From: gregkh@linuxfoundation.org
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Oliver Hartkopp <socketcan@hartkopp.net>,
Andre Naujoks <nautsch2@gmail.com>,
Eric Dumazet <edumazet@google.com>,
Oleksij Rempel <o.rempel@pengutronix.de>,
Marc Kleine-Budde <mkl@pengutronix.de>
Subject: [PATCH 4.9 06/78] can: skb: can_skb_set_owner(): fix ref counting if socket was closed before setting skb ownership
Date: Mon, 15 Mar 2021 14:51:29 +0100 [thread overview]
Message-ID: <20210315135212.274978979@linuxfoundation.org> (raw)
In-Reply-To: <20210315135212.060847074@linuxfoundation.org>
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
From: Oleksij Rempel <o.rempel@pengutronix.de>
commit e940e0895a82c6fbaa259f2615eb52b57ee91a7e upstream.
There are two ref count variables controlling the free()ing of a socket:
- struct sock::sk_refcnt - which is changed by sock_hold()/sock_put()
- struct sock::sk_wmem_alloc - which accounts the memory allocated by
the skbs in the send path.
In case there are still TX skbs on the fly and the socket() is closed,
the struct sock::sk_refcnt reaches 0. In the TX-path the CAN stack
clones an "echo" skb, calls sock_hold() on the original socket and
references it. This produces the following back trace:
| WARNING: CPU: 0 PID: 280 at lib/refcount.c:25 refcount_warn_saturate+0x114/0x134
| refcount_t: addition on 0; use-after-free.
| Modules linked in: coda_vpu(E) v4l2_jpeg(E) videobuf2_vmalloc(E) imx_vdoa(E)
| CPU: 0 PID: 280 Comm: test_can.sh Tainted: G E 5.11.0-04577-gf8ff6603c617 #203
| Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)
| Backtrace:
| [<80bafea4>] (dump_backtrace) from [<80bb0280>] (show_stack+0x20/0x24) r7:00000000 r6:600f0113 r5:00000000 r4:81441220
| [<80bb0260>] (show_stack) from [<80bb593c>] (dump_stack+0xa0/0xc8)
| [<80bb589c>] (dump_stack) from [<8012b268>] (__warn+0xd4/0x114) r9:00000019 r8:80f4a8c2 r7:83e4150c r6:00000000 r5:00000009 r4:80528f90
| [<8012b194>] (__warn) from [<80bb09c4>] (warn_slowpath_fmt+0x88/0xc8) r9:83f26400 r8:80f4a8d1 r7:00000009 r6:80528f90 r5:00000019 r4:80f4a8c2
| [<80bb0940>] (warn_slowpath_fmt) from [<80528f90>] (refcount_warn_saturate+0x114/0x134) r8:00000000 r7:00000000 r6:82b44000 r5:834e5600 r4:83f4d540
| [<80528e7c>] (refcount_warn_saturate) from [<8079a4c8>] (__refcount_add.constprop.0+0x4c/0x50)
| [<8079a47c>] (__refcount_add.constprop.0) from [<8079a57c>] (can_put_echo_skb+0xb0/0x13c)
| [<8079a4cc>] (can_put_echo_skb) from [<8079ba98>] (flexcan_start_xmit+0x1c4/0x230) r9:00000010 r8:83f48610 r7:0fdc0000 r6:0c080000 r5:82b44000 r4:834e5600
| [<8079b8d4>] (flexcan_start_xmit) from [<80969078>] (netdev_start_xmit+0x44/0x70) r9:814c0ba0 r8:80c8790c r7:00000000 r6:834e5600 r5:82b44000 r4:82ab1f00
| [<80969034>] (netdev_start_xmit) from [<809725a4>] (dev_hard_start_xmit+0x19c/0x318) r9:814c0ba0 r8:00000000 r7:82ab1f00 r6:82b44000 r5:00000000 r4:834e5600
| [<80972408>] (dev_hard_start_xmit) from [<809c6584>] (sch_direct_xmit+0xcc/0x264) r10:834e5600 r9:00000000 r8:00000000 r7:82b44000 r6:82ab1f00 r5:834e5600 r4:83f27400
| [<809c64b8>] (sch_direct_xmit) from [<809c6c0c>] (__qdisc_run+0x4f0/0x534)
To fix this problem, only set skb ownership to sockets which have still
a ref count > 0.
