Wierd core dump on sparc-1

[Bill Gunshannon]

In article <14976 at smoke.brl.mil>, gwyn at smoke.brl.mil (Doug Gwyn) writes:
> to a function.  Since SPARC tends to be less tolerant of such errors than
> some other C environments, it is possible that your code has such a problem.

Actually, I wouldn't lay the blame at the feet of SPARC.  I have a MIPS
that suffers the same problems.  And I have seen it in the past on 68K
systems with un-satisfactory memory management hardware.  It is a rather
wide spread problem.  I have reached the point where the first thing I
look for is an uninitialized string/char[] variable.

> Is "lint" happy with the code?

Unfortunately, passing "lint" doesn't guarantee that the hardware will
do what the compiler (or language) thinks is proper.

bill


-- 

     Bill Gunshannon          |        If this statement wasn't here,
     bill at platypus.uofs.edu   |  This space would be left intentionally blank