Re: 2.6.0-test11-mm1

[Andrew Morton <akpm@osdl.org>]

Diego Calleja García <aradorlinux@yahoo.es> wrote:
>
> local_bh_enable() was called in hard irq context.   This is probably a bug
> Call Trace:
>  [<c0127b96>] local_bh_enable+0x96/0xa0
>  [<e08fc4d8>] ppp_asynctty_receive+0x78/0xd0 [ppp_async]
>  [<c01cec6c>] flush_to_ldisc+0xdc/0x130
>  [<c01ecc17>] receive_chars+0x227/0x240
>  [<c01eccd5>] transmit_chars+0xa5/0xe0
>  [<c01ecf7c>] serial8250_interrupt+0x12c/0x130
>  [<c010c9f9>] handle_IRQ_event+0x49/0x80
>  [<c010cdc8>] do_IRQ+0xb8/0x180
>  [<c02ca760>] common_interrupt+0x18/0x20


ppp_asynctty_receive() is called from hard IRQ context and hence may not use
spin_unlock_bh().  The patch converts ppp to use an IRQ-safe spinlock.


 25-akpm/drivers/net/ppp_async.c |    9 +++++----
 1 files changed, 5 insertions(+), 4 deletions(-)

diff -puN drivers/net/ppp_async.c~ppp-locking-fix drivers/net/ppp_async.c
--- 25/drivers/net/ppp_async.c~ppp-locking-fix	Wed Dec 17 13:57:33 2003
+++ 25-akpm/drivers/net/ppp_async.c	Wed Dec 17 13:58:27 2003
@@ -321,12 +321,13 @@ ppp_asynctty_receive(struct tty_struct *
 		  char *flags, int count)
 {
 	struct asyncppp *ap = ap_get(tty);
+	unsigned long flags;
 
 	if (ap == 0)
 		return;
-	spin_lock_bh(&ap->recv_lock);
+	spin_lock_irqsave(&ap->recv_lock, flags);
 	ppp_async_input(ap, buf, flags, count);
-	spin_unlock_bh(&ap->recv_lock);
+	spin_unlock_irqrestore(&ap->recv_lock, flags);
 	ap_put(ap);
 	if (test_and_clear_bit(TTY_THROTTLED, &tty->flags)
 	    && tty->driver->unthrottle)
@@ -396,9 +397,9 @@ ppp_async_ioctl(struct ppp_channel *chan
 		if (get_user(val, (int *) arg))
 			break;
 		ap->flags = val & ~SC_RCV_BITS;
-		spin_lock_bh(&ap->recv_lock);
+		spin_lock_irq(&ap->recv_lock);
 		ap->rbits = val & SC_RCV_BITS;
-		spin_unlock_bh(&ap->recv_lock);
+		spin_unlock_irq(&ap->recv_lock);
 		err = 0;
 		break;
 

_