Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!panix!newsfeed.internetmci.com!raven.eva.net!bighorn.accessnv.com!jca From: jca@bighorn.accessnv.com (J.C. Archambeau) Newsgroups: comp.unix.bsd.freebsd.misc Subject: Re: 3Com Card Config Date: 19 Nov 1996 19:51:42 GMT Organization: Access Nevada Inc. Lines: 22 Message-ID: <56t34e$3ha@raven.eva.net> References: <56rl1n$20m@que.iphase.com> NNTP-Posting-Host: bighorn.accessnv.com X-Newsreader: TIN [version 1.2 PL2] Mark Veteikis (mark@iphase.com) wrote: : While configuring a 3Com 3C509 10BaseT card I found the following. : I used a custom kernel with NO IO Port or IRQ conflicts and found : that this card (OS?) behaves badly when the IRQ is set to 9. : Looking at sniffer traces during simple ping tests, it looks like using IRQ 9 : does not seem to generate receive interrupts consistently. It appears that : configuring this card at 9 might be a FreeBSD driver/OS problem, a hw bug, : or some old IRQ2/9 foolishness?? I think it's the IRQ 2/9 foolishness which for all practical purposes is a hardware bug (design flaw) of the ISA bus. I had the same problem with devices under Linux. They'd work fine until they get put onto IRQ 2/9. I also found a really interesting bug with Archive VP-402 tape controllers. They won't work at IRQ 2/9 at all (at least with my motherboard and particular card) under any OS including Archive's QIC-Stream backup software under MS-DOS. -- /* ** Internet: jca@accessnv.com | Don't blame me, I didn't vote for Clinton. ** jca@anv.net */