PCMCIA and APM suspend

Russell King rmk+pcmcia at arm.linux.org.uk
Wed Oct 22 00:19:38 BST 2003


On Mon, Oct 20, 2003 at 05:12:53PM +0300, Jari Tenhunen wrote:
> > - with yenta_socket loaded, cards in the sockets, but execute cardctl eject
> >   before suspending?
> 
> "cardctl eject" produces the following log (Orinoco wlan card plugged):
> Oct 20 16:00:45 t20 kernel: unregister_netdevice: waiting for eth1 to become free. Usage count = 5
> 
> However, this message can be avoided by doing "ifconfig eth1 down" 
> before eject. Trying to suspend results in exactly the same kind of 
> behaviour as earlier (doesn't matter if the net interface is down or up).

Ok.  So we know that suspend isn't doing something which unloading the
module does.

One thing which is immediately obvious is this:

        exca_writeb(sock, I365_CSCINT, 0);

which occurs in yenta_close() but not yenta_sock_suspend().  Can you
try this patch please?

--- linux/drivers/pcmcia/yenta_socket.c.old	Sun Sep 28 09:54:57 2003
+++ linux/drivers/pcmcia/yenta_socket.c	Tue Oct 21 23:18:47 2003
@@ -493,6 +493,7 @@
 
 	/* Disable CSC interrupts */
 	cb_writel(socket, CB_SOCKET_MASK, 0x0);
+	exca_writeb(socket, I365_CSCINT, 0);
 
 	return 0;
 }

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 PCMCIA      - http://pcmcia.arm.linux.org.uk/
                 2.6 Serial core



More information about the linux-pcmcia mailing list