From owner-freebsd-hackers Fri Mar 23 12:48:22 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from crotchety.newsbastards.org (netcop.newsbastards.org [193.162.153.124]) by hub.freebsd.org (Postfix) with ESMTP id 6AA4937B726 for ; Fri, 23 Mar 2001 12:48:00 -0800 (PST) (envelope-from news@news-feed.inet.tele.dk) Received: (from news@localhost) by crotchety.newsbastards.org (8.11.1/8.11.1) id f2NKlwU92669; Fri, 23 Mar 2001 21:47:58 +0100 (CET) (envelope-from news@news-feed.inet.tele.dk) Date: Fri, 23 Mar 2001 21:47:58 +0100 (CET) Message-Id: <200103232047.f2NKlwU92669@crotchety.newsbastards.org> X-Authentication-Warning: crotchety.newsbastards.org: news set sender to news@news-feed.inet.tele.dk using -f References: <200103221833.f2MIXQR22409@prism.flugsvamp.com> <200103231508.f2NF8di91831@crotchety.newsbastards.org> <20010323103842.D82645@prism.flugsvamp.com> In-Reply-To: <20010323103842.D82645@prism.flugsvamp.com> Cc: hackers@FreeBSD.ORG, usenet@tdk.net Reply-To: Tele Danmark Team of Newsbastards To: jlemon@flugsvamp.com From: Olibert Obdachlos Subject: Re: Intel driver doc's Take 2. Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > > restricted by the NDA, does your driver support the newer Intel > > Pro/1000 F cards, which had been mentioned here a couple months > I'm admittedly not familiar with Intels marketing nomenclature. But > if by "1000 F", you mean the 82543/Livingood chip, then yes, this > driver should work on that board. Thanks! That's exactly what I was hoping to know before getting too deep over my head. Consider yourself lucky -- I spent a while navigating the Intel website to see if much had changed. I did dig up the old message I had sent which isn't worth reposting, but may contain pointers to marketroid data and the like. It's from 19.Jan in -hardware, Message-ID: <200101192209.f0JM9pD93464@crotchety.newsbastards.org> or news:200101192209.f0JM9pD93464_crotchety.newsbastards.org@ns.sol.net but it's probably not worth digging out of the archive now. > In fact, the driver was designed > around the 82543, and then backported to 82542 (wiseman). I haven't had my sweaty hands over one of these cards, but I'll assume the 82543GC discussed in these messages is equivalent. I'll also try to get a close look at a card to know what is in use. > > the impossible. Or consider this a possible problem report on the > > driver with the Pro 1000 F... > > Hmm. Send me details, the more information, the better. I'll wait a second before passing on details I don't have at hand, because I can't say I trust the test machine it was placed in, which spontaneously rebooted on me at reboot time, and then wedged up at a following reboot. I think it needs a high-speed lesson through the machine room to meet Mister Floor. Also, I don't trust the configuration that dmesg reveals at boot -- I got stellar performance from a similar machine when it assigned IRQs like 20 and 21 to the fxp ethernet card, while on a different HP machine, I had hell trying to get both an ethernet card and a pile of sound cards to work, when IRQs were given similar to what this test machine wants to give, and I needed to do some juggling of cards to get anything to happen. Everything worked great when I got the high IRQs though. So it's quite possible that this machine needs to have the gigabit ethernet card swapped around -- it came up at irq5. This card is also installed in another machine currently doing production service and was given irq21 or something, so there is definitely some weird BIOS difference between the two machines. I hate PC hardware. This will probably wait til monday, and I'll also ask our techie who posted some details last time if he's seeing exactly the same thing now. I made some tests I didn't try last time, but before I make any claims, I want to eliminate the possibility of flaky hardware at this end and gather relevant details. Thanks, I'll get back to you about this... barry bouwsma, resident TDC internet netmangler To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message