Message-Id:	<199908091553.KAA03679@darwin.sfbr.org>
Date:	Mon, 9 Aug 1999 10:53:05 -0500 (CDT)
From:	Jeff Williams <jeffw AT darwin DOT sfbr DOT org>
Subject:	Re: CPU identification (Was: Re: uname -m ?)
To:	djgpp-workers AT delorie DOT com
MIME-Version:	1.0
X-Mailer:	dtmail 1.3.0 @(#)CDE Version 1.3.4 SunOS 5.7 sun4u sparc
Reply-To:	djgpp-workers AT delorie DOT com

-: Also, if detecting x87 could be dangerous (i.e. risk crashing the
-: program), I would vote against it.

On a related note, I have seen lots of ways to *detect* an FPU
(a 387 to be exact), but what about *testing* the FPU for functionality?

I remember installing lots of 387's; it was easy enough to install the
chip and then adjust a DIP switch, add/remove a jumper, then run some
utility (e.g., Norton's SI) just to make sure it was detected, but
afterwards you never really knew if the chip was really working as it
should.  If a program would run with or without a 387, would a faulty
387 ever have been detected after installation?  This still bothers me.

Does anyone know of a test suite for testing FPU functionality?

TIA---jtw

- Raw text -