Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!convex!convex!cs.utexas.edu!zaphod.mps.ohio-state.edu!uwm.edu!spool.mu.edu!agate!agate.berkeley.edu!cgd
From: cgd@erewhon.CS.Berkeley.EDU (Chris G. Demetriou)
Newsgroups: comp.os.386bsd.bugs
Subject: Re: The patchkit (was Re: Excessive Interrupt Latencies)
Date: 27 Mar 93 12:17:02
Organization: Kernel Hackers 'r' Us
Lines: 80
Message-ID: <CGD.93Mar27121702@erewhon.CS.Berkeley.EDU>
References: <GENE.93Mar15115756@stark.stark.uucp> <1993Mar16.093636.29923@gmd.de>
<g89r4222.732307558@kudu> <C409MC.n1D@agora.rain.com>
<1obts0$doq@hal.gnu.ai.mit.edu> <C4H1M1.3L5@sleeper.apana.org.au>
NNTP-Posting-Host: erewhon.cs.berkeley.edu
In-reply-to: raz@sleeper.apana.org.au's message of Fri, 26 Mar 1993 01:03:36 GMT
I OWN TOO MANY HATS, BUT NONE OF THEM KEEP MY HEAD WARM!
In article <C4H1M1.3L5@sleeper.apana.org.au> raz@sleeper.apana.org.au (Roland Turner) writes:
>Re: comp.os.386bsd.patches
>
>Given the circus that surrounded creation of comp.os.386bsd.*, I am
>not optimistic about a new group (although even if I was, we are looking
>at approx 2 months), how about comp.os.386bsd.announce? Not strictly in
>its charter I know, but it IS moderated and a patch release IS a "launch"
>of general interest to the 386bsd community at large.
(wearing my group-creator's hat...)
umm, yah, a new group would take at least 2 months, even given
that all went smoothly. and there wasn't too much weirdness when
somebody (me... 8-) finally settled down to DOING IT.
>>Obviously, this is a fair amount of work, and I can't do it all by
>>myself. Anyone wishing to help out (or even just say `Go for it!') is
>>welcome to send me mail.
>
>Go for it! (Chris Demetriou (sp?): care to comment?)
(wearing my moderator's hat...)
GO FOR IT... but wait 2 weeks. the world as we know it
will have changed by then -- (at least one) interim release should
have hit the fan by then, so the "base" will have moved...
also, i *do* consider .announce a good place to post patches, or
announcements of patches. (WELL TESTED patches, that is. or calls-for-
testers of major patches/packages.)
i'd like to see .patches, but a need has to exist first.
as it is, i'd like to see more stuff on .announce...
>Silly question time: Why isn't cgd's com driver distributed as a patch?
>(Is it?)
(wearing my author's hat...)
because it's a complete and total piece of CRAP.
the driver that it's a hack on was a piece of crap, too.
bruce evans' driver is wonderful, apparently, and i'd like
to get my thoughts integraged into it, when i have the chance...
however, i don't currently have the chance.
>>c) If the patchkit mechanism weren't *braindead*, I wouldn't have to
>>back anything out.
>
>If you apply a bad patch - you have to back it out one day :-)
if you apply a bad patch, then you can simply "back it out" by
applying _another_ patch which does the equivalent of
removing the first.
while it's ugly, it does guarantee a monotonically-increasing set
of patch numbers... 8-)
things to look forward soon, from various fronts:
a list of lots of bugs for you folks to bang on...
some sort of interim release. (if it's one that i do, it'll
definitely have "weekly source patches" and probably
monthly binary sets...)
<sigh>
chris
who's like to crawl into a big hat, and sleep... 8-)
--
Chris G. Demetriou cgd@cs.berkeley.edu
MENTALLY CONTAMINATED and proud of it!