Xref: utzoo comp.sys.ibm.pc:12786 comp.sys.att:2693 comp.unix.questions:5949 comp.unix.xenix:1613 comp.unix.microport:130
Path: utzoo!mnetor!uunet!lll-winken!lll-tis!ames!nrl-cmf!cmcl2!phri!manhat!mancol!samperi
From: samperi@mancol.UUCP (Dominick Samperi)
Newsgroups: comp.sys.ibm.pc,comp.sys.att,comp.unix.questions,comp.unix.xenix,comp.unix.microport
Subject: floppy disk I/O
Message-ID: <287@mancol.UUCP>
Date: 3 Mar 88 21:07:24 GMT
Organization: Manhattan College, NYC, USA
Lines: 22
Keywords: floppy I/O, blocking factor


Can somebody tell me what is a 'safe' blocking factor to use for floppy
diskette I/O, in particular, for 1.2 Meg floppies? Since these floppies
are formatted with 30 sectors per cylinder, I've been using a blocking
factor of 30, for maximum speed (I always write sequentially, starting
from the first cylinder, so this means there is minimum head movement from
one read/write to the next). Although this has worked nicely in the MS-DOS
environment, I occasionally experience data corruption when I use this
blocking factor in the UNIX environment (under both Microport's System V/AT
and SCO Xenix). Using a blocking factor of 10 seems to solve the problem,
so far...

Floppy I/O in the UNIX environment is about 4-5 times slower than it is
in the MS-DOS environment (I'm using the BIOS low-level sector read/write
calls in the MS-DOS environment). Is there a way I can give a process
exclusive access to a floppy drive in the UNIX environment, in such a
way that I can get faster and more reliable floppy I/O?

-- 
Dominick Samperi, Manhattan College, NYC
    manhat!samperi@NYU.EDU           ihnp4!rutgers!nyu.edu!manhat!samperi
    philabs!cmcl2!manhat!samperi     ihnp4!rutgers!hombre!samperi
              (^ that's an ell)      uunet!swlabs!mancol!samperi