From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Eric Dumazet <edumazet@google.com>,
syzbot+5498a510ff9de39d37da@syzkaller.appspotmail.com,
Eulgyu Kim <eulgyukim@snu.ac.kr>,
Mat Martineau <martineau@kernel.org>,
"Matthieu Baerts (NGI0)" <matttbe@kernel.org>,
Jakub Kicinski <kuba@kernel.org>
Subject: [PATCH 6.6 24/25] mptcp: fix race in mptcp_pm_nl_flush_addrs_doit()
Date: Fri, 13 Feb 2026 14:48:50 +0100 [thread overview]
Message-ID: <20260213134704.760568317@linuxfoundation.org> (raw)
In-Reply-To: <20260213134703.882698935@linuxfoundation.org>
6.6-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <edumazet@google.com>
commit e2a9eeb69f7d4ca4cf4c70463af77664fdb6ab1d upstream.
syzbot and Eulgyu Kim reported crashes in mptcp_pm_nl_get_local_id()
and/or mptcp_pm_nl_is_backup()
Root cause is list_splice_init() in mptcp_pm_nl_flush_addrs_doit()
which is not RCU ready.
list_splice_init_rcu() can not be called here while holding pernet->lock
spinlock.
Many thanks to Eulgyu Kim for providing a repro and testing our patches.
Fixes: 141694df6573 ("mptcp: remove address when netlink flushes addrs")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: syzbot+5498a510ff9de39d37da@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/all/6970a46d.a00a0220.3ad28e.5cf0.GAE@google.com/T/
Reported-by: Eulgyu Kim <eulgyukim@snu.ac.kr>
Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/611
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260124-net-mptcp-race_nl_flush_addrs-v3-1-b2dc1b613e9d@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ Conflicts because the code has been moved from pm_netlink.c to
pm_kernel.c later on in commit 8617e85e04bd ("mptcp: pm: split
in-kernel PM specific code"). The same modifications can be applied
in pm_netlink.c with one exception, because 'pernet->local_addr_list'
has been renamed to 'pernet->endp_list' in commit 35e71e43a56d
("mptcp: pm: in-kernel: rename 'local_addr_list' to 'endp_list'"). The
previous name is then still being used in this version.
Also, another conflict is caused by commit 7bcf4d8022f9 ("mptcp: pm:
rename helpers linked to 'flush'") which is not in this version:
mptcp_nl_remove_addrs_list() has been renamed to
mptcp_nl_flush_addrs_list(). The previous name has then been kept. ]
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/mptcp/pm_netlink.c | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -1859,16 +1859,26 @@ static void __reset_counters(struct pm_n
static int mptcp_nl_cmd_flush_addrs(struct sk_buff *skb, struct genl_info *info)
{
struct pm_nl_pernet *pernet = genl_info_pm_nl(info);
- LIST_HEAD(free_list);
+ struct list_head free_list;
spin_lock_bh(&pernet->lock);
- list_splice_init(&pernet->local_addr_list, &free_list);
+ free_list = pernet->local_addr_list;
+ INIT_LIST_HEAD_RCU(&pernet->local_addr_list);
__reset_counters(pernet);
pernet->next_id = 1;
bitmap_zero(pernet->id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
spin_unlock_bh(&pernet->lock);
- mptcp_nl_remove_addrs_list(sock_net(skb->sk), &free_list);
+
+ if (free_list.next == &pernet->local_addr_list)
+ return 0;
+
synchronize_rcu();
+
+ /* Adjust the pointers to free_list instead of pernet->local_addr_list */
+ free_list.prev->next = &free_list;
+ free_list.next->prev = &free_list;
+
+ mptcp_nl_remove_addrs_list(sock_net(skb->sk), &free_list);
__flush_addrs(&free_list);
return 0;
}
next prev parent reply other threads:[~2026-02-13 13:57 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-13 13:48 [PATCH 6.6 00/25] 6.6.125-rc1 review Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 01/25] smb: client: split cached_fid bitfields to avoid shared-byte RMW races Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 02/25] ksmbd: fix infinite loop caused by next_smb2_rcv_hdr_off reset in error paths Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 03/25] smb: server: fix leak of active_num_conn in ksmbd_tcp_new_connection() Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 04/25] driver core: enforce device_lock for driver_match_device() Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 05/25] Bluetooth: btusb: Add USB ID 7392:e611 for Edimax EW-7611UXB Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 06/25] crypto: octeontx - Fix length check to avoid truncation in ucode_load_store Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 07/25] crypto: omap - Allocate OMAP_CRYPTO_FORCE_COPY scatterlists correctly Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 08/25] crypto: virtio - Add spinlock protection with virtqueue notification Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 09/25] crypto: virtio - Remove duplicated virtqueue_kick in virtio_crypto_skcipher_crypt_req Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 10/25] nilfs2: Fix potential block overflow that cause system hang Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 11/25] wifi: rtw88: Fix alignment fault in rtw_core_enable_beacon() Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 12/25] scsi: qla2xxx: Validate sp before freeing associated memory Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 13/25] scsi: qla2xxx: Allow recovery for tape devices Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 14/25] scsi: qla2xxx: Delay module unload while fabric scan in progress Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 15/25] scsi: qla2xxx: Free sp in error path to fix system crash Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 16/25] scsi: qla2xxx: Query FW again before proceeding with login Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 17/25] net: sfp: Fix quirk for Ubiquiti U-Fiber Instant SFP module Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 18/25] nfsd: dont ignore the return code of svc_proc_register() Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 19/25] netfilter: nf_tables: missing objects with no memcg accounting Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 20/25] netfilter: nft_set_pipapo: prevent overflow in lookup table allocation Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 21/25] vsock/test: verify socket options after setting them Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 22/25] spi: cadence-quadspi: Implement refcount to handle unbind during busy Greg Kroah-Hartman
2026-02-13 13:48 ` [PATCH 6.6 23/25] selftests: mptcp: pm: ensure unknown flags are ignored Greg Kroah-Hartman
2026-02-13 13:48 ` Greg Kroah-Hartman [this message]
2026-02-13 13:48 ` [PATCH 6.6 25/25] gpio: omap: do not register driver in probe() Greg Kroah-Hartman
2026-02-13 18:53 ` [PATCH 6.6 00/25] 6.6.125-rc1 review Florian Fainelli
2026-02-13 19:20 ` Jon Hunter
2026-02-14 1:05 ` Peter Schneider
2026-02-14 10:49 ` Ron Economos
2026-02-14 16:02 ` Brett A C Sheffield
2026-02-15 0:03 ` Miguel Ojeda
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=20260213134704.760568317@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=edumazet@google.com \
--cc=eulgyukim@snu.ac.kr \
--cc=kuba@kernel.org \
--cc=martineau@kernel.org \
--cc=matttbe@kernel.org \
--cc=patches@lists.linux.dev \
--cc=stable@vger.kernel.org \
--cc=syzbot+5498a510ff9de39d37da@syzkaller.appspotmail.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