I have written a simple example to use the new cuFFT callback feature of CUDA 6.5, but it is not working. The full code is the following:
#include "cuda_runtime.h"
#include "device_launch_parameters.h"
#include <stdio.h>
#include <assert.h>
#include <cufft.h>
#include <cufftXt.h>
//#define DEBUG
#define BLOCKSIZE 256
#define NN 16
/**********/
/* iDivUp */
/**********/
int iDivUp(int a, int b) { return ((a % b) != 0) ? (a / b + 1) : (a / b); }
/********************/
/* CUDA ERROR CHECK */
/********************/
#define gpuErrchk(ans) { gpuAssert((ans), __FILE__, __LINE__); }
inline void gpuAssert(cudaError_t code, const char *file, int line, bool abort=true)
{
if (code != cudaSuccess)
{
fprintf(stderr,"GPUassert: %s %s %d\n", cudaGetErrorString(code), file, line);
if (abort) exit(code);
}
}
/*********************/
/* CUFFT ERROR CHECK */
/*********************/
// See http://stackoverflow.com/questions/16267149/cufft-error-handling
#ifdef _CUFFT_H_
// cuFFT API errors
static const char *_cudaGetErrorEnum(cufftResult error)
{
switch (error)
{
case CUFFT_SUCCESS:
return "CUFFT_SUCCESS";
case CUFFT_INVALID_PLAN:
return "CUFFT_INVALID_PLAN";
case CUFFT_ALLOC_FAILED:
return "CUFFT_ALLOC_FAILED";
case CUFFT_INVALID_TYPE:
return "CUFFT_INVALID_TYPE";
case CUFFT_INVALID_VALUE:
return "CUFFT_INVALID_VALUE";
case CUFFT_INTERNAL_ERROR:
return "CUFFT_INTERNAL_ERROR";
case CUFFT_EXEC_FAILED:
return "CUFFT_EXEC_FAILED";
case CUFFT_SETUP_FAILED:
return "CUFFT_SETUP_FAILED";
case CUFFT_INVALID_SIZE:
return "CUFFT_INVALID_SIZE";
case CUFFT_UNALIGNED_DATA:
return "CUFFT_UNALIGNED_DATA";
}
return "<unknown>";
}
#endif
#define cufftSafeCall(err) __cufftSafeCall(err, __FILE__, __LINE__)
inline void __cufftSafeCall(cufftResult err, const char *file, const int line)
{
if( CUFFT_SUCCESS != err) {
fprintf(stderr, "CUFFT error in file '%s', line %d\n %s\nerror %d: %s\nterminating!\n",__FILE__, __LINE__,err, \
_cudaGetErrorEnum(err)); \
cudaDeviceReset(); assert(0); \
}
}
/****************************************/
/* FFTSHIFT 1D INPLACE MEMORY MOVEMENTS */
/****************************************/
__global__ void fftshift_1D_inplace_memory_movements(float2* d_inout, unsigned int N)
{
unsigned int tid = threadIdx.x + blockIdx.x * blockDim.x;
if (tid < N/2)
{
printf("%f %f\n", 2.f*d_inout[tid].x, 3.f*d_inout[tid].x);
float2 temp = d_inout[tid];
d_inout[tid] = d_inout[tid + (N / 2)];
d_inout[tid + (N / 2)] = temp;
}
}
/**********************************************/
/* FFTSHIFT 1D INPLACE CHESSBOARD - VERSION 1 */
/**********************************************/
__device__ float2 fftshift_1D_chessboard_callback_v1(void *d_in, size_t offset, void *callerInfo, void *sharedPtr) {
float a = (float)(1-2*(offset%2));
float2 out = ((float2*)d_in)[offset];
out.x = out.x * a;
out.y = out.y * a;
return out;
}
__device__ cufftCallbackLoadC fftshift_1D_chessboard_callback_v1_Ptr = fftshift_1D_chessboard_callback_v1;
/********/
/* MAIN */
/********/
int main()
{
const int N = 16;
// --- Host side input array
float2 *h_vect = (float2 *)malloc(N*sizeof(float2));
for (int i=0; i<N; i++) {
h_vect[i].x = (float)rand() / (float)RAND_MAX;
h_vect[i].y = (float)rand() / (float)RAND_MAX;
}
// --- Host side output arrays
float2 *h_out1 = (float2 *)malloc(N*sizeof(float2));
float2 *h_out2 = (float2 *)malloc(N*sizeof(float2));
// --- Device side input arrays
float2 *d_vect1; gpuErrchk(cudaMalloc((void**)&d_vect1, N*sizeof(float2)));
