From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Karsten Graul <kgraul@linux.ibm.com>,
Ursula Braun <ubraun@linux.ibm.com>,
"David S. Miller" <davem@davemloft.net>,
Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.19 79/88] net/smc: wait for pending work before clcsock release_sock
Date: Wed, 15 Nov 2023 15:36:31 -0500 [thread overview]
Message-ID: <20231115191430.830506616@linuxfoundation.org> (raw)
In-Reply-To: <20231115191426.221330369@linuxfoundation.org>
4.19-stable review patch. If anyone has any objections, please let me know.
------------------
From: Karsten Graul <kgraul@linux.ibm.com>
[ Upstream commit fd57770dd198f5b2ddd5b9e6bf282cf98d63adb9 ]
When the clcsock is already released using sock_release() and a pending
smc_listen_work accesses the clcsock than that will fail. Solve this
by canceling and waiting for the work to complete first. Because the
work holds the sock_lock it must make sure that the lock is not hold
before the new helper smc_clcsock_release() is invoked. And before the
smc_listen_work starts working check if the parent listen socket is
still valid, otherwise stop the work early.
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Stable-dep-of: 5211c9729484 ("net/smc: fix dangling sock under state SMC_APPFINCLOSEWAIT")
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
net/smc/af_smc.c | 14 ++++++++------
net/smc/smc_close.c | 25 +++++++++++++++++++++----
net/smc/smc_close.h | 1 +
3 files changed, 30 insertions(+), 10 deletions(-)
diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 6f342f6cc4876..6b30bec54b624 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -158,10 +158,9 @@ static int smc_release(struct socket *sock)
if (sk->sk_state == SMC_CLOSED) {
if (smc->clcsock) {
- mutex_lock(&smc->clcsock_release_lock);
- sock_release(smc->clcsock);
- smc->clcsock = NULL;
- mutex_unlock(&smc->clcsock_release_lock);
+ release_sock(sk);
+ smc_clcsock_release(smc);
+ lock_sock(sk);
}
if (!smc->use_fallback)
smc_conn_free(&smc->conn);
@@ -1014,13 +1013,13 @@ static void smc_listen_out(struct smc_sock *new_smc)
struct smc_sock *lsmc = new_smc->listen_smc;
struct sock *newsmcsk = &new_smc->sk;
- lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING);
if (lsmc->sk.sk_state == SMC_LISTEN) {
+ lock_sock_nested(&lsmc->sk, SINGLE_DEPTH_NESTING);
smc_accept_enqueue(&lsmc->sk, newsmcsk);
+ release_sock(&lsmc->sk);
} else { /* no longer listening */
smc_close_non_accepted(newsmcsk);
}
- release_sock(&lsmc->sk);
/* Wake up accept */
lsmc->sk.sk_data_ready(&lsmc->sk);
@@ -1216,6 +1215,9 @@ static void smc_listen_work(struct work_struct *work)
int rc = 0;
u8 ibport;
+ if (new_smc->listen_smc->sk.sk_state != SMC_LISTEN)
+ return smc_listen_out_err(new_smc);
+
if (new_smc->use_fallback) {
smc_listen_out_connected(new_smc);
return;
diff --git a/net/smc/smc_close.c b/net/smc/smc_close.c
index 3e7858793d485..cac0773f5ebd9 100644
--- a/net/smc/smc_close.c
+++ b/net/smc/smc_close.c
@@ -21,6 +21,22 @@
#define SMC_CLOSE_WAIT_LISTEN_CLCSOCK_TIME (5 * HZ)
+/* release the clcsock that is assigned to the smc_sock */
+void smc_clcsock_release(struct smc_sock *smc)
+{
+ struct socket *tcp;
+
+ if (smc->listen_smc && current_work() != &smc->smc_listen_work)
+ cancel_work_sync(&smc->smc_listen_work);
+ mutex_lock(&smc->clcsock_release_lock);
+ if (smc->clcsock) {
+ tcp = smc->clcsock;
+ smc->clcsock = NULL;
+ sock_release(tcp);
+ }
+ mutex_unlock(&smc->clcsock_release_lock);
+}
+
static void smc_close_cleanup_listen(struct sock *parent)
{
struct sock *sk;
@@ -331,6 +347,7 @@ static void smc_close_passive_work(struct work_struct *work)
close_work);
struct smc_sock *smc = container_of(conn, struct smc_sock, conn);
struct smc_cdc_conn_state_flags *rxflags;
+ bool release_clcsock = false;
struct sock *sk = &smc->sk;
int old_state;
@@ -417,13 +434,13 @@ static void smc_close_passive_work(struct work_struct *work)
if ((sk->sk_state == SMC_CLOSED) &&
(sock_flag(sk, SOCK_DEAD) || !sk->sk_socket)) {
smc_conn_free(conn);
- if (smc->clcsock) {
- sock_release(smc->clcsock);
- smc->clcsock = NULL;
- }
+ if (smc->clcsock)
+ release_clcsock = true;
}
}
release_sock(sk);
+ if (release_clcsock)
+ smc_clcsock_release(smc);
sock_put(sk); /* sock_hold done by schedulers of close_work */
}
diff --git a/net/smc/smc_close.h b/net/smc/smc_close.h
index 19eb6a211c23c..e0e3b5df25d24 100644
--- a/net/smc/smc_close.h
+++ b/net/smc/smc_close.h
@@ -23,5 +23,6 @@ void smc_close_wake_tx_prepared(struct smc_sock *smc);
int smc_close_active(struct smc_sock *smc);
int smc_close_shutdown_write(struct smc_sock *smc);
void smc_close_init(struct smc_sock *smc);
+void smc_clcsock_release(struct smc_sock *smc);
#endif /* SMC_CLOSE_H */
--
2.42.0
next prev parent reply other threads:[~2023-11-15 20:46 UTC|newest]
