I have a CUDA core that I compile into a Cuban file without any special flags:
nvcc text.cu -cubin
It compiles, although with this post:
Consultation: I can not indicate which pointer points, taking the global memory space
and a link to a line in some temporary cpp file. I can get this to work by commenting on some seemingly arbitrary code that makes no sense to me.
The core is as follows:
__global__ void string_search(char** texts, int* lengths, char* symbol, int* matches, int symbolLength)
{
int localMatches = 0;
int blockId = blockIdx.x + blockIdx.y * gridDim.x;
int threadId = threadIdx.x + threadIdx.y * blockDim.x;
int blockThreads = blockDim.x * blockDim.y;
__shared__ int localMatchCounts[32];
bool breaking = false;
for(int i = 0; i < (lengths[blockId] - (symbolLength - 1)); i += blockThreads)
{
if(texts[blockId][i] == symbol[0])
{
for(int j = 1; j < symbolLength; j++)
{
if(texts[blockId][i + j] != symbol[j])
{
breaking = true;
break;
}
}
if (breaking) continue;
localMatches++;
}
}
localMatchCounts[threadId] = localMatches;
__syncthreads();
if(threadId == 0)
{
int sum = 0;
for(int i = 0; i < 32; i++)
{
sum += localMatchCounts[i];
}
matches[blockId] = sum;
}
}
If I replace the string
localMatchCounts[threadId] = localMatches;
after the first loop for this line
localMatchCounts[threadId] = 5;
It compiles without notice. This can also be achieved by commenting on seemingly random parts of the loop above the line. I also tried replacing the local memory array with a normal array for no effect. Can someone tell me what the problem is?
Vista 64bit, .
: , , . , , , ( ).