All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Hurley <peter@hurleysoftware.com>
To: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.cz>,
	linux-kernel@vger.kernel.org,
	Peter Hurley <peter@hurleysoftware.com>
Subject: [PATCH v3 17/19] tty: Destroy ldisc instance on hangup
Date: Sun, 10 Jan 2016 22:41:06 -0800	[thread overview]
Message-ID: <1452494468-21359-18-git-send-email-peter@hurleysoftware.com> (raw)
In-Reply-To: <1452494468-21359-1-git-send-email-peter@hurleysoftware.com>

Currently, when the tty is hungup, the ldisc is re-instanced; ie., the
current instance is destroyed and a new instance is created. The purpose
of this design was to guarantee a valid, open ldisc for the lifetime of
the tty.

However, now that tty buffers are owned by and have lifetime equivalent
to the tty_port (since v3.10), any data received immediately after the
ldisc is re-instanced may cause continued driver i/o operations
concurrently with the driver's hangup() operation. For drivers that
shutdown h/w on hangup, this is unexpected and usually bad. For example,
the serial core may free the xmit buffer page concurrently with an
in-progress write() operation (triggered by echo).

With the existing stable and robust ldisc reference handling, the
cleaned-up tty_reopen(), the straggling unsafe ldisc use cleaned up, and
the preparation to properly handle a NULL tty->ldisc, the ldisc instance
can be destroyed and only re-instanced when the tty is re-opened.

If the tty was opened as /dev/console or /dev/tty0, the original behavior
of re-instancing the ldisc is retained (the 'reinit' parameter to
tty_ldisc_hangup() is true). This is required since those file descriptors
are never hungup.

This patch has neglible impact on userspace; the tty file_operations ptr
is changed to point to the hungup file operations _before_ the ldisc
instance is destroyed, so only racing file operations might now retrieve
a NULL ldisc reference (which is simply handled as if the hungup file
operation had been called instead -- see "tty: Prepare for destroying
line discipline on hangup").

This resolves a long-standing FIXME and several crash reports.

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
---
 drivers/tty/tty_io.c    | 12 ++++++------
 drivers/tty/tty_ldisc.c | 40 +++++++++++++++++-----------------------
 include/linux/tty.h     |  3 ++-
 3 files changed, 25 insertions(+), 30 deletions(-)

diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
index f6f559c..4ad8454 100644
--- a/drivers/tty/tty_io.c
+++ b/drivers/tty/tty_io.c
@@ -728,7 +728,7 @@ static void __tty_hangup(struct tty_struct *tty, int exit_session)
 	while (refs--)
 		tty_kref_put(tty);
 
-	tty_ldisc_hangup(tty);
+	tty_ldisc_hangup(tty, cons_filp != NULL);
 
 	spin_lock_irq(&tty->ctrl_lock);
 	clear_bit(TTY_THROTTLED, &tty->flags);
@@ -753,10 +753,9 @@ static void __tty_hangup(struct tty_struct *tty, int exit_session)
 	} else if (tty->ops->hangup)
 		tty->ops->hangup(tty);
 	/*
-	 * We don't want to have driver/ldisc interactions beyond
-	 * the ones we did here. The driver layer expects no
-	 * calls after ->hangup() from the ldisc side. However we
-	 * can't yet guarantee all that.
+	 * We don't want to have driver/ldisc interactions beyond the ones
+	 * we did here. The driver layer expects no calls after ->hangup()
+	 * from the ldisc side, which is now guaranteed.
 	 */
 	set_bit(TTY_HUPPED, &tty->flags);
 	tty_unlock(tty);
@@ -1480,7 +1479,8 @@ static int tty_reopen(struct tty_struct *tty)
 
 	tty->count++;
 
