From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Desmond Cheong Zhi Xi <desmondcheongzx@gmail.com>,
syzbot+2f6d7c28bb4bf7e82060@syzkaller.appspotmail.com,
Luiz Augusto von Dentz <luiz.von.dentz@intel.com>,
Sasha Levin <sashal@kernel.org>,
linux-bluetooth@vger.kernel.org, netdev@vger.kernel.org
Subject: [PATCH AUTOSEL 5.13 123/219] Bluetooth: schedule SCO timeouts with delayed_work
Date: Thu, 9 Sep 2021 07:44:59 -0400 [thread overview]
Message-ID: <20210909114635.143983-123-sashal@kernel.org> (raw)
In-Reply-To: <20210909114635.143983-1-sashal@kernel.org>
From: Desmond Cheong Zhi Xi <desmondcheongzx@gmail.com>
[ Upstream commit ba316be1b6a00db7126ed9a39f9bee434a508043 ]
struct sock.sk_timer should be used as a sock cleanup timer. However,
SCO uses it to implement sock timeouts.
This causes issues because struct sock.sk_timer's callback is run in
an IRQ context, and the timer callback function sco_sock_timeout takes
a spin lock on the socket. However, other functions such as
sco_conn_del and sco_conn_ready take the spin lock with interrupts
enabled.
This inconsistent {SOFTIRQ-ON-W} -> {IN-SOFTIRQ-W} lock usage could
lead to deadlocks as reported by Syzbot [1]:
CPU0
----
lock(slock-AF_BLUETOOTH-BTPROTO_SCO);
<Interrupt>
lock(slock-AF_BLUETOOTH-BTPROTO_SCO);
To fix this, we use delayed work to implement SCO sock timouts
instead. This allows us to avoid taking the spin lock on the socket in
an IRQ context, and corrects the misuse of struct sock.sk_timer.
As a note, cancel_delayed_work is used instead of
cancel_delayed_work_sync in sco_sock_set_timer and
sco_sock_clear_timer to avoid a deadlock. In the future, the call to
bh_lock_sock inside sco_sock_timeout should be changed to lock_sock to
synchronize with other functions using lock_sock. However, since
sco_sock_set_timer and sco_sock_clear_timer are sometimes called under
the locked socket (in sco_connect and __sco_sock_close),
cancel_delayed_work_sync might cause them to sleep until an
sco_sock_timeout that has started finishes running. But
sco_sock_timeout would also sleep until it can grab the lock_sock.
Using cancel_delayed_work is fine because sco_sock_timeout does not
change from run to run, hence there is no functional difference
between:
1. waiting for a timeout to finish running before scheduling another
timeout
2. scheduling another timeout while a timeout is running.
Link: https://syzkaller.appspot.com/bug?id=9089d89de0502e120f234ca0fc8a703f7368b31e [1]
Reported-by: syzbot+2f6d7c28bb4bf7e82060@syzkaller.appspotmail.com
Tested-by: syzbot+2f6d7c28bb4bf7e82060@syzkaller.appspotmail.com
Signed-off-by: Desmond Cheong Zhi Xi <desmondcheongzx@gmail.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
net/bluetooth/sco.c | 35 +++++++++++++++++++++++++++++------
1 file changed, 29 insertions(+), 6 deletions(-)
diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c
index 3bd41563f118..f909577932ca 100644
--- a/net/bluetooth/sco.c
+++ b/net/bluetooth/sco.c
@@ -48,6 +48,8 @@ struct sco_conn {
spinlock_t lock;
struct sock *sk;
+ struct delayed_work timeout_work;
+
unsigned int mtu;
};
@@ -74,9 +76,20 @@ struct sco_pinfo {
#define SCO_CONN_TIMEOUT (HZ * 40)
#define SCO_DISCONN_TIMEOUT (HZ * 2)
-static void sco_sock_timeout(struct timer_list *t)
+static void sco_sock_timeout(struct work_struct *work)
{
- struct sock *sk = from_timer(sk, t, sk_timer);
+ struct sco_conn *conn = container_of(work, struct sco_conn,
+ timeout_work.work);
+ struct sock *sk;
+
+ sco_conn_lock(conn);
+ sk = conn->sk;
+ if (sk)
+ sock_hold(sk);
+ sco_conn_unlock(conn);
+
+ if (!sk)
+ return;
BT_DBG("sock %p state %d", sk, sk->sk_state);
@@ -91,14 +104,21 @@ static void sco_sock_timeout(struct timer_list *t)
static void sco_sock_set_timer(struct sock *sk, long timeout)
{
+ if (!sco_pi(sk)->conn)
+ return;
+
BT_DBG("sock %p state %d timeout %ld", sk, sk->sk_state, timeout);
- sk_reset_timer(sk, &sk->sk_timer, jiffies + timeout);
+ cancel_delayed_work(&sco_pi(sk)->conn->timeout_work);
+ schedule_delayed_work(&sco_pi(sk)->conn->timeout_work, timeout);
}
static void sco_sock_clear_timer(struct sock *sk)
{
+ if (!sco_pi(sk)->conn)
+ return;
+
BT_DBG("sock %p state %d", sk, sk->sk_state);
- sk_stop_timer(sk, &sk->sk_timer);
+ cancel_delayed_work(&sco_pi(sk)->conn->timeout_work);
}
/* ---- SCO connections ---- */
