github/ROCm
1
1
Fork 0
mirror of https://github.com/ROCm/ROCm.git synced 2026-04-05 03:01:17 -04:00
Code Issues Packages Projects Releases Wiki Activity

[BACKEND] Optimize wgmma with accumulator source equal to 0 (#2343)

Browse Source 
Also add a test for MMA v3 reduction.
This commit is contained in:
Thomas Raoux
2023-09-20 14:05:12 -07:00
committed by GitHub
parent ed5a53057d
commit 9cab885dff
3 changed files with 35 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
python/test/unit/language/test_core.py
View File

@@ -1831,7 +1831,8 @@ layouts = [
BlockedLayout([1, 4], [8, 4], [4, 1], [0, 1], [1, 1], [1, 1], [0, 1]),
BlockedLayout([4, 4], [2, 16], [4, 1], [1, 0], [1, 1], [1, 1], [0, 1]),
MmaLayout(version=(2, 0), warps_per_cta=[4, 1], ctas_per_cga=[1, 1], cta_split_num=[1, 1], cta_order=[0, 1], instr_shape=[16, 8]),
MmaLayout(version=(2, 0), warps_per_cta=[2, 2], ctas_per_cga=[1, 1], cta_split_num=[1, 1], cta_order=[0, 1], instr_shape=[16, 8])
MmaLayout(version=(2, 0), warps_per_cta=[2, 2], ctas_per_cga=[1, 1], cta_split_num=[1, 1], cta_order=[0, 1], instr_shape=[16, 8]),
MmaLayout(version=(3, 0), warps_per_cta=[4, 1], ctas_per_cga=[1, 1], cta_split_num=[1, 1], cta_order=[1, 0], instr_shape=[16, 16, 16]),
]