Add necessary compile flag

ROCm · Jul 10, 2024 · 3b62d48 · 3b62d48
1 parent b7ffdbf
commit 3b62d48
Showing 1 changed file with 1 addition and 0 deletions.
diff --git a/setup.py b/setup.py
@@ -317,6 +317,7 @@ def validate_and_update_archs(archs):
         "nvcc":
             [
                 "-O3","-std=c++17",
+                "-mllvm", "-enable-post-misched=0",
                 "-DCK_TILE_FMHA_FWD_FAST_EXP2=1",
                 "-fgpu-flush-denormals-to-zero",
                 "-DCK_ENABLE_BF16",