From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, "Michael S. Tsirkin" <mst@redhat.com>,
Jason Wang <jasowang@redhat.com>,
Vladislav Yasevich <vyasevic@redhat.com>,
"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 3.16 32/55] macvtap: Fix race between device delete and open.
Date: Mon, 13 Oct 2014 04:24:46 +0200 [thread overview]
Message-ID: <20141013022445.143717016@linuxfoundation.org> (raw)
In-Reply-To: <20141013022443.729870634@linuxfoundation.org>
3.16-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vlad Yasevich <vyasevich@gmail.com>
[ Upstream commit 40b8fe45d1f094e3babe7b2dc2b71557ab71401d ]
In macvtap device delete and open calls can race and
this causes a list curruption of the vlan queue_list.
The race intself is triggered by the idr accessors
that located the vlan device. The device is stored
into and removed from the idr under both an rtnl and
a mutex. However, when attempting to locate the device
in idr, only a mutex is taken. As a result, once cpu
perfoming a delete may take an rtnl and wait for the mutex,
while another cput doing an open() will take the idr
mutex first to fetch the device pointer and later take
an rtnl to add a queue for the device which may have
just gotten deleted.
With this patch, we now hold the rtnl for the duration
of the macvtap_open() call thus making sure that
open will not race with delete.
CC: Michael S. Tsirkin <mst@redhat.com>
CC: Jason Wang <jasowang@redhat.com>
Signed-off-by: Vladislav Yasevich <vyasevic@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/net/macvtap.c | 18 ++++++++----------
1 file changed, 8 insertions(+), 10 deletions(-)
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -112,17 +112,15 @@ out:
return err;
}
+/* Requires RTNL */
static int macvtap_set_queue(struct net_device *dev, struct file *file,
struct macvtap_queue *q)
{
struct macvlan_dev *vlan = netdev_priv(dev);
- int err = -EBUSY;
- rtnl_lock();
if (vlan->numqueues == MAX_MACVTAP_QUEUES)
- goto out;
+ return -EBUSY;
- err = 0;
rcu_assign_pointer(q->vlan, vlan);
rcu_assign_pointer(vlan->taps[vlan->numvtaps], q);
sock_hold(&q->sk);
@@ -136,9 +134,7 @@ static int macvtap_set_queue(struct net_
vlan->numvtaps++;
vlan->numqueues++;
-out:
- rtnl_unlock();
- return err;
+ return 0;
}
static int macvtap_disable_queue(struct macvtap_queue *q)
@@ -454,11 +450,12 @@ static void macvtap_sock_destruct(struct
static int macvtap_open(struct inode *inode, struct file *file)
{
struct net *net = current->nsproxy->net_ns;
- struct net_device *dev = dev_get_by_macvtap_minor(iminor(inode));
+ struct net_device *dev;
struct macvtap_queue *q;
- int err;
+ int err = -ENODEV;
- err = -ENODEV;
+ rtnl_lock();
+ dev = dev_get_by_macvtap_minor(iminor(inode));
if (!dev)
goto out;
@@ -498,6 +495,7 @@ out:
if (dev)
dev_put(dev);
+ rtnl_unlock();
return err;
}
next prev parent reply other threads:[~2014-10-13 2:24 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-13 2:24 [PATCH 3.16 00/55] 3.16.6-stable review Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 01/55] netlink: reset network header before passing to taps Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 02/55] rtnetlink: fix VF info size Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 03/55] net: Always untag vlan-tagged traffic on input Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 04/55] myri10ge: check for DMA mapping errors Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 05/55] Revert "macvlan: simplify the structure port" Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 06/55] tcp: dont use timestamp from repaired skb-s to calculate RTT (v2) Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 07/55] sit: Fix ipip6_tunnel_lookup device matching criteria Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 08/55] tcp: fix tcp_release_cb() to dispatch via address family for mtu_reduced() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 09/55] tcp: fix ssthresh and undo for consecutive short FRTO episodes Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 10/55] net: phy: smsc: move smsc_phy_config_init reset part in a soft_reset function Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 11/55] tipc: fix message importance range check Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 12/55] packet: handle too big packets for PACKET_V3 Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 13/55] bnx2x: Revert UNDI flushing mechanism Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 14/55] net: ipv6: fib: dont sleep inside atomic lock Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 15/55] openvswitch: fix panic with multiple vlan headers Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 16/55] vxlan: fix incorrect initializer in union