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, Mikulas Patocka <mpatocka@redhat.com>,
	Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [PATCH 3.14 05/59] iscsi-target: fix iscsit_del_np deadlock on unload
Date: Fri,  4 Jul 2014 15:18:54 -0700	[thread overview]
Message-ID: <20140704221539.793500419@linuxfoundation.org> (raw)
In-Reply-To: <20140704221539.544876024@linuxfoundation.org>

3.14-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Mikulas Patocka <mpatocka@redhat.com>

commit 81a9c5e72bdf7109a65102ca61d8cbd722cf4021 upstream.

On uniprocessor preemptible kernel, target core deadlocks on unload. The
following events happen:
* iscsit_del_np is called
* it calls send_sig(SIGINT, np->np_thread, 1);
* the scheduler switches to the np_thread
* the np_thread is woken up, it sees that kthread_should_stop() returns
  false, so it doesn't terminate
* the np_thread clears signals with flush_signals(current); and goes back
  to sleep in iscsit_accept_np
* the scheduler switches back to iscsit_del_np
* iscsit_del_np calls kthread_stop(np->np_thread);
* the np_thread is waiting in iscsit_accept_np and it doesn't respond to
  kthread_stop

The deadlock could be resolved if the administrator sends SIGINT signal to
the np_thread with killall -INT iscsi_np

The reproducible deadlock was introduced in commit
db6077fd0b7dd41dc6ff18329cec979379071f87, but the thread-stopping code was
racy even before.

This patch fixes the problem. Using kthread_should_stop to stop the
np_thread is unreliable, so we test np_thread_state instead. If
np_thread_state equals ISCSI_NP_THREAD_SHUTDOWN, the thread exits.

Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/target/iscsi/iscsi_target_login.c |   13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

--- a/drivers/target/iscsi/iscsi_target_login.c
+++ b/drivers/target/iscsi/iscsi_target_login.c
@@ -1196,7 +1196,7 @@ old_sess_out:
 static int __iscsi_target_login_thread(struct iscsi_np *np)
 {
 	u8 *buffer, zero_tsih = 0;
-	int ret = 0, rc, stop;
+	int ret = 0, rc;
 	struct iscsi_conn *conn = NULL;
 	struct iscsi_login *login;
 	struct iscsi_portal_group *tpg = NULL;
@@ -1210,6 +1210,9 @@ static int __iscsi_target_login_thread(s
 	if (np->np_thread_state == ISCSI_NP_THREAD_RESET) {
 		np->np_thread_state = ISCSI_NP_THREAD_ACTIVE;
 		complete(&np->np_restart_comp);
+	} else if (np->np_thread_state == ISCSI_NP_THREAD_SHUTDOWN) {
+		spin_unlock_bh(&np->np_thread_lock);
+		goto exit;
 	} else {
 		np->np_thread_state = ISCSI_NP_THREAD_ACTIVE;
 	}
@@ -1402,10 +1405,8 @@ old_sess_out:
 	}
 
 out:
-	stop = kthread_should_stop();
-	/* Wait for another socket.. */
-	if (!stop)
-		return 1;
+	return 1;
+
 exit:
 	iscsi_stop_login_thread_timer(np);
 	spin_lock_bh(&np->np_thread_lock);
@@ -1422,7 +1423,7 @@ int iscsi_target_login_thread(void *arg)
 
 	allow_signal(SIGINT);
 
-	while (!kthread_should_stop()) {
+	while (1) {
 		ret = __iscsi_target_login_thread(np);
 		/*
 		 * We break and exit here unless another sock_accept() call



