Archive-only list for patches
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev,
	Zero Day Initiative <zdi-disclosures@trendmicro.com>,
	Eric Dumazet <edumazet@google.com>,
	Ido Schimmel <idosch@nvidia.com>, Paolo Abeni <pabeni@redhat.com>,
	Denis Arefev <arefev@swemel.ru>
Subject: [PATCH 5.10 15/43] ipv4: igmp: Fix potential UAF in igmp_gq_start_timer()
Date: Mon, 31 Aug 2026 15:35:23 +0200	[thread overview]
Message-ID: <20260831133359.197749902@linuxfoundation.org> (raw)
In-Reply-To: <20260831133358.571886287@linuxfoundation.org>

5.10-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <edumazet@google.com>

commit 7b19c0f81ed1fdaec6bc522569be367199a9edf3 upstream.

A race condition exists between device teardown (inetdev_destroy) and
incoming IGMP query processing (igmp_rcv), leading to a Use-After-Free
in the IGMP timer callback.

During device destruction, inetdev_destroy() drops the primary reference
to in_device, which can drop its refcount to 0. The actual freeing of
in_device memory is deferred via RCU (using call_rcu()).

Concurrently, igmp_rcv() runs under RCU read lock and obtains the
in_device pointer. Because the memory is RCU-protected, CPU-0 can safely
dereference in_device even if its refcount has hit 0.

However, if CPU-0 calls igmp_gq_start_timer() and re-arms the timer, it
attempts to acquire a reference using in_dev_hold(). This increments the
refcount from 0 to 1, triggering a "refcount_t: addition on 0" warning.
Since the in_device memory is still scheduled to be freed after the RCU
grace period (as the free callback does not check the refcount again),
the device is freed while the timer is still armed. When the timer
expires, it accesses the freed memory, causing a kernel panic.

Fix this by using refcount_inc_not_zero() (via a new helper
in_dev_hold_safe()) to prevent acquiring a reference if the device is
already being destroyed. If the refcount is 0, we do not arm the timer.

A similar issue in IPv6 MLD is fixed in a subsequent patch.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Reported-by: Zero Day Initiative <zdi-disclosures@trendmicro.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20260705181756.963063-2-edumazet@google.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
[Denis Arefev: adapted for 5.10/5.15: keep prandom_u32(),
get_random_u32_below() not used here]
Signed-off-by: Denis Arefev <arefev@swemel.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 include/linux/inetdevice.h |    5 +++++
 net/ipv4/igmp.c            |   14 +++++++++-----
 2 files changed, 14 insertions(+), 5 deletions(-)

--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -277,6 +277,11 @@ static inline void in_dev_put(struct in_
 #define __in_dev_put(idev)  refcount_dec(&(idev)->refcnt)
 #define in_dev_hold(idev)   refcount_inc(&(idev)->refcnt)
 
+static inline bool in_dev_hold_safe(struct in_device *idev)
+{
+	return refcount_inc_not_zero(&idev->refcnt);
+}
+
 #endif /* __KERNEL__ */
 
 static __inline__ __be32 inet_make_mask(int logmask)
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -232,16 +232,20 @@ static void igmp_gq_start_timer(struct i
 		return;
 
 	in_dev->mr_gq_running = 1;
-	if (!mod_timer(&in_dev->mr_gq_timer, exp))
-		in_dev_hold(in_dev);
+	if (in_dev_hold_safe(in_dev)) {
+		if (mod_timer(&in_dev->mr_gq_timer, exp))
+			in_dev_put(in_dev);
+	}
 }
 
 static void igmp_ifc_start_timer(struct in_device *in_dev, int delay)
 {
-	int tv = prandom_u32() % delay;
+	if (in_dev_hold_safe(in_dev)) {
+		int tv = prandom_u32() % delay;
 