Fixes: 0ae89beb283a ("can: add destructor for self generated skbs")
Cc: Oliver Hartkopp <socketcan@hartkopp.net>
Cc: Andre Naujoks <nautsch2@gmail.com>
Link: https://lore.kernel.org/r/20210226092456.27126-1-o.rempel@pengutronix.de
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Oliver Hartkopp <socketcan@hartkopp.net>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
include/linux/can/skb.h | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
--- a/include/linux/can/skb.h
+++ b/include/linux/can/skb.h
@@ -48,8 +48,12 @@ static inline void can_skb_reserve(struc
static inline void can_skb_set_owner(struct sk_buff *skb, struct sock *sk)
{
- if (sk) {
- sock_hold(sk);
+ /* If the socket has already been closed by user space, the
+ * refcount may already be 0 (and the socket will be freed
+ * after the last TX skb has been freed). So only increase
+ * socket refcount if the refcount is > 0.
+ */
+ if (sk && atomic_inc_not_zero(&sk->sk_refcnt)) {
skb->destructor = sock_efree;
skb->sk = sk;
}
next prev parent reply other threads:[~2021-03-15 13:53 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-15 13:51 [PATCH 4.9 00/78] 4.9.262-rc1 review gregkh
2021-03-15 13:51 ` [PATCH 4.9 01/78] uapi: nfnetlink_cthelper.h: fix userspace compilation error gregkh
2021-03-15 13:51 ` [PATCH 4.9 02/78] ethernet: alx: fix order of calls on resume gregkh
2021-03-15 13:51 ` [PATCH 4.9 03/78] ath9k: fix transmitting to stations in dynamic SMPS mode gregkh
2021-03-15 13:51 ` [PATCH 4.9 04/78] net: Fix gro aggregation for udp encaps with zero csum gregkh
2021-03-15 13:51 ` [PATCH 4.9 05/78] net: avoid infinite loop in mpls_gso_segment when mpls_hlen == 0 gregkh
2021-03-15 13:51 ` gregkh [this message]
2021-03-15 13:51 ` [PATCH 4.9 07/78] can: flexcan: assert FRZ bit in flexcan_chip_freeze() gregkh
2021-03-15 13:51 ` [PATCH 4.9 08/78] can: flexcan: enable RX FIFO after FRZ/HALT valid gregkh
2021-03-15 13:51 ` [PATCH 4.9 09/78] netfilter: x_tables: gpf inside xt_find_revision() gregkh
2021-03-15 13:51 ` [PATCH 4.9 10/78] cifs: return proper error code in statfs(2) gregkh
2021-03-15 13:51 ` [PATCH 4.9 11/78] Revert "mm, slub: consider rest of partial list if acquire_slab() fails" gregkh
2021-03-15 13:51 ` [PATCH 4.9 12/78] net/mlx4_en: update moderation when config reset gregkh
2021-03-15 13:51 ` [PATCH 4.9 13/78] net: sched: avoid duplicates in classes dump gregkh
2021-03-15 13:51 ` [PATCH 4.9 14/78] net: lapbether: Remove netif_start_queue / netif_stop_queue gregkh
2021-03-15 13:51 ` [PATCH 4.9 15/78] net: davicom: Fix regulator not turned off on failed probe gregkh
2021-03-15 13:51 ` [PATCH 4.9 16/78] net: davicom: Fix regulator not turned off on driver removal gregkh
2021-03-15 13:51 ` [PATCH 4.9 17/78] media: usbtv: Fix deadlock on suspend gregkh
2021-03-15 13:51 ` [PATCH 4.9 18/78] udf: fix silent AED tagLocation corruption gregkh
2021-03-15 13:51 ` [PATCH 4.9 19/78] mmc: mxs-mmc: Fix a resource leak in an error handling path in mxs_mmc_probe() gregkh
2021-03-15 13:51 ` [PATCH 4.9 20/78] mmc: mediatek: fix race condition between msdc_request_timeout and irq gregkh
