nvcc compilation under linux nvopencc command not found

Accelerated Computing CUDA CUDA Programming and Performance
1

Hello ,
i’m compiling a .cu file with nvcc under linux , however i get an error:
“sh: nvopencc: command not found”

the nvcc.profile is as follow:

TOP = /usr/lib/cuda-toolkit/…

LD_LIBRARY_PATH += $(TOP)/lib:$(TOP)/extools/lib:
PATH += $(TOP)/open64/bin:$(TOP)/bin:

INCLUDES += “-I$(TOP)/include” “-I$(TOP)/include/cudart” $(SPACE)

LIBRARIES =+ $(SPACE) “-L$(TOP)/lib” -lcudart

CUDAFE_FLAGS +=
OPENCC_FLAGS +=
PTXAS_FLAGS +=

on the linux machine i’m a normal user.
i read some lines on the nvidia_forums and also the “CUDA Compiler Driver NVCC” document, but i did’nt find the solution.
tankx for the help
egal

2

Which Linux distribution are you using?
Did you add the path to where the CUDA toolkit was installed to your $PATH ?

3

Hi netllama,

u have Right. it was the pfad problem.

tank u very much

Hi netllama,

u have Right. it was the PATH problem.

tank u very much,

egal

4

The command to compile is not ‘nvopencc’, but ‘nvcc’ (installed in /usr/local/cuda/bin)