#include "fft_cufft.h"
-#include <cublas_v2.h>
cuFFT::cuFFT()
{
- CublasErrorCheck(cublasCreate(&cublas));
+ CudaSafeCall(cudaSetDeviceFlags(cudaDeviceMapHost));
+ cudaErrorCheck(cublasCreate(&cublas));
+ cudaErrorCheck(cublasSetStream(cublas, cudaStreamPerThread));
}
+cufftHandle cuFFT::create_plan_fwd(uint howmany) const
+{
+ int rank = 2;
+ int n[] = {(int)m_height, (int)m_width};
+ int idist = m_height * m_width, odist = m_height * (m_width / 2 + 1);
+ int istride = 1, ostride = 1;
+ int *inembed = n, onembed[] = {(int)m_height, (int)m_width / 2 + 1};
+
+ cufftHandle plan;
+ cudaErrorCheck(cufftPlanMany(&plan, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_R2C, howmany));
+ cudaErrorCheck(cufftSetStream(plan, cudaStreamPerThread));
+ return plan;
+}
+
+cufftHandle cuFFT::create_plan_inv(uint howmany) const
+{
+ int rank = 2;
+ int n[] = {(int)m_height, (int)m_width};
+ int idist = m_height * (m_width / 2 + 1), odist = m_height * m_width;
+ int istride = 1, ostride = 1;
+ int inembed[] = {(int)m_height, (int)m_width / 2 + 1}, *onembed = n;
+
+ cufftHandle plan;
+ cudaErrorCheck(cufftPlanMany(&plan, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_C2R, howmany));
+ cudaErrorCheck(cufftSetStream(plan, cudaStreamPerThread));
+ return plan;
+}
+
+
void cuFFT::init(unsigned width, unsigned height, unsigned num_of_feats, unsigned num_of_scales)
{
Fft::init(width, height, num_of_feats, num_of_scales);
std::cout << "FFT: cuFFT" << std::endl;
- // FFT forward
- {
- int rank = 2;
- int n[] = {int(m_height), int(m_width)};
- int howmany = IF_BIG_BATCH(m_num_of_scales, 1);
- int idist = m_height * m_width, odist = m_height * (m_width / 2 + 1);
- int istride = 1, ostride = 1;
- int *inembed = n, onembed[] = {int(m_height), int(m_width) / 2 + 1};
-
- CufftErrorCheck(cufftPlanMany(&plan_f, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_R2C, howmany));
- CufftErrorCheck(cufftSetStream(plan_f, cudaStreamPerThread));
- }
+ plan_f = create_plan_fwd(1);
+ plan_fw = create_plan_fwd(m_num_of_feats);
+ plan_i_1ch = create_plan_inv(1);
- // FFT forward window
- {
- int rank = 2;
- int n[] = {int(m_height), int(m_width)};
- int howmany = m_num_of_feats * IF_BIG_BATCH(m_num_of_scales, 1);
- int idist = m_height * m_width, odist = m_height * (m_width / 2 + 1);
- int istride = 1, ostride = 1;
- int *inembed = n, onembed[] = {int(m_height), int(m_width) / 2 + 1};
-
- CufftErrorCheck(cufftPlanMany(&plan_fw, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_R2C, howmany));
- CufftErrorCheck(cufftSetStream(plan_fw, cudaStreamPerThread));
- }
- // FFT inverse all channels
- {
- int rank = 2;
- int n[] = {int(m_height), int(m_width)};
- int howmany = m_num_of_feats * IF_BIG_BATCH(m_num_of_scales, 1);
- int idist = int(m_height * (m_width / 2 + 1)), odist = 1;
- int istride = 1, ostride = m_num_of_feats * IF_BIG_BATCH(m_num_of_scales, 1);
- int inembed[] = {int(m_height), int(m_width / 2 + 1)}, *onembed = n;
-
- CufftErrorCheck(cufftPlanMany(&plan_i_features, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_C2R, howmany));
- CufftErrorCheck(cufftSetStream(plan_i_features, cudaStreamPerThread));
- }
- // FFT inverse one channel
- {
- int rank = 2;
- int n[] = {int(m_height), int(m_width)};
- int howmany = IF_BIG_BATCH(m_num_of_scales, 1);
- int idist = m_height * (m_width / 2 + 1), odist = 1;
- int istride = 1, ostride = IF_BIG_BATCH(m_num_of_scales, 1);
- int inembed[] = {int(m_height), int(m_width / 2 + 1)}, *onembed = n;
-
- CufftErrorCheck(cufftPlanMany(&plan_i_1ch, rank, n, inembed, istride, idist, onembed, ostride, odist, CUFFT_C2R, howmany));
- CufftErrorCheck(cufftSetStream(plan_i_1ch, cudaStreamPerThread));
- }
+#ifdef BIG_BATCH
+ plan_f_all_scales = create_plan_fwd(m_num_of_scales);
+ plan_fw_all_scales = create_plan_fwd(m_num_of_scales * m_num_of_feats);
