leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 30 Packages Projects Releases Wiki Activity

Support ElementD to be void for tma (#1153)

Browse Source 
* Support void D with AuxStore

* refine get_element_aux
This commit is contained in:
Chengquan Jiang
2024-01-16 18:15:42 -05:00
committed by GitHub
parent 751eb9a885
commit 362abbf274
6 changed files with 805 additions and 45 deletions
Download Patch File Download Diff File Expand all files Collapse all files
test/unit/gemm/device/CMakeLists.txt
+1
View File
@@ -327,6 +327,7 @@ cutlass_test_unit_add_executable(
sm90_gemm_f16_f16_f16_tensor_op_f32_cluster_warpspecialized_pingpong_reduce.cu
sm90_gemm_f16_f16_f16_tensor_op_f32_cluster_warpspecialized_cooperative_dag.cu
sm90_gemm_f16_f16_f16_tensor_op_f32_cluster_warpspecialized_pingpong_dag.cu
sm90_gemm_f16_f16_f16_tensor_op_f32_cluster_warpspecialized_cooperative_aux_store.cu
)
cutlass_test_unit_add_executable(
cutlass_test_unit_gemm_device_tensorop_cluster_multicast_sm90