[mercury-users] gcc errors (was Programming "contests").

[Fergus Henderson]

On 30-Sep-1998, Tyson Dowd <trd at cs.mu.OZ.AU> wrote:
> On 29-Sep-1998, Gustavo A. Ospina <g-ospina at uniandes.edu.co> wrote:
> > What really happens when upon installing Mercury compiler occurs this 
> > error?:
> > 
> > 	Internal Compiler Error: program cc1 got fatal signal 4
> > 
> > It is a bug of gcc? of Linux kernel? I don't know...
> > 
> > This error occurs me frequently upon installing Mercury compiler (version 
> > 0.6 and 0.7.3 in Linux 2.0.0 using gcc 2.7.2). I had to restart several
> > times 'make' to continue and finish with the installation.
> 
> If it is not repeatable, it is often bad hardware.
> 
> gcc (and Mercury) use quite a bit of memory, and tend to exercise
> bad memory or overheated CPUs quite a bit.  fatal signal 11 is a
> very common error with bad memory.  I'm not sure what fatal signal 4
> is about.

Fatal signal 4 is "illegal instruction", typically caused by
the program jumping into data space instead of code space due
to the return addresses in the stack being overwritten.

But I agree with Tyson's diagnosis here: if simply typing `make' again
is enough to fix the problem, then hardware failure is the most likely
cause.  This can be due to problems with RAM, CPU, or cache, or it can
sometimes be caused by incorrect (overambitious) CMOS settings.
Check the GCC SIG11 FAQ (on my Zuse Linux installation, it is
in /usr/doc/faq/faq/GCC-SIG11-FAQ.gz) for more details.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
WWW: <http://www.cs.mu.oz.au/~fjh>  |  of excellence is a lethal habit"
PGP: finger fjh at 128.250.37.3        |     -- the last words of T. S. Garp.