Re: [PATCH 1/3] Char: mxser_new, fix recursive locking

[Jan Yenya Kasprzak <kas@fi.muni.cz>]

Alan Cox wrote:
: > Acquire a port lock only if not in_interrupt in some places, because ISR
: > holds the lock yet (and ldisc calls some of driver's routines which tries to
: > acquire it again due to tty->low_latency).
: 
: NAK
: 
: This is the wrong way to do it. If you don't support recursive entry then
: don't use ->low_latency. If you do then ensure you drop the lock before
: you call tty_flip_buffer_push().

	I did as you suggested, and I am not able to reproduce the problem
now. The patch is attached. I think it is quite minimal, so it should be
safe to apply it. What do you think, Jiri?

Signed-off-by: Jan "Yenya" Kasprzak <kas@fi.muni.cz>
Cc: Jiri Slaby <jirislaby@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>

--- linux-2.6.21-rc2/drivers/char/mxser_new.c.orig	2007-04-20 15:41:46.000000000 +0200
+++ linux-2.6.21-rc2/drivers/char/mxser_new.c	2007-04-20 15:46:46.000000000 +0200
@@ -2230,7 +2230,14 @@
 	port->mon_data.rxcnt += cnt;
 	port->mon_data.up_rxcnt += cnt;
 
+	/*
+	 * We are called from an interrupt context with &port->slock
+	 * being held. Drop it temporarily in order to prevent
+	 * recursive locking.
+	 */
+	spin_unlock(&port->slock);
 	tty_flip_buffer_push(tty);
+	spin_lock(&port->slock);
 }
 
 static void mxser_transmit_chars(struct mxser_port *port)


-Yenya

-- 
| Jan "Yenya" Kasprzak  <kas at {fi.muni.cz - work | yenya.net - private}> |
| GPG: ID 1024/D3498839      Fingerprint 0D99A7FB206605D7 8B35FCDE05B18A5E |
| http://www.fi.muni.cz/~kas/    Journal: http://www.fi.muni.cz/~kas/blog/ |
> I will never go to meetings again because I think  face to face meetings <
> are the biggest waste of time you can ever have.        --Linus Torvalds <