Optimisation of DTRSM and ZTRSM

Browse Source

1. Extract instruction replaced with cast when accessing first 128bit,
   as cast inst needs no cycle but extract takes few cycles
2. Added prefetch of A buffer when computing gemm operation
3. Added prefetch of C11 buffer before TRSM operation, with offset of 7 to cs_c

With above changes performance improvements observed in case of Single thread

Change-Id: Id377c490ddac8b06384acfa9a6d89dbe11bbc7be

...

This commit is contained in:

Mangala.V

2022-07-22 14:52:24 +05:30

committed by Mangala V

parent 737e08cd7a

commit 8504ef013d

1 changed files with 252 additions and 297 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

549

kernels/zen/3/bli_trsm_small.c

File diff suppressed because it is too large Load Diff