svn commit: r568012 - head/net/tightvnc

[Mathieu Arnold]

On Wed, Mar 17, 2021 at 09:51:00AM +0100, Gerald Pfeifer wrote:
> On Mon, 15 Mar 2021, Mathieu Arnold wrote:
> >> Do you intentionally focus on USE_GCC=any?  On riscv64 this is always
> >> equivalent to USE_GCC=yes, and even on powerpc - do we want to use the
> >> the age old GCC 4.2 in base *and*, more importantly, use different
> >> compilers on 11.x and 13.x?
> > When I saw that commit, I wondered if USE_GCC=any should not be 
> > retired, it probably serves no purpose nowadays.
> 
> Yes, please.
> 
> I'd be happy to get rid of it (and note that I had not added or
> promoted it, though I understand the original rationale).
> 
> How do we go about that?

Well, the easiest is to change gcc.mk to make USE_GCC=any work like
USE_GCC=yes, and add a bit to sanity.mk saying `any` use is deprecated
and should be using `yes` (and when all ports have been converted, or
only a few remain, move them to yes and garbage collecte the any->yes
and make it invalid.

The less easy way, but probably better, is to make it invalid directly
and change the 50 something ports using it to USE_GCC=yes (or nothing,
or using USE_CSTD or whatever else works.)

First step could probably be to make a patch for `any` mean `yes`,
build stuff and look at the size of the explosion from afar.

-- 
Mathieu Arnold
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 963 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-ports-head/attachments/20210317/89b566b6/attachment.sig>