[comp.os.minix] math lib needed

meulenbr@cstw01.UUCP (Frans Meulenbroeks) (10/03/88)

Hi!

Last weekend I decided to try to port PERL to Minix.
I'm now in the state that all modules compile. 
However, when linking, I get unresolved references due to the fact that 
Minix does not contain a math lib.
Does anyone have the source of such a beast (most likely in C or
68000 assembly). If so, a copy would be appreciated greatly!

Many thanks!
-- 
Frans Meulenbroeks
	Centre for Software Technology
	...!mcvax!philmds!prle!cst!meulenbr   or   ...!uunet!prlb2!cst!meulenbr
        or perhaps   meulenbr@cst.prl.philips.nl

SJONES%HAMPVMS.BITNET@mitvma.mit.edu (Steve Jones) (10/05/88)

Greetings,

        In regards to a math library, I'd remind everyone of MATH(3M) in
4.3 BSD The following appears in the Programmers Reference ...

        ...But the [MATH(3M)] codes in 4.3 BSD are intended for the
        public domain; they may be copied freely provided their
        provenance is always acknowledged, and provided users assist
        the authors ... by reporting experience with the code.

        Now while most of this stuff seems to depend on or at least be
geared towards VAX D_floating-point internal representations, it is a
source of proven routines which are available to the ambitious soul
who undertakes this endeavor.

Good Luck!

--Steve.

BITNET:  sjones@hampvms         INET:  sjones%hampvms.bitnet@mitvma.mit.edu
CSNET:   sjones%hamp@umass-cs   UUCP:  ...uunet!hampvms.bitnet!sjones

USnail:  box 753; Hampshire College; Amherst, MA  01002

DEDOUREK%UNB.CA@cornellc.ccs.cornell.edu (10/05/88)

Could I suggest that if someone decides to hack a math library
and place it in the public domain, a good guide might be
the book "Software Manual for the Elementary Functions", Cody and
Waite, published by Prentice-Hall, 1980, ISBN 0-13-822064-6.
This book is intended for compiler writers and such who must
make a library of functions containing sqrt, sin, cos, etc.  It
gives full algorithms along with discussion of implementation
techniques.  A math library following these guidelines might
not be the last word in math libraries, but it should be relatively
robust and correct.
John DeDourek

henry@utzoo.uucp (Henry Spencer) (10/07/88)

In article <4657@louie.udel.EDU> DEDOUREK%UNB.CA@cornellc.ccs.cornell.edu writes:
>..."Software Manual for the Elementary Functions", Cody and
>Waite, published by Prentice-Hall, 1980, ISBN 0-13-822064-6.
>... A math library following these guidelines might
>not be the last word in math libraries, but it should be relatively
>robust and correct.

The 4.3BSD math library, which as I recall it is nearly public domain,
would be a better place to start for most modern machines.  Kahan et al,
the authors of it (and major contributors to the IEEE floating-point
standard), observe that Cody&Waite was written in the Bad Old Days and
is not ideal for relatively well-behaved floating-point hardware like
the IEEE-compatible chips (8087, 68881, etc.).
-- 
The meek can have the Earth;    |    Henry Spencer at U of Toronto Zoology
the rest of us have other plans.|uunet!attcan!utzoo!henry henry@zoo.toronto.edu