Re: [PATCH 20/28] libfcoe: add implementation to support fcoe transport

[Mike Christie <michaelc@cs.wisc.edu>]

On 01/28/2011 06:04 PM, Robert Love wrote:
> +static int fcoe_transport_create(const char *buffer, struct kernel_param *kp)
> +{
> +	int rc = -ENODEV;
> +	struct net_device *netdev = NULL;
> +	struct fcoe_transport *ft = NULL;
> +	enum fip_state fip_mode = (enum fip_state)(long)kp->arg;
> +
> +	if (!mutex_trylock(&ft_mutex))
> +		return restart_syscall();

I was just wondering what is the reason for this. I see it in other 
network related sysfs code but no where else. Is there some networky reason?

> +
> +#ifdef CONFIG_LIBFCOE_MODULE
> +	/*
> +	 * Make sure the module has been initialized, and is not about to be
> +	 * removed.  Module parameter sysfs files are writable before the
> +	 * module_init function is called and after module_exit.
> +	 */
> +	if (THIS_MODULE->state != MODULE_STATE_LIVE)
> +		goto out_nodev;
> +#endif

It seems that this would be a problem for every driver that uses module 
params. Should the module code be getting a module refcount or doing 
this test?