  parent reply	other threads:[~2014-07-04 22:18 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-04 22:18 [PATCH 3.14 00/59] 3.14.11-stable review Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 01/59] ARM: dts: disable MDMA1 node for exynos5420 Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 02/59] target: Fix left-over se_lun->lun_sep pointer OOPs Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 03/59] iscsi-target: Avoid rejecting incorrect ITT for Data-Out Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 04/59] iscsi-target: Explicily clear login response PDU in exception path Greg Kroah-Hartman
2014-07-04 22:18 ` Greg Kroah-Hartman [this message]
2014-07-04 22:18 ` [PATCH 3.14 06/59] Input: synaptics - fix resolution for manually provided min/max Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 07/59] Input: elantech - deal with clickpads reporting right button events Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 08/59] Input: elantech - dont set bit 1 of reg_10 when the no_hw_res quirk is set Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 09/59] PCI: Add new ID for Intel GPU "spurious interrupt" quirk Greg Kroah-Hartman
2014-07-04 22:18 ` [PATCH 3.14 10/59] PCI: Fix incorrect vgaarb conditional in WARN_ON() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 11/59] mtip32xx: Fix ERO and NoSnoop values in PCIe upstream on AMD systems Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 12/59] mtip32xx: Increase timeout for STANDBY IMMEDIATE command Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 13/59] mtip32xx: Remove dfs_parent after pci unregister Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 14/59] recordmcount/MIPS: Fix possible incorrect mcount_loc table entries in modules Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 15/59] MIPS: MSC: Prevent out-of-bounds writes to MIPS SC ioremapd region Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 16/59] UBIFS: fix an mmap and fsync race condition Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 17/59] UBIFS: Remove incorrect assertion in shrink_tnc() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 18/59] watchdog: sp805: Set watchdog_device->timeout from ->set_timeout() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 19/59] watchdog: ath79_wdt: avoid spurious restarts on AR934x Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 20/59] watchdog: kempld-wdt: Use the correct value when configuring the prescaler with the watchdog Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 21/59] kernel/watchdog.c: remove preemption restrictions when restarting lockup detector Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 22/59] IB/mlx5: add missing padding at end of struct mlx5_ib_create_cq Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 23/59] IB/mlx5: add missing padding at end of struct mlx5_ib_create_srq Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 24/59] IB/qib: Fix port in pkey change event Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 25/59] IB/ipath: Translate legacy diagpkt into newer extended diagpkt Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 26/59] IB/srp: Fix a sporadic crash triggered by cable pulling Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 27/59] IB/umad: Fix error handling Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 28/59] IB/umad: Fix use-after-free on close Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 29/59] SUNRPC: Fix a module reference leak in svc_handle_xprt Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 30/59] pNFS: Handle allocation errors correctly in filelayout_alloc_layout_hdr() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 31/59] nfsd4: fix FREE_STATEID lockowner leak Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 32/59] nfsd: getattr for FATTR4_WORD0_FILES_AVAIL needs the statfs buffer Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 33/59] NFS: Dont declare inode uptodate unless all attributes were checked Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 34/59] NFS: Use raw_write_seqcount_begin/end int nfs4_reclaim_open_state Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 35/59] NFS: populate ->net in mount data when remounting Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 36/59] nfs: Fix cache_validity check in nfs_write_pageuptodate() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 37/59] powerpc/pseries: Fix overwritten PE state Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 38/59] powerpc/mm: Check paca psize is up to date for huge mappings Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 39/59] powerpc/serial: Use saner flags when creating legacy ports Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 40/59] powerpc: 64bit sendfile is capped at 2GB Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 41/59] powerpc: fix typo CONFIG_PMAC Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 42/59] powerpc/perf: Ensure all EBB register state is cleared on fork() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 43/59] powerpc: fix typo CONFIG_PPC_CPU Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 44/59] powerpc: Dont setup CPUs with bad status Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 45/59] powerpc: Add AT_HWCAP2 to indicate V.CRYPTO category support Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 46/59] powerpc: Dont skip ePAPR spin-table CPUs Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 47/59] xfs: xfs_readsb needs to check for magic numbers Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 48/59] reiserfs: call truncate_setsize under tailpack mutex Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 49/59] MIPS: KVM: Remove redundant NULL checks before kfree() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 50/59] MIPS: KVM: Fix memory leak on VCPU Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 51/59] ipvs: Fix panic due to non-linear skb Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 52/59] ptrace,x86: force IRET path after a ptrace_stop() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 53/59] lz4: add overrun checks to lz4_uncompress_unknownoutputsize() Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 54/59] Documentation/SubmittingPatches: describe the Fixes: tag Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 55/59] tracing: Try again for saved cmdline if failed due to locking Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 56/59] tracing: Fix syscall_*regfunc() vs copy_process() race Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 57/59] ALSA: usb-audio: Fix races at disconnection and PCM closing Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 58/59] ALSA: hda - hdmi: call overridden init on resume Greg Kroah-Hartman
2014-07-04 22:19 ` [PATCH 3.14 59/59] ALSA: hda - Adjust speaker HPF and add LED support for HP Spectre 13 Greg Kroah-Hartman
2014-07-04 23:20   ` Holger Hoffstätte
2014-07-05 12:58   ` Takashi Iwai
2014-07-05 19:10     ` Greg Kroah-Hartman
2014-07-05  0:47 ` [PATCH 3.14 00/59] 3.14.11-stable review Holger Hoffstätte
2014-07-05  5:45 ` Guenter Roeck
2014-07-05  7:00   ` Satoru Takeuchi
2014-07-05  7:08     ` Guenter Roeck
2014-07-05  9:21     ` Satoru Takeuchi
2014-07-05 12:56       ` Takashi Iwai
2014-07-05 19:10         ` Greg Kroah-Hartman
2014-07-05 21:53           ` Satoru Takeuchi

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=20140704221539.793500419@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mpatocka@redhat.com \
    --cc=nab@linux-iscsi.org \
    --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;
as well as URLs for NNTP newsgroup(s).