mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Arnd Bergmann <arnd@arndb.de>
To: linux-kernel@vger.kernel.org
Cc: Arnd Bergmann <arnd@arndb.de>,
	Alan Cox <alan@lxorguk.ukuu.org.uk>, Greg KH <gregkh@suse.de>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	John Kacur <jkacur@redhat.com>, Al Viro <viro@zeniv.linux.org.uk>,
	Ingo Molnar <mingo@elte.hu>
Subject: [PATCH 07/10] tty: reorder ldisc locking
Date: Sat, 15 May 2010 22:59:53 +0200	[thread overview]
Message-ID: <1273957196-13768-8-git-send-email-arnd@arndb.de> (raw)
In-Reply-To: <1273957196-13768-1-git-send-email-arnd@arndb.de>

We need to release the BTM in paste_selection() when
sleeping in tty_ldisc_ref_wait to avoid deadlocks
with tty_ldisc_enable.

In tty_set_ldisc, we now always grab the BTM before
taking the ldisc_mutex in order to avoid AB-BA
deadlocks between the two.

tty_ldisc_halt potentially blocks on a workqueue
function that takes the BTM, so we must release
the BTM before calling it.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/char/selection.c |    9 +++++++--
 drivers/char/tty_ldisc.c |   24 ++++++++++++++++++++----
 2 files changed, 27 insertions(+), 6 deletions(-)

diff --git a/drivers/char/selection.c b/drivers/char/selection.c
index 85211a3..75889cd 100644
--- a/drivers/char/selection.c
+++ b/drivers/char/selection.c
@@ -319,8 +319,13 @@ int paste_selection(struct tty_struct *tty)
 	poke_blanked_console();
 	release_console_sem();
 
-	ld = tty_ldisc_ref_wait(tty);
-	
+	ld = tty_ldisc_ref(tty);
+	if (!ld) {
+		tty_unlock();
+		ld = tty_ldisc_ref_wait(tty);
+		tty_lock();
+	}
+
 	add_wait_queue(&vc->paste_wait, &wait);
 	while (sel_buffer && sel_buffer_lth > pasted) {
 		set_current_state(TASK_INTERRUPTIBLE);
diff --git a/drivers/char/tty_ldisc.c b/drivers/char/tty_ldisc.c
index 97681ff..0f49479 100644
--- a/drivers/char/tty_ldisc.c
+++ b/drivers/char/tty_ldisc.c
@@ -582,6 +582,7 @@ int tty_set_ldisc(struct tty_struct *tty, int ldisc)
 
 	tty_wait_until_sent(tty, 0);
 
+	tty_lock();
 	mutex_lock(&tty->ldisc_mutex);
 
 	/*
@@ -591,13 +592,13 @@ int tty_set_ldisc(struct tty_struct *tty, int ldisc)
 
 	while (test_bit(TTY_LDISC_CHANGING, &tty->flags)) {
 		mutex_unlock(&tty->ldisc_mutex);
+		tty_unlock();
 		wait_event(tty_ldisc_wait,
 			test_bit(TTY_LDISC_CHANGING, &tty->flags) == 0);
+		tty_lock();
 		mutex_lock(&tty->ldisc_mutex);
 	}
 
-	tty_lock();
-
 	set_bit(TTY_LDISC_CHANGING, &tty->flags);
 
 	/*
@@ -634,8 +635,8 @@ int tty_set_ldisc(struct tty_struct *tty, int ldisc)
 
 	flush_scheduled_work();
 
-	mutex_lock(&tty->ldisc_mutex);
 	tty_lock();
+	mutex_lock(&tty->ldisc_mutex);
 	if (test_bit(TTY_HUPPED, &tty->flags)) {
 		/* We were raced by the hangup method. It will have stomped
 		   the ldisc data and closed the ldisc down */
@@ -782,7 +783,20 @@ void tty_ldisc_hangup(struct tty_struct *tty)
 	 * Avoid racing set_ldisc or tty_ldisc_release
 	 */
 	mutex_lock(&tty->ldisc_mutex);
-	tty_ldisc_halt(tty);
+
+	/*
+	 * this is like tty_ldisc_halt, but we need to give up
+	 * the BTM before calling cancel_delayed_work_sync,
+	 * which may need to wait for another function taking the BTM
+	 */
+	clear_bit(TTY_LDISC, &tty->flags);
+	tty_unlock();
+	cancel_delayed_work_sync(&tty->buf.work);
+	mutex_unlock(&tty->ldisc_mutex);
+
+	tty_lock();
+	mutex_lock(&tty->ldisc_mutex);
+
 	/* At this point we have a closed ldisc and we want to
 	   reopen it. We could defer this to the next open but
 	   it means auditing a lot of other paths so this is
@@ -853,8 +867,10 @@ void tty_ldisc_release(struct tty_struct *tty, struct tty_struct *o_tty)
 	 * race with the set_ldisc code path.
 	 */
 
