From owner-freebsd-hackers Mon Apr 28 12:35:32 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA03478 for hackers-outgoing; Mon, 28 Apr 1997 12:35:32 -0700 (PDT) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA03469 for ; Mon, 28 Apr 1997 12:35:12 -0700 (PDT) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id PAA02221; Mon, 28 Apr 1997 15:40:27 -0400 (EDT) Message-Id: <3.0.32.19970428153420.00baa100@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Mon, 28 Apr 1997 15:34:24 -0400 To: John Hay From: dennis Subject: Re: pci probes with multiple "units" (MORE) Cc: hackers@FreeBSD.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk At 08:36 PM 4/28/97 +0200, John Hay wrote: >> > >> >Why don't you call your cards ethc0..X and then the devices eth0...X? >> >Something like the Arnet (ar(4)), SDL RISCom N2(pci) or Cronyx-Sigma >> >(cx(4)) drivers? >> We *could*, but its just a matter of symantics and there shouldnt be different >> methods for different cards when it could easily be done if a few folks >> get together on it. The separation of "controller" and "device" is getting >> a bit fogged because so many are making up their own standards. >> Its *easy* to do without the kludge you suggest...I call it a kludge because, >> unlike SCSI, there is no controller (as your nomenclature suggests) just >> multiple *devices* on a card. Since there is little or no distinction between >> 2 single port cards and a dual card, I would not only consider calling it >> a controller confusing but also explicitly wrong. To suggest that it be >> done that way because of a simple symantic problem in the pci probe >> code seems extreme. >> > >Well it depends on how you look at it. The Arnet and SDL cards at least >do share the IO ports and MEM mapped area and the interrupt between the >different serial ports on the card. Also there is two serial ports in >the Hitachi chip and some of those io ports are common. :-) Yes, but you're talking about particulars internal to the driver, which should be transparent to the user. Perhaps you can make a case that all hardware probe-able devices be considered "controllers" and all network interfaces logical entities, but that is not how its modelled now.... Dennis >John >-- >John Hay -- John.Hay@mikom.csir.co.za > >