Pre-Patch RFC: proposed changes to option-lookup

[Steven Bosscher]

On Wed, Oct 30, 2013 at 9:39 PM, David Malcolm wrote:

> I want to eliminate hidden use of the preprocessor in our code, in favor
> of using block caps to signal to people reading the code that macro
> magic is happening.

Good idea. In the past this kind of change would be sort-of
controversial (large number of changes for small gain) but if the
whole source code base is being overhauled anyway, it's a good time to
get rid of this black magic at the same time...


> My idea is to introduce a GCC_OPTION macro, and replace the above with:
>
>   static bool
>   gate_vrp (void)
>   {
>     return GCC_OPTION (flag_tree_vrp) != 0;
>   }
>
> thus signaling to humans that macros are present.
>
> Is such a patch likely to be accepted?   Should I try to break the
> options up into logical groups e.g. with separate macros for warnings vs
> optimizations, or some other scheme?

Why not just expose the different "classes" of options in some struct or object?

struct {
  unsigned tree_vrp : 1;
  ...
} opt_flags;

struct {
  ...
} warn_flags;

static bool gate_vrp (void) { return opt_flags.tree_vrp != 0; }

This would also make streaming the flags settings easier for LTO.

Ciao!
Steven