home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!cs.utexas.edu!qt.cs.utexas.edu!yale.edu!ira.uka.de!Sirius.dfn.de!tubsibr!ramz.ing.tu-bs.de!ruediger
- From: ruediger@ramz.ing.tu-bs.de (Ruediger Helsch)
- Subject: Re: Variable-length device major/minor (was Re: SCSI Autodetect?)
- Message-ID: <1993Jan22.160111.23975@ibr.cs.tu-bs.de>
- Sender: postnntp@ibr.cs.tu-bs.de (Mr. Nntp Inews Entry)
- Organization: Mechanikzentrum, Technische Universitaet Braunschweig, Germany
- References: <C0x6JC.1Bw@ra.nrl.navy.mil> <1993Jan17.161247.19574@gremlin.muug.mb.ca> <1993Jan18.123421.2395@klaava.Helsinki.FI> <1993Jan18.194407.19389@klaava.Helsinki.FI> <2B5C2074.13F36@tct.com>
- Date: Fri, 22 Jan 1993 16:01:11 GMT
- Lines: 16
-
- chip@tct.com (Chip Salzenberg) wrote:
- >According to torvalds@klaava.Helsinki.FI (Linus Torvalds):
- >>So, the only remaining reason for not making dev_t a 32-bit integer is
- >>my lazyness (the win/work ratio is way too low).
- >
- >If the goal is to make MINOR(dev) greater than eight bits wide, why
- >not consider making MAJOR(dev) smaller than eight bits?
-
- Why should anybody do this? He could just as good make the dev_t
- 32 bit wide. HP-UX does this, taking 8 bit for major and 24 bit
- for minor. Works very good.
- If you invest the work to change something, please don't waste your time
- counting bits like the IBM engineers did it with all those *&$% PC
- interfaces.
-
- Ruediger Helsch <ruediger@ramz.ing.tu-bs.de>
-