USE_GCC=any -> USE_GCC=yes (was: svn commit: r504198 - head/Mk)

[Alexey Dokuchaev]

On Thu, Jul 04, 2019 at 11:16:53PM +0200, Gerald Pfeifer wrote:
> On Thu, 4 Jul 2019, Mark Linimon wrote:
> > But ... what I understand is that "any" means "it's ok to use base
> > gcc".  That was always the intended difference to "yes" which means
> > "you *must* use a recent gcc" [e.g. from ports].
> > 
> > I don't really see why we need to add the dependency on a newer gcc
> > if the version in base suffices.  What am I missing here?
> 
> I think we agree that nearly all FreeBSD users and developers are on
> amd64 (or i386).

Well... :-)

> When a port maintainer or committer sets USE_GCC=any she and (nearly)
> everyone else therefore uses GCC 8 as of today.
> 
> When that port is then built on powerpc64, USE_GCC=any implies the age
> old GCC 4.2 in base, and thus the build is attempted with a compiler
> very, very different from what the original maintainer/committer/testers
> and users had.

I'd have to agree with Mark here: the literal reading of "any" means
"it's OK to use base GCC" if it's available.  While I understand that
it's aged and on x86 it's the same as USE_GCC=yes, it's still a bad
idea to change the meaning of words.

./danfe