Easiest way to test for existence of cuda-capable GPU from cmake?

Question

We have some nightly build machines that have the cuda libraries installed, but which do not have a cuda-capable GPU installed. These machines are capable of building cuda-enabled programs, but they are not capable of running these programs.

In our automated nightly build process, our cmake scripts use the cmake command

find_package(CUDA)

to determine whether the cuda software is installed. This sets the cmake variable CUDA_FOUND on platforms that have cuda software installed. This is great and it works perfectly. When CUDA_FOUND is set, it is OK to build cuda-enabled programs. Even when the machine has no cuda-capable GPU.

But cuda-using test programs naturally fail on the non-GPU cuda machines, causing our nightly dashboards look "dirty". So I want cmake to avoid running those tests on such machines. But I still want to build the cuda software on those machines.

After getting a positive CUDA_FOUND result, I would like to test for the presence of an actual GPU, and then set a variable, say CUDA_GPU_FOUND, to reflect this.

What is the simplest way to get cmake to test for the presence of a cuda-capable gpu?

This needs to work on three platforms: Windows with MSVC, Mac, and Linux. (That's why we use cmake in the first place)

EDIT: There are a couple of good looking suggestions in the answers for how write a program to test for the presence of a GPU. What is still missing is the means of getting CMake to compile and run this program at configuration time. I suspect that the TRY_RUN command in CMake will be critical here, but unfortunately that command is nearly undocumented, and I cannot figure out how to make it work. This CMake part of the problem might be a much more difficult question. Perhaps I should have asked this as two separate questions...

Answer 1

The answer to this question consists of two parts:

1. A program to detect the presence of a cuda-capable GPU.
2. CMake code to compile, run, and interpret the result of that program at configuration time.

For part 1, the gpu sniffing program, I started with the answer provided by fabrizioM because it is so compact. I quickly discovered that I needed many of the details found in unknown's answer to get it to work well. What I ended up with is the following C source file, which I named has_cuda_gpu.c:

#include <stdio.h>
#include <cuda_runtime.h>

int main() {
    int deviceCount, device;
    int gpuDeviceCount = 0;
    struct cudaDeviceProp properties;
    cudaError_t cudaResultCode = cudaGetDeviceCount(&deviceCount);
    if (cudaResultCode != cudaSuccess) 
        deviceCount = 0;
    /* machines with no GPUs can still report one emulation device */
    for (device = 0; device < deviceCount; ++device) {
        cudaGetDeviceProperties(&properties, device);
        if (properties.major != 9999) /* 9999 means emulation only */
            ++gpuDeviceCount;
    }
    printf("%d GPU CUDA device(s) found\n", gpuDeviceCount);

    /* don't just return the number of gpus, because other runtime cuda
       errors can also yield non-zero return values */
    if (gpuDeviceCount > 0)
        return 0; /* success */
    else
        return 1; /* failure */
}

Notice that the return code is zero in the case where a cuda-enabled GPU is found. This is because on one of my has-cuda-but-no-GPU machines, this program generates a runtime error with non-zero exit code. So any non-zero exit code is interpreted as "cuda does not work on this machine".

You might ask why I don't use cuda emulation mode on non-GPU machines. It is because emulation mode is buggy. I only want to debug my code, and work around bugs in cuda GPU code. I don't have time to debug the emulator.

The second part of the problem is the cmake code to use this test program. After some struggle, I have figured it out. The following block is part of a larger CMakeLists.txt file:

find_package(CUDA)
if(CUDA_FOUND)
    try_run(RUN_RESULT_VAR COMPILE_RESULT_VAR
        ${CMAKE_BINARY_DIR} 
        ${CMAKE_CURRENT_SOURCE_DIR}/has_cuda_gpu.c
        CMAKE_FLAGS 
            -DINCLUDE_DIRECTORIES:STRING=${CUDA_TOOLKIT_INCLUDE}
            -DLINK_LIBRARIES:STRING=${CUDA_CUDART_LIBRARY}
        COMPILE_OUTPUT_VARIABLE COMPILE_OUTPUT_VAR
        RUN_OUTPUT_VARIABLE RUN_OUTPUT_VAR)
    message("${RUN_OUTPUT_VAR}") # Display number of GPUs found
    # COMPILE_RESULT_VAR is TRUE when compile succeeds
    # RUN_RESULT_VAR is zero when a GPU is found
    if(COMPILE_RESULT_VAR AND NOT RUN_RESULT_VAR)
        set(CUDA_HAVE_GPU TRUE CACHE BOOL "Whether CUDA-capable GPU is present")
    else()
        set(CUDA_HAVE_GPU FALSE CACHE BOOL "Whether CUDA-capable GPU is present")
    endif()
endif(CUDA_FOUND)

This sets a CUDA_HAVE_GPU boolean variable in cmake that can subsequently be used to trigger conditional operations.

It took me a long time to figure out that the include and link parameters need to go in the CMAKE_FLAGS stanza, and what the syntax should be. The try_run documentation is very light, but there is more information in the try_compile documentation, which is a closely related command. I still needed to scour the web for examples of try_compile and try_run before getting this to work.

Another tricky but important detail is the third argument to try_run, the "bindir". You should probably always set this to ${CMAKE_BINARY_DIR}. In particular, do not set it to ${CMAKE_CURRENT_BINARY_DIR} if you are in a subdirectory of your project. CMake expects to find the subdirectory CMakeFiles/CMakeTmp within bindir, and spews errors if that directory does not exist. Just use ${CMAKE_BINARY_DIR}, which is one location where those subdirectories seem to naturally reside.

Answer 2

Write a simple program like

#include<cuda.h>

int main (){
    int deviceCount;
    cudaError_t e = cudaGetDeviceCount(&deviceCount);
    return e == cudaSuccess ? deviceCount : -1;
}

and check the return value.