From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev,
Shardul Bankar <shardul.b@mpiricsoftware.com>,
"Matthieu Baerts (NGI0)" <matttbe@kernel.org>,
Jakub Kicinski <kuba@kernel.org>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.1 36/92] mptcp: pm: fix memory leak from alloc-during-teardown race
Date: Mon, 31 Aug 2026 15:34:34 +0200 [thread overview]
Message-ID: <20260831133401.575712492@linuxfoundation.org> (raw)
In-Reply-To: <20260831133359.482388899@linuxfoundation.org>
6.1-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shardul Bankar <shardul.b@mpiricsoftware.com>
[ Upstream commit efc33b5102ff859bacd390a5f30112d8e0c084c0 ]
mptcp_pm_destroy() empties msk->pm.anno_list and
msk->pm.userspace_pm_local_addr_list under msk->pm.lock during socket
teardown, dropping the lock between the two.
A concurrent userspace PM genl ANNOUNCE on the same msk holds a sock
reference via mptcp_token_get_sock() and, in
mptcp_pm_nl_announce_doit(), calls
mptcp_userspace_pm_append_new_local_addr() and
mptcp_pm_announced_alloc(). Both take msk->pm.lock briefly to add to
their respective lists. Because the genl handler holds a sock reference,
mptcp_pm_destroy() may run on the same msk via mptcp_disconnect(), which
invokes mptcp_destroy_common() without dropping the sock refcount,
before the handler completes.
If the lock acquisitions interleave such that mptcp_pm_destroy() empties
a list first, the later alloc adds its entry to a list head that nothing
else iterates for this msk, and the entry leaks. kmemleak reports both
mptcp_pm_add_addr objects (from mptcp_pm_announced_alloc()) and
mptcp_pm_addr_entry objects (from
mptcp_userspace_pm_append_new_local_addr()) under sustained concurrent
ANNOUNCE + close load against the userspace PM.
Add an MPTCP_PM_DESTROYING bit in msk->pm.status, set by
mptcp_pm_destroy() under pm.lock before the lists are emptied and
checked under pm.lock by the alloc paths. Either the alloc takes pm.lock
first, in which case its entry is on the list when mptcp_pm_destroy()
frees it; or mptcp_pm_destroy() takes pm.lock first, in which case the
later alloc observes the bit and refuses.
Found by an MPTCP protocol-flow harness extending BRF (arXiv:2305.08782).
Fixes: 9ab4807c84a4 ("mptcp: netlink: Add MPTCP_PM_CMD_ANNOUNCE")
Cc: stable@vger.kernel.org
Signed-off-by: Shardul Bankar <shardul.b@mpiricsoftware.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Signed-off-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Link: https://patch.msgid.link/20260803-net-mptcp-misc-fixes-7-2-rc6-v2-6-b8f496d71664@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
[ set MPTCP_PM_DESTROYING in mptcp_destroy_common() since mptcp_pm_destroy() doesn't exist yet ]
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/mptcp/pm_netlink.c | 3 +++
net/mptcp/pm_userspace.c | 10 +++++++---
net/mptcp/protocol.c | 5 +++++
net/mptcp/protocol.h | 7 ++++---
4 files changed, 19 insertions(+), 6 deletions(-)
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -408,6 +408,9 @@ bool mptcp_pm_alloc_anno_list(struct mpt
lockdep_assert_held(&msk->pm.lock);
+ if (msk->pm.status & BIT(MPTCP_PM_DESTROYING))
+ return false;
+
add_entry = mptcp_lookup_anno_list_by_saddr(msk, addr);
if (add_entry) {
--- a/net/mptcp/pm_userspace.c
+++ b/net/mptcp/pm_userspace.c
@@ -13,9 +13,9 @@ void mptcp_free_local_addr_list(struct m
struct sock *sk = (struct sock *)msk;
LIST_HEAD(free_list);
- if (!mptcp_pm_is_userspace(msk))
- return;
-
+ /* Free the userspace local address list unconditionally: the socket
+ * can be reused (mptcp_disconnect()) and re-selected to a different PM
+ */
spin_lock_bh(&msk->pm.lock);
list_splice_init(&msk->pm.userspace_pm_local_addr_list, &free_list);
spin_unlock_bh(&msk->pm.lock);
@@ -53,6 +53,10 @@ static int mptcp_userspace_pm_append_new
bitmap_zero(id_bitmap, MPTCP_PM_MAX_ADDR_ID + 1);
spin_lock_bh(&msk->pm.lock);
+ if (msk->pm.status & BIT(MPTCP_PM_DESTROYING)) {
+ spin_unlock_bh(&msk->pm.lock);
+ return -EINVAL;
+ }
list_for_each_entry(e, &msk->pm.userspace_pm_local_addr_list, list) {
addr_match = mptcp_addresses_equal(&e->addr, &entry->addr, true);
if (addr_match && entry->addr.id == 0 && needs_id)
--- a/net/mptcp/protocol.c
+++ b/net/mptcp/protocol.c
@@ -3540,6 +3540,11 @@ void mptcp_destroy_common(struct mptcp_s
sk_forward_alloc_add(sk, msk->rmem_fwd_alloc);
WRITE_ONCE(msk->rmem_fwd_alloc, 0);
mptcp_token_destroy(msk);
+
+ spin_lock_bh(&msk->pm.lock);
+ msk->pm.status |= BIT(MPTCP_PM_DESTROYING);
+ spin_unlock_bh(&msk->pm.lock);
+
mptcp_pm_free_anno_list(msk);
mptcp_free_local_addr_list(msk);
}
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -180,9 +180,10 @@ enum mptcp_pm_status {
MPTCP_PM_ESTABLISHED,
MPTCP_PM_SUBFLOW_ESTABLISHED,
MPTCP_PM_ALREADY_ESTABLISHED, /* persistent status, set after ESTABLISHED event */
- MPTCP_PM_MPC_ENDPOINT_ACCOUNTED /* persistent status, set after MPC local address is
- * accounted int id_avail_bitmap
- */
+ MPTCP_PM_MPC_ENDPOINT_ACCOUNTED, /* persistent status, set after MPC local address is
+ * accounted int id_avail_bitmap
+ */
+ MPTCP_PM_DESTROYING, /* To fence out PM list allocs */
};
enum mptcp_pm_type {
next prev parent reply other threads:[~2026-08-31 13:59 UTC|newest]
Thread overview: 102+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-31 13:33 [PATCH 6.1 00/92] 6.1.187-rc1 review Greg Kroah-Hartman
