Date: Thu, 06 Jun 1996 14:25:59 -0700 From: Jim Binkley <jrb@cs.pdx.edu> To: hosokawa@mt.cs.keio.ac.jp (HOSOKAWA Tatsumi) Cc: matt@lkg.dec.com, mobile@FreeBSD.org Subject: Re: Wavelan PCMCIA Message-ID: <199606062126.OAA25431@sirius.cs.pdx.edu> In-Reply-To: Your message of "Fri, 07 Jun 1996 02:38:05 %2B0900." <199606061738.CAA19844@frig.mt.cs.keio.ac.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
Hosokawa Tatsumi, I suppose this is your way of telling me that you ported the wavelan driver into the pccard setup. Thank you. Given that we have about 12 laptops and N more "base stations" using the current wavelan drivers + a fair number of kernel hacks, we are still mired in 2.1R. However over the summer, we want to upgrade our kernel environment (at least) to speak pccard. There are a number of pccards that we would like to be able to use at the same time as wavelan cards, basically gps and/or various crypto cards (a des card would be nice). If there are cards along those lines (crypto/gps) that anyone has used, I would appreciate hearing about it. We are still using my non-pccard wavelan drivers (isa/pcmcia). (The pcmcia driver is basically like David Greenman's if_ze.c driver). I have put a new version out of both isa/pcmcia drivers available at zymurgy.cs.pdx.edu via anon ftp in the mobility directory. ftp://zymurgy.cs.pdx.edu/pub/mobility/wavelan.freebsd.tar.gz Changes include the following: .support for multicast (minimal testing, but it works with nv) in both drivers .fixed a bug in how promiscious mode was done on the pcmcia driver, so you can get into/out of tcpdump faster .put a workaround in for a hang problem that shows up on reception of frames (apparently due to "overruns") ONLY with 2.0 cards. This does not happen on 1.0 and 1.1 wavelan cards. It doesn't happen on all 2.0 cards either, but it does happen on some. If the overrun occurs, a hw reset is done, and an input error is logged. I hope we can get up to sync with the pccard later this summer. Re dec stuff. We are doing mobile-ip (and in fact it is beginning to work), so we aren't interested in "roaming" (basically link layer access point switches) per se. The way att seems to do that, it involves having a different NWID per access point (bridge). The unix drivers won't understand that of course. I understand from Kannan A., formerly at DEC, that DEC has a form of mobile-ip. Are your access points running that Matt? regards, Jim Binkley jrb@cs.pdx.edu Portland State CS dept. Your message <199606061738.CAA19844@frig.mt.cs.keio.ac.jp>: >In article <199606061247.MAA16354@whydos.lkg.dec.com> >matt@lkg.dec.com writes: > >>> I'm using the 960602 version of pccard-test with a Digital RoamAbout/DS. >>> Happily I can say that it seems to working fine after I added >>> >>> card "Digital" "RoamAbout/DS" # version 2.00 >>> config default "wlp0" any >>> ether wavelan >>> >>> to /etc/pccard.conf. > >Wow! > >Pls tell me the detailed configuration of your card (NetID, MAC addr, >etc.) and your environment (how many cars? are you using the >station?). > >I've not succeeded to use this card under our package yet. > >-- > HOSOKAWA, Tatsumi E-mail: hosokawa@mt.cs.keio.ac.jp > WWW homepage: http://www.mt.cs.keio.ac.jp/person/hosokawa.html > Department of Computer Science, Keio University, Yokohama, Japan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199606062126.OAA25431>