@@ -179,6 +199,9 @@ static void sco_conn_del(struct hci_conn *hcon, int err)
bh_unlock_sock(sk);
sco_sock_kill(sk);
sock_put(sk);
+
+ /* Ensure no more work items will run before freeing conn. */
+ cancel_delayed_work_sync(&conn->timeout_work);
}
hcon->sco_data = NULL;
@@ -193,6 +216,8 @@ static void __sco_chan_add(struct sco_conn *conn, struct sock *sk,
sco_pi(sk)->conn = conn;
conn->sk = sk;
+ INIT_DELAYED_WORK(&conn->timeout_work, sco_sock_timeout);
+
if (parent)
bt_accept_enqueue(parent, sk, true);
}
@@ -500,8 +525,6 @@ static struct sock *sco_sock_alloc(struct net *net, struct socket *sock,
sco_pi(sk)->setting = BT_VOICE_CVSD_16BIT;
- timer_setup(&sk->sk_timer, sco_sock_timeout, 0);
-
bt_sock_link(&sco_sk_list, sk);
return sk;
}
--
2.30.2
next prev parent reply other threads:[~2021-09-09 12:15 UTC|newest]
Thread overview: 64+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20210909114635.143983-1-sashal@kernel.org>
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 011/219] libbpf: Fix reuse of pinned map on older kernel Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 016/219] tipc: keep the skb in rcv queue until the whole data is read Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 017/219] net: phy: Fix data type in DP83822 dp8382x_disable_wol() Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 019/219] iavf: do not override the adapter state in the watchdog task Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 020/219] iavf: fix locking of critical sections Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 024/219] netlink: Deal with ESRCH error in nlmsg_notify() Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 025/219] net: ipa: fix IPA v4.11 interconnect data Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 031/219] igc: Check if num of q_vectors is smaller than max before array access Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 049/219] nfp: fix return statement in nfp_net_parse_meta() Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 052/219] ethtool: improve compat ioctl handling Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 055/219] bpf/tests: Fix copy-and-paste error in double word test Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 056/219] bpf/tests: Do not PASS tests without actually testing the result Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 062/219] net: ipa: fix ipa_cmd_table_valid() Sasha Levin
2021-09-09 11:43 ` [PATCH AUTOSEL 5.13 063/219] ipv4: ip_output.c: Fix out-of-bounds warning in ip_copy_addrs() Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 064/219] flow_dissector: Fix out-of-bounds warnings Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 072/219] samples: bpf: Fix tracex7 error raised on the missing argument Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 073/219] libbpf: Fix race when pinning maps in parallel Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 077/219] Bluetooth: skip invalid hci_sync_conn_complete_evt Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 082/219] netfilter: nft_compat: use nfnetlink_unicast() Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 083/219] bonding: 3ad: fix the concurrency between __bond_release_one() and bond_3ad_state_machine_handler() Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 087/219] bpf: Fix off-by-one in tail call count limiting Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 089/219] Bluetooth: Fix not generating RPA when required Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 090/219] dpaa2-switch: do not enable the DPSW at probe time Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 100/219] net: ipa: fix IPA v4.9 interconnects Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 111/219] net: ethernet: stmmac: Do not use unreachable() in ipq806x_gmac_probe() Sasha Levin
2021-09-09 11:44 ` [PATCH AUTOSEL 5.13 120/219] selftests/bpf: Fix xdp_tx.c prog section name Sasha Levin
2021-09-09 11:44 ` Sasha Levin [this message]
