Re: [gnugo-devel] Platform dependencies

[Trevor Morris]

At 12:28 AM 3/28/2002 +0100, Arend Bayer wrote:

>On Mon, 25 Mar 2002, Gunnar Farneback wrote:
>
>> Arend wrote:
>> > Platform depencies (it would be quite good if we could solve these before 
>> > 3.2):
>> > ./regress.sh . ld_owl.tst
>> > 301 unexpected FAIL: Correct '3 B18', got '0'
>> > ./regress.sh . 13x13.tst
>> > 27 unexpected PASS!
>> 
>> I'd even say they are showstoppers unless we absolutely cannot track
>> them down.
>The second was trevorc:280, not 13x13:27 as above.
>
>However, both tests behave as usual if I compile without optimization.
>So it is possibly a compiler bug (gcc --version says egcs-2.91.66). 
>
>I'd conclude that it is not worth spending much more time looking into that.
>Any opinions?
>
>Arend

Seems to me it's more likely an array bounds problem than a compiler
bug.  Is that a known buggy compiler version?  On the other hand, I don't
think it must hold up 3.2, but it's probably worth keeping an eye on,
especially if GG_FULL_ASSERT is adopted.

-Trevor