Re: [PATCH] Gigabit Ethernet driver of Topcliff PCH

[Joe Perches]

On Thu, 2010-08-26 at 09:29 -0700, Stephen Hemminger wrote:
> On Thu, 26 Aug 2010 09:16:12 -0700
> Joe Perches <joe@xxxxxxxxxxx> wrote:
> > My message logging order preference:
> > if there's a netif_msg_<type> and a struct net_device
> > 	netif_<level>		(struct private *, netif_msg_<type>,
> > 				 struct net_device *, fmt, args...)
> > if there's a struct net_device
> > 	netdev_<level>		(struct net_device *, fmt, args...)
> > if there's a struct device
> > 	dev_<level>		(struct device *, fmt, args...)
> > otherwise
> > 	pr_<level>		(fmt, args...)
> At this point in process, netdevice isn't registered so it
> doesn't have a ethX name. so dev_info would be best.

Yes, likely true.

Ideally netdev_<level> is used after register_netdev.
The default "eth%d" is printed after alloc_netdev
but before register_netdev.

Before commmit 256df2f3879efdb2e9808bdb1b54b16fbb11fa38
when netdev_<level>s were converted to functions, they
would also dereference a NULL pointer if used before
register_netdev.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/