2026-08-31 13:33 ` [PATCH 6.1 01/92] RDMA/rxe: Fix OOB in free_rd_atomic_resources() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 02/92] inet: RAW sockets using IPPROTO_RAW MUST drop incoming ICMP Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 03/92] KVM: x86/mmu: Check write tracking in all address spaces Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 04/92] ext4: dont enable DAX on new encrypted files Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 05/92] io_uring/io-wq: fix worker accounting when canceling creation callbacks Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 06/92] ipvs: reload ip header after head reallocation Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 07/92] Revert "usb: phy: fsl-usb: Fix use-after-free in delayed work during device removal" Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 08/92] wifi: ath11k: Add missing hw_ops->get_ring_selector() for IPQ5018 Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 09/92] drm/nouveau/kms/nv50-: init hpd_irq_lock for PIOR DP Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 10/92] bpf: Remove tst_run from lwt_seg6local_prog_ops Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 11/92] jfs: add check read-only before truncation in jfs_truncate_nolock() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 12/92] jfs: add check read-only before txBeginAnon() call Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 13/92] ibmvnic: Use kernel helpers for hex dumps Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 14/92] jfs: Fix null-ptr-deref in jfs_ioc_trim Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 15/92] exfat: fix double free in delayed_free Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 16/92] media: platform: exynos4-is: Add hardware sync wait to fimc_is_hw_change_mode() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 17/92] mISDN: hfcpci: Fix warning when deleting uninitialized timer Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 18/92] can: j1939: implement NETDEV_UNREGISTER notification handler Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 19/92] can: j1939: add missing calls in " Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 20/92] can: j1939: make j1939_sk_bind() fail if device is no longer registered Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 21/92] smc: Fix use-after-free in __pnet_find_base_ndev() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 22/92] KVM: arm64: Prevent access to vCPU events before init Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 23/92] smc: Use __sk_dst_get() and dst_dev_rcu() in in smc_clc_prfx_set() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 24/92] smc: Use __sk_dst_get() and dst_dev_rcu() in smc_clc_prfx_match() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 25/92] ASoC: nau8821: Cancel delayed work on component remove Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 26/92] bpf: Fix use-after-free in offloaded map/prog info fill Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 27/92] riscv: Fix register corruption from uninitialized cregs on error Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 28/92] Revert "PM: sleep: Use complete() in device_pm_sleep_init()" Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 29/92] ASoC: nau8821: Cancel pending work before suspend Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 30/92] smc: Use __sk_dst_get() and dst_dev_rcu() in smc_vlan_by_tcpsk() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 31/92] selinux: switch two allocations to use kzalloc_objs() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 32/92] Revert "mtd: maps: vmu-flash: fix fault in unaligned fixup" Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 33/92] Revert "smb: client: use kvzalloc() for megabyte buffer in simple fallocate" Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 34/92] ASoC: tegra: Fix Master Volume Control Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 35/92] block: make bio_check_eod work for zero sized devices Greg Kroah-Hartman
2026-08-31 13:34 ` Greg Kroah-Hartman [this message]
2026-08-31 13:34 ` [PATCH 6.1 37/92] ext4: propagate errors from fast commit range replay Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 38/92] nilfs2: correct return value kernel-doc descriptions for ioctl functions Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 39/92] nilfs2: reject invalid block index in GC ioctl Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 40/92] nfc: nci: add data_len bound checks to activation parameter extractors Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 41/92] HID: magicmouse: fix battery reporting for Bluetooth Magic Trackpad USB-C Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 42/92] HID: magicmouse: prevent unbounded recursion in magicmouse_raw_event() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 43/92] HID: magicmouse: re-enable multitouch after reset-resume Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 44/92] nvme: rename CDR/MORE/DNR to NVME_STATUS_* Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 45/92] nvmet-tcp: bound SGL data length before allocating command buffers Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 46/92] HID: nintendo: stop device IO before hid_hw_stop on probe failure Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 47/92] HID: uclogic: fix use-after-free of inrange_timer on remove Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 48/92] HID: ft260: improve i2c write performance Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 49/92] HID: ft260: improve i2c large reads performance Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 50/92] HID: ft260: skip unexpected HID input reports Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 