Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 7 Nov 1996 13:27:20 -0700 (MST)
From:      Terry Lambert <terry@lambert.org>
To:        jehamby@lightside.com (Jake Hamby)
Cc:        hackers@freebsd.org
Subject:   Re: Welcome to POSIX...
Message-ID:  <199611072027.NAA10568@phaeton.artisoft.com>
In-Reply-To: <Pine.AUX.3.94.961107104602.19524A-100000@covina.lightside.com> from "Jake Hamby" at Nov 7, 96 10:54:27 am

next in thread | previous in thread | raw e-mail | index | archive | help
> I now have a copy of the NIST POSIX conformance test suite, or
> NIST-PCTS:151-2.  Right now I'm running it on a Solaris box so I can be
> familiar with it before I try FreeBSD.
> 
> I should note, though, two things:  First, this is a test for FIPS 151-2,
> and not strictly POSIX.1 (it's POSIX.1 with a few additional tidbits).
> Second, this test suite does _NOT_ require TET to be installed, as Terry
> had mentioned.  Perhaps I have a different test suite from what Terry
> was thinking of?  If so, is that other test free as well?

I already answered this.

I made a mistake; the TET is required for my FABIO test suite which
I have been working on.  I plopped the thing into the same framework
and call it from a TET environment.  It was my setup, not the test
itself, which required TET.

I made the mistake because I was running both PCTS and SVID against
UnixWare back at Novell.  SVID requires TET (so does the X11 validation
suite).


> I received the test suite from Martha Gray at NIST, and as Terry
> mentioned, it does not yet have the proper legal notices for a more
> wide-spread distribution.  However, as I've already discovered at least
> one Linux distribution which claims to be POSIX.1 and FIPS 151-2
> conformant already (Linux-FT from Lasermoon), I'm going to hurry and post
> up my finding ASAP (it's only a matter of time before RedHat, e.g., get
> tested for POSIX, and we don't want to be left in the dust :-)

The Linux distributor bought the suite.

They ran the suite and hacked the OS iteratively to make it conform.

Then they paid over $50,000 to an accredited testing lab to run the
suite for them, and sign off on a conformance certificate.


You personally running the test is meaningless.  You can claim compliance
(in a sort of carefully worded way), but not conformance.

To be able to claim conformance requires certification, which requires $$$.


The only thing this changes is that now you can do precertification
testing without purchasing the suite.


					Regards,
					Terry Lambert
					terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611072027.NAA10568>