-	WARN_ON(!tty->ldisc);
+	if (!tty->ldisc)
+		return tty_ldisc_reinit(tty, tty->termios.c_line);
 
 	return 0;
 }
diff --git a/drivers/tty/tty_ldisc.c b/drivers/tty/tty_ldisc.c
index 527fc5b..199e4b4 100644
--- a/drivers/tty/tty_ldisc.c
+++ b/drivers/tty/tty_ldisc.c
@@ -257,6 +257,9 @@ const struct file_operations tty_ldiscs_proc_fops = {
  *	reference to it. If the line discipline is in flux then
  *	wait patiently until it changes.
  *
+ *	Returns: NULL if the tty has been hungup and not re-opened with
+ *		 a new file descriptor, otherwise valid ldisc reference
+ *
  *	Note: Must not be called from an IRQ/timer context. The caller
  *	must also be careful not to hold other locks that will deadlock
  *	against a discipline change, such as an existing ldisc reference
@@ -644,14 +647,15 @@ static void tty_reset_termios(struct tty_struct *tty)
  *	@disc: line discipline to reinitialize
  *
  *	Completely reinitialize the line discipline state, by closing the
- *	current instance and opening a new instance. If an error occurs opening
- *	the new non-N_TTY instance, the instance is dropped and tty->ldisc reset
- *	to NULL. The caller can then retry with N_TTY instead.
+ *	current instance, if there is one, and opening a new instance. If
+ *	an error occurs opening the new non-N_TTY instance, the instance
+ *	is dropped and tty->ldisc reset to NULL. The caller can then retry
+ *	with N_TTY instead.
  *
  *	Returns 0 if successful, otherwise error code < 0
  */
 
-static int tty_ldisc_reinit(struct tty_struct *tty, int disc)
+int tty_ldisc_reinit(struct tty_struct *tty, int disc)
 {
 	struct tty_ldisc *ld;
 	int retval;
@@ -695,11 +699,9 @@ static int tty_ldisc_reinit(struct tty_struct *tty, int disc)
  *	tty itself so we must be careful about locking rules.
  */
 
-void tty_ldisc_hangup(struct tty_struct *tty)
+void tty_ldisc_hangup(struct tty_struct *tty, bool reinit)
 {
 	struct tty_ldisc *ld;
-	int reset = tty->driver->flags & TTY_DRIVER_RESET_TERMIOS;
-	int err = 0;
 
 	tty_ldisc_debug(tty, "%p: hangup\n", tty->ldisc);
 
@@ -727,25 +729,17 @@ void tty_ldisc_hangup(struct tty_struct *tty)
 	 */
 	tty_ldisc_lock(tty, MAX_SCHEDULE_TIMEOUT);
 
-	if (tty->ldisc) {
+	if (tty->driver->flags & TTY_DRIVER_RESET_TERMIOS)
+		tty_reset_termios(tty);
 
-		/* At this point we have a halted ldisc; we want to close it and
-		   reopen a new ldisc. We could defer the reopen to the next
-		   open but it means auditing a lot of other paths so this is
-		   a FIXME */
-		if (reset == 0)
-			err = tty_ldisc_reinit(tty, tty->termios.c_line);
-
-		/* If the re-open fails or we reset then go to N_TTY. The
-		   N_TTY open cannot fail */
-		if (reset || err < 0)
-			tty_ldisc_reinit(tty, N_TTY);
+	if (tty->ldisc) {
+		if (reinit) {
+			if (tty_ldisc_reinit(tty, tty->termios.c_line) < 0)
+				tty_ldisc_reinit(tty, N_TTY);
+		} else
+			tty_ldisc_kill(tty);
 	}
 	tty_ldisc_unlock(tty);
-	if (reset)
-		tty_reset_termios(tty);
-
-	tty_ldisc_debug(tty, "%p: re-opened\n", tty->ldisc);
 }
 
 /**
diff --git a/include/linux/tty.h b/include/linux/tty.h
index ee73220..56d1133 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -493,7 +493,8 @@ extern int tty_set_termios(struct tty_struct *tty, struct ktermios *kt);
 extern struct tty_ldisc *tty_ldisc_ref(struct tty_struct *);
 extern void tty_ldisc_deref(struct tty_ldisc *);
 extern struct tty_ldisc *tty_ldisc_ref_wait(struct tty_struct *);
-extern void tty_ldisc_hangup(struct tty_struct *tty);
+extern void tty_ldisc_hangup(struct tty_struct *tty, bool reset);
+extern int tty_ldisc_reinit(struct tty_struct *tty, int disc);
 extern const struct file_operations tty_ldiscs_proc_fops;
 