Thread overview: 92+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-15 20:35 [PATCH 4.19 00/88] 4.19.299-rc1 review Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 01/88] vfs: fix readahead(2) on block devices Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 02/88] genirq/matrix: Exclude managed interrupts in irq_matrix_allocated() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 03/88] i40e: fix potential memory leaks in i40e_remove() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 04/88] tcp_metrics: add missing barriers on delete Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 05/88] tcp_metrics: properly set tp->snd_ssthresh in tcp_init_metrics() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 06/88] tcp_metrics: do not create an entry from tcp_init_metrics() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 07/88] wifi: rtlwifi: fix EDCA limit set by BT coexistence Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 08/88] can: dev: move driver related infrastructure into separate subdir Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 09/88] can: dev: can_restart(): dont crash kernel if carrier is OK Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 10/88] can: dev: can_restart(): fix race condition between controller restart and netif_carrier_on() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 11/88] thermal: core: prevent potential string overflow Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 12/88] chtls: fix tp->rcv_tstamp initialization Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 13/88] ACPI: sysfs: Fix create_pnp_modalias() and create_of_modalias() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 14/88] ipv6: avoid atomic fragment on GSO packets Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 15/88] macsec: Fix traffic counters/statistics Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 16/88] macsec: use DEV_STATS_INC() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 17/88] net: add DEV_STATS_READ() helper Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 18/88] ipvlan: properly track tx_errors Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 19/88] regmap: debugfs: Fix a erroneous check after snprintf() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 20/88] clk: qcom: clk-rcg2: Fix clock rate overflow for high parent frequencies Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 21/88] clk: keystone: pll: fix a couple NULL vs IS_ERR() checks Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 22/88] clk: npcm7xx: Fix incorrect kfree Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 23/88] clk: mediatek: clk-mt6797: Add check for mtk_alloc_clk_data Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 24/88] clk: mediatek: clk-mt2701: " Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 25/88] platform/x86: wmi: Fix probe failure when failing to register WMI devices Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 26/88] platform/x86: wmi: remove unnecessary initializations Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 27/88] platform/x86: wmi: Fix opening of char device Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 28/88] hwmon: (coretemp) Fix potentially truncated sysfs attribute name Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 29/88] drm/rockchip: vop: Fix reset of state in duplicate state crtc funcs Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 30/88] drm/radeon: possible buffer overflow Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 31/88] drm/rockchip: cdn-dp: Fix some error handling paths in cdn_dp_probe() Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 32/88] ARM: dts: qcom: mdm9615: populate vsdcc fixed regulator Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 33/88] firmware: ti_sci: Mark driver as non removable Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 34/88] clk: scmi: Free scmi_clk allocated when the clocks with invalid info are skipped Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 35/88] hwrng: geode - fix accessing registers Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 36/88] sched/rt: Provide migrate_disable/enable() inlines Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 37/88] nd_btt: Make BTT lanes preemptible Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 38/88] HID: cp2112: Use irqchip template Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 39/88] hid: cp2112: Fix duplicate workqueue initialization Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 40/88] ARM: 9321/1: memset: cast the constant byte to unsigned char Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 41/88] ext4: move ix sanity check to corrent position Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 42/88] RDMA/hfi1: Workaround truncation compilation error Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 43/88] sh: bios: Revive earlyprintk support Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 44/88] ASoC: Intel: Skylake: Fix mem leak when parsing UUIDs fails Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 45/88] ASoC: ams-delta.c: use component after check Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 46/88] mfd: dln2: Fix double put in dln2_probe Greg Kroah-Hartman
