Date: Sat, 01 Jun 1996 10:49:28 -0700 From: "Jordan K. Hubbard" <jkh@time.cdrom.com> To: "Matthew N. Dodd" <winter@jurai.net> Cc: freebsd-stable@freebsd.org Subject: Re: stable breakage Message-ID: <17980.833651368@time.cdrom.com> In-Reply-To: Your message of "Sat, 01 Jun 1996 12:39:42 CDT." <Pine.BSI.3.93.960601123634.28731B-100000@sasami>
next in thread | previous in thread | raw e-mail | index | archive | help
> Considering that -stable is compiling right now... I'd say it was that > easy. Have YOU got stable to compile all the way through? > > Tell you what, when this one finishes, I'll re-sup, make that one change > and remake world again. I think you're somehow missing the point. We're trying to merge the tree here, and if something is in 2.2 then we want to make it look as much like that in 2.1 so that the *second* pass I do doesn't show up the same diffs again. NOW, if it turns out that something is broken about 2.2 then that needs to be pointed out and fixed there, then merged. If something is simply broken because it was merged insufficiently, then the additional merging needs to be done (the number of mutual dependencies in our tree is now somewhat insane, I'm quickly learning). As a matter of fact, I'm now thinking about merging libc wholesale and stop trying to dance around the thread changes. The thread changes appear to be benign, by putting them into 2.1 we'd also be giving people a chance to work with them (since so many hack on the latest releases and do NOT run -current, for one reason or another [usually network lossage]). Anyway, I've got several boxes now coming on line to start regression testing in earnest, and I'm confident that I'll be able to get the tree back into shape (along with additional merging - might as well merge and fix in one pass) by late tonite.. Jordan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?17980.833651368>