 extern void tty_wakeup(struct tty_struct *tty);
-- 
2.7.0

  parent reply	other threads:[~2016-01-11  6:42 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-27 21:38 [PATCH 00/19] Fix driver crashes on hangup Peter Hurley
2015-11-27 21:38 ` [PATCH 01/19] staging: digi: Replace open-coded tty_wakeup() Peter Hurley
2015-11-27 21:38 ` [PATCH 02/19] serial: 68328: Remove bogus ldisc reset Peter Hurley
2015-11-27 21:39 ` [PATCH 03/19] bluetooth: hci_ldisc: Remove dead code Peter Hurley
2015-12-02  7:47   ` Marcel Holtmann
2015-11-27 21:39 ` [PATCH 04/19] NFC: nci: " Peter Hurley
2015-11-27 21:39 ` [PATCH 05/19] tty: Remove chars_in_buffer() line discipline method Peter Hurley
2015-11-27 21:39 ` [PATCH 06/19] tty: Fix unsafe ldisc reference via ioctl(TIOCGETD) Peter Hurley
2015-11-27 21:39 ` [PATCH 07/19] n_tty: Fix unsafe reference to "other" ldisc Peter Hurley
2015-11-27 21:39 ` [PATCH 08/19] tty: Reset c_line from driver's init_termios Peter Hurley
2015-11-27 21:39 ` [PATCH 09/19] staging/speakup: Use tty_ldisc_ref() for paste kworker Peter Hurley
2015-11-27 21:39 ` [PATCH 10/19] tty: Fix comments for tty_ldisc_get() Peter Hurley
2015-11-27 21:39 ` [PATCH 11/19] tty: Fix comments for tty_ldisc_release() Peter Hurley
2015-11-27 21:39 ` [PATCH 12/19] tty: Prepare for destroying line discipline on hangup Peter Hurley
2015-11-27 21:39 ` [PATCH 13/19] tty: Handle NULL tty->ldisc Peter Hurley
2015-11-27 21:39 ` [PATCH 14/19] tty: Move tty_ldisc_kill() Peter Hurley
2015-11-27 21:39 ` [PATCH 15/19] tty: Use 'disc' for line discipline index name Peter Hurley
2015-11-27 21:39 ` [PATCH 16/19] tty: Refactor tty_ldisc_reinit() for reuse Peter Hurley
2015-11-27 21:39 ` [PATCH 17/19] tty: Destroy ldisc instance on hangup Peter Hurley
2015-11-27 21:39 ` [PATCH 18/19] tty: Document c_line == N_TTY initial condition Peter Hurley
2015-11-27 21:39 ` [PATCH 19/19] tty: Touch up style issues in ldisc core Peter Hurley
2016-01-10  4:40 ` [PATCH v2 00/19] Fix driver crashes on hangup Peter Hurley
2016-01-10  4:40   ` [PATCH v2 01/19] staging: digi: Replace open-coded tty_wakeup() Peter Hurley
2016-01-10  4:40   ` [PATCH v2 02/19] serial: 68328: Remove bogus ldisc reset Peter Hurley
2016-01-10  4:40   ` [PATCH v2 03/19] bluetooth: hci_ldisc: Remove dead code Peter Hurley
2016-01-10  4:40   ` [PATCH v2 04/19] NFC: nci: " Peter Hurley
2016-01-10  4:40   ` [PATCH v2 05/19] tty: Remove chars_in_buffer() line discipline method Peter Hurley
2016-01-10  4:40   ` [PATCH v2 06/19] tty: Fix unsafe ldisc reference via ioctl(TIOCGETD) Peter Hurley
2016-01-10  5:24     ` Peter Hurley
2016-01-10  4:40   ` [PATCH v2 07/19] n_tty: Fix unsafe reference to "other" ldisc Peter Hurley
2016-01-10  5:26     ` Peter Hurley
2016-01-10  4:40   ` [PATCH v2 08/19] tty: Reset c_line from driver's init_termios Peter Hurley
2016-01-10  4:41   ` [PATCH v2 09/19] staging/speakup: Use tty_ldisc_ref() for paste kworker Peter Hurley
2016-01-10 23:16     ` Ben Hutchings
2016-01-11  0:25       ` Peter Hurley
2016-01-11  5:40         ` Peter Hurley
2016-01-11 10:37           ` Ben Hutchings
2016-01-10  4:41   ` [PATCH v2 10/19] tty: Fix comments for tty_ldisc_get() Peter Hurley
2016-01-10  4:41   ` [PATCH v2 11/19] tty: Fix comments for tty_ldisc_release() Peter Hurley
2016-01-10  4:41   ` [PATCH v2 12/19] tty: Prepare for destroying line discipline on hangup Peter Hurley
2016-01-10  4:41   ` [PATCH v2 13/19] tty: Handle NULL tty->ldisc Peter Hurley
2016-01-10  4:41   ` [PATCH v2 14/19] tty: Move tty_ldisc_kill() Peter Hurley
2016-01-10  4:41   ` [PATCH v2 15/19] tty: Use 'disc' for line discipline index name Peter Hurley
2016-01-10  4:41   ` [PATCH v2 16/19] tty: Refactor tty_ldisc_reinit() for reuse Peter Hurley
2016-01-10  4:41   ` [PATCH v2 17/19] tty: Destroy ldisc instance on hangup Peter Hurley
2016-01-10  6:24     ` kbuild test robot
2016-01-10  6:24     ` [PATCH] tty: fix badzero.cocci warnings kbuild test robot
2016-01-10  7:02       ` Peter Hurley
2016-01-10  4:41   ` [PATCH v2 18/19] tty: Document c_line == N_TTY initial condition Peter Hurley
2016-01-10  4:41   ` [PATCH v2 19/19] tty: Avoid unnecessary temporaries for tty->ldisc Peter Hurley
2016-01-11  6:40   ` [PATCH v3 00/19] Fix driver crashes on hangup Peter Hurley
2016-01-11  6:40     ` [PATCH v3 01/19] staging: digi: Replace open-coded tty_wakeup() Peter Hurley
2016-01-11  6:40     ` [PATCH v3 02/19] serial: 68328: Remove bogus ldisc reset Peter Hurley
2016-01-11 14:12       ` One Thousand Gnomes
2016-01-11  6:40     ` [PATCH v3 03/19] bluetooth: hci_ldisc: Remove dead code Peter Hurley
2016-01-11  6:40     ` [PATCH v3 04/19] NFC: nci: " Peter Hurley
2016-01-11  6:40     ` [PATCH v3 05/19] tty: Remove chars_in_buffer() line discipline method Peter Hurley
2016-01-11  6:40     ` [PATCH v3 06/19] tty: Fix unsafe ldisc reference via ioctl(TIOCGETD) Peter Hurley
2016-01-11  6:40     ` [PATCH v3 07/19] n_tty: Fix unsafe reference to "other" ldisc Peter Hurley
2016-01-11  6:40     ` [PATCH v3 08/19] tty: Reset c_line from driver's init_termios Peter Hurley
2016-01-11  6:40     ` [PATCH v3 09/19] staging/speakup: Use tty_ldisc_ref() for paste kworker Peter Hurley
2016-01-11  6:40     ` [PATCH v3 10/19] tty: Fix comments for tty_ldisc_get() Peter Hurley
2016-01-11  6:41     ` [PATCH v3 11/19] tty: Fix comments for tty_ldisc_release() Peter Hurley
2016-01-11  6:41     ` [PATCH v3 12/19] tty: Prepare for destroying line discipline on hangup Peter Hurley
2016-01-11  6:41     ` [PATCH v3 13/19] tty: Handle NULL tty->ldisc Peter Hurley
2016-01-11  6:41     ` [PATCH v3 14/19] tty: Move tty_ldisc_kill() Peter Hurley
2016-01-11  6:41     ` [PATCH v3 15/19] tty: Use 'disc' for line discipline index name Peter Hurley
2016-01-11  6:41     ` [PATCH v3 16/19] tty: Refactor tty_ldisc_reinit() for reuse Peter Hurley
2016-01-11  6:41     ` Peter Hurley [this message]
2016-01-11  6:41     ` [PATCH v3 18/19] tty: Document c_line == N_TTY initial condition Peter Hurley
2016-01-11  6:41     ` [PATCH v3 19/19] tty: Avoid unnecessary temporaries for tty->ldisc Peter Hurley

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=1452494468-21359-18-git-send-email-peter@hurleysoftware.com \
    --to=peter@hurleysoftware.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jslaby@suse.cz \
    --cc=linux-kernel@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.