Re: fortran code for 'cp' / subroutine file_copy
- From: Janne Blomqvist <foo@xxxxxxxxxxx>
- Date: Fri, 31 Mar 2006 20:39:43 +0300 (EEST)
In article <442D5A5A.3B67C375@xxxxxxxxxxxxx>, Walter Spector wrote:
Joost wrote:
but some, even very recent systems, really don't allow for a system
call, at the OS level (e.g. CRAY XT3).
Don't they still support CALL PXFSYSTEM?
IIRC the compiler provided with the XT3 is Portland, and according
to
http://www.polyhedron.com/pb05/linux/language.html
it doesn't support the PXF* stuff. Actually, except for ifort none of
the tested compilers support it.
Again, according to the above link all the compilers seem to support
some kind of "Invoke External command" feature, so I don't know why it
wouldn't work on the XT3. Yes, the XT3 is special in that most
syscalls on the compute nodes are not handled by the compute node
kernel but rather forwarded to one of the IO nodes, but Cray claims
some level of POSIX support so I guess it should work. But apparently
it doesn't. Perhaps Joost, who apparently has actual experience with
the thing, can explain?
--
Janne Blomqvist
.
- Follow-Ups:
- Re: fortran code for 'cp' / subroutine file_copy
- From: Joost
- Re: fortran code for 'cp' / subroutine file_copy
- References:
- fortran code for 'cp' / subroutine file_copy
- From: Carsten Lemmen
- RE: fortran code for 'cp' / subroutine file_copy
- From: meek
- Re: fortran code for 'cp' / subroutine file_copy
- From: Joost
- Re: fortran code for 'cp' / subroutine file_copy
- From: Walter Spector
- fortran code for 'cp' / subroutine file_copy
- Prev by Date: Re: O/T: Patenting a computational scheme?
- Next by Date: Re: Structure of large link libraries in f95
- Previous by thread: Re: fortran code for 'cp' / subroutine file_copy
- Next by thread: Re: fortran code for 'cp' / subroutine file_copy
- Index(es):