I'm using mpif90 with Intel 10 fortran complier:
% mpif90 -compile_info
ifort -I/usr/mpi/qlogic//include/mpich/intel10/x86_64 -c -I/usr/mpi/qlogic//include
If I don't specify any compiler options, the
compiler vectorises some loops:
% mpif90 p-grains1.f90
p-grains1.f90(123): (col. 1) remark: LOOP WAS VECTORIZED.
p-grains1.f90(250): (col. 35) remark: LOOP WAS VECTORIZED.
Is loop vectorisation/unrolling safe for MPI logic?
I presume it is, but are there situations where
loop vectorisation could e.g. violate the order
of execution of MPI calls?
many thanks
anton
--
Anton Shterenlikht
Room 2.6, Queen's Building
Mech Eng Dept
Bristol University
University Walk, Bristol BS8 1TR, UK
Tel: +44 (0)117 331 5944
Fax: +44 (0)117 929 4423
|