Stack Overflow
Stack Overflow

Questions tagged [cufft]

cuFFT is a FFT library for CUDA enabled GPUs. Capabilities are similar to the FFTW library.

cuFFT is a FFT library for CUDA enabled GPUs. cuFFT provides functions to do various kinds of forward and reverse Fast Fourier Transforms including multidimensional transforms and batched transforms.

146 questions
1
vote
1 answer

Porting R2R FFT from FFTW to cuFFT

i'm trying to port some code from CPU to GPU that includes some FFTs. So, on CPU code some complex array is transformed using fftw_plan_many_r2r for both real and imag parts of it separately. Function foo represents R2R transform routine and called…
aleks
  • 77
  • 4
1
vote
1 answer

Multi-GPU batched 1D FFTs: only a single GPU seems to work

I have three Tesla V100s on RHEL 8 with CUDA toolkit version 10.2.89. I'm attempting to compute a batch of 1D FFTs of the columns of a row-major matrix. In the example below, the matrix is 16x8, so with three GPUs I'd expect GPU 0 to perform the…
mcfenn
  • 21
  • 3
1
vote
1 answer

cuFFT in column direction

I have a Complex matrix of nx * ny. I want to perform FFT in only column direction. One way is to transpose the entire matrix and then use cufftPlan1d to obtain FFT. Is there any other efficient way to obtain FFT without taking transpose of matrix.…
Shubham Gupta
  • 29
  • 5
1
vote
1 answer

How to schedule multiple 1d FFTs using Scikit-cuda FFT?

I'm looking to parallelize multiple 1d FFTs using CUDA. I'm working on a GTX 1050Ti with CUDA 6.1. For instance in the code I attached, I have a 3d input array 'data', and I want to do 1d FFTs over the second dimension of this array. The purpose…
Ankit_85
  • 31
  • 1
  • 4
1
vote
3 answers

skcuda.fft not the same as numpy.fft.rfft?

I was trying to test the output of an fft against a numpy fft for unit testing, I realized soon after when it failed, it wasn't because I had done something wrong, but skcuda literally doesn't produce the same answer. I knew they were going to be…
Krupip
  • 4,404
  • 2
  • 32
  • 54
1
vote
1 answer

Why the same cufft code of the following program takes different amount of time?

I ran the following code in cufft (cuda 9) (Nvidia 1080). The code is same for all execution. However, the execution time (below the code) varies a lot. Can anyone please describe how to get the lowest time always and the reason behind this…
Morshed
  • 23
  • 4
1
vote
1 answer

cuFFT batched vs single transformss

NVIDIA cuda documentation for cuFFT says "These batched transforms have higher performance than single transforms" (Read more at: http://docs.nvidia.com/cuda/cufft/index.html#ixzz57haP0Mtz Follow us: @GPUComputing on Twitter | NVIDIA on Facebook)…
JimBamFeng
  • 709
  • 1
  • 4
  • 20
1
vote
1 answer

openACC pgc++ vs pgcc, error at linking

I downloaded the example from https://www.olcf.ornl.gov/tutorials/mixing-openacc-with-gpu-libraries/ The codes are given in the above mwntioned links 1) using pgcc pgc++ -c cuFFT.cu pgcc -acc -Mcudalib=cufft fft.c cufft.o works perfectly fine 2)…
JimBamFeng
  • 709
  • 1
  • 4
  • 20
1
vote
1 answer

Result of 3D FFT using pyculib is wrong

I use pyculib to perform 3D FFT on a matrix in Anaconda 3.5. I just followed the example code posted in the website. But I found something interesting and don't understand why. Performing a 3D FFT on matrix with pyculib is correct only when using…
billinair
  • 93
  • 1
  • 11
1
vote
0 answers

pyculib fft using gpu: speed up

I am a beginner trying to learn how to use a GPU to perform high speed calculations.I am trying to implement a simple FFT program using GPU. Below is the program I used for calculating FFT using the CPU core. from time import time as timer import…
Kanmani
  • 479
  • 7
  • 21
1
vote
2 answers

Solve the Poisson equation using FFT with CUDA

I'm following a tutorial on using the cuFFT library here: http://gpgpu.org/static/sc2007/SC07_CUDA_3_Libraries.pdf After following line by line of its code, I'm getting really strange results. I have input data that is an NxN array of floats. The…
flylord1234
  • 41
  • 7
1
vote
0 answers

Building projects with cuFFT callbacks and C++11 using CMake

Building a CUDA 8.0 project with cuFFT callbacks requires using the statically linked cuFFT library and compile the code as relocatable device code using (-dc compiler option). I've been unable to make this happen with CMake v3.7.0 using…
Andreas Yankopolus
  • 852
  • 9
  • 22
1
vote
1 answer

Incorrect output when transforming from complex to real number using cuda cuFFT

I am using cuda version 7.5 cufft to perform some FFT and inverse FFT. I have a problem when performing inverse FFT using cufftExecC2R(.,.) function. Actually, when I use a batch_size = 1 in the cufftPlan1d(,) I get correct result. However, when I…
kcc__
  • 1,638
  • 4
  • 30
  • 59
1
vote
0 answers

Scikit cuda FFT large data

I recently downloaded the newest scikit for work with FFTs. However, I have run into a problem. I have data size and window size of 2^19. The size of array going into the fft function is 524288, which is far below the 2^27 element limit listed in…
SanticL
  • 56
  • 2
  • 8
1
vote
1 answer

CUDA FFT does not return the values I expect

I am currently debugging my code, where I use the CUDA FFT routines. I have something like this (please see comments for my thoughts on what I do): #include #include #include #include void…
FreddyKay
  • 275
  • 1
  • 4
  • 13
1 2
3
…
9 10