Printing from CUDA code

[gpu2024]

Hi
I am new to vllm and I am trying to learn about the CUDA implementation.
I added a print statement to the host side code of attention_kernel.cu file.

To be specific, I added the following line of code:

py::print("simple message\n");

However, running a test with this change did not print the message I added to the .cu file.
Can someone let me know what I am missing?

Thanks