Age | Commit message (Collapse) | Author |
|
case of fp16 arguments are cast on host and in kernel
|
|
verbose print statements to the cache
|
|
|
|
|
|
|
|
declspec(dllimport) when not building the library
|
|
|
|
|
|
|
|
|
|
|
|
and/or transposing
|
|
routines
|
|
|
|
using 'make test'
|
|
|
|
|
|
|
|
|
|
|
|
HGEMV/HGBMV/HHEMV/HHBMV/HHPMV/HSYMV/HSBMV/HSPMV/HTRMV/HTBMV/HTPMV
|
|
HSWAP/HSCAL/HCOPY/HAXPY/HDOT/HNRM2/HASUM/HSUM/iHAMAX/iHMAX/iHMIN
|
|
|
|
|
|
|
|
|
|
|
|
|
|
for large power-of-2 kernels on AMD GPUs
|
|
|
|
|
|
|
|
CPU BLAS library
|
|
|
|
|
|
|
|
and IxAMAX
|
|
ClearCompiledProgramCache function to clear the cache
|
|
|
|
|
|
library
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|