[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [MIT-Scheme-devel] releases
|
From: |
Taylor R Campbell |
|
Subject: |
Re: [MIT-Scheme-devel] releases |
|
Date: |
Wed, 12 Jun 2013 15:10:04 +0000 |
|
User-agent: |
IMAIL/1.21; Edwin/3.116; MIT-Scheme/9.1.1 |
Date: Wed, 12 Jun 2013 07:55:51 -0700
From: Matt Birkholz <address@hidden>
> If so, let's please revert them after 9.2, not before, so that we can
> build 9.2 from 9.1.1.
The patches train a 9.1 host to compile 9.2, but the resulting 9.2ish
system doesn't need them. They can be removed before 9.2ish compiles
the release thus removing all remnants of (ucode-type sequence-2) and
the training required to grok the new (ucode-type sequence).
What's the harm in keeping the patches in? I would like to observe a
policy that release X must be able to build release X+1. If for some
reason we badly want to do something that breaks release X's ability
to build the current development version, we can always just first
make release X+1 and then break it.