mkadavil a5c4a8c7e0 Int4 B matrix reordering support in LPGEMM. ...

Support for reordering B matrix of datatype int4 as per the pack schema
requirements of u8s8s32 kernel. Vectorized int4_t -> int8_t conversion
implemented via leveraging the vpmultishiftqb instruction. The reordered
B matrix will then be used in the u8s8s32o<s32|s8> api.

AMD-Internal: [SWLCSG-2390]
Change-Id: I3a8f8aba30cac0c4828a31f1d27fa1b45ea07bba

2024-06-24 07:55:34 -04:00

..

lpgemm_packa.h

Added Trans A feature for all INT8 LPGEMM APIs

2024-01-30 03:40:56 -05:00

lpgemm_packb.h

Int4 B matrix reordering support in LPGEMM.

2024-06-24 07:55:34 -04:00