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 124/219] Bluetooth: avoid circular locks in sco_sock_connect Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 128/219] net/mlx5: Fix variable type to match 64bit Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 131/219] mac80211: Fix monitor MTU limit so that A-MSDUs get through Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 142/219] Bluetooth: Fix handling of LE Enhanced Connection Complete Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 146/219] tcp: enable data-less, empty-cookie SYN with TFO_SERVER_COOKIE_NOT_REQD Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 148/219] rpc: fix gss_svc_init cleanup on failure Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 149/219] iavf: use mutexes for locking of critical sections Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 150/219] selftests/bpf: Correctly display subtest skip status Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 151/219] selftests/bpf: Fix flaky send_signal test Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 154/219] selftests: nci: Fix the code for next nlattr offset Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 155/219] selftests: nci: Fix the wrong condition Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 156/219] net: Fix offloading indirect devices dependency on qdisc order creation Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 163/219] rtl8xxxu: Fix the handling of TX A-MPDU aggregation Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 164/219] rtw88: use read_poll_timeout instead of fixed sleep Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 165/219] rtw88: wow: build wow function only if CONFIG_PM is on Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 166/219] rtw88: wow: fix size access error of probe request Sasha Levin
2021-09-09 11:45 ` [PATCH AUTOSEL 5.13 167/219] octeontx2-pf: Fix NIX1_RX interface backpressure Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 184/219] mptcp: fix ADD_ADDR and RM_ADDR maybe flush addr_signal each other Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 185/219] selftests/bpf: Enlarge select() timeout for test_maps Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 201/219] iwlwifi: pcie: free RBs during configure Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 202/219] iwlwifi: mvm: Do not use full SSIDs in 6GHz scan Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 203/219] iwlwifi: mvm: fix a memory leak in iwl_mvm_mac_ctxt_beacon_changed Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 204/219] iwlwifi: mvm: avoid static queue number aliasing Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 205/219] iwlwifi: mvm: Fix umac scan request probe parameters Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 206/219] iwlwifi: mvm: fix access to BSS elements Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 207/219] iwlwifi: fw: correctly limit to monitor dump Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 208/219] iwlwifi: mvm: Fix scan channel flags settings Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 209/219] net/mlx5: DR, fix a potential use-after-free bug Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 210/219] net/mlx5: DR, Enable QP retransmission Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 212/219] selftests/bpf: Fix potential unreleased lock Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 213/219] wcn36xx: Fix missing frame timestamp for beacon/probe-resp Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 214/219] ath9k: fix OOB read ar9300_eeprom_restore_internal Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 215/219] ath9k: fix sleeping in atomic context Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 216/219] net: fix NULL pointer reference in cipso_v4_doi_free Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 217/219] fix array-index-out-of-bounds in taprio_change Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 218/219] net: w5100: check return value after calling platform_get_resource() Sasha Levin
2021-09-09 11:46 ` [PATCH AUTOSEL 5.13 219/219] net: hns3: clean up a type mismatch warning Sasha Levin
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=20210909114635.143983-123-sashal@kernel.org \
--to=sashal@kernel.org \
--cc=desmondcheongzx@gmail.com \
--cc=linux-bluetooth@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luiz.von.dentz@intel.com \
--cc=netdev@vger.kernel.org \
--cc=stable@vger.kernel.org \
--cc=syzbot+2f6d7c28bb4bf7e82060@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).