Quantcast
Channel: Community : All Content - OpenCL
Viewing all articles
Browse latest Browse all 2400

VGPR usage question (code attached)

$
0
0

I have a simple test kernel which I compile in kernelanalyzer2. It says on Tahiti it will use 110 VGPR. I tried to put #pragma unroll 1 and it does not have any effect at all. Is there a known way to avoid compiler from using so many registers? ( keep in mind this is a dummy test kernel, but this seems to effect an actual kernel and reduce occupancy)

 

__kernel void test (__global double *distrValues, __global double *distrValuesOut) {    __private const int id = get_global_id(0);    __private double den2;        __private int i;   for (i=0;i<53;i++) {       den2 += distrValues[i];   }    distrValuesOut[id]=den2;
}

 

 

Thanks!


Viewing all articles
Browse latest Browse all 2400

Latest Images

Trending Articles