Date: Mon, 06 May 2002 23:37:35 -0600 (MDT) From: "M. Warner Losh" <imp@village.org> To: grog@FreeBSD.org Cc: brooks@one-eyed-alien.net, chris@masto.com, jhay@icomtek.csir.co.za, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/share/man/man4 wi.4 Message-ID: <20020506.233735.118306874.imp@village.org> In-Reply-To: <20020507145834.P63106@wantadilla.lemis.com> References: <20020506.211945.111478471.imp@village.org> <20020506222616.B1977@Odin.AC.HMC.Edu> <20020507145834.P63106@wantadilla.lemis.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <20020507145834.P63106@wantadilla.lemis.com>
"Greg 'groggy' Lehey" <grog@FreeBSD.org> writes:
: On Monday, 6 May 2002 at 22:26:17 -0700, Brooks Davis wrote:
: > On Mon, May 06, 2002 at 09:19:45PM -0600, M. Warner Losh wrote:
: >> In message: <20020507115643.N75198@wantadilla.lemis.com>
: >>> Interesting question. You only need to create an IBSS once per IBSS
: >>> network. I now have definitive proof that the station which creates
: >>> the IBSS doesn't do very much: I've taken that station out of the net,
: >>> and the other two machines can still talk to each other. But I can
: >>> see issues when more than one station creates the IBSS: the net could
: >>> partition itself into two different IBSSs, so I suspect we should keep
: >>> the distinction, though the term "master" seems less appropriate.
: >>
: >> Yes, but you need to keep recreating it from time to time, as these
: >> things time out. "master" will be there until you convince OpenBSD to
: >> change, so quit harping on that.
: >
: > My intented implementation is that adhoc will mean ibss-master in
: > 5.0.
:
: I think this would be *really* confusing. I can see people setting up
: ad-hoc interfaces on different releases with identical commands, and
: wondering why they can't interoperate. If we change adhoc, I think we
: should get rid of it altogether, not change the meaning.
You just totally contradicted what you said in email yesterday where
you approved of this idea. The reason that we want to harmonize what
ad-hoc means in current is that it means different things to different
drivers in -stable right now. For lucent cards, that have the old
firmware, and many prism2 cards it means "demo ad-hoc mode". For
lucent cards with new firmware and some prism and symbol cards, it
means "ibss mode". And for all cisco cards (supported by the an
driver), it means "ibss mode". It is a horrible confusion of rats
nets of differing meanings, none of which seem to make sense (and I
may have gotten at least one of them wrong). Having it be an alias
for "ibss-master" in current is the only sane thing to do.
: > If there are devices out there that are so totally broken that they
: > screw things up when they are allowed to create an IBSS and someone
: > else has done it first, we'll deal with the problem then.
:
: I don't think that's the issue. I've been using IBSS for some time,
: and once I knew what I was doing, I had no problems.
You must not be using symbol cards then, because it is an issue with
them. The wi driver in stable does support symbol cards in ibss mode,
but doesn't support them creating an ibss network when they need to.
Warner
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020506.233735.118306874.imp>