v4.2 tag release. (#2638)

2025-09-16 00:21:53 +08:00
parent 56f0718a97
commit 6a35b4d22f
161 changed files with 14056 additions and 3793 deletions
--- a/python/CuTeDSL/base_dsl/runtime/cuda.py
+++ b/python/CuTeDSL/base_dsl/runtime/cuda.py
@@ -309,7 +309,7 @@ def get_kernel_function(module, kernel_name):
    return kernel


-def launch_kernel(kernel, grid_dims, block_dims, stream, smem_size=0, kernel_args=None):
+def launch_kernel(kernel, grid_dims, block_dims, stream, smem_size, kernel_args=None):
    """
    Launches the CUDA kernel.
    """