Home > Archive > Fortran > May 2005 > cygwin now works OK with large g77 arrays
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
cygwin now works OK with large g77 arrays
|
|
| Charles Russell 2005-05-10, 8:58 pm |
| According to recent messages in the cygwin mailing list archives, the
problem with large fortran arrays has been fixed.
| |
| Charles Russell 2005-05-11, 4:00 am |
| Charles Russell wrote:
> According to recent messages in the cygwin mailing list archives, the
> problem with large fortran arrays has been fixed.
But in a development snapshot, not the official release. See
http://sources.redhat.com/ml/cygwin...5/msg00495.html
| |
| Mark Hadfield 2005-05-26, 9:00 pm |
| Charles Russell wrote:
> Charles Russell wrote:
>
>
>
> But in a development snapshot, not the official release. See
>
> http://sources.redhat.com/ml/cygwin...5/msg00495.html
And now there has been a release of the Cygwin DLL (1.5.17-1) in which
the fix is made.
The problem (array size limited to 160 MiB or so) affected Cygwin
executables generated by the Cygwin g77 and gfortran compilers. (It did
not affect non-Cygwin executables generated with the -mno-cygwin switch.
Nor did it affect executables generated by the MinGW versions of the
same compilers.) The new version of the Cygwin DLL fixes it in both g77
and gfortran. Presumably it's the version used at *run time* that matters.
This problem did *not* affect g95.
Now the size of programs generated by Cygwin Fortran compilers is
limited only by the heap_chunk_in_mb registry setting, plus the obvious
things like the amount of virtual memory on the machine and the limits
assocated with 32-bit addressing in Windows and Cygwin. I have run test
programs with array sizes up to 1000 MiB or so.
--
Mark Hadfield "Ka puwaha te tai nei, Hoea tatou"
m.hadfield@niwa.co.nz
National Institute for Water and Atmospheric Research (NIWA)
|
|
|
|
|