stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Alan Stern <stern@rowland.harvard.edu>,
	syzbot+76629376e06e2c2ad626@syzkaller.appspotmail.com
Subject: [PATCH 5.4 20/64] USB: core: Fix hang in usb_kill_urb by adding memory barriers
Date: Mon, 31 Jan 2022 11:56:05 +0100	[thread overview]
Message-ID: <20220131105216.343049836@linuxfoundation.org> (raw)
In-Reply-To: <20220131105215.644174521@linuxfoundation.org>

From: Alan Stern <stern@rowland.harvard.edu>

commit 26fbe9772b8c459687930511444ce443011f86bf upstream.

The syzbot fuzzer has identified a bug in which processes hang waiting
for usb_kill_urb() to return.  It turns out the issue is not unlinking
the URB; that works just fine.  Rather, the problem arises when the
wakeup notification that the URB has completed is not received.

The reason is memory-access ordering on SMP systems.  In outline form,
usb_kill_urb() and __usb_hcd_giveback_urb() operating concurrently on
different CPUs perform the following actions:

CPU 0					CPU 1
----------------------------		---------------------------------
usb_kill_urb():				__usb_hcd_giveback_urb():
  ...					  ...
  atomic_inc(&urb->reject);		  atomic_dec(&urb->use_count);
  ...					  ...
  wait_event(usb_kill_urb_queue,
	atomic_read(&urb->use_count) == 0);
					  if (atomic_read(&urb->reject))
						wake_up(&usb_kill_urb_queue);

Confining your attention to urb->reject and urb->use_count, you can
see that the overall pattern of accesses on CPU 0 is:

	write urb->reject, then read urb->use_count;

whereas the overall pattern of accesses on CPU 1 is:

	write urb->use_count, then read urb->reject.

This pattern is referred to in memory-model circles as SB (for "Store
Buffering"), and it is well known that without suitable enforcement of
the desired order of accesses -- in the form of memory barriers -- it
is entirely possible for one or both CPUs to execute their reads ahead
of their writes.  The end result will be that sometimes CPU 0 sees the
old un-decremented value of urb->use_count while CPU 1 sees the old
un-incremented value of urb->reject.  Consequently CPU 0 ends up on
the wait queue and never gets woken up, leading to the observed hang
in usb_kill_urb().

The same pattern of accesses occurs in usb_poison_urb() and the
failure pathway of usb_hcd_submit_urb().

The problem is fixed by adding suitable memory barriers.  To provide
proper memory-access ordering in the SB pattern, a full barrier is
required on both CPUs.  The atomic_inc() and atomic_dec() accesses
themselves don't provide any memory ordering, but since they are
present, we can use the optimized smp_mb__after_atomic() memory
barrier in the various routines to obtain the desired effect.

This patch adds the necessary memory barriers.

CC: <stable@vger.kernel.org>
Reported-and-tested-by: syzbot+76629376e06e2c2ad626@syzkaller.appspotmail.com
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Link: https://lore.kernel.org/r/Ye8K0QYee0Q0Nna2@rowland.harvard.edu
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/usb/core/hcd.c |   14 ++++++++++++++
 drivers/usb/core/urb.c |   12 ++++++++++++
 2 files changed, 26 insertions(+)

--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1567,6 +1567,13 @@ int usb_hcd_submit_urb (struct urb *urb,
 		urb->hcpriv = NULL;
 		INIT_LIST_HEAD(&urb->urb_list);
 		atomic_dec(&urb->use_count);
+		/*
+		 * Order the write of urb->use_count above before the read
+		 * of urb->reject below.  Pairs with the memory barriers in
+		 * usb_kill_urb() and usb_poison_urb().
+		 */
+		smp_mb__after_atomic();
+
 		atomic_dec(&urb->dev->urbnum);
 		if (atomic_read(&urb->reject))
 			wake_up(&usb_kill_urb_queue);
@@ -1662,6 +1669,13 @@ static void __usb_hcd_giveback_urb(struc
 