vxlan_addr Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 17/55] net: fix checksum features handling in netif_skb_features() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 18/55] ipv6: fix rtnl locking in setsockopt for anycast and multicast Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 19/55] l2tp: fix race while getting PMTU on PPP pseudo-wire Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 20/55] ipv6: restore the behavior of ipv6_sock_ac_drop() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 21/55] bonding: fix div by zero while enslaving and transmitting Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 22/55] net: filter: fix possible use after free Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 23/55] bridge: Check if vlan filtering is enabled only once Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 24/55] bridge: Fix br_should_learn to check vlan_enabled Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 25/55] net: allow macvlans to move to net namespace Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 26/55] macvlan: allow to enqueue broadcast pkt on virtual device Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 27/55] tg3: Work around HW/FW limitations with vlan encapsulated frames Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 28/55] tg3: Allow for recieve of full-size 8021AD frames Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 29/55] xfrm: Generate blackhole routes only from route lookup functions Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 30/55] xfrm: Generate queueing " Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 31/55] ip_tunnel: Dont allow to add the same tunnel multiple times Greg Kroah-Hartman
2014-10-13 2:24 ` Greg Kroah-Hartman [this message]
2014-10-13 2:24 ` [PATCH 3.16 33/55] Revert "net/macb: add pinctrl consumer support" Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 34/55] net/mlx4_core: Allow not to specify probe_vf in SRIOV IB mode Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 35/55] net/mlx4: Correctly configure single ported VFs from the host Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 36/55] gro: fix aggregation for skb using frag_list Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 37/55] ipv6: remove rt6i_genid Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 38/55] hyperv: Fix a bug in netvsc_start_xmit() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 39/55] ip6_gre: fix flowi6_proto value in xmit path Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 40/55] net: systemport: fix bcm_sysport_insert_tsb() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 41/55] team: avoid race condition in scheduling delayed work Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 42/55] hyperv: Fix a bug in netvsc_send() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 43/55] sctp: handle association restarts when the socket is closed Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 44/55] net_sched: copy exts->type in tcf_exts_change() Greg Kroah-Hartman
2014-10-13 2:24 ` [PATCH 3.16 45/55] uas: Add a quirk for rejecting ATA_12 and ATA_16 commands Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 46/55] uas: Add no-report-opcodes quirk Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 47/55] uas: Add US_FL_NO_ATA_1X quirk for Seagate (0bc2:ab20) drives Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 48/55] uas: Add another ASM1051 usb-id to the uas blacklist Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 49/55] USB: Add device quirk for ASUS T100 Base Station keyboard Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 50/55] USB: serial: cp210x: added Ketra N1 wireless interface support Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 51/55] USB: cp210x: add support for Seluxit USB dongle Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 52/55] usb: musb: dsps: kill OTG timer on suspend Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 53/55] crypto: caam - fix addressing of struct member Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 54/55] driver/base/node: remove unnecessary kfree of node struct from unregister_one_node Greg Kroah-Hartman
2014-10-13 2:25 ` [PATCH 3.16 55/55] serial: 8250: Add Quark X1000 to 8250_pci.c Greg Kroah-Hartman
2014-10-13 15:19 ` [PATCH 3.16 00/55] 3.16.6-stable review Guenter Roeck
2014-10-13 20:32 ` Shuah Khan
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=20141013022445.143717016@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=davem@davemloft.net \
--cc=jasowang@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@redhat.com \
--cc=stable@vger.kernel.org \
--cc=vyasevic@redhat.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).