Tag Archives: clang

OpenCL Cookbook: Listing all devices and their critical attributes

Last time, in our OpenCL Cookbook series, we looked at how to list all platforms and their attributes. This time, we take the next step and list all devices that a platform provides access to and their critical attributes. This program is very useful in that it provides a quick and easy way of introspecting a given system’s OpenCL capabilities. Note that error handling has been omitted for brevity and the host language as before is C.

To recap on terminology: a platform is an OpenCL SDK such as an Apple, Intel, NVidia or AMD SDK. A device, on the other hand, may be a cpu, gpu or accelerator and as a result it’s highly likely a system will have multiple devices. For each device we list its critical attributes: hardware OpenCL version, software driver version, opencl c version supported by compiler for device and finally the number of parallel compute units (cores) it possesses which symbolises the extent of task based parallelism that we can achieve.

#include <stdio.h>                                                                                                                                               
#include <stdlib.h>
#ifdef __APPLE__
#include <OpenCL/opencl.h>
#else
#include <CL/cl.h>
#endif

int main() {

    int i, j;
    char* value;
    size_t valueSize;
    cl_uint platformCount;
    cl_platform_id* platforms;
    cl_uint deviceCount;
    cl_device_id* devices;
    cl_uint maxComputeUnits;

    // get all platforms
    clGetPlatformIDs(0, NULL, &platformCount);
    platforms = (cl_platform_id*) malloc(sizeof(cl_platform_id) * platformCount);
    clGetPlatformIDs(platformCount, platforms, NULL);

    for (i = 0; i < platformCount; i++) {

        // get all devices
        clGetDeviceIDs(platforms[i], CL_DEVICE_TYPE_ALL, 0, NULL, &deviceCount);
        devices = (cl_device_id*) malloc(sizeof(cl_device_id) * deviceCount);
        clGetDeviceIDs(platforms[i], CL_DEVICE_TYPE_ALL, deviceCount, devices, NULL);

        // for each device print critical attributes
        for (j = 0; j < deviceCount; j++) {

            // print device name
            clGetDeviceInfo(devices[j], CL_DEVICE_NAME, 0, NULL, &valueSize);
            value = (char*) malloc(valueSize);
            clGetDeviceInfo(devices[j], CL_DEVICE_NAME, valueSize, value, NULL);
            printf("%d. Device: %sn", j+1, value);
            free(value);

            // print hardware device version
            clGetDeviceInfo(devices[j], CL_DEVICE_VERSION, 0, NULL, &valueSize);
            value = (char*) malloc(valueSize);
            clGetDeviceInfo(devices[j], CL_DEVICE_VERSION, valueSize, value, NULL);
            printf(" %d.%d Hardware version: %sn", j+1, 1, value);
            free(value);

            // print software driver version
            clGetDeviceInfo(devices[j], CL_DRIVER_VERSION, 0, NULL, &valueSize);
            value = (char*) malloc(valueSize);
            clGetDeviceInfo(devices[j], CL_DRIVER_VERSION, valueSize, value, NULL);
            printf(" %d.%d Software version: %sn", j+1, 2, value);
            free(value);

            // print c version supported by compiler for device
            clGetDeviceInfo(devices[j], CL_DEVICE_OPENCL_C_VERSION, 0, NULL, &valueSize);
            value = (char*) malloc(valueSize);
            clGetDeviceInfo(devices[j], CL_DEVICE_OPENCL_C_VERSION, valueSize, value, NULL);
            printf(" %d.%d OpenCL C version: %sn", j+1, 3, value);
            free(value);

            // print parallel compute units
            clGetDeviceInfo(devices[j], CL_DEVICE_MAX_COMPUTE_UNITS,
                    sizeof(maxComputeUnits), &maxComputeUnits, NULL);
            printf(" %d.%d Parallel compute units: %dn", j+1, 4, maxComputeUnits);

        }

        free(devices);

    }

    free(platforms);
    return 0;

}

Compile and run on the Mac as follows. If you don’t have the clang, g++ or gcc commands install them. Any of those commands should work.

$ clang -framework OpenCL devices.c -o devices && ./devices

The output produced on my machine is as follows but may differ on your system.

1. Device: Intel(R) Core(TM)2 Duo CPU     U9600  @ 1.60GHz
 1.1 Hardware version: OpenCL 1.2 
 1.2 Software version: 1.1
 1.3 OpenCL C version: OpenCL C 1.2 
 1.4 Parallel compute units: 2
2. Device: GeForce 320M
 2.1 Hardware version: OpenCL 1.0 
 2.2 Software version: CLH 1.0
 2.3 OpenCL C version: OpenCL C 1.1 
 2.4 Parallel compute units: 6

As you can see my Macbook Air shows rather feeble and outdated metadata being an old slimline laptop. As always, the code is only tested on my Macbook Air but, in theory, should run on Windows and Linux though the way you compile and run will differ from above slightly. If you have any issues or would like to critique and improve my code (given I’m not a C programmer) by all means leave a comment.

Did this help you? Let me know in the comments!