Re: [RFC PATCH] livepatch/module: Do not patch modules that are not ready

[Josh Poimboeuf <jpoimboe@redhat.com>]

On Tue, Mar 03, 2015 at 01:31:28PM -0600, Josh Poimboeuf wrote:
> @@ -89,16 +89,29 @@ static bool klp_is_object_loaded(struct klp_object *obj)
>  /* sets obj->mod if object is not vmlinux and module is found */
>  static void klp_find_object_module(struct klp_object *obj)
>  {
> -	if (!klp_is_module(obj))
> +	struct module *mod;
> +
> +	if (!klp_is_module(obj) || obj->mod)
>  		return;
>  
>  	mutex_lock(&module_mutex);
> +
>  	/*
>  	 * We don't need to take a reference on the module here because we have
>  	 * the klp_mutex, which is also taken by the module notifier.  This
>  	 * prevents any module from unloading until we release the klp_mutex.
>  	 */
> -	obj->mod = find_module(obj->name);
> +	mod = find_module(obj->name);
> +
> +	/*
> +	 * MODULE_STATE_COMING means we got to the module first before the
> +	 * notifier did.  ftrace is already initialized, so it's fine to go
> +	 * ahead and start using it.
> +	 */

This comment should probably be improved to say:

"MODULE_STATE_COMING and !obj->mod means..."

--
Josh