Date: Wed, 03 Apr 96 23:40:32 PST From: "Brett Glass" <Brett_Glass@ccgate.infoworld.com> To: Michael Smith <msmith@atrad.adelaide.edu.au> Cc: msmith@atrad.adelaide.edu.au, jkh@time.cdrom.com, hardware@freebsd.org Subject: Re: Some solutions to disk problems.... I think. Message-ID: <9603038286.AA828600274@ccgate.infoworld.com>
next in thread | raw e-mail | index | archive | help
> *sigh* The whole point is that _nothing_ other than the 'wd' driver > should know this; Why not? When a tape drive or CD-ROM is attached to a SCSI interface, there are exposed flags for both the SCSI adapter and the drive. This isn't "special case" code. > You want to be able to match a vendor string and a model number. If the > textual content changes, you duplicate the entry in the table. Thus making the table larger than intelligent recognition code. Even wildcards aren't enough to solve the problem. (Regular expressions WOULD be, but I don't think there's much justification for adding a regular expression matcher to the kernel!) >> Different things will be searched for in different cases. It's easy to >> break this out into functions that search for the right thing in each >>case. > Huh? You can "search for" the contents of the ID string. The contents of the ID string may need to be parsed further. At which point, one will need code anyway. > No, egregious monsterism aside, the majority of drives still follow the > spec. All you want is a list of drives that behave outside the envelope > that's required for conformance with the driver. This is not 'all > drives', as is demonstrated by the lack of such a table to date. The lack of such a table to date only indicates that the problem has not been addressed.... --Brett
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9603038286.AA828600274>