Date: Tue, 19 Jan 1999 23:21:28 +0100 From: Stefan Esser <se@mi.uni-koeln.de> To: David Kelly <dkelly@hiwaay.net> Cc: freebsd-scsi@FreeBSD.ORG, Stefan Esser <se@FreeBSD.ORG> Subject: Re: Symbios 875 activity LED? Message-ID: <19990119232128.D363@dialup124.mi.uni-koeln.de> In-Reply-To: <199901180014.SAA11049@nospam.hiwaay.net>; from David Kelly on Sun, Jan 17, 1999 at 06:14:11PM -0600 References: <se@mi.uni-koeln.de> <199901180014.SAA11049@nospam.hiwaay.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On 1999-01-17 18:14 -0600, David Kelly <dkelly@hiwaay.net> wrote: > Stefan Esser writes: > > It is simple to positively detect the Tekram SCSI cards, and I > > know those don't follow the SDMS convention (and because of > > Tekram's policy to support free operating systems, we even know > > exactly how to make best use of their cards!). > > That's nice to know. Have heard Tekram cards have formatting and bad > block scanning support in their BIOS config much like Adaptec and > unlike my Asus SC875. Street prices for Tekram 390F's are lower than > I've seen for an SC875. Problem is I'm not in the market for another > SCSI card at the moment. :-( Excuses, excuses, that didn't stop me > from buying an extra Matrox Millenium II 4MB yesterday for $45. I'd love to be able to get them at that price! > While we can positively ID a Tekram card, just how else is one > superior to other Symbios based cards? There is not much of a difference, actually. All NCR cards currently come with active terminators, AFAIK, and there is not much else on the card (besides the actual SCSI chip). I don't think that firmware differences are that important, actually. The Tekram cards ar every high quality, as are the Symbios cards. (I have never used one from any other vendor.) > > But as long as I'm not able to *positively* detect SDMS compatible > > cards, it's just to dangerous to make any assumptions about what > > GPIO<0> does or doesn't ... ;-) > > I felt pretty confident that the #ifdef was there for a good reason. But > the only way I found it was via mention on these mail lists. Once Upon A > Time thought my SCSI card was broken because the LED didn't work. Yes. I understand that you had wanted to find some hint at the option, but in fact, I do not want to take the chance, that there is a card from some vendor I never heard of, that does some very nasty things if GPIO<0> is toggled ... I do not understand, why the LED is not connected to the SCSI bus BUSY signal. (Ok, it takes an external driver, while the GPIO output drivers are able to sink the LED current ....) > Is this option too petty or trivial to be added to LINT and config(8)? It is trivial to make this a supported option. > Config complains about it but kernels compile anyhow. The option is not in /sys/conf/options (nor in /sys/i386/options.i386). Just add it to the end of the other SCSI_NCR_* options in the "options" file, and config will stop to complain. As long as the option is not in that file, config will add a -D<option> to the kernel Makefile. But this will not cause files that depend on this option to be rebuilt, and that is what config warns about ... Regards, STefan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990119232128.D363>