This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: cygwin specific internal error with multiple subroutines in source file - postscripit
- From: "Paul Thomas" <paulthomas2 at wanadoo dot fr>
- To: "Andrew Pinski" <pinskia at physics dot uc dot edu>,"Tim Prince" <tprince at computer dot org>
- Cc: <fortran at gcc dot gnu dot org>
- Date: Sun, 10 Apr 2005 17:06:34 +0200
- Subject: Re: cygwin specific internal error with multiple subroutines in source file - postscripit
- References: <000601c53da3$61554930$0400000a@Paul> <6.1.2.0.0.20050410060814.01f8d258@imap.myrealbox.com>
I don't know where to find the f90 version you are using. As I got
assistance with this one on c.l.f a few years ago, it's possible there
could be a similarity.
This problem has crept in since the end of January:
$ d:/irun/bin/gfortran --version
GNU Fortran 95 (GCC 4.0.0 20050129 (experimental))
compiles and gives results comparable to the Digital compiler:
Maximum Rate = 3136.7856 Mega-Flops/Sec.
Quartile Q3 = 721.5527 Mega-Flops/Sec.
Average Rate = 567.1037 Mega-Flops/Sec.
Geometric Mean = 394.9383 Mega-Flops/Sec.
Median Q2 = 343.6634 Mega-Flops/Sec.
Harmonic Mean = 278.5250 Mega-Flops/Sec.
Quartile Q1 = 208.5226 Mega-Flops/Sec.
Minimum Rate = 55.3533 Mega-Flops/Sec.
$ d:/base/bin/gfortran --version
GNU Fortran 95 (GCC 4.1.0 20050405 (experimental))
on the other hand bombs out, as reported by Tim.
Paul T