RANLIB_TEST in gcc/Makefile?

[DJ Delorie]

> The only reason I can think of is that someone might want to run make
> RANLIB=/path/to/ranlib and except RANLIB_TEST to reflect the change.
> But I wouldn't worry too much about it.  Presumably, we should have
> RANLIB looked for are SUBSTed in the Makefile by configure, just like
> the rest of the world does.
> 
> DJ (and anybody else), does this match your opinion?

The only thing I can think of is to use a just-built ranlib, if
available, but I can't see why ranlib is special.