float2 *d_vect2; gpuErrchk(cudaMalloc((void**)&d_vect2, N*sizeof(float2)));
gpuErrchk(cudaMemcpy(d_vect1, h_vect, N*sizeof(float2), cudaMemcpyHostToDevice));
gpuErrchk(cudaMemcpy(d_vect2, h_vect, N*sizeof(float2), cudaMemcpyHostToDevice));
// --- Device side output arrays
float2 *d_out1; gpuErrchk(cudaMalloc((void**)&d_out1, N*sizeof(float2)));
float2 *d_out2; gpuErrchk(cudaMalloc((void**)&d_out2, N*sizeof(float2)));
float time;
cudaEvent_t start, stop;
cudaEventCreate(&start);
cudaEventCreate(&stop);
/*******************************************/
/* cuFFT + MEMORY MOVEMENTS BASED FFTSHIFT */
/*******************************************/
cufftHandle planinverse; cufftSafeCall(cufftPlan1d(&planinverse, N, CUFFT_C2C, 1));
cudaEventRecord(start, 0);
cufftSafeCall(cufftExecC2C(planinverse, d_vect1, d_vect1, CUFFT_INVERSE));
fftshift_1D_inplace_memory_movements<<<iDivUp(N/2, BLOCKSIZE), BLOCKSIZE>>>(d_vect1, N);
#ifdef DEBUG
gpuErrchk(cudaPeekAtLastError());
gpuErrchk(cudaDeviceSynchronize());
#endif
cudaEventRecord(stop, 0);
cudaEventSynchronize(stop);
cudaEventElapsedTime(&time, start, stop);
printf("Memory movements elapsed time: %3.3f ms \n", time);
gpuErrchk(cudaMemcpy(h_out1, d_vect1, N*sizeof(float2), cudaMemcpyDeviceToHost));
/****************************************/
/* CHESSBOARD MULTIPLICATION V1 + cuFFT */
/****************************************/
cufftCallbackLoadC hfftshift_1D_chessboard_callback_v1_Ptr;
gpuErrchk(cudaMemcpyFromSymbol(&hfftshift_1D_chessboard_callback_v1_Ptr, fftshift_1D_chessboard_callback_v1_Ptr, sizeof(hfftshift_1D_chessboard_callback_v1_Ptr)));
cufftHandle planinverse_v1; cufftSafeCall(cufftPlan1d(&planinverse_v1, N, CUFFT_C2C, 1));
cufftResult status = cufftXtSetCallback(planinverse_v1, (void **)&hfftshift_1D_chessboard_callback_v1_Ptr, CUFFT_CB_LD_COMPLEX, 0);
if (status == CUFFT_LICENSE_ERROR) {
printf("This sample requires a valid license file.\n");
printf("The file was either not found, out of date, or otherwise invalid.\n");
exit(EXIT_FAILURE);
} else {
cufftSafeCall(status);
}
cudaEventRecord(start, 0);
cufftSafeCall(cufftExecC2C(planinverse_v1, d_vect2, d_out2, CUFFT_INVERSE));
cudaEventRecord(stop, 0);
cudaEventSynchronize(stop);
cudaEventElapsedTime(&time, start, stop);
printf("Chessboard v1 elapsed time: %3.3f ms \n", time);
gpuErrchk(cudaMemcpy(h_out2, d_out2, N*sizeof(float2), cudaMemcpyDeviceToHost));
// --- Checking the results
for (int i=0; i<N; i++) if ((h_out1[i].x != h_out2[i].x)||(h_out1[i].y != h_out2[i].y)) { printf("Chessboard v1 test failed!\n"); return 0; }
printf("Chessboard v1 test passed!\n");
return 0;
}
The test is not passed. If I check the calculated values, they are completely messy.
I’m working on a Linux machine with Ubuntu 12.04 equipped with Kepler K20c cards and, of course, CUDA 6.5. I’m compiling with the following lines
nvcc -ccbin g++ -dc -m64 -o kernel.o -c kernel.cu
nvcc -ccbin g++ -m64 -o cufft_callbacks kernel.o -lcufft_static -lculibos
The strange thing that I notice is that if I use printf within the device callback functions, then any floating point operation I’m performing is returning messy values. Opposite to that, printf returns correct values for offset and integer operations on offset (e.g., 2*offset), but returns always 0s for offset&1. I have already tried to specify the architecture within the compilation line, but the situation didn’t change.
Any clue?