[832] in netbsd-help mailing list archive
Re: 'weather' somewhat broken
daemon@ATHENA.MIT.EDU (Eric Ding)
Tue Mar 5 23:54:03 1996
To: Daniel P Kamalic <pocky@MIT.EDU>
Cc: netbsd-help@MIT.EDU
In-Reply-To: Your message of Tue, 05 Mar 1996 15:27:24 -0500.
<199603052027.PAA00695@discordia.MIT.EDU>
Date: Tue, 05 Mar 1996 23:49:53 EST
From: Eric Ding <ericding@MIT.EDU>
Hi Daniel,
right you are, I believe there's a problem with the FPU simulation (I have
no 387 math coprocessor installed). When I run the sequence you asked me
to, I get
;Value:
;Hardware trap SIGSEGV
;Type D to debug error, Q to quit back to REP loop:
My question is, why would a program like "weather" need any FPU ops?
Eric
>>>>> Daniel P Kamalic <pocky@MIT.EDU> writes:
> Eric:
> You may be having the same problem that I'm having on my 486-DLC
> 40. You probably have no floating point unit (unless you've installed a
> separate math coprocessor). Perhaps that's why you've been getting the
> math errors. I've been hoping that my math coprocessor problem (with
> Java and Scheme) will be no more once I upgrade to 1.1.
> Eric, would you do me a favor and try running MIT Scheme:
> add 6.001
> scheme001
> (edwin)
> And then
> (/ 5 2)
> and press ctrl-x ctrl-c to evaluate.
> If you get 2.5, then upgrading to 1.1 should fix my problem.
> If you get some scary error, then I'm worried.
> (Also, if there's anyone out there running 1.0A without an FPU,
> please try that and let me know what happens.)
