Re: [RESEND][RFC PATCH v2] waitfd

[Ingo Molnar <mingo@elte.hu>]

* Casey Dahlin <cdahlin@redhat.com> wrote:

> >> +asmlinkage long sys_waitfd(int which, pid_t upid, int options, int unused)
> >> +{
> >> +	int ufd;
> >> +	struct waitfd_ctx *ctx;
> >> +
> >> +	/* Just to make sure we don't end up with a sys_waitfd4 */
> >> +	(void)unused;
> > 
> > looks a bit silly ...
> 
> 
> Do you mean the principle of having an unused argument around for future 
> use or the cast to void? The cast to void is there to suppress the 
> "Waning: unused argument" messages and make gcc happy.

gcc will not warn about unused function arguments - only about unused 
local variables. The 'unused' argument should either be removed 
altogether, or replaced with a properly named parameter and a check 
returning -ENOSYS if the argument is not zero (or something like that). 

(It's generally better to keep such syscalls extensible via such trivial 
means, if there's a remote chance for ever needing to extend that 
syscall.)

	Ingo