Date: Wed, 04 Oct 2000 01:35:21 -0500 From: "Jeffrey J. Mountin" <jeff-ml@mountin.net> To: "Daniel C. Sobral" <dcs@newsguy.com>, Kris Kennaway <kris@FreeBSD.ORG> Cc: Warner Losh <imp@village.org>, Paul Richards <paul@originative.co.uk>, cvs-committers@FreeBSD.ORG, freebsd-security@FreeBSD.ORG Subject: Re: cvs commit: src/usr.bin/finger finger.c Message-ID: <4.3.2.20001004010120.00b1cb50@207.227.119.2> In-Reply-To: <39DAC368.C6C213B7@newsguy.com> References: <39DA182C.C70ED553@originative.co.uk> <39D98B55.126DAFC4@originative.co.uk> <200010022227.PAA62603@freefall.freebsd.org> <39D92E08.E00CF2E4@owp.csus.edu> <20001002180303.A40584@freefall.freebsd.org> <39D98B55.126DAFC4@originative.co.uk> <200010031530.JAA26493@harmony.village.org> <20001003124008.A4892@netmonger.net> <39DA182C.C70ED553@originative.co.uk> <200010031800.MAA27859@harmony.village.org> <20001003162720.D51546@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
At 02:43 PM 10/4/00 +0900, Daniel C. Sobral wrote: >Kris Kennaway wrote: > > > > I think a formal MFC process may be too stifling, unless we have a > > VERY responsive MFC team. Consider that we don't want the same thing > > to happen as did with 3.x, where 4.0-CURRENT was allowed to diverge so > > much that merging bugfixes became difficult. > >I don't think the comparision is appropriate. The divergence between 3.x >and 4.x came at the very beginning of 3.x's life, and it was not merged >back because it was too big a change. Aren't the SMP changes in -current similar and will reduce the amount of code that can be backported to -stable with relative ease. "Allowed to diverge" smacks of halting progress and some changes are painful. I'm not one to get into detail or give examples, which would be better suited on -arch and -hackers, but will say that divergence in userland is more likely to slow/stunt/make a PITA out of MFC'ing than kernel changes. Don't think the pain of change can be avoided, but in general some do put more thought into changes than others and take the time to consider if -stable can benefit. Simply are the gains worth the cost. Can't really come up with a blanket to wrap the ideas here simply. Can say that after the SMP changes in -current the ratio of -current to -stable commits dropped. In and of itself that doesn't say much. There are quite a few commits to -stable (even since 4.1.1) and even 3.x and lest we forget that -current is the development branch. Some things just have to wait. I recall waiting for 3.x and SMP, many improvements again in 4.x, and am waiting once more (not sure for what just yet, but I'm waaay behind with my -current mail. Hardly expect more then some of the bigger changes (if any) will make it back from -current to -stable. Almost seems like to solve the divergence issue we must merge -current and -stable, making the latter a well tested snapshot of the former. Not one of them, but there were cases for a production environment where -current was needed. Prior to 4.0R NFS was probably the best example. It is also quite possible that the divergence and Jordan's observation that it's hard to get -current developers to MFC, that with the fewer changes filtering back that more time was spent working on the "new and improved" to make 4.0 such an excellent release. Just may not be possible to have our cake and eat it too. Personally I think endless discussions that go nowhere have been hurting the amount of comitting. Recently the amount of commits seems to have slowed quite a bit. This may be a good thing if the developers are wading through their mail. Perhaps those of us that don't code should be a bit more quiet. This thread is an exception, but -security has gone to pot and yes this is CC'd to -security, but no longer belongs there. Nor does it's spin-off, which I plan to wade through in the hope of a glimmer. <sigh> Jeff Mountin - jeff@mountin.net Systems/Network Administrator FreeBSD - the power to serve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-security" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.3.2.20001004010120.00b1cb50>