joshua.hahnjy@gmail.com, rakie.kim@sk.com,
ying.huang@linux.alibaba.com, akpm@linux-foundation.org,
arnd@arndb.de, larips <larips950@gmail.com>
Subject: Re: [PATCH] mm/mempolicy: limit compat bitmap reads to the source nodemask
Date: Wed, 18 Nov 2026 10:20:05 +0000 [thread overview]
Message-ID: <sashiko-outbox-220674@kernel.org> (raw)
In-Reply-To: <20261117101910.1820066-1-gourry@gourry.net>
Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Medium] mm/mempolicy: compat buffer overflow in copy_nodes_to_user()
when maxnode=65
--
commit f6a3a5979382343bb522ac36af5733f2e67acb24
Author: Gregory Price (Meta) <gourry@gourry.net>
mm/mempolicy: limit compat bitmap reads to the source nodemask
This commit bounds the bitmap size passed to compat_put_bitmap() in
copy_nodes_to_user() by MAX_NUMNODES to prevent reading past the kernel
nodemask on the stack.
> diff --git a/mm/mempolicy.c b/mm/mempolicy.c
> index fb627b37afcbb..7cfcb6b49c542 100644
> --- a/mm/mempolicy.c
> +++ b/mm/mempolicy.c
> @@ -1738,7 +1738,8 @@ static int copy_nodes_to_user(unsigned long __user *mask, unsigned long maxnode,
>
> if (compat)
> return compat_put_bitmap((compat_ulong_t __user *)mask,
> - nodes_addr(*nodes), maxnode);
> + nodes_addr(*nodes),
> + min(maxnode, MAX_NUMNODES));
>
> return copy_to_user(mask, nodes_addr(*nodes), copy) ? -EFAULT : 0;
> }
[Severity: Medium]
Can this still write 4 bytes past the caller's userspace nodemask buffer
when MAX_NUMNODES > 64?
In the mempolicy ABI, maxnode is 1-based, so a 32-bit caller with a
64-bit (8-byte) buffer passes maxnode = 65:
mm/mempolicy.c:copy_nodes_to_user() {
unsigned long copy = ALIGN(maxnode-1, 64) / 8;
unsigned int nbytes = BITS_TO_LONGS(nr_node_ids) * sizeof(long);
bool compat = in_compat_syscall();
if (compat)
nbytes = BITS_TO_COMPAT_LONGS(nr_node_ids) * sizeof(compat_long_t);
if (copy > nbytes) {
...
maxnode = nr_node_ids;
}
On a kernel with CONFIG_NODES_SHIFT > 6 (for example MAX_NUMNODES = 1024)
running on a machine where 33 <= nr_node_ids <= 64, both copy and nbytes
are 8, so the copy > nbytes clamp is skipped and min(maxnode, MAX_NUMNODES)
evaluates to 65. compat_put_bitmap() then aligns bitmap_size = 65 up to
96 bits (three compat_ulong_t words) and writes 12 bytes to mask,
overwriting 4 bytes past the 8-byte userspace buffer or failing with
-EFAULT at a page boundary.
Since kernel_get_mempolicy() already rejects maxnode < nr_node_ids and the
copy > nbytes block zeroes any trailing bytes up to copy, passing
nr_node_ids (or maxnode - 1 after clamping) to compat_put_bitmap() would
bound both the kernel stack read and the userspace write.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261117101910.1820066-1-gourry@gourry.net?part=1
prev parent reply other threads:[~2026-11-18 10:20 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-17 10:19 [PATCH] mm/mempolicy: limit compat bitmap reads to the source nodemask Gregory Price (Meta)
2026-11-17 11:10 ` David Hildenbrand (Arm)
2026-11-17 12:44 ` Arnd Bergmann
2026-11-17 13:51 ` Arnd Bergmann
2026-11-18 10:20 ` sashiko-bot [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=sashiko-outbox-220674@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=arnd@arndb.de \
--cc=david@kernel.org \
--cc=gourry@gourry.net \
--cc=joshua.hahnjy@gmail.com \
--cc=kernel-team@meta.com \
--cc=larips950@gmail.com \
--cc=linux-mm@kvack.org \
--cc=rakie.kim@sk.com \
--cc=sashiko-reviews@lists.linux.dev \
--cc=ying.huang@linux.alibaba.com \
--cc=ziy@nvidia.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox