How to pass a vector parameter to the OpenCL core in C?

Question

I am unable to pass a vector type parameter (uint8) to the OpenCL kernel function from host code in C.

In the host, I have data in the array:

cl_uint dataArr[8] = { 1, 2, 3, 4, 5, 6, 7, 8 }; 

(My real data is more than just [1, 8], this is just for convenience of explanation.)

Then I transfer the data to a buffer that should be passed to the kernel:

 cl_mem kernelInputData = clCreateBuffer(context, CL_MEM_READ_ONLY | CL_MEM_COPY_HOST_PTR, sizeof(cl_uint)*8, dataArr, NULL); 

Then I pass this buffer to the kernel:

 clSetKernelArg(kernel, 0, sizeof(cl_mem), &kernelInputData); 

And the kernel function signature looks something like this:

 kernel void kernelFunction(constant uint8 *vectorPtr) 

However, the kernel does not seem to receive the correct input from the pointer to kernelInputData . When I pass values ​​from inside the kernel, I see that vectorPtr points to something with this structure: ( 1, 2, 3, 4, 5, ?, ?, ? ) , Where the question marks are usually 4293848814 , but sometimes 0 . In any case, not what they should be.

What am I doing wrong?

---

EDIT:

I switched from using an array to cl_uint8 on the host side. Now I have:

cl_uint8 dataVector = { 1, 2, 3, 4, 5, 6, 7, 8 };

And I pass this vector to the kernel like this:

clSetKernelArg(kernel, 0, sizeof(cl_uint8), &dataVector);

And the kernel function signature looks something like this:

kernel void kernelFunction(constant uint8 *vectorPtr)

However, running this code gives me the CL_INVALID_ARG_SIZE error on clSetKernelArg() . This error disappears if I switch the ARG_SIZE parameter to sizeof(cl_uint8 *) , but then I get the EXC_BAD_ACCESS error in __dynamic_cast inside clSetKernelArg() .

My device:

Apple Macbook Pro (Mid-2009)
OSX 10.8 Mountain Lion
NVIDIA GeForce 9400M
Opencl 1.0
CLH 1.0

Answer 1

You define a cl_uint array of size 8. Creating cl_mem and setting the kernel argument is correct. But your kernel argument is incorrect: you are trying to read the cl_uint8 array instead of cl_uint.

If you want to use a vector data type, you must declare: cl_uint8 dataArr of size 1. Or if you want to use an array of size 8: kernel void kernelFunction(constant uint *vectorPtr, uint size):

Edit:

The kernel parameter for cl_uint8 dataVector not a pointer. So the correct code is:

 cl_uint8 dataVector = { 1, 2, 3, 4, 5, 6, 7, 8 }; clSetKernelArg(kernel, 0, sizeof(cl_uint8), &dataVector); 

and

 kernel void kernelFunction(constant uint8 vectorPtr) 

Answer 2

Minimal runnable example

int2 is passed to the kernel. It is initially created as a cl_int array.

 #include <assert.h> #include <stdlib.h> #include <CL/cl.h> int main(void) { const char *source = "__kernel void main(__global int2 *out) {\n" " out[get_global_id(0)]++;\n" "}\n"; cl_command_queue command_queue; cl_context context; cl_device_id device; cl_int input[] = {0, 1, 2, 3}; const size_t global_work_size = sizeof(input) / sizeof(cl_int2); cl_kernel kernel; cl_mem buffer; cl_platform_id platform; cl_program program; clGetPlatformIDs(1, &platform, NULL); clGetDeviceIDs(platform, CL_DEVICE_TYPE_ALL, 1, &device, NULL); context = clCreateContext(NULL, 1, &device, NULL, NULL, NULL); command_queue = clCreateCommandQueue(context, device, 0, NULL); buffer = clCreateBuffer(context, CL_MEM_READ_WRITE | CL_MEM_COPY_HOST_PTR, sizeof(input), &input, NULL); program = clCreateProgramWithSource(context, 1, &source, NULL, NULL); clBuildProgram(program, 1, &device, "", NULL, NULL); kernel = clCreateKernel(program, "main", NULL); clSetKernelArg(kernel, 0, sizeof(cl_mem), &buffer); clEnqueueNDRangeKernel(command_queue, kernel, 1, NULL, &global_work_size, NULL, 0, NULL, NULL); clFlush(command_queue); clFinish(command_queue); clEnqueueReadBuffer(command_queue, buffer, CL_TRUE, 0, sizeof(input), &input, 0, NULL, NULL); assert(input[0] == 1); assert(input[1] == 2); assert(input[2] == 3); assert(input[3] == 4); return EXIT_SUCCESS; } 

Tested on Ubuntu 15.10 OpenCL 1.2 NVIDIA 352.53.