2021-03-15 13:51 ` [PATCH 4.9 21/78] powerpc/perf: Record counter overflow always if SAMPLE_IP is unset gregkh
2021-03-15 13:51 ` [PATCH 4.9 22/78] PCI: xgene-msi: Fix race in installing chained irq handler gregkh
2021-03-15 13:51 ` [PATCH 4.9 23/78] s390/smp: __smp_rescan_cpus() - move cpumask away from stack gregkh
2021-03-15 13:51 ` [PATCH 4.9 24/78] scsi: libiscsi: Fix iscsi_prep_scsi_cmd_pdu() error handling gregkh
2021-03-15 13:51 ` [PATCH 4.9 25/78] ALSA: hda/hdmi: Cancel pending works before suspend gregkh
2021-03-15 13:51 ` [PATCH 4.9 26/78] ALSA: hda: Avoid spurious unsol event handling during S3/S4 gregkh
2021-03-15 13:51 ` [PATCH 4.9 27/78] ALSA: usb-audio: Fix "cannot get freq eq" errors on Dell AE515 sound bar gregkh
2021-03-15 13:51 ` [PATCH 4.9 28/78] s390/dasd: fix hanging DASD driver unbind gregkh
2021-03-15 13:51 ` [PATCH 4.9 29/78] mmc: core: Fix partition switch time for eMMC gregkh
2021-03-15 13:51 ` [PATCH 4.9 30/78] scripts/recordmcount.{c,pl}: support -ffunction-sections .text.* section names gregkh
2021-03-15 13:51 ` [PATCH 4.9 31/78] Goodix Fingerprint device is not a modem gregkh
2021-03-15 13:51 ` [PATCH 4.9 32/78] usb: gadget: f_uac2: always increase endpoint max_packet_size by one audio slot gregkh
2021-03-15 13:51 ` [PATCH 4.9 33/78] usb: renesas_usbhs: Clear PIPECFG for re-enabling pipe with other EPNUM gregkh
2021-03-15 13:51 ` [PATCH 4.9 34/78] xhci: Improve detection of device initiated wake signal gregkh
2021-03-15 13:51 ` [PATCH 4.9 35/78] USB: serial: io_edgeport: fix memory leak in edge_startup gregkh
2021-03-15 13:51 ` [PATCH 4.9 36/78] USB: serial: ch341: add new Product ID gregkh
2021-03-15 13:52 ` [PATCH 4.9 37/78] USB: serial: cp210x: add ID for Acuity Brands nLight Air Adapter gregkh
2021-03-15 13:52 ` [PATCH 4.9 38/78] USB: serial: cp210x: add some more GE USB IDs gregkh
2021-03-15 13:52 ` [PATCH 4.9 39/78] usbip: fix stub_dev to check for stream socket gregkh
2021-03-15 13:52 ` [PATCH 4.9 40/78] usbip: fix vhci_hcd " gregkh
2021-03-15 13:52 ` [PATCH 4.9 41/78] usbip: fix vudc " gregkh
2021-03-15 13:52 ` [PATCH 4.9 42/78] usbip: fix stub_dev usbip_sockfd_store() races leading to gpf gregkh
2021-03-15 13:52 ` [PATCH 4.9 43/78] usbip: fix vhci_hcd attach_store() " gregkh
2021-03-15 13:52 ` [PATCH 4.9 44/78] staging: rtl8192u: fix ->ssid overflow in r8192_wx_set_scan() gregkh
2021-03-15 13:52 ` [PATCH 4.9 45/78] staging: rtl8188eu: prevent ->ssid overflow in rtw_wx_set_scan() gregkh
2021-03-15 13:52 ` [PATCH 4.9 46/78] staging: rtl8712: unterminated string leads to read overflow gregkh
2021-03-15 13:52 ` [PATCH 4.9 47/78] staging: rtl8188eu: fix potential memory corruption in rtw_check_beacon_data() gregkh
2021-03-15 13:52 ` [PATCH 4.9 48/78] staging: ks7010: prevent buffer overflow in ks_wlan_set_scan() gregkh
2021-03-15 13:52 ` [PATCH 4.9 49/78] staging: rtl8712: Fix possible buffer overflow in r8712_sitesurvey_cmd gregkh
2021-03-15 13:52 ` [PATCH 4.9 50/78] staging: rtl8192e: Fix possible buffer overflow in _rtl92e_wx_set_scan gregkh
2021-03-15 13:52 ` [PATCH 4.9 51/78] staging: comedi: addi_apci_1032: Fix endian problem for COS sample gregkh
2021-03-15 13:52 ` [PATCH 4.9 52/78] staging: comedi: addi_apci_1500: Fix endian problem for command sample gregkh
