FK20 CUDA
|
Go to the source code of this file.
Functions | |
__global__ void | fk20_poly2hext_fft (g1p_t *hext_fft, const fr_t *polynomial, const g1p_t xext_fft[8192]) |
polynomial + xext_fft -> hext_fft More... | |
__global__ void fk20_poly2hext_fft | ( | g1p_t * | hext_fft, |
const fr_t * | polynomial, | ||
const g1p_t | xext_fft[8192] | ||
) |
polynomial + xext_fft -> hext_fft
Grid must be 1-D, 256 threads per block. Dynamic shared memory: fr_sharedmem (16384 Bytes) shared memory is used both in MSM loop and FFTs, without conflict
[out] | hext_fft | array with dimensions [gridDim.x * 16 * 512] |
[in] | polynomial | array with dimensions [gridDim.x * 16 * 512] |
[in] | xext_fft | array with dimensions [16 * 512], computed with fk20_setup2xext_fft() |
Definition at line 24 of file fk20_poly2hext_fft.cu.