could not locate deviceQuery on my installation

My system details:
OS: OpenSuse 12.3
gcc: 4.7.2

I have installed the CUDA-5.0.
Here are the steps which I followed.

  1. DRIVER INSTALLATION
    Downloaded the latest driver from NVIDIA (NVIDIA-Linux-x86_64-310.40.run) and installed.

  2. SAMPLES & TOOLKIT
    sh cuda_5.0.35_linux_64_suse12.1-1.run -toolkit -samples -override

(Followed the steps from http://www.blogger.com/comment.g?blogID=7541893048189678829&postID=4613747566058581441&page=1&token=1363964696435)

Then I compiled one of the samples located in /usr/local/cuda-5.0/samples/0_Simple/matrixMul

Compilation was successful.

Then I wanted to verify my CUDA installation.
I followed http://docs.nvidia.com/cuda/cuda-getting-started-guide-for-linux/index.html

There is a instruction to run the binaries.
“After compilation, go to ~/NVIDIA_CUDA-5.0_Samples/C/bin/linux/release and run deviceQuery.”

I could not locate deviceQuery on my installation. Am I missing something?

I have added LIB path
#CUDA-5.0
export CUDA_HOME=/usr/local/cuda-5.0
export PATH=$CUDA_HOME/bin:$PATH
export LD_LIBRARY_PATH=$CUDA_HOME/lib64:/lib:$LD_LIBRARY_PATH

Kindly suggest.

ls -l ~/NVIDIA_CUDA-5.0_Samples/1_Utilities/deviceQuery
total 76
-rwxrwxr-x 1 cva cva 22588 Feb 13 23:01 deviceQuery
-rw-r--r-- 1 cva cva  9954 Feb 13 23:00 deviceQuery.cpp
-rw-rw-r-- 1 cva cva 20968 Feb 13 23:01 deviceQuery.o
-rw-r--r-- 1 cva cva  4269 Feb 13 23:00 Makefile
-rw-r--r-- 1 cva cva  1894 Feb 13 23:00 NsightEclipse.xml
-rw-r--r-- 1 cva cva   172 Feb 13 23:00 readme.txt

Couple of things to note here.

One, the samples installed in /usr/local/cuda#.#/samples are intended to be a read-only snapshot of the samples. We install a writable copy of the samples to ~/NVIDIA_CUDA-#.#_Samples for compilation, playing around, etc. They are the same, and you can really use each one as you wish.

Two, deviceQuery is just another sample. You’ll need to compile it before running it. Go to ~/NVIDIA_CUDA-5.0_Samples/1_Utilities/deviceQuery (as listed above) and run make. Then run it from the bin location where you couldn’t find it before.

After you install the cuda toolkit and the samples (assuming all requirement are fulfilled) you can go to the NVDIA_CUDA folder and compile by typing make. Same goes for each example. If you go inside the fodler and type make it will compile it.

Thanks Pasoleatis and Tamarzan for the guidance. I could locate and execute deviceQuery. I have also executed the bandwidth test also. And I suppose the installation was smooth and succesfull. Thanks

Hi !!

Upon execution of nvidia-smi command, following output is produced

±-----------------------------------------------------+
| NVIDIA-SMI 4.310.40 Driver Version: 310.40 |
|-------------------------------±---------------------±---------------------+
| GPU Name | Bus-Id Disp. | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 NVS 300 | 0000:03:00.0 N/A | N/A |
| N/A 48C N/A N/A / N/A | 3% 16MB / 511MB | N/A Default |
±------------------------------±---------------------±---------------------+
| 1 Tesla K20c | 0000:04:00.0 Off | 0 |
| 30% 34C P8 16W / 225W | 0% 13MB / 4799MB | 0% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Compute processes: GPU Memory |
| GPU PID Process name Usage |
|=============================================================================|
| 0 Not Supported |
±----------------------------------------------------------------------------+

Does the last line mean that the GPU card is nor supported.

That last line is in regards to giving you information of processes that are actively using CUDA GPU resources. I believe it is only active/supported in certain scenarios/configurations that I can’t remember offhand. Your installation is fine. BTW, you have ECC enabled on the Tesla K20c, you can disable it via nvidia-smi if you also wish.

Thanks

I have the same problem. I compiled in the NVDIA_CUDA directly but I still cannot run deviceQuery