 	usb_anchor_resume_wakeups(anchor);
 	atomic_dec(&urb->use_count);
+	/*
+	 * Order the write of urb->use_count above before the read
+	 * of urb->reject below.  Pairs with the memory barriers in
+	 * usb_kill_urb() and usb_poison_urb().
+	 */
+	smp_mb__after_atomic();
+
 	if (unlikely(atomic_read(&urb->reject)))
 		wake_up(&usb_kill_urb_queue);
 	usb_put_urb(urb);
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -691,6 +691,12 @@ void usb_kill_urb(struct urb *urb)
 	if (!(urb && urb->dev && urb->ep))
 		return;
 	atomic_inc(&urb->reject);
+	/*
+	 * Order the write of urb->reject above before the read
+	 * of urb->use_count below.  Pairs with the barriers in
+	 * __usb_hcd_giveback_urb() and usb_hcd_submit_urb().
+	 */
+	smp_mb__after_atomic();
 
 	usb_hcd_unlink_urb(urb, -ENOENT);
 	wait_event(usb_kill_urb_queue, atomic_read(&urb->use_count) == 0);
@@ -732,6 +738,12 @@ void usb_poison_urb(struct urb *urb)
 	if (!urb)
 		return;
 	atomic_inc(&urb->reject);
+	/*
+	 * Order the write of urb->reject above before the read
+	 * of urb->use_count below.  Pairs with the barriers in
+	 * __usb_hcd_giveback_urb() and usb_hcd_submit_urb().
+	 */
+	smp_mb__after_atomic();
 
 	if (!urb->dev || !urb->ep)
 		return;



  parent reply	other threads:[~2022-01-31 10:59 UTC|newest]

