How about the torch.compile in TransformerEngine

[south-ocean]

In PyTorch, we know that Torch.Compile will bring us a lot of benefits, and the TransformerEngine also brings performance improvements through strategies such as Transformer fusion optimization, so does the Transformer Engine also support Torch.compile? Is there any documentation on whether it is possible to get better benefits by using torch.compile in TE mode compared to non-TE mode?
Do you have suggestions for me to use torch.compile in TransformerEngine?

In llama2, we found that torch.compile can make better profits on rmsnorm and swiglu, but in TE, it is not possible to directly add torch.compile to rmsnorm and swig;u, is there any good way?