In pytorch/pytorch#52695 pytorch added a new signature for clamp

static at::Tensor clamp(const at::Tensor & self, const c10::optional<at::Tensor> & min, const c10::optional<at::Tensor> & max);

Pytorch / XLA already supported

static at::Tensor clamp(const at::Tensor & self, const c10::optional<at::Scalar> & min, const c10::optional<at::Scalar> & max);

and we generate clamp_out based on our implementation of the clamp. Since the new signature is added, gen script will try to generate the out_fn but it failed to find the corresponding base kernel.

The workaround would be remove the clamp_out from the gen script but that will make clamp_out with scalar fallback to cpu. I will look into lower the clamp with min and max as tensor.