Thread overview: 73+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-31 10:55 [PATCH 5.4 00/64] 5.4.176-rc1 review Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 01/64] Bluetooth: refactor malicious adv data check Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 02/64] s390/hypfs: include z/VM guests with access control group set Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 03/64] scsi: zfcp: Fix failed recovery on gone remote port with non-NPIV FCP devices Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 04/64] udf: Restore i_lenAlloc when inode expansion fails Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 05/64] udf: Fix NULL ptr deref when converting from inline format Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 06/64] PM: wakeup: simplify the output logic of pm_show_wakelocks() Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 07/64] tracing/histogram: Fix a potential memory leak for kstrdup() Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 08/64] tracing: Dont inc err_log entry count if entry allocation fails Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 09/64] fsnotify: fix fsnotify hooks in pseudo filesystems Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 10/64] drm/etnaviv: relax submit size limits Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 11/64] arm64: errata: Fix exec handling in erratum 1418040 workaround Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 12/64] netfilter: nft_payload: do not update layer 4 checksum when mangling fragments Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 13/64] serial: 8250: of: Fix mapped region size when using reg-offset property Greg Kroah-Hartman
2022-01-31 10:55 ` [PATCH 5.4 14/64] serial: stm32: fix software flow control transfer Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 15/64] tty: n_gsm: fix SW flow control encoding/handling Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 16/64] tty: Add support for Brainboxes UC cards Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 17/64] usb-storage: Add unusual-devs entry for VL817 USB-SATA bridge Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 18/64] usb: common: ulpi: Fix crash in ulpi_match() Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 19/64] usb: gadget: f_sourcesink: Fix isoc transfer for USB_SPEED_SUPER_PLUS Greg Kroah-Hartman
2022-01-31 10:56 ` Greg Kroah-Hartman [this message]
2022-01-31 10:56 ` [PATCH 5.4 21/64] usb: typec: tcpm: Do not disconnect while receiving VBUS off Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 22/64] ucsi_ccg: Check DEV_INT bit only when starting CCG4 Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 23/64] net: sfp: ignore disabled SFP node Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 24/64] powerpc/32: Fix boot failure with GCC latent entropy plugin Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 25/64] i40e: Increase delay to 1 s after global EMP reset Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 26/64] i40e: Fix issue when maximum queues is exceeded Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 27/64] i40e: Fix queues reservation for XDP Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 28/64] i40e: fix unsigned stat widths Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 29/64] rpmsg: char: Fix race between the release of rpmsg_ctrldev and cdev Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 30/64] rpmsg: char: Fix race between the release of rpmsg_eptdev " Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 31/64] scsi: bnx2fc: Flush destroy_work queue before calling bnx2fc_interface_put() Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 32/64] ipv6_tunnel: Rate limit warning messages Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 33/64] ARM: 9170/1: fix panic when kasan and kprobe are enabled Greg Kroah-Hartman
2022-01-31 12:14   ` Ard Biesheuvel
2022-01-31 12:21     ` Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 34/64] net: fix information leakage in /proc/net/ptype Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 35/64] hwmon: (lm90) Mark alert as broken for MAX6646/6647/6649 Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 36/64] hwmon: (lm90) Mark alert as broken for MAX6680 Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 37/64] ping: fix the sk_bound_dev_if match in ping_lookup Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 38/64] ipv4: avoid using shared IP generator for connected sockets Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 39/64] hwmon: (lm90) Reduce maximum conversion rate for G781 Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 40/64] NFSv4: Handle case where the lookup of a directory fails Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 41/64] NFSv4: nfs_atomic_open() can race when looking up a non-regular file Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 42/64] net-procfs: show net devices bound packet types Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 43/64] drm/msm: Fix wrong size calculation Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 44/64] drm/msm/dsi: Fix missing put_device() call in dsi_get_phy Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 45/64] drm/msm/dsi: invalid parameter check in msm_dsi_phy_enable Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 46/64] ipv6: annotate accesses to fn->fn_sernum Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 47/64] NFS: Ensure the server has an up to date ctime before hardlinking Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 48/64] NFS: Ensure the server has an up to date ctime before renaming Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 49/64] netfilter: conntrack: dont increment invalid counter on NF_REPEAT Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 50/64] net: phy: broadcom: hook up soft_reset for BCM54616S Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 51/64] phylib: fix potential use-after-free Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 52/64] rxrpc: Adjust retransmission backoff Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 53/64] hwmon: (lm90) Mark alert as broken for MAX6654 Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 54/64] ibmvnic: init ->running_cap_crqs early Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 55/64] ibmvnic: dont spin in tasklet Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 56/64] drm/msm/hdmi: Fix missing put_device() call in msm_hdmi_get_phy Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 57/64] yam: fix a memory leak in yam_siocdevprivate() Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 58/64] net: hns3: handle empty unknown interrupt for VF Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 59/64] ipv4: raw: lock the socket in raw_bind() Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 60/64] ipv4: tcp: send zero IPID in SYNACK messages Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 61/64] ipv4: remove sparse error in ip_neigh_gw4() Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 62/64] dt-bindings: can: tcan4x5x: fix mram-cfg RX FIFO config Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 63/64] fsnotify: invalidate dcache before IN_DELETE event Greg Kroah-Hartman
2022-01-31 10:56 ` [PATCH 5.4 64/64] block: Fix wrong offset in bio_truncate() Greg Kroah-Hartman
2022-01-31 14:06 ` [PATCH 5.4 00/64] 5.4.176-rc1 review Jon Hunter
2022-01-31 19:49 ` Florian Fainelli
2022-01-31 22:17 ` Shuah Khan
2022-02-01  4:24 ` Guenter Roeck
2022-02-01  8:17 ` Naresh Kamboju
2022-02-01 15:39 ` Sudip Mukherjee

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=20220131105216.343049836@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=stern@rowland.harvard.edu \
    --cc=syzbot+76629376e06e2c2ad626@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;
as well as URLs for NNTP newsgroup(s).