Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: Notesfiles $Revision: 1.7.0.10 $; site ndm20
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!inuxc!pur-ee!uiucdcs!convex!ndm20!lvs
From: lvs@ndm20
Newsgroups: net.unix-wizards
Subject: Byte at a time disk IO
Message-ID: <28800001@ndm20>
Date: Fri, 7-Mar-86 15:58:00 EST
Article-I.D.: ndm20.28800001
Posted: Fri Mar  7 15:58:00 1986
Date-Received: Sat, 15-Mar-86 03:29:02 EST
Lines: 23
Nf-ID: #N:ndm20:28800001:004:932
Nf-From: ndm20!lvs    Mar  7 14:58:00 1986


Since I have no access to any UNIX source code, and this aspect is
not very well documented, I will ask "Those who know where from they speak".

The point in question is byte at a time disk i/o.  Does it really work.
Can n processes be reading/writting from/to different bytes in the same
locality (block) without wiping each other out.  In another guise, if you
set a file mode to "unbuffered", is it REALLY unbuffered to the point of
being able to write individual bytes without possibly stomping on another
process reading the byte next to it.

I would like information about as many different versions of UNIX as I can
get.  Please mail to me directly, and I will summarize the relevant details.
If any are received.

Thanks in advance.

Larry V. Streepy Jr.
Nathan D. Maier Consulting Engineers
VOICE:  (214)739-4741
Usenet: ...!{allegra|ihnp4}!convex!smu!ndm20!lvs
CSNET:  ndm20!lvs@smu
ARPA:   ndm20!lvs%smu@csnet-relay.ARPA