51/92] HID: ft260: wake up device from power saving mode Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 52/92] HID: ft260: missed NACK from busy device Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 53/92] HID: ft260: validate i2c input report length Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 54/92] HID: ft260: fix stack-use-after-return write in I2C read race Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 55/92] Bluetooth: hci_sync: Use bt_dev_err() to log error message in hci_update_event_filter_sync() Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 56/92] Bluetooth: hci_sync: Fix accept list UAF during suspend Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 57/92] Bluetooth: hci_conn: Fix not matching by CIS ID Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 58/92] Bluetooth: ISO: use correct CIS order in Set CIG Parameters event Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 59/92] Bluetooth: hci_event: fix Set CIG Parameters error status handling Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 60/92] Bluetooth: hci_event: validate LE Set CIG Parameters response Greg Kroah-Hartman
2026-08-31 13:34 ` [PATCH 6.1 61/92] ipv4: igmp: Fix potential UAF in igmp_gq_start_timer() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 62/92] fpga: dfl: fme: add error handling Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 63/92] accessibility: speakup: unregister tty ldisc on later init failures Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 64/92] usb: xhci: Handle USB3 port events when there is one roothub Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 65/92] xhci: dbgtty: Fix unregister on tty_register_driver() failure Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 66/92] xhci: dbgtty: Fix unregister on tty_alloc_driver() failure Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 67/92] fuse: fix invalidate lock leak on setattr writeback failure Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 68/92] fuse: fix invalidate lock leak on open O_TRUNC DAX failure Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 69/92] usb: usbtest: disable dynamic ID support Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 70/92] usb: gadget: f_tcm: keep port count until LUN teardown completes Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 71/92] tls: device: fix out-of-bounds write in tls_append_frag() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 72/92] xfrm: espintcp: fix UAF during close Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 73/92] xfrm: drop ESP-in-TCP packets with no ingress device Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 74/92] xfrm: ah6: validate routing header segments_left Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 75/92] xfrm: fix xfrm_state_construct() auth-trunc leak Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 76/92] net: bridge: mcast: fix use-after-free of a master VLANs multicast context Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 77/92] ipv6: seg6: clear IPv4 control block on IPIP decapsulation Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 78/92] mm/swap: reject swapon() on filesystem-level encrypted files Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 79/92] crypto: atmel-tdes - use scatterlist length before DMA mapping Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 80/92] crypto: qce - fix CCM AAD buffer underallocation Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 81/92] crypto: mxs-dcp - fix source scatterlist length access Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 82/92] crypto: qce - Remove unsafe/deprecated algorithms Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 83/92] KVM: s390: vsie: zero stale crypto bits Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 84/92] Bluetooth: hci_core: Fix hci_conn_hash_lookup_cis Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 85/92] usb: core: Add lock to usb_wakeup_notification() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 86/92] usb: core: Strengthen error handling in hub_hub_status() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 87/92] ALSA: usb-audio: fix OOB write in snd_usbmidi_novation_output() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 88/92] ALSA: usb-audio: Complete cleanup after system-resume errors Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 89/92] USB: serial: option: fix slab OOB read in interrupt URB callback Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 90/92] USB: serial: spcp8x5: drop broken carrier detect support Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 91/92] USB: c67x00: fix use-after-free in c67x00_add_iso_urb() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 6.1 92/92] usb: usbfs: fix use-after-free of usb_device in usbdev_release() Greg Kroah-Hartman
2026-08-31 17:50 ` [PATCH 6.1 00/92] 6.1.187-rc1 review Florian Fainelli
2026-08-31 18:15 ` Francesco Dolcini
2026-08-31 19:12 ` Brett A C Sheffield
2026-08-31 19:41 ` Peter Schneider
2026-09-01 8:45 ` Pavel Machek
2026-09-01 11:36 ` Barry K. Nathan
2026-09-01 16:46 ` Shuah Khan
2026-09-01 22:25 ` Miguel Ojeda
2026-09-01 23:26 ` Ron Economos
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=20260831133401.575712492@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=kuba@kernel.org \
--cc=matttbe@kernel.org \
--cc=patches@lists.linux.dev \
--cc=sashal@kernel.org \
--cc=shardul.b@mpiricsoftware.com \
--cc=stable@vger.kernel.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.