amd/blis
1
0
Fork 0
mirror of https://github.com/amd/blis.git synced 2026-05-03 22:11:12 +00:00
Code Issues Packages Projects Releases Wiki Activity

Added low precision GEMM - bf16bf16f32of32

Browse Source 
Feature Addition: Added a new variant of low precision GEMM to addon - BFloat16. The kernel takes bf16 type inputs and perform BF16 GEMM operations. The intermediate accumulation and output are in float.

1. Compute kernels will perform computations only if B matrix is reordered in accordance with the usage of AVX-512 BF16 instruction - dpbf16_ps
2. Kernel for packing B matrix is provided

Change-Id: If5d08213068869eff060c9998596d2d2703a6793
This commit is contained in:
eashdash
2022-08-17 08:25:30 +00:00
committed by Eashan Dash
parent 219c41ded9
commit 4e3e00fb7e
23 changed files with 7942 additions and 734 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
addon/aocl_gemm/frame/lpgemm_5loop_interface_apis.h
View File

@@ -37,6 +37,7 @@
#include "lpgemm_types.h"
#include "lpgemm_post_ops.h"
#include "aocl_bf16_type.h"
#define LPGEMM_5LOOP(A_type,B_type,C_type,LP_SFX) \
void lpgemm_rowvar_ ## LP_SFX \
@@ -64,4 +65,5 @@ void lpgemm_rowvar_ ## LP_SFX \
LPGEMM_5LOOP(uint8_t,int8_t,int32_t,u8s8s32o32);
LPGEMM_5LOOP(uint8_t,int8_t,int16_t,u8s8s16o16);
LPGEMM_5LOOP(float,float,float,f32f32f32of32);
LPGEMM_5LOOP(bfloat16,bfloat16,float,bf16bf16f32of32);
#endif // LPGEMM_5LOOP_INTF_H