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
0
votes
1 answer

CUDA cuFFT API behavior in concurrent streams

I'm using CUDA 7.0 with a nVidia 980 GTX for some image processing. In a particular iteration, multiple tiles are processed independently via 15-20 kernel calls and multiple cuFFT FFT/IFFT API calls. Because of this, I've placed each tile within…
mfeuling
  • 51
  • 6
0
votes
0 answers

How to compute the throughput of cuFFT in GFLOPs

I'd like to compute the throughput of FFT and IFFT using cuFFT. I think that it should be O(B n log n) where n is the signal length and B is the batch size. The throughput in GFLOPs should be = 10^-9 * B * N * log N * c / runtime. The thing is how…
Sullivan Risk
  • 319
  • 1
  • 4
  • 21
0
votes
1 answer

cufftcomplex.h Programmers reference/Documentation

I am working on an cufft implementation and can't find any reference to the cufftcomplex functions. I found cucomplex.h through google, though, but that doesn't help me. Specifically i want to know, how to read out the imaginary part and the real…
Konno23
  • 3
  • 2
0
votes
1 answer

Recursively use of self-implemented cuIDFT.cu leads to changing output every time when re-runing the code

I have implemented a CUDA version of inverse discrete cosine transform (IDCT), by "translating" the MATLAB built-in function idct.m into CUDA: My implementation is cuIDCT.cu, works when m = n and both m and n are even numbers. cuIDCT.cu #include…
WDC
  • 334
  • 1
  • 3
  • 14
0
votes
1 answer

Debugging CUFFTW interface plan creation

I am begining to port an existing fftw3 application to make use of the cuda fftw libraries. The initial stage is to simply replace the fftw3.h header with the cufft.h header and link the cufft libraries instead of the fftw3 libraries. That is simple…
inJeans
  • 199
  • 1
  • 9
0
votes
1 answer

Wrong results in cufft

I need help with cufft, my results are wrong and I have no idea why. Here is my code: #include #include #include __global__ void print(cufftDoubleComplex *c, int h, int w){ for(int i=0; i<1; i++){ for (int j=0;…
protas
  • 617
  • 1
  • 5
  • 10
0
votes
1 answer

What is the meaning of CUFFT_ALLOC_FAILED return value when calling cufftGetSize*()?

cufftGetSize*() is not supposed to allocate any memory, and it doesn't (I checked available memory before and after calling cufftGetSize*). Does it return CUFFT_ALLOC_FAILED if a later allocation would fail? Example code: #include…
0
votes
1 answer

2D cufft transform

My first question on stackoverflow. I'm new to cuda. I simply want to perform a 2D complex-to-complex FFT. My input data is treated and no padding is needed. I just cann't get the expected result. Here's my code: #include #include…
Zonglin Gu
  • 15
  • 4
0
votes
1 answer

no cuda compatible device detected on nsight eclipse. why?

i'm writing a simple code for fast fourier transform with cufft cuda library. My source file work well with visual studio in windows7 but with eclipse nsight, in ubuntu 14.04, not work! i've installed nvidia 346.72 driver and cuda toolkit 7.0 and my…
0
votes
2 answers

about CUFFT input sizes

It's written that CUFFT library supports algorithms that higly optimized for input sizes can be written in the folowing form: 2^a X 3^b X 5^c X 7^d. How could they managed to do that? For as far as I know, FFT must provide best perfomance only…
Aleksandr Ianevski
  • 1,894
  • 1
  • 18
  • 22
0
votes
2 answers

FFT is slower on Jetson TK1?

I have written a CUDA program for Synthetic Aperture Radar Image processing. The significant portion of the computation involves finding FFTs and iFFTs and I have used cuFFT library for it. I ran my CUDA code on Jetson TK1 and on a laptop having…
Raj
  • 31
  • 3
0
votes
0 answers

CUDA CUFFT on Shield Tablet

I have tried to modify the CUDA samples on Shield Tablet by using CUFFT. However, when I build the program, I got the link errors: undefined reference to 'cufftExecR2C'; undefined reference to 'cufftExecC2R'. I thought the errors may cause by…
Yang Yang
  • 1
  • 3
0
votes
1 answer

Is it possible to overlap batched FFTs with CUDA's cuFFT library and cufftPlanMany?

I am trying to parallelize the FFT transforms of an acoustic fingerprinting library known as Chromaprint. It works by "splitting the original audio into many overlapping frames and applying the Fourier transform on them." Chromaprint uses a frame…
Harry Hedger
  • 371
  • 1
  • 2
  • 12
0
votes
1 answer

1D batched FFTs of real arrays

I have heard/read that we can use the batch mode of cuFFT if we have some n FFTs to perform of some m vectors each. So to test it, I made a sample program and ran it. The data I used was a file with some 1024 floating-point numbers as the same 1024…
Developer by Blood
  • 155
  • 1
  • 3
  • 11
0
votes
1 answer

Can we use cuFFT for processing multiple files of different sizes?

I am an amateur, working on parallelizing FFT operation/execution of multiple files together. I have, say, 1000 files each having real data of different sizes i.e. if one file has some 22000 values, other file can have 15000 values, the next one can…
Developer by Blood
  • 155
  • 1
  • 3
  • 11