2023-11-15 20:35 ` [PATCH 4.19 47/88] leds: pwm: simplify if condition Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 48/88] leds: pwm: convert to atomic PWM API Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 49/88] leds: pwm: Dont disable the PWM when the LED should be off Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 50/88] ledtrig-cpu: Limit to 8 CPUs Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 51/88] leds: trigger: ledtrig-cpu:: Fix output may be truncated issue for cpu Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 52/88] tty: tty_jobctrl: fix pid memleak in disassociate_ctty() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 53/88] usb: dwc2: fix possible NULL pointer dereference caused by driver concurrency Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 54/88] dmaengine: ti: edma: handle irq_of_parse_and_map() errors Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 55/88] misc: st_core: Do not call kfree_skb() under spin_lock_irqsave() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 56/88] tools: iio: privatize globals and functions in iio_generic_buffer.c file Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 57/88] tools: iio: iio_generic_buffer: Fix some integer type and calculation Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 58/88] tools: iio: iio_generic_buffer ensure alignment Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 59/88] USB: usbip: fix stub_dev hub disconnect Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 60/88] dmaengine: pxa_dma: Remove an erroneous BUG_ON() in pxad_free_desc() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 61/88] f2fs: fix to initialize map.m_pblk in f2fs_precache_extents() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 62/88] pcmcia: cs: fix possible hung task and memory leak pccardd() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 63/88] pcmcia: ds: fix refcount leak in pcmcia_device_add() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 64/88] pcmcia: ds: fix possible name leak in error path " Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 65/88] media: bttv: fix use after free error due to btv->timeout timer Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 66/88] media: s3c-camif: Avoid inappropriate kfree() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 67/88] media: dvb-usb-v2: af9035: fix missing unlock Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 68/88] pwm: sti: Avoid conditional gotos Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 69/88] pwm: sti: Reduce number of allocations and drop usage of chip_data Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 70/88] pwm: brcmstb: Utilize appropriate clock APIs in suspend/resume Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 71/88] Input: synaptics-rmi4 - fix use after free in rmi_unregister_function() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 72/88] llc: verify mac len before reading mac header Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 73/88] tipc: Change nla_policy for bearer-related names to NLA_NUL_STRING Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 74/88] dccp: Call security_inet_conn_request() after setting IPv4 addresses Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 75/88] dccp/tcp: Call security_inet_conn_request() after setting IPv6 addresses Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 76/88] r8169: improve rtl_set_rx_mode Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 77/88] net: r8169: Disable multicast filter for RTL8168H and RTL8107E Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 78/88] net/smc: postpone release of clcsock Greg Kroah-Hartman
2023-11-15 20:36 ` Greg Kroah-Hartman [this message]
2023-11-15 20:36 ` [PATCH 4.19 80/88] net/smc: fix dangling sock under state SMC_APPFINCLOSEWAIT Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 81/88] tg3: power down device only on SYSTEM_POWER_OFF Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 82/88] r8169: respect userspace disabling IFF_MULTICAST Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 83/88] netfilter: xt_recent: fix (increase) ipv6 literal buffer length Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 84/88] fbdev: imsttfb: Fix error path of imsttfb_probe() Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 85/88] fbdev: imsttfb: fix a resource leak in probe Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 86/88] fbdev: fsl-diu-fb: mark wr_reg_wa() static Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 87/88] Revert "mmc: core: Capture correct oemid-bits for eMMC cards" Greg Kroah-Hartman
2023-11-15 20:36 ` [PATCH 4.19 88/88] btrfs: use u64 for buffer sizes in the tree search ioctls Greg Kroah-Hartman
2023-11-16 11:13 ` [PATCH 4.19 00/88] 4.19.299-rc1 review Naresh Kamboju
2023-11-17 4:24 ` Guenter Roeck
2023-11-17 17:00 ` Pavel Machek
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=20231115191430.830506616@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=davem@davemloft.net \
--cc=kgraul@linux.ibm.com \
--cc=patches@lists.linux.dev \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.org \
--cc=ubraun@linux.ibm.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).