clock.cu:28:30: fatal error: helper_functions.h: No such file or directory #include <helper_func

Hi, I am very new to cuda, I have installed a clean install of ubuntu 14.04.1 64bit and and GT 730 GPU

Wed Mar 9 17:46:08 2016
±-----------------------------------------------------+
| NVIDIA-SMI 352.63 Driver Version: 352.63 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce GT 730 Off | 0000:01:00.0 N/A | N/A |
| 40% 36C P8 N/A / N/A | 193MiB / 1023MiB | N/A Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 Not Supported |
±----------------------------------------------------------------------------+
mark@mark-K30AD-M31AD-M51AD:~/Downloads$ lspci | grep -i nvidia
01:00.0 VGA compatible controller: NVIDIA Corporation GK208 [GeForce GT 730] (rev a1)
01:00.1 Audio device: NVIDIA Corporation GK208 HDMI/DP Audio Controller (rev a1

I have installed the CUDA Toolkit 7.5 and used the default settings for the samples, all installed with no error’s.

The problems I have is when i try to compile the samples:

Example…

mark@mark-K30AD-M31AD-M51AD:~/NVIDIA_CUDA-7.5_Samples/0_Simple/clock$ nvcc clock.cu
clock.cu:28:30: fatal error: helper_functions.h: No such file or directory
#include <helper_functions.h>

so i checked the location of the includes in the make file:

INCLUDES  := -I../../common/inc

so i had a look to make sure the path was correct:

mark@mark-K30AD-M31AD-M51AD:~/NVIDIA_CUDA-7.5_Samples/0_Simple/clock$ cd …/…/common/inc
mark@mark-K30AD-M31AD-M51AD:~/NVIDIA_CUDA-7.5_Samples/common/inc$ ls -l
total 432
-rw-r–r-- 1 paddy paddy 20993 Mar 9 17:41 cuda_drvapi_dynlink.c
-rw-r–r-- 1 paddy paddy 11015 Mar 9 17:41 drvapi_error_string.h
drwxr-xr-x 2 paddy paddy 4096 Mar 9 17:41 dynlink
-rw-r–r-- 1 paddy paddy 14883 Mar 9 17:41 dynlink_d3d10.h
-rw-r–r-- 1 paddy paddy 6299 Mar 9 17:41 dynlink_d3d11.h
-rw-r–r-- 1 paddy paddy 5106 Mar 9 17:41 exception.h
drwxr-xr-x 2 paddy paddy 4096 Mar 9 17:41 GL
-rw-r–r-- 1 paddy paddy 15894 Mar 9 17:41 helper_cuda_drvapi.h
-rw-r–r-- 1 paddy paddy 4675 Mar 9 17:41 helper_cuda_gl.h
-rw-r–r-- 1 paddy paddy 36774 Mar 9 17:41 helper_cuda.h
-rw-r–r-- 1 paddy paddy 4159 Mar 9 17:41 helper_cusolver.h
-rw-r–r-- 1 paddy paddy 1168 Mar 9 17:41 helper_functions.h

The list goes on… but sure enough my ‘helper_functions.h’ is where I believe it should be, I am missing something but I have looked high and low on forums and the only info i find is to check that the includes files are in the right location.

any help is really appreciated.

mark

This:

nvcc clock.cu

is not how you compile the samples.

Use the make file provided in /usr/local/cuda/samples. Probably the same as the one in your NVIDIA_CUDA-7.5_Samples directory.

Hi, thanks for the reply, I tried the make…

make clock.cu
make: Nothing to be done for `clock.cu’

and I get the above message, is there something else?

mark

Hi,

Pls try just “make”(or sudo make) to see whether it can work.

thanks,

1 Like

I came across the same issue, and digged into the Makefile. If you have copied the example folder from the source CUDA installation folder to somewhere else on your machine (or for whatever other reason that the paths are not what the Makefile expects), then the include folder is not well recognized. To fix it, you need to edit the variable INCLUDES in your Makefile; To find the relevant line, do

$> grep -n INCLUDES Makefile

After the fix, the build went OK for me.
I hope this helped.

2 Likes

Very good, solved my problem.