Re: Beyond GCC 3.0: Summing Up

[Phil Edwards <pedwards at disaster dot jaj dot com>]

On Tue, Jul 10, 2001 at 11:44:03AM -0700, Joe Buck wrote:
> > Which processors do you say no longer exist?  It's true that 1750a,
> > clipper, convex, dsp16xx, elxsi, i860, romp, we32k have little or no sign
> > of active work lately and probably are bitrotten, but of which of these
> > processors are there no longer any in working order?  (I think all other
> > CPUs have had at least the odd fix in the past few years, even if lacking
> > any listed maintainer or regular testing.)
> 
> What alternative do you propose?  Not allow any checkin that might break
> the elxsi port?  (Mike Stump said a while back that the last elxsi was
> taken out of service, so it really, really doesn't exist).
> 
> If we want to forbid checkins that cause breakage, we have to distinguish
> somehow between ports that must be kept working and ports that can be
> broken (until some interested party fixes them).

Maybe put up a survey on the webpage saying, "these are the ports which
are not obviously in use, if you use them, tell us before 3.1 or they go
into the museum"?


A few of us were hoping that GCC 3.0 would be something of a "break from
the past" release.  That this would be the last "we support the entire
universe; if your host has a CPU consisting of only twelve NAND gates and
a Bourne shell from the 15th century we will by damn jump through hoops
to keep it working" release.

Originally it started out as a small discussion about dropping the K&R-only
requirement (not too onerous a change).  Then it sort of migrated into a
discussion about requiring a Bourne shell that doesn't suck (big change).
Just that single step -- although a big one -- would help us, although
that's a different topic.

The rationale was that if users wish to use one of those machines, they can
download GCC 3.0, and then use it to build GCC 3.1 (or whatever).  If we
made the Bourne shell change, they'd have to download ash or something,
and we would be ethically obliged to give pointers to ways they can do that.


That was unrelated to what I meant to originally say, sorry.


Phil

-- 
Would I had phrases that are not known, utterances that are strange, in
new language that has not been used, free from repetition, not an utterance
which has grown stale, which men of old have spoken.
                                     - anonymous Egyptian scribe, c.1700 BC