2021-03-15 13:52 ` [PATCH 4.9 53/78] staging: comedi: adv_pci1710: Fix endian problem for AI command data gregkh
2021-03-15 13:52 ` [PATCH 4.9 54/78] staging: comedi: das6402: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 55/78] staging: comedi: das800: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 56/78] staging: comedi: dmm32at: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 57/78] staging: comedi: me4000: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 58/78] staging: comedi: pcl711: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 59/78] staging: comedi: pcl818: " gregkh
2021-03-15 13:52 ` [PATCH 4.9 60/78] sh_eth: fix TRSCER mask for R7S72100 gregkh
2021-03-15 13:52 ` [PATCH 4.9 61/78] NFSv4.2: fix return value of _nfs4_get_security_label() gregkh
2021-03-15 13:52 ` [PATCH 4.9 62/78] block: rsxx: fix error return code of rsxx_pci_probe() gregkh
2021-03-15 13:52 ` [PATCH 4.9 63/78] configfs: fix a use-after-free in __configfs_open_file gregkh
2021-03-15 13:52 ` [PATCH 4.9 64/78] prctl: fix PR_SET_MM_AUXV kernel stack leak gregkh
2021-03-15 13:52 ` [PATCH 4.9 65/78] alpha: add $(src)/ rather than $(obj)/ to make source file path gregkh
2021-03-15 13:52 ` [PATCH 4.9 66/78] alpha: merge build rules of division routines gregkh
2021-03-15 13:52 ` [PATCH 4.9 67/78] alpha: make short build log available for " gregkh
2021-03-15 13:52 ` [PATCH 4.9 68/78] alpha: Package string routines together gregkh
2021-03-15 13:52 ` [PATCH 4.9 69/78] alpha: switch __copy_user() and __do_clean_user() to normal calling conventions gregkh
2021-03-15 13:52 ` [PATCH 4.9 70/78] powerpc/64s: Fix instruction encoding for lis in ppc_function_entry() gregkh
2021-03-15 13:52 ` [PATCH 4.9 71/78] binfmt_misc: fix possible deadlock in bm_register_write gregkh
2021-03-15 13:52 ` [PATCH 4.9 72/78] hwmon: (lm90) Fix max6658 sporadic wrong temperature reading gregkh
2021-03-15 13:52 ` [PATCH 4.9 73/78] KVM: arm64: Fix exclusive limit for IPA size gregkh
2021-03-15 13:52 ` [PATCH 4.9 74/78] iio: imu: adis16400: release allocated memory on failure gregkh
2021-03-15 13:52 ` [PATCH 4.9 75/78] iio: imu: adis16400: fix memory leak gregkh
2021-03-15 13:52 ` [PATCH 4.9 76/78] xen/events: reset affinity of 2-level event when tearing it down gregkh
2021-03-15 13:52 ` [PATCH 4.9 77/78] xen/events: dont unmask an event channel when an eoi is pending gregkh
2021-03-15 13:52 ` [PATCH 4.9 78/78] xen/events: avoid handling the same event on two cpus at the same time gregkh
2021-03-15 17:07 ` [PATCH 4.9 00/78] 4.9.262-rc1 review Jon Hunter
2021-03-15 20:42 ` Florian Fainelli
2021-03-17 15:18 ` Greg KH
2021-03-28 20:37 ` Ben Hutchings
2021-03-15 21:29 ` Guenter Roeck
2021-03-15 22:57 ` Jason Self
2021-03-16 12:02 ` Naresh Kamboju
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=20210315135212.274978979@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=edumazet@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mkl@pengutronix.de \
--cc=nautsch2@gmail.com \
--cc=o.rempel@pengutronix.de \
--cc=socketcan@hartkopp.net \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox