From: George Spelvin <linux@horizon.com>
Date: Sun, 10 Feb 2013 09:43:41 +0000 (-0500)
Subject: pps: Don't crash the machine when exiting will do
X-Git-Tag: v3.9-rc1~125^2~13
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=ce3da1a654c83c6c9cb0b33477815e5d1293cc00;p=profile%2Fivi%2Fkernel-x86-ivi.git

pps: Don't crash the machine when exiting will do

PPS is not really the must-have subsystem that warrants crashing
the machine if the ldisc interface is broken.

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: George Spelvin <linux@horizon.com>
Acked-by: Rodolfo Giometti <giometti@enneenne.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---

diff --git a/drivers/pps/clients/pps-ldisc.c b/drivers/pps/clients/pps-ldisc.c
index 60cee9e0ecb..a94f73e1480 100644
--- a/drivers/pps/clients/pps-ldisc.c
+++ b/drivers/pps/clients/pps-ldisc.c
@@ -25,6 +25,7 @@
 #include <linux/serial_core.h>
 #include <linux/tty.h>
 #include <linux/pps_kernel.h>
+#include <linux/bug.h>
 
 #define PPS_TTY_MAGIC		0x0001
 
@@ -33,7 +34,12 @@ static void pps_tty_dcd_change(struct tty_struct *tty, unsigned int status,
 {
 	struct pps_device *pps = pps_lookup_dev(tty);
 
-	BUG_ON(pps == NULL);
+	/*
+	 * This should never fail, but the ldisc locking is very
+	 * convoluted, so don't crash just in case.
+	 */
+	if (WARN_ON_ONCE(pps == NULL))
+		return;
 
 	/* Now do the PPS event report */
 	pps_event(pps, ts, status ? PPS_CAPTUREASSERT :
@@ -93,6 +99,9 @@ static void pps_tty_close(struct tty_struct *tty)
 
 	alias_n_tty_close(tty);
 
+	if (WARN_ON(!pps))
+		return;
+
 	dev_info(pps->dev, "removed\n");
 	pps_unregister_source(pps);
 }