Merge pull request codeplaysoftware#164 from t4c1/online_softmax

Add cutlass 3 implementation of online softmax that works on sm80 to example 35. --------- Co-authored-by: Finlay <[email protected]> Co-authored-by: Alejandro Acosta <[email protected]>
aacostadiaz · Dec 11, 2024 · c76cf17 · c76cf17
2 parents 6fe64d2 + 561a82c
commit c76cf17
Show file tree

Hide file tree

Showing 8 changed files with 1,655 additions and 1 deletion.
diff --git a/examples/35_gemm_softmax/CMakeLists.txt b/examples/35_gemm_softmax/CMakeLists.txt
@@ -29,8 +29,14 @@
 
 
 
+if (NOT CUTLASS_ENABLE_SYCL)
 cutlass_example_add_executable(
   35_gemm_softmax
   gemm_softmax.cu
   )
-
+else()
+cutlass_example_add_executable(
+  35_gemm_online_softmax
+  gemm_online_softmax.cpp
+  )
+endif()