develooper Front page | perl.perl5.porters | Postings from February 2000

t/op/numconvert.t

Thread Next
From:
Jarkko Hietaniemi
Date:
February 27, 2000 10:31
Subject:
t/op/numconvert.t
Message ID:
200002271830.UAA04814@alpha.hut.fi

Ilya, *what* is t/op/numconvert.t trying to test, really?

Especially, why is the test bailing out (because "Unsigned arithmethic
is not sane") on 64-bit builds?  (Either natively 64-bit or gcc's long
long).

I'm asking because it seems there is no reason to bail out.

If I remove the exit, the test completely succeeds in Digital UNIX
(alpha, natively 64-bit), in Solaris 2.7 (sparc, using gcc's long
long), and Linux 2.2.13 (ix86 using gcc's long long), and I have a
report from Steven Hirsch of IBM that a natively 64-bit AIX tests out
fine too if he just removes the exit.  Here's what the variables
controlling bailing out are:

# max_uv1 = 4294967295, max_uv2 = 4294967295, big_iv = 68719476720

-- 
$jhi++; # http://www.iki.fi/jhi/
        # There is this special biologist word we use for 'stable'.
        # It is 'dead'. -- Jack Cohen

Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About