Code coverage tool for CUDA device code

Is there any code coverage tool/method available to analyze CUDA kernel code with the latest version?
I found the following info in the given link,
NVCC compiled source code now works with the code coverage tool Bullseye. The code coverage is only for the CPU or the host functions. Code coverage for device function is not supported through bullseye.

To check code coverage for device functions and kernel functions, anything tool is available with nvcc?

a google search turned this up. I know nothing about it.

As a semiautomatic tool: Nsight Compute records and shows for each source code line (and separately for PTX/SASS) the number of instructions executed.