>Date: Tue, 10 Sep 1996 08:38:46 -0500
>From: Jon Buller <jonb@metronet.com>

>"Wiker, Raymond" <etorwi@eto.ericsson.se> writes:
>Raymond> 	Um. I had some trouble with Elk (though I can't remember if it
>Raymond> was 3.0 or the previous version). Anyway, the problem turned out to be
>Raymond> a compiler optimisation - something to do with picking the wrong
>Raymond> address from a cons cell (could actually be the 'cdr' function...). I
>Raymond> can check this out if necessary. 
>
>That may well be it.  I'll recompile without the -O2 flag given to gcc and
>see what happens.  Do you know which file had the problem?

Another thing to try is to apply the patches for gcc-2.7.2.1.  I've
done that on my (i386) machine, and have not experienced any problems.
Supposedly 2.7.2.1 fixes the problem that required the use of
-fno-strength-reduce to avoid miscompiled code with -O2 on the i386.

I could send a PR, but there really wasn't much work involved in
applying the FSF patch.
-- 
Mike Long <mike.long@analog.com>     <URL:http://www.shore.net/~mikel>
VLSI Design Engineer         finger mikel@shore.net for PGP public key
Analog Devices, CPD Division          CCBF225E7D3F7ECB2C8F7ABB15D9BE7B
Norwood, MA 02062 USA       (eq (opinion 'ADI) (opinion 'mike)) -> nil