Re: [PATCH v4 14/18] static_call: Add static_cond_call()

[Rasmus Villemoes]

On 01/05/2020 22.29, Peter Zijlstra wrote:
> Extend the static_call infrastructure to optimize the following common
> pattern:
> 
> 	if (func_ptr)
> 		func_ptr(args...)
> 
> +
>  #define static_call(name)	__static_call(name)
> +#define static_cond_call(name)	(void)__static_call(name)
>  
> +
>  #define static_call(name)	__static_call(name)
> +#define static_cond_call(name)	(void)__static_call(name)
>  

> +#define static_cond_call(name)						\
> +	if (STATIC_CALL_KEY(name).func)					\
> +		((typeof(STATIC_CALL_TRAMP(name))*)(STATIC_CALL_KEY(name).func))
> +

This addresses neither the READ_ONCE issue nor the fact that, AFAICT,
the semantics of

  static_cond_call(foo)(i++)

will depend on CONFIG_HAVE_STATIC_CALL. Also, I'd have appreciated being
cc'ed on new revisions instead of stumbling on it by chance.

Rasmus