This is the mail archive of the
mailing list for the GCC project.
Re: Success report on Linux/PPC, small Ada problem
- From: dewar at gnat dot com (Robert Dewar)
- To: mrs at windriver dot com, rth at redhat dot com
- Cc: dewar at gnat dot com, fw at deneb dot enyo dot de, gcc at gcc dot gnu dot org,jsm28 at cam dot ac dot uk, minyard at acm dot org
- Date: Tue, 5 Feb 2002 22:36:44 -0500 (EST)
- Subject: Re: Success report on Linux/PPC, small Ada problem
<<And so I stand by my assertion: Ada will continue to languish
in the FSF repository until such time as the folks working on
it believe that the FSF repository is the master.
Well I see always some distance between the trees in the following sense.
The ACT tree has to be very conservative in accepting patches, and we always
ensure that the ACT tree will pass all the internal regression tests (the
test suite that is proprietary and cannot be made public). We don't want
to restrict updates to the FSF repository in the same way. For updates
to the ACT tree, we insist that the full regression suite be run before
any change, no matter how minor, is made, because we need to ensure that
on a day to day basis, builds from our tree are suitable for providing to
supported customers in terms of reliability.
The changes to the FSF repository cannot possibly meet this criterion, and
in any case I don't think they should have to. I actually see the result
that the internal ACT tree will eventually lag behind the most recent
changes in the FSF tree, and I don't think that's a big problem, because
obviously we will want to acquire changes and keep the trees up to date.
<<Not really. When you believe you're working out of the master
repository, you have no particular incentive to perform merges
against a secondary repository. Merges may happen, but they will
not be regular without extreme discipline.
We are used to working in an environment of extreme discipline :-)