From mboxrd@z Thu Jan  1 00:00:00 1970
From: Lee Revell <rlrevell@joe-job.com>
Subject: [PATCH] Make netif_rx_ni preempt-safe
Date: Tue, 19 Oct 2004 19:55:33 -0400
Sender: netdev-bounce@oss.sgi.com
Message-ID: <1098230132.23628.28.camel@krustophenia.net>
Mime-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Cc: linux-kernel <linux-kernel@vger.kernel.org>,
        "David S. Miller" <davem@davemloft.net>, herbert@gondor.apana.org.au,
        vda@port.imtp.ilyichevsk.odessa.ua, linux-kernel@oss.sgi.com,
        maxk@qualcomm.com, irda-users@lists.sourceforge.net,
        Linux Network Development <netdev@oss.sgi.com>,
        Alain Schroeder <alain@parkautomat.net>
Return-path: <netdev-bounce@oss.sgi.com>
To: Andrew Morton <akpm@osdl.org>
Errors-to: netdev-bounce@oss.sgi.com
List-Id: netdev.vger.kernel.org

This patch makes netif_rx_ni() preempt-safe.  The problem was reported
by Alain Schroeder.  Here are the users:

drivers/s390/net/ctcmain.c
drivers/s390/net/netiucv.c
drivers/net/irda/vlsi_ir.c
drivers/net/tun.c

As David S. Miller explained, the do_softirq (and therefore the preempt
dis/enable) is required because there is no softirq check on the return
path when netif_rx is called from non-interrupt context.

Signed-Off-By: Lee Revell <rlrevell@joe-job.com>

--- include/linux/netdevice.h~	2004-10-19 18:50:18.000000000 -0400
+++ include/linux/netdevice.h	2004-10-19 18:51:01.000000000 -0400
@@ -696,9 +696,11 @@
  */
 static inline int netif_rx_ni(struct sk_buff *skb)
 {
+       preempt_disable();
        int err = netif_rx(skb);
        if (softirq_pending(smp_processor_id()))
                do_softirq();
+       preempt_enable();
        return err;
 }