Re: [PATCH 1b/7] dlm: core locking

[Daniel Phillips <phillips@istop.com>]

On Monday 25 April 2005 18:27, Nikita Danilov wrote:
>  > >  > +
>  > >  > +static int is_remote(struct dlm_rsb *r)
>  > >  > +{
>  > >  > + DLM_ASSERT(r->res_nodeid >= 0, dlm_print_rsb(r););
>  > >  > + return r->res_nodeid ? TRUE : FALSE;
>  > >  > +}
>  > >
>  > > This can be simply
>  > >
>  > >       return r->res_nodeid;
>  >
>  > Not quite the same.  Perhaps you meant:
>  >
>  >         return !!r->res_nodeid;
>
> Strictly speaking yes (assuming TRUE is defined as 1), but name
> is_remote() implies usages like
>
>          if (is_remote(r)) {
>                  do_something();
>          }
>
> in such contexts !! is not necessary.

Any objection to making it inline and let the compiler delete the redundant 
code?  The princple is: it's better to spell out "!!" when that's intended, 
rather than build in a nasty surprise for later.  The inline code will be 
smaller than a function call anyway.

Regards,

Daniel