[Wien] MPI bug in lapw1c_mpi

[L. D. Marks]

I tracked down that lapw1_mpi is dying at line 618 of modules.F with
complex data (the benchmark) where the lines:

!_COMPLEX          A(1:HSROWS,1:LINDEX)=DCMPLX(A1R(1:HSROWS,1:LINDEX), &
!_COMPLEX                                     -A1I(1:HSROWS,1:LINDEX))
!_COMPLEX          CALL COPYTOABCLM(TYPE,LINDEX,NV,NLO,A,ABCLM1)
!_COMPLEX          A(1:HSROWS,1:LINDEX)=DCMPLX(A2R(1:HSROWS,1:LINDEX), &
!_COMPLEX                                     -A2I(1:HSROWS,1:LINDEX))

are. A2I and A1I are allocated elsewhere as (11d,NSL2) and HSROWS > 11d.
Any suggestions?

Compilation:
Red Hat Linux 3.2.3-34 (rocks), ifort 8.1, /opt/mpich/intel/bin/mpif90
FPOPT =  -FR -mp1 -prec_div -w -r8 -pc80 -pad -fpconstant -traceback -fminshared -g -check

-----------------------------------------------
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60201, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L - marks @ northwestern . edu
http://www.numis.northwestern.edu
-----------------------------------------------