Date: Mon, 21 Jul 1997 16:41:09 +0200 (CEST) From: Mikael Karpberg <karpen@ocean.campus.luth.se> To: kuku@gilberto.physik.rwth-aachen.de (Christoph Kukulies) Cc: current@FreeBSD.ORG Subject: Re: I am contemplating the following change... Message-ID: <199707211441.QAA09641@ocean.campus.luth.se> In-Reply-To: <19970721154922.61073@gil.physik.rwth-aachen.de> from Christoph Kukulies at "Jul 21, 97 03:49:22 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
According to Christoph Kukulies: [...] > 280/5 was one of the most common setting I was using in the past > for the WD8003 8 bit cards. I adopted this setting also > for those 16 bit cards (WD8013) the latter requiring to be programmed > for this value pair. > > Another common setting was 300/10 for the 16 bit cards which is a jumper > selectable option on the WD8013 (phased out) and SMC Ultra 8216 cheapo > cards of these days. > > I vote for keeping 280/5 because this doesn't rule out 8 bit cards. [...] > > 10, every time. Or 11 if you really want to, but not 5 and > > certainly not 3, for the very good reasons already pointed out. I > > would guess that anyone who still uses an 8-bit network card has been > > long enough at this game to know how to tweak the default value. > > > > Dermot Ok... This just might seem like a silly question, but Jordan wanted to take away the double devices. Why? ed0 == 300 / 10 ed1 == 280 / 5 This seems to be a really great solution, since it satisfies both sides with a good default. It will make installing easier for everyone. Now what could possibly be good about removing the double entry? Ofcourse it should be completely dynamic with having no ed* if there are no cards, and have one for each detected card if there are any. But before ISA probing code is fixed so it can test and autodetect where things are, like it should, why not have a double entry? /Mikael
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199707211441.QAA09641>