+	tty_unlock();
 	tty_ldisc_halt(tty);
 	flush_scheduled_work();
+	tty_lock();
 
 	mutex_lock(&tty->ldisc_mutex);
 	/*
-- 
1.7.0.4


  parent reply	other threads:[~2010-05-15 21:00 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-05-15 20:59 [PATCH v3 00/10] BKL conversion in tty layer Arnd Bergmann
2010-05-15 20:59 ` [PATCH 01/10] tty: replace BKL with a new tty_lock Arnd Bergmann
2010-05-15 20:59 ` [PATCH 02/10] tty: never hold BTM while getting tty_mutex Arnd Bergmann
2010-05-15 20:59 ` [PATCH 03/10] tty: fix console_sem lock order Arnd Bergmann
2010-05-15 20:59 ` [PATCH 04/10] cdc-acm: remove dead code Arnd Bergmann
2010-05-15 20:59 ` [PATCH 05/10] tty: introduce wait_event_interruptible_tty Arnd Bergmann
2010-05-15 20:59 ` [PATCH 06/10] tty: annotate tty_write_lock Arnd Bergmann
2010-05-15 20:59 ` Arnd Bergmann [this message]
2010-05-15 20:59 ` [PATCH 08/10] tty: untangle locking of wait_until_sent Arnd Bergmann
2010-05-16  3:12   ` Daniel K.
2010-05-15 20:59 ` [PATCH 09/10] tty: remove tty_lock_nested Arnd Bergmann
2010-05-15 20:59 ` [PATCH 10/10] tty: implement BTM as mutex instead of BKL Arnd Bergmann
2010-05-16  3:33   ` Daniel K.
2010-05-16 12:58     ` Arnd Bergmann
2010-05-17 13:41 ` [PATCH v3 00/10] BKL conversion in tty layer Alan Cox
2010-05-17 15:30   ` Greg KH
2010-05-17 18:30     ` Arnd Bergmann
2010-05-18  4:27       ` Greg KH
2010-05-18 21:52         ` Arnd Bergmann
2010-05-19  1:50           ` Greg KH
2010-05-22 13:54             ` Arnd Bergmann
2010-05-17 18:49   ` Arnd Bergmann
2010-06-17 19:13   ` Tony Luck
2010-06-17 19:40     ` Arnd Bergmann
2010-06-17 20:15       ` Tony Luck
2010-06-17 21:48         ` Arnd Bergmann
2010-06-17 22:09           ` Frederic Weisbecker
2010-06-18 12:58             ` [PATCH] tty: avoid recursive BTM in pty_close Arnd Bergmann
2010-06-18 16:21               ` Alan Cox
2010-06-18 16:52                 ` Tony Luck
2010-06-18 18:35                   ` Arnd Bergmann
2010-06-18 20:25                     ` Tony Luck
2010-06-19 12:32                       ` Arnd Bergmann
2010-06-19 20:29                         ` [PATCH] serial: revert "Use block_til_ready helper" Arnd Bergmann
2010-06-19 21:57                           ` Alan Cox
2010-06-20 20:54                             ` Arnd Bergmann
2010-06-21 14:13                               ` Alan Cox
2010-06-21 20:57                                 ` Arnd Bergmann
2010-06-21 20:58                                   ` Arnd Bergmann
2010-06-21 17:11                           ` Tony Luck
2010-06-22 23:01                           ` Greg KH
2010-06-28 17:17                     ` [PATCH] tty: avoid recursive BTM in pty_close Tony Luck
2010-06-28 19:03                       ` Arnd Bergmann

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=1273957196-13768-8-git-send-email-arnd@arndb.de \
    --to=arnd@arndb.de \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=fweisbec@gmail.com \
    --cc=gregkh@suse.de \
    --cc=jkacur@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    --cc=viro@zeniv.linux.org.uk \
    /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

Powered by JetHome