| Subject: | Fails on nvsize!=8 (epsilon too small) |
Sample fail reports:
http://www.cpantesters.org/cpan/report/1198574
http://www.cpantesters.org/cpan/report/35293204
Changing the three epsilon values that are already in the test files to:
t/06_4ptsW1-D.t:my $epsilon = 1.0e-11;
t/07_4ptsW2-D.t:my $epsilon = 1.0e-11;
t/13_long_array.t:my $epsilon = 1.0e-08;
lets all tests pass on my smokers that are compiled with -Duselongdouble on linux amd64.
Would it be a good idea to just change the three values as I did?
HTH && Thanks && Regards,