-	if (!mod_timer(&in_dev->mr_ifc_timer, jiffies+tv+2))
-		in_dev_hold(in_dev);
+		if (mod_timer(&in_dev->mr_ifc_timer, jiffies + tv + 2))
+			in_dev_put(in_dev);
+	}
 }
 
 static void igmp_mod_timer(struct ip_mc_list *im, int max_delay)



  parent reply	other threads:[~2026-08-31 14:06 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-31 13:35 [PATCH 5.10 00/43] 5.10.269-rc1 review Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 01/43] RDMA/rxe: Fix OOB in free_rd_atomic_resources() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 02/43] ipv6: mcast: Fix use-after-free when processing MLD queries Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 03/43] ext4: dont enable DAX on new encrypted files Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 04/43] ipvs: reload ip header after head reallocation Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 05/43] io_uring/io-wq: fix worker accounting when canceling creation callbacks Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 06/43] bpf: Remove tst_run from lwt_seg6local_prog_ops Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 07/43] jfs: add check read-only before truncation in jfs_truncate_nolock() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 08/43] jfs: add check read-only before txBeginAnon() call Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 09/43] KVM: arm64: Prevent access to vCPU events before init Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 10/43] bpf: Fix use-after-free in offloaded map/prog info fill Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 11/43] Revert "PM: sleep: Use complete() in device_pm_sleep_init()" Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 12/43] selinux: switch two allocations to use kzalloc_objs() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 13/43] Revert "mtd: maps: vmu-flash: fix fault in unaligned fixup" Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 14/43] Revert "smb: client: use kvzalloc() for megabyte buffer in simple fallocate" Greg Kroah-Hartman
2026-08-31 13:35 ` Greg Kroah-Hartman [this message]
2026-08-31 13:35 ` [PATCH 5.10 16/43] kcov: fix data corruption and race conditions on PREEMPT_RT Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 17/43] ext4: propagate errors from fast commit range replay Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 18/43] nilfs2: reject invalid block index in GC ioctl Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 19/43] nfc: nci: fix uninit-value in the RF discover/activated NTF handlers Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 20/43] HID: magicmouse: do not keep a stale msc->input if no input is claimed Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 21/43] nfc: nci: add data_len bound checks to activation parameter extractors Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 22/43] nvmet-tcp: bound SGL data length before allocating command buffers Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 23/43] HID: input: read battery capacity from its actual report offset Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 24/43] fpga: dfl: fme: add error handling Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 25/43] accessibility: speakup: unregister tty ldisc on later init failures Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 26/43] usb: usbtest: disable dynamic ID support Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 27/43] usb: gadget: f_tcm: keep port count until LUN teardown completes Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 28/43] xfrm: espintcp: fix UAF during close Greg Kroah-Hartman
2026-09-04  9:51   ` D. Wythe
2026-09-04 12:15     ` Greg Kroah-Hartman
2026-09-05 10:17       ` D. Wythe
2026-09-05 11:16         ` Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 29/43] xfrm: drop ESP-in-TCP packets with no ingress device Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 30/43] xfrm: ah6: validate routing header segments_left Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 31/43] xfrm: fix xfrm_state_construct() auth-trunc leak Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 32/43] ipv6: seg6: clear IPv4 control block on IPIP decapsulation Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 33/43] mm/swap: reject swapon() on filesystem-level encrypted files Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 34/43] crypto: atmel-tdes - use scatterlist length before DMA mapping Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 35/43] crypto: mxs-dcp - fix source scatterlist length access Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 36/43] KVM: s390: vsie: zero stale crypto bits Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 37/43] usb: core: Add lock to usb_wakeup_notification() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 38/43] usb: core: Strengthen error handling in hub_hub_status() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 39/43] ALSA: usb-audio: fix OOB write in snd_usbmidi_novation_output() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 40/43] USB: serial: option: fix slab OOB read in interrupt URB callback Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 41/43] USB: serial: spcp8x5: drop broken carrier detect support Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 42/43] USB: c67x00: fix use-after-free in c67x00_add_iso_urb() Greg Kroah-Hartman
2026-08-31 13:35 ` [PATCH 5.10 43/43] usb: usbfs: fix use-after-free of usb_device in usbdev_release() Greg Kroah-Hartman
2026-08-31 17:01 ` [PATCH 5.10 00/43] 5.10.269-rc1 review Florian Fainelli
2026-08-31 19:23 ` Brett A C Sheffield
2026-08-31 20:58 ` Woody Suwalski
2026-09-01  2:00 ` Dominique Martinet
2026-09-01  8:45 ` Pavel Machek
2026-09-01 16:48 ` Shuah Khan
2026-09-02  0:11 ` Barry K. Nathan

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=20260831133359.197749902@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=arefev@swemel.ru \
    --cc=edumazet@google.com \
    --cc=idosch@nvidia.com \
    --cc=pabeni@redhat.com \
    --cc=patches@lists.linux.dev \
    --cc=stable@vger.kernel.org \
    --cc=zdi-disclosures@trendmicro.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