Date: Sun, 5 May 1996 18:42:58 +0800 (GMT+0800)
From: Orlando Andico <orly AT abigail DOT eee DOT upd DOT edu DOT ph>
To: Paul Derbyshire <ao950 AT FreeNet DOT Carleton DOT CA>
cc: djgpp AT delorie DOT com
Subject: Re: libm.a/linker *BUG* with test program
In-Reply-To: <4me6om$5md@freenet-news.carleton.ca>
Message-ID: <Pine.LNX.3.91.960505184203.5127B-100000@abigail.eee.upd.edu.ph>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

I think it's 'cause libc.a contains a (subset? full set?) of the math 
routines in libm.a, albeit I think the accuracy is less. So things aren't 
supposed to crash or anything, just lessened number of significant digits 
in your answer.

On 4 May 1996, Paul Derbyshire wrote:

> 
> 
> This may be a stupid question but: why is it that when the math library is
> not on the command line, it not only links and resolves the calls to pow
> and ldexp, but it *works* too???
> And, of course, if this program works without the library and crashes with
> it, maybe you might try just not using the library for this situation...
> 
> --
>     .*.  "Clouds are not spheres, mountains are not cones, coastlines are not
>  -()  <  circles, and bark is not smooth, nor does lightning travel in a
>     `*'  straight line."    ,------------------------------------------------
>          -- B. Mandelbrot  |  Paul Derbyshire (PGD) ao950 AT freenet DOT carleton DOT ca
>