[AMBER] building error of multi-GPU pmemd.cuda

From: Masakazu SEKIJIMA <sekijima.gsic.titech.ac.jp>
Date: Mon, 6 Dec 2010 16:30:19 +0900

Dear Amber-users

I'm just building multi-GPU pmemd.cuda. I compiled single-GPU
pmemd.cuda successfully.
But I am getting below errors:
Could you give me some advise on this problem?

make[3]: Entering directory
`/home0/usr0/sekijima-m-aa/amber11/src/pmemd/src/cuda'
cpp -traditional -DMPI -I/include -P -DBINTRAJ -DDIRFRC_EFS
-DDIRFRC_COMTRANS -DDIRFRC_NOVEC -DFFTLOADBAL_2PROC -DPUBFFT -DCUDA
-DMPI -DMPICH_IGNORE_CXX_SEEK cuda_info.fpp cuda_info.f90
mpif90 -ip -O3 -xHost -DCUDA -DMPI -DMPICH_IGNORE_CXX_SEEK
-I/opt/cuda/3.1/include -I/include -c cuda_info.f90
mpicc -ip -O3 -xHost -DMPICH_IGNORE_CXX_SEEK -D_FILE_OFFSET_BITS=64
-D_LARGEFILE_SOURCE -DBINTRAJ -DMPI -I/include -DCUDA -DMPI
-DMPICH_IGNORE_CXX_SEEK -I/opt/cuda/3.1/include -I/include -c gpu.cpp
mpicc -ip -O3 -xHost -DMPICH_IGNORE_CXX_SEEK -D_FILE_OFFSET_BITS=64
-D_LARGEFILE_SOURCE -DBINTRAJ -DMPI -I/include -DCUDA -DMPI
-DMPICH_IGNORE_CXX_SEEK -I/opt/cuda/3.1/include -I/include -c
gputypes.cpp
/opt/cuda/3.1/bin/nvcc -use_fast_math -O3 -gencode
arch=compute_13,code=sm_13 -gencode arch=compute_20,code=sm_20 -DCUDA
-DMPI -DMPICH_IGNORE_CXX_SEEK -I/opt/cuda/3.1/include -I/include -c
kForcesUpdate.cu
In file included from gpu.h:15,
                 from kForcesUpdate.cu:14:
gputypes.h:25:17: error: mpi.h: No such file or directory
make[3]: *** [kForcesUpdate.o] Error 1

-- 
_____
Associate Professor Masakazu SEKIJIMA , Ph.D.
Global Scientific Information and Computing Center(GSIC)
Tokyo Institute of Technology
Mail: sekijima.gsic.titech.ac.jp
Web: http://www.bio.gsic.titech.ac.jp/index.html
2-12-1-i7-5, Ookayama, Meguro-ku, TOKYO
152-8552 JAPAN
Tel: +81-3-5734-3325
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Dec 06 2010 - 00:00:03 PST
Custom Search