CODEWITHSUNDEEP
linuxcudamakefile
Nikhil
Nikhil

Reputation: 1309

Library error when building a make file for a CUDA program

I am trying to build a makefile in linux. The code which goes into the makefile is as follows:

NVCC = /usr/local/cuda/bin/nvcc 
CUDAPATH = /usr/local/cuda 
NVCCFLAGS = -I$(CUDAPATH)/include 
LFLAGS = -L$(CUDAPATH)/lib64 -lcuda -lcudart -lm

VectorAdd:
    $(NVCC) $(NVCCFLAGS) $(LFLAGS) -o VectorAdd VectorAdd.cu

So when I type "make"

I get the following error:

enter image description here

How do i get rid of it and build my make file?

Upvotes: 0

Views: 321

Answers (1)

Robert Crovella
Robert Crovella

Reputation: 151799

Since you're using NVCC you don't really need those LFLAGS and NVCCFLAGS definitions. nvcc knows how to find all that automatically.

But if you want to fix it, get rid of the space at the end of your cuda path definition:

CUDAPATH = /usr/local/cuda 
                          ^ there is a space here, delete it

Upvotes: 4

Related Questions