example: using nvrtc kernel for aot plugin

[bowang007]

Description

Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.

Fixes # (issue)

Type of change

Please delete options that are not relevant and/or add your own.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes
• I have added the relevant labels to my PR in so that relevant reviewers are notified

[github-actions]

There are some changes that do not conform to Python style guidelines:

--- /home/runner/work/TensorRT/TensorRT/examples/dynamo/nvrtc_aot_plugin.py	2025-10-31 23:30:56.561066+00:00
+++ /home/runner/work/TensorRT/TensorRT/examples/dynamo/nvrtc_aot_plugin.py	2025-10-31 23:31:29.376720+00:00
@@ -43,11 +43,13 @@

_cuda_device = _CudaDevice()
_cuda_device.set_current()
_cuda_stream = _cuda_device.create_stream()
_program_options = _CudaProgramOptions(
-    std="c++17", arch=f"sm_{_cuda_device.arch}", include_path=["/usr/local/cuda/include"]
+    std="c++17",
+    arch=f"sm_{_cuda_device.arch}",
+    include_path=["/usr/local/cuda/include"],
)
_program = _CudaProgram(cu_code, code_type="c++", options=_program_options)
_module = _program.compile("ptx", name_expressions=("pointwise_sigmoid_kernel_nvrtc",))
_kernel = _module.get_kernel("pointwise_sigmoid_kernel_nvrtc")

@@ -121,26 +123,24 @@
    outputs: Tuple[trtp.TensorDesc],
) -> List[trtp.AutoTuneCombination]:
    return [trtp.AutoTuneCombination("FP16, FP16", "LINEAR")]


-
@trtp.aot_impl("pointwise_sigmoid_ops::pointwise_sigmoid")
def sigmoid_aot_nvrtc_impl(
    input: trtp.TensorDesc,
    outputs: Tuple[trtp.TensorDesc],
    tactic: int,
) -> Tuple[
    Union[str, bytes], Union[str, bytes], trtp.KernelLaunchParams, trtp.SymExprs
]:

-    compiled_kernel= _module.code.decode("utf-8")
+    compiled_kernel = _module.code.decode("utf-8")
    print(type(compiled_kernel))
    print(compiled_kernel)

    # import pdb; pdb.set_trace()
-

    N = input.shape_expr.numel()
    launch_params = trtp.KernelLaunchParams()
    block = 256
    launch_params.grid_x = trtp.cdiv(N, block)

[narendasan]

remove

[narendasan]

Can't we use the autogenerated registration for this?