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 NEW LPGEMM TYPE- S8S8S32/S8

Browse Source 
1. New LPGEMM type - S8S8S32/S8 is added.
2. New interface, frame and kernel files are added.
3. Frame and kernel files added/modified for S8S8S32/S8 have
   2 operations - Pack B and Mat Mul
4. Pack B kernel routines to pack B matrix for VNNI and compute the sum
   of every column of B matrix to implement the S8S8S32 operation using
   the VNNI instructions.
5. Mat Mul Kernel files to compute the GEMM output using the VNNI.
   Here the A matrix elements are converted from int8 to uint8 (VNNI
   works with A matrix type uint8 only).
6. Post GEMM computation, additional operations are performed on the
   accumulated outputs to get the correct results.
7. With this change, two new LPGEMM APIs are introduced in LPGEMM -
   s8s8s32os32 and s8s8s32os8.
8. All previously added post-ops are supported on S8S8S32/S8 also.

AMD-Internal: [CPUPL-3154]
Change-Id: Ib18f82bde557ea4a815a63adc7870c4234bfb9d3
This commit is contained in:
eashdash
2023-03-24 11:36:20 +05:30
committed by Eashan Dash
parent 58f85bb8f1
commit bd8cd763ff
28 changed files with 18575 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
addon/aocl_gemm/frame/lpgemm_5loop_interface_apis.h
View File

@@ -69,4 +69,5 @@ 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);
LPGEMM_5LOOP(int8_t,int8_t,int32_t,s8s8s32o32);
#endif // LPGEMM_5LOOP_INTF_H