+ plan_i_all_scales = create_plan_inv(m_num_of_scales);
+#endif
}
void cuFFT::set_window(const MatDynMem &window)
m_window = window;
}
-void cuFFT::forward(const MatDynMem &real_input, ComplexMat &complex_result)
+void cuFFT::forward(const MatScales &real_input, ComplexMat &complex_result)
{
Fft::forward(real_input, complex_result);
- auto in = static_cast<cufftReal *>(const_cast<MatDynMem&>(real_input).deviceMem());
-
- CufftErrorCheck(cufftExecR2C(plan_f, in, complex_result.get_p_data()));
+ auto in = static_cast<cufftReal *>(const_cast<MatScales&>(real_input).deviceMem());
+
+ if (real_input.size[0] == 1)
+ cudaErrorCheck(cufftExecR2C(plan_f, in, complex_result.get_dev_data()));
+#ifdef BIG_BATCH
+ else
+ cudaErrorCheck(cufftExecR2C(plan_f_all_scales, in, complex_result.get_dev_data()));
+#endif
}
-void cuFFT::forward_window(MatDynMem &feat, ComplexMat &complex_result, MatDynMem &temp)
+void cuFFT::forward_window(MatScaleFeats &feat, ComplexMat &complex_result, MatScaleFeats &temp)
{
Fft::forward_window(feat, complex_result, temp);
- uint n_channels = feat.size[0];
cufftReal *temp_data = temp.deviceMem();
-
- assert(feat.dims == 3);
- assert(n_channels == m_num_of_feats || n_channels == m_num_of_feats * m_num_of_scales);
-
- for (uint i = 0; i < n_channels; ++i) {
- cv::Mat feat_plane(feat.dims - 1, feat.size + 1, feat.cv::Mat::type(), feat.ptr<void>(i));
- cv::Mat temp_plane(temp.dims - 1, temp.size + 1, temp.cv::Mat::type(), temp.ptr(i));
- temp_plane = feat_plane.mul(m_window);
+ uint n_scales = feat.size[0];
+
+ for (uint s = 0; s < n_scales; ++s) {
+ for (uint ch = 0; ch < uint(feat.size[1]); ++ch) {
+ cv::Mat feat_plane = feat.plane(s, ch);
+ cv::Mat temp_plane = temp.plane(s, ch);
+ temp_plane = feat_plane.mul(m_window);
+ }
}
- CufftErrorCheck(cufftExecR2C(plan_fw, temp_data, complex_result.get_p_data()));
+
+ if (n_scales == 1)
+ cudaErrorCheck(cufftExecR2C(plan_fw, temp_data, complex_result.get_dev_data()));
+#ifdef BIG_BATCH
+ else
+ cudaErrorCheck(cufftExecR2C(plan_fw_all_scales, temp_data, complex_result.get_dev_data()));
+#endif
}
-void cuFFT::inverse(ComplexMat &complex_input, MatDynMem &real_result)
+void cuFFT::inverse(ComplexMat &complex_input, MatScales &real_result)
{
Fft::inverse(complex_input, real_result);
uint n_channels = complex_input.n_channels;
- cufftComplex *in = reinterpret_cast<cufftComplex *>(complex_input.get_p_data());
+ cufftComplex *in = reinterpret_cast<cufftComplex *>(complex_input.get_dev_data());
cufftReal *out = real_result.deviceMem();
float alpha = 1.0 / (m_width * m_height);
- if (n_channels == 1) {
- CufftErrorCheck(cufftExecC2R(plan_i_1ch, in, out));
- } else {
- CufftErrorCheck(cufftExecC2R(plan_i_features, in, out));
- }
- // TODO: Investigate whether this scalling is needed or not
- CublasErrorCheck(cublasSscal(cublas, real_result.total(), &alpha, out, 1));
+ if (n_channels == 1)
+ cudaErrorCheck(cufftExecC2R(plan_i_1ch, in, out));
+#ifdef BIG_BATCH
+ else
+ cudaErrorCheck(cufftExecC2R(plan_i_all_scales, in, out));
+#endif
+ cudaErrorCheck(cublasSscal(cublas, real_result.total(), &alpha, out, 1));
+ // The result is a cv::Mat, which will be accesses by CPU, so we
+ // must synchronize with the GPU here
+ CudaSafeCall(cudaStreamSynchronize(cudaStreamPerThread));
}
cuFFT::~cuFFT()
{
- CublasErrorCheck(cublasDestroy(cublas));
+ cudaErrorCheck(cublasDestroy(cublas));
+
+ cudaErrorCheck(cufftDestroy(plan_f));
+ cudaErrorCheck(cufftDestroy(plan_fw));
+ cudaErrorCheck(cufftDestroy(plan_i_1ch));
- CufftErrorCheck(cufftDestroy(plan_f));
- CufftErrorCheck(cufftDestroy(plan_fw));
- CufftErrorCheck(cufftDestroy(plan_i_1ch));
- CufftErrorCheck(cufftDestroy(plan_i_features));
+#ifdef BIG_BATCH
+ cudaErrorCheck(cufftDestroy(plan_f_all_scales));
+ cudaErrorCheck(cufftDestroy(plan_fw_all_scales));
+ cudaErrorCheck(cufftDestroy(plan_i_all_scales));
+#endif
}