On Fri, Oct 04, 2019 at 02:26:26PM -0600, Jeff Law wrote:
> Same objections as before. As long as we're using macros like this,
> we're going to have increased potential for shadowing problems and
> macros which touch implementation details that just happen to be
> available in the context where the macro is used.
>
> Convert to real functions. It avoids the shadowing problem and avoids
> macros touching/referencing things they shouldn't. Code in macros may
> have been reasonable in the 80s/90s, but we should know better by now.
>
> I'm not ranting against you Bernd, it's more a rant against the original
> coding style for GCC. Your changes just highlight how bad of an idea
> this kind of macro usage really is. We should take the opportunity to
> fix this stuff for real.
To get 95% of the benefit for only 2% of the pain, you can make an inline
function where there was a macro before, and define that same macro to
just call the function.
Like
#define DEFAULT_SOME_MACRO(PARMS) { lots of code }
becomes
#define DEFAULT_SOME_MACRO(PARMS) default_some_macro(PARMS)
static inline int
default_some_macro (int parm, long another)
{
lots of code;
}
The point is that all this is completely *local* changes.
Segher
