Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 24 Feb 1998 07:01:10 -0500
From:      "Evan Champion" <evanc@synapse.net>
To:        "John Saunders" <john.saunders@scitec.com.au>, <freebsd-stable@FreeBSD.ORG>
Subject:   Re: stable branch funnies
Message-ID:  <007001bd411b$e6673d90$c9252fce@cello.synapse.net>

next in thread | raw e-mail | index | archive | help
>1) -stable is very volatile and whole programs coming and going, and
>   failed builds, is a part of life. (I can't really believe this).

Right now it is slightly volitile because people are moving large amounts of
stable stuff from -current in to -stable in preparation for the imminent
2.2.6 release.

I don't think there have been that many mistakes, and they were quickly
corrected.  jkh and co. have done a pretty good job considering how many
patches have been put in.  In general, however, stable is basically stagnant
change-wise, with maybe a few patches each day to non-critical components.

If you look at what is being patched, almost none of the patches are to
critical components like the kernel or libc.  Most are to bring applications
up to date with -current (from your perspective, upgrading a stock 2.2.5
system, you might see a huge number of patches, but that's because you're 6
months behind -stable :-)

The only thing that -stable consistently seems to break is make -j, however
this has been reported a few times.  The problem is that it seems none of
the commiters use -j so there isn't much incentive for them to make sure
things still work after they commit...  Unless you have a multiprocessor
machine or are compiling over a relatively high-delay channel (ie: NFS)
there isn't any advantage to make -j, and -stable doesn't support SMP anyway
so there goes one of the two reasons to use it :-)

[Of course, I do compile over NFS, so I'd certainly appreciate it if people
stopped breaking it :-)]

>2) Somehow cvsup managed to reverse the sup it did a few days ago
>   and removed things it shouldn't have (evidence boggle going). Or
>   maybe it just went plain bonkers :-)

It was very intentional that boggle went.  See the huge discussion that just
happened in the freebsd-current and cvs-all...  I won't rehash it here as
everyone has had enough of it, but it had to do with trademark infringement.


Evan



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?007001bd411b$e6673d90$c9252fce>