# /* ----------------------------------------------------------------------   
#    LAMMPS - Large-scale Atomic/Molecular Massively Parallel Simulator       
#    http://lammps.sandia.gov, Sandia National Laboratories                   
#    Steve Plimpton, sjplimp@sandia.gov                                       
#                                                                             
#    Copyright (2003) Sandia Corporation.  Under the terms of Contract        
#    DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains   
#    certain rights in this software.  This software is distributed under      
#    the GNU General Public License.                                          
#                                                                             
#    See the README file in the top-level LAMMPS directory.                   
# ------------------------------------------------------------------------- */
#                                                                             
# /* ----------------------------------------------------------------------   
#    Contributing authors: Mike Brown (SNL), wmbrown@sandia.gov               
#                          Peng Wang (Nvidia), penwang@nvidia.com             
#                          Paul Crozier (SNL), pscrozi@sandia.gov             
# ------------------------------------------------------------------------- */

BIN_DIR = .
OBJ_DIR = .
AR = ar
CUDA_CPP  = nvcc -I/usr/local/cuda/include -DUNIX -O3 -Xptxas -v --use_fast_math
CUDA_ARCH = -arch=sm_13
CUDA_PREC = -D_SINGLE_SINGLE
CUDA_LINK = -L/usr/local/cuda/lib -lcudart $(CUDA_LIB)

CUDA = $(CUDA_CPP) $(CUDA_ARCH) $(CUDA_PREC)

CUDA_LIB = $(OBJ_DIR)/libgpu.a

# Headers for CUDA Stuff
NVC_H  = nvc_macros.h nvc_device.h nvc_timer.h nvc_memory.h nvc_traits.h
# Headers for Pair Stuff
PAIR_H  = pair_gpu_texture.h pair_gpu_atom.h pair_gpu_nbor.h pair_gpu_cell.h
# Dependencies for the Texture Tar
TAR_H = $(NVC_H) $(PAIR_H) pair_gpu_atom.cu lj_gpu_memory.h lj_gpu_memory.cu \
        lj_gpu_kernel.h lj_gpu.cu gb_gpu_memory.h gb_gpu_memory.cu \
        gb_gpu_extra.h gb_gpu_kernel.h gb_gpu.cu

ALL_H = $(NVC_H) $(PAIR_H)

EXECS = $(BIN_DIR)/nvc_get_devices
OBJS = $(OBJ_DIR)/nvc_device.o $(OBJ_DIR)/pair_gpu_nbor.cu_o \
       $(OBJ_DIR)/pair_tex_tar.cu_o $(OBJ_DIR)/pair_gpu_cell.cu_o 

all: $(CUDA_LIB) $(EXECS)

$(OBJ_DIR)/nvc_device.o: nvc_device.cu $(NVC_H)
	$(CUDA) -o $@ -c nvc_device.cu

$(OBJ_DIR)/pair_gpu_nbor.cu_o: pair_gpu_nbor.cu pair_gpu_texture.h pair_gpu_nbor.h $(NVC_H)
	$(CUDA) -o $@ -c pair_gpu_nbor.cu

$(OBJ_DIR)/pair_tex_tar.cu_o: $(TAR_H)
	$(CUDA) -o $@ -c pair_tex_tar.cu

$(OBJ_DIR)/pair_gpu_cell.cu_o: pair_gpu_cell.cu pair_gpu_cell.h lj_gpu_memory.h
	$(CUDA) -o $@ -c pair_gpu_cell.cu

$(BIN_DIR)/nvc_get_devices: nvc_get_devices.cu $(NVC_H) $(OBJ_DIR)/nvc_device.o
	$(CUDA) -o $@ nvc_get_devices.cu $(CUDALNK) $(OBJ_DIR)/nvc_device.o 

$(CUDA_LIB): $(OBJS)
	$(AR) -crusv $(CUDA_LIB) $(OBJS)

clean:
	rm -rf $(EXECS) $(CUDA_LIB) $(OBJS) *.linkinfo

veryclean: clean
	rm -rf *~ *.linkinfo