[Cbe-oss-dev] [PATCH] spidernet: enable poll() before registering interrupts

[Arnd Bergmann]

We must not call netif_poll_enable after enabling interrupts,
because an interrupt might come in and set the __LINK_STATE_RX_SCHED
bit before we get to clear that bit again. If that happens,
the next call to the ->poll() function will oops.

Signed-off-by: Arnd Bergmann <arnd.bergmann at de.ibm.com>
---
This was found during testing with the fedora kernel,
with all patches from netdev-2.6.git applied.

It may not be the right fix, but this is currently the
only way I can get that kernel to boot.

One part I don't understand at the moment is that Christian
Krafft reported the same problem with tg3, but that driver
has all interrupts disabled at the device while calling
the request_irq() function, which seems to be the best
solution for avoiding the bug in the first place.

Index: linux-2.6/drivers/net/spider_net.c
===================================================================
--- linux-2.6.orig/drivers/net/spider_net.c
+++ linux-2.6/drivers/net/spider_net.c
@@ -1997,6 +1997,7 @@ spider_net_open(struct net_device *netde
 	/* further enhancement: setup hw vlan, if needed */
 
 	result = -EBUSY;
+	netif_poll_enable(netdev);
 	if (request_irq(netdev->irq, spider_net_interrupt,
 			     IRQF_SHARED, netdev->name, netdev))
 		goto register_int_failed;
@@ -2005,11 +2006,11 @@ spider_net_open(struct net_device *netde
 
 	netif_start_queue(netdev);
 	netif_carrier_on(netdev);
-	netif_poll_enable(netdev);
 
 	return 0;
 
 register_int_failed:
+	netif_poll_disable(netdev);
 	spider_net_free_rx_chain_contents(card);
 alloc_skbs_failed:
 	spider_net_free_chain(card, &card->rx_chain);