gcc versions later than 10 are not supported

[astronerdF]

The nvcc compiler used by torch-radon doesn’t support GCC versions greater than 10. Unfortunately, I’m working on a system (Debian 12 "Bookworm") that only has GCC versions 11 and 12 available. Despite trying to use the -allow-unsupported-compiler flag, the compilation continues to fail due to the following error:

#error -- unsupported GNU version! gcc versions later than 10 are not supported! The nvcc flag '-allow-unsupported-compiler' can be used to override this version check; however, using an unsupported host compiler may cause compilation failure or incorrect run time execution. Use at your own risk.
Steps Taken:
I cloned the torch-radon repository and attempted to build it using the instructions provided.
I’ve modified the setup.py file to pass the -allow-unsupported-compiler flag to nvcc to bypass the GCC version check.
Even after doing this, the compilation fails with the error above.
I also attempted to install Docker to use the prebuilt image of torch-radon, but due to my system being on Debian 12, Docker’s official repository does not yet support this version.
System Information:
OS: Debian 12 (Bookworm)
GCC Versions Available: 11 and 12 (unable to downgrade easily to GCC 10)
CUDA Version: 11.x
PyTorch Version: 1.7.1 (as recommended for torch-radon)

Could you kindly provide an update or suggestion to resolve the issue? I understand that the precompiled version supports older GCC versions, but with the widespread adoption of newer GCC versions (11 and 12), it would be beneficial if the torch-radon library could support these as well.