Date: Wed, 11 Jul 2001 11:55:14 -0700 (PDT) From: Matthew Jacob <mjacob@feral.com> To: Joerg Wunsch <joerg_wunsch@uriah.heep.sax.de> Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: Trouble with SCSI Streamer Message-ID: <Pine.BSF.4.21.0107111150300.7306-100000@beppo> In-Reply-To: <20010711204434.J1948@uriah.heep.sax.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 11 Jul 2001, Joerg Wunsch wrote: > As Matthew Jacob wrote: > > > > We've been there before, Matt. /All/ current tape drives have to > > > > Yes, and I didn't agree with some of the conclusions. Yes, all > > tape drives 'emulate'. The QIC manufacturers, from my experience, > > don't do a good job of it. > > Which is contradictionary to my experience (and i use them a lot). > > Can you elaborate a bit why you think the QIC manufacturer do a bad > job on it? Btw., there's currently probably only a single one left > (Tandberg), and their SCSI implementations usually belong to the Yes. Wangtek was the worst. The variable mode for the 51000H was awful and just plain wrong. > better ones. (Let's leave out stuff like Travan here.) The HP-Travan SCSI (T20) has been okay for me. But bad for others. > > > Let's cut this short. If you're willing to take over SA maintenance, > > please say so. > > Well no, i don't wanna piss you off, nor do i have more time at hand No, no, it's just at least on this topic I can't seem to please anyone. > than you. I'll be happy to eventually finish some long-standing > floppy driver work... Some of the things there have been in my brain > pipeline for years now. > > I once thought about re-implementing `programmable' tape devices, like > they did exist pre-CAM, and are still reported as `available modes' in > the mt stat output. This would allow people to pre-assign variable > length for, say /dev/sa0.0, and fixed-length (for QIC-150 media) to > /dev/sa0.1, and then use the appropriate subdevice. Actually, I think these are a bit of a crock- for every Nth option you'd want to set, you'd need a different matrix name. What you need to do instead is to establish persistent settings for specific drives that re-establish after reboots and even if devices change address. This is likely part of the 'hints/quirks' stuff that the loader can than bring in. My major problem with implementing this is that I really don't have as yet a clever idea for how to make the quirk syntax look good and easy to use and sufficiently broad to get what we want. -matt 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?Pine.BSF.4.21.0107111150300.7306-100000>