Date: Tue, 9 Apr 1996 12:42:34 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: joerg_wunsch@uriah.heep.sax.de Cc: freebsd-current@FreeBSD.org Subject: Re: floppy format detection [was Re: devfs questions] Message-ID: <199604091942.MAA05667@phaeton.artisoft.com> In-Reply-To: <199604090801.KAA08669@uriah.heep.sax.de> from "J Wunsch" at Apr 9, 96 10:01:01 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > IMO, the tradeoff between full automatic format detection, semi-automatic > > detection (as in the 16/6000) and manual specification, is a question of > > driver complexity plus how much delay you are willing to experience on each > > open of the drive, plus whether you can determine certain things without > > damage to the hardware. I personally prefer having an semi-automatic > > dev plus manual devs or an ioctl available for explicit settings. > > I fully agree with you here. That's why i would like to have a part > of the minor number available for several densities, where only > density 0 is autodetect. > > > In the PC/AT world, the CMOS provides the "hints" needed to perform > > semi-automatic operations and get the test time down to an > > acceptable level. > > But only for the first two drives. :-) I really do not see the problem with time penalizing people for old hardware. How are you going to "make things equitable" so that my 65ms MFM drive goes as fast as possible? By penalizing everyone with 8.5ms drives, who have paid a premium for the speed? The same argument applies to penalizing me by only allowing me to store 120M on my MFM drive, while Jordan with his SCSI drive gets to store 2G -- aren't all platters equivalent? Shouldn't we put in compression so as to make things "more equitable", even though the consequence is a system Jordan feels is unusable? The penalty for hardware without media change notification is the need to check on access after the drive motoro is stopped. You can take the last active format as a "hint" of where to start checking. You can also take the CMOS drive type information as a hint as to a range restriction on the list that must be checked. You run old hardware? You pay in time. You run more floppy devices than are in CMOS? You pay in time. The problem with density hints is that there is no such thing as an automount. I can stuff a floppy in a drive on a DOS box, do a "dir" and have it list the floppy contents *regardless* of the format. And you guys are saying FreeBSD can never do this thing that DOS does because it needs to have the device identified for a manual mount instead of a mount as a result of automatic device identification. For comparison, see the SCSI floppy drives and the multiple density Sony optical drive on NeXT machines -- which automount removable media to a "folder" (directory) in the user's home directory when the media is inserted. Next you'll be suggesting that PCMCIA cards (insertable media, much like floppies, which require a complex procedure to identify the media following insertion) should have a manual component to their insertion, based on card type (a weenie "mount" command, if you will). People with crappy ardware will have to live with it. Just like Bruce lives with typing ^C to flush the FCB's on his DOS drive without change notification when changing media. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199604091942.MAA05667>