Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.cs.su.oz.au!inferno.mpx.com.au!news.unimelb.EDU.AU!munnari.OZ.AU!news.ecn.uoknor.edu!solace!nntp.uio.no!news.cais.net!newsfeed.internetmci.com!in2.uu.net!news.artisoft.com!usenet
From: Terry Lambert <terry@lambert.org>
Newsgroups: comp.unix.bsd.netbsd.misc
Subject: Re: Historic Opportunity facing Free Unix (
Date: Tue, 23 Apr 1996 21:21:51 -0700
Organization: Me
Lines: 52
Message-ID: <317DAC5F.63699063@lambert.org>
References: <31784ED6.1047B01F@lambert.org> <4ljn7q$orr@news.ossi.com>
NNTP-Posting-Host: hecate.artisoft.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.01 (X11; I; Linux 1.1.76 i486)
Mitchell Erblich wrote:
] ] "Self-verification will be used to assure conformance to the ABI+
] ] specification. The ABI+ group does not wish to officially certify
] ] either systems or applications. However, the ABI+ group will be
] ] making available a family of verification tests that can be used
] ] by software developers[...]"
]
] Would all the "EOPNOTSUPP" returned error codes be part
] of the API when the operation is NOT supported and NOT
] the intended "Operation not supported on socket".
]
] I believe that both freebsd and netbsd should change these
] errors back to EINVAL. This is not proper behavior for
] a SYSV.3 or a SYSV.4 OS to behave.
]
] In addition what about the naming of a "pcfs" as "msdosfs".
]
] And .... :)
You mailed this without context, as well as posting it; look
for my confused reply in your inbox. 8-).
I was quoting the ABI+ documents that he pointed at, not
agreeing with them. This is, of course, a problem with any
self-verification situation, which makes me believe ABI+
will be pretty useless (though we could subvert it by
turning it into something useful against its will).
As far as error return standardization, all properly functioning
code must handle any error subcode for any error return; this
is mandated by POSIX (assuming POSIX is considered a good
arbiter -- I think it is).
A -1 return is "operation has failed", and you can not expect
to get a fully enumerated error list from any documentation
whatsoever. For one thing, there error from a library call
to a routine in a shared library in a pageable mapped text
area accessed over a network may very well be a network
access error. 8-).
FS naming is irrelevant; a proper OS needs no FS type
identification for anything but a high level format program.
The FS type is implied from the media being accessed's on
disk format (at least, it *can* be, so it probably *should*
be...).
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.