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

Question

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 <iostream>
#include <stdio.h>
#include <cuda.h>
#include <cufft.h>

int main() {
  for (int N=1; N<1800; ++N) {
    std::cerr << "N = "<< N << " ";

    cufftResult r;
    cufftHandle planR2C;

    cudaDeviceReset();

    r = cufftCreate(&planR2C);
    if(r) return 1;
    r = cufftSetCompatibilityMode(planR2C, CUFFT_COMPATIBILITY_FFTW_PADDING);
    if(r) return 1;
    r = cufftSetAutoAllocation(planR2C, 0);
    if(r) return 1;

    size_t workSize;
    r = cufftGetSize3d(planR2C, 1800, 1800, N, CUFFT_R2C, &workSize);
    if(r==CUFFT_ALLOC_FAILED) std::cerr << "CUFFT_ALLOC_FAILED\n";

    std::cerr << " Estimated workSize: "
              << workSize / ( 1024 * 1024 )
              << " MB" << std::endl;

    cudaDeviceReset();
  }
  std::cerr << "****** Done.\n";
  return 0;
}

On a GPU with 4693 MB free memory at the start of the process, above code produces the following output:

N = 1  Estimated workSize: 197 MB
N = 2  Estimated workSize: 395 MB
...
N = 15  Estimated workSize: 791 MB
N = 16  Estimated workSize: 197 MB
N = 17 CUFFT_ALLOC_FAILED
N = 18  Estimated workSize: 222 MB
...

From N=73 on all odd N fail and even N pass. From N=166 all N fail.

Since required memory would not grow linearly with N, I assume (!) that the answer to my question indeed is: "it return[s] CUFFT_ALLOC_FAILED if a later allocation would fail" Although, a prove of that statement would be nice.

(My problem arises under CUDA 5.5.22, I have not checked any other version)

Answer 1

To mark this question answered:

Confidence among readers is high that "CUFFT_ALLOC_FAILED return value when calling cufftGetSize*()" actually means "CUFFT_ALLOC_WOULD_FAIL".