From owner-freebsd-hackers  Mon Oct 13 09:28:00 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id JAA04328
          for hackers-outgoing; Mon, 13 Oct 1997 09:28:00 -0700 (PDT)
          (envelope-from owner-freebsd-hackers)
Received: from consys.com (consys.com [209.60.202.194])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id JAA04310
          for <freebsd-hackers@FreeBSD.ORG>; Mon, 13 Oct 1997 09:27:57 -0700 (PDT)
          (envelope-from rcarter@consys.com)
Received: from dnstoo.consys.com (dnstoo.ConSys.COM [209.60.202.195])
	by consys.com (8.8.6/8.8.6) with ESMTP id JAA24501;
	Mon, 13 Oct 1997 09:27:55 -0700 (MST)
Received: from dnstoo.consys.com (localhost [127.0.0.1])
	by dnstoo.consys.com (8.8.5/8.8.6) with ESMTP id JAA03837;
	Mon, 13 Oct 1997 09:27:54 -0700 (MST)
Message-Id: <199710131627.JAA03837@dnstoo.consys.com>
X-Mailer: exmh version 2.0zeta 7/24/97
To: Mike Smith <mike@smith.net.au>
cc: "Russell L. Carter" <rcarter@consys.com>, gjohnson@nola.srrc.usda.gov,
        tlambert@primenet.com, freebsd-hackers@FreeBSD.ORG
Subject: Re: Floating point exceptions 
In-reply-to: Your message of "Sat, 11 Oct 1997 17:01:47 +0930."
             <199710110731.RAA00733@word.smith.net.au> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 13 Oct 1997 09:27:54 -0700
From: "Russell L. Carter" <rcarter@consys.com>
Sender: owner-freebsd-hackers@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

}> |> FreeBSD (so I can replace Linux). My understanding is that FreeBSD traps 
}> |> FPE's. Is there a way for me to change the FreeBSD source code to mask FPE's 
}> |> rather than trap them? Better yet, can I do that for a specific application at
}> |> compile time? Your assistance is greatly appreciated.
}> |
}> |Fix:		Correct the code to not generate exceptions
}> 
}> Hmm, I don't think so.  There are a variety of reasons that NA people
}> want to use those carefully thought out exceptions.  Find
}> 'em yerself.  
}
}In the original case the NA people weren't using the exceptions; they 
}were ignoring them.
}
}IMHO FreeBSD does the correct thing; it notifies you of the exception.  
}It is your responsibility to decide what to do about it.

I agree, as long as it is understood that some codes are correctly
written to generate exceptions that are best masked.  fpsetmask()
appears to fit the bill.

}
}I have spent not a little time carefully explaining to various 
}scientific programmers about arithmetic exceptions and why they should 
}care about them, and so far have had a 100% conversion rate.  If you 
}look a physicist in the eye and say "unquantifiable error", they will 
}usually listen.

What do physicists know about NA?  :-) :-) :-)

Russell