Pre-compute masks for 4xn kernels

Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
This commit is contained in:
Rafael Ravedutti
2023-03-28 22:30:30 +02:00
parent 04ade6bcec
commit 5c000444a4
3 changed files with 58 additions and 26 deletions

View File

@@ -126,6 +126,8 @@ typedef struct {
MD_FLOAT *diagonal_2xnn_j_minus_i;
unsigned int masks_2xnn_hn[8];
unsigned int masks_2xnn_fn[8];
unsigned int masks_4xn_hn[16];
unsigned int masks_4xn_fn[16];
} Atom;
extern void initAtom(Atom*);