Added support for column major B matrix in BF16S4F32F32 reorder API.

Browse Source

-Added new pack kernels that packs/reorders B matrix (odd strides) from
 column-major input format. This also supports the transB scenario if
 input B matrix is row major.

Change-Id: Ia0fe7e5f19ae9eba5c418f4089c7e6df11091853

...

This commit is contained in:

Deepak Negi

2024-10-04 03:06:43 +05:30

committed by sireesha.sanga

parent 064803e069

commit bd8bc77381

1 changed files with 1150 additions and 410 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

1560

kernels/zen4/lpgemm/bf16bf16f32/lpgemm_packb_bf16_s4_amd512vnni.c

File diff suppressed because it is too large Load Diff