Hi, and thanks for reply.
On Wed, 19 Nov 2003, Craig A. Berry wrote:
> Since no one stepped forward to do any VAX testing during the very
> long release candidate cycle for 5.8.2, it's not that surprising we'd
> encounter problems. However, it looks like you did get a successful
> build, so that's good news.
Surprising, I thought at least the comp.os.vms newsgroup is quite active
with many vax users.
Anyway I have a bunch of vaxen available, but I'm not a programmer,
so my knowledge is rather limited.
>
> This would very likely have something to do with a test that expects
> a floating point range outside of what can be done with D_FLOAT,
> which is the default on VAX. Details of the test failure would be
> helpful, which you can obtain by running this from the top of your
> build tree:
>
> $ @[.vms]test .exe "" -"v" [.base]num.t
if ($^O eq 'ultrix') {
# Ultrix enters looong nirvana over this.
print "ok 47\n";
} else {
$a =
0.00000000000000000000000000000000000000000000000000000000000000000001;
print $a > 0 ? "ok 47\n" : "not ok 47\n";
}
<clip>
ok 44
ok 45
ok 46
not ok 47
t/BASE/NUM....FAILED at test 47
Can't run 47.
<clip>
>
> You may have better results building with G_FLOAT rather than
> D_FLOAT. See the section entitled "Floating Point Considerations" in
> README.vms.
I'll do a rebuild.
>
> Thanks for the report.
> --
> ________________________________________
> Craig A. Berry
> mailto:[EMAIL PROTECTED]
>
> "... getting out of a sonnet is much more
> difficult than getting in."
> Brad Leithauser
>
Regards,
--Saku
