On Mon, 2007-10-08 at 13:28 +0200, Angelo Graziosi wrote:
I have tried the test gfc_gamma_test_c302m.F
(http://gcc.gnu.org/ml/fortran/2007-10/msg00096.html) on GNU/Linux
kubuntu
Feisty 7.04 with GFortran 20071008-129121 and it still fails:
We are now using the gamma functions provided by the respective
C libraries. Looking at the gcc-testresults failures, and at
your report, it seems as if the quality of existing gamma functions
is rather low.
At the least, we should
- relax the tolerances in gamma_5.f90 to something that doesn't fail
on as many systems
- note in the documentation that the GAMMA functions are sometimes
inaccurate, and people may be better off using an external
gamma function.
Maybe we should also provide our own, higher-quality gamma function,
but I don't see that in the 4.3 timeframe (not if I have to do
it, at least :-)
Thomas