floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add tunings for baseline AArch64 

Previously this was left without `SWITCH_RATIO` or `GEMM_PREFERED_SIZE` and older default values, but it can be seen across other cores that these values seem to work for many devices.
This commit is contained in:
Chris Sidebottom 2024-08-01 17:08:37 +01:00
parent 22d8803593
commit 5f8744d4e4
1 changed files with 12 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
param.h
View File

@ -3708,6 +3708,14 @@ Until then, just keep it different than DGEMM_DEFAULT_UNROLL_N to keep copy rout
#else /* Other/undetected ARMv8 cores */
#if defined(XDOUBLE) || defined(DOUBLE)
#define SWITCH_RATIO 8
#define GEMM_PREFERED_SIZE 4
#else
#define SWITCH_RATIO 16
#define GEMM_PREFERED_SIZE 8
#endif
#define SGEMM_DEFAULT_UNROLL_M 16
#define SGEMM_DEFAULT_UNROLL_N 4
@ -3720,13 +3728,13 @@ Until then, just keep it different than DGEMM_DEFAULT_UNROLL_N to keep copy rout
#define ZGEMM_DEFAULT_UNROLL_M 4
#define ZGEMM_DEFAULT_UNROLL_N 4
#define SGEMM_DEFAULT_P 128
#define DGEMM_DEFAULT_P 160
#define SGEMM_DEFAULT_P 240
#define DGEMM_DEFAULT_P 240
#define CGEMM_DEFAULT_P 128
#define ZGEMM_DEFAULT_P 128
#define SGEMM_DEFAULT_Q 352
#define DGEMM_DEFAULT_Q 128
#define SGEMM_DEFAULT_Q 640
#define DGEMM_DEFAULT_Q 320
#define CGEMM_DEFAULT_Q 224
#define ZGEMM_DEFAULT_Q 112