Xu YANG

please merge this PR first https://github.com/apptainer/apptainer/pull/607

> This is only one piece of #412. There's still testing needed for using squashfuse and fuse2fs. It's fine to do this in multiple PRs but then don't say in...

@DrDaveD I squashed previous commits. but some tests failed. It looks like these tests failed due to some changes in the remote image. I can also reproduce this fatal locally...

@DrDaveD just to confirm, are these tests still for `apptainer build`? Besides, it looks like that putting `--fakeroot --writable-tmpfs` together is not expected? https://github.com/apptainer/apptainer/blob/22520a1f1cda6d419a8ea700720b4c93cec826be/cmd/internal/cli/build_linux.go#L188

looks like GPU-related tests are skipped in the GitHub ci build. also, I do not have a local environment for running the Nvidia GPU test.

@anderbubble It looks like it has one progress bar during push? 

@anderbubble yes, oras protocol does not implement the progress bar function for push, let me handle it. I will update the latest progress here. Thanks so much for the feature...

@johanneskoester the only thing is that LPMX's dependency, i.e. libfakechroot (https://github.com/jasonyangshadow/fakechroot), requires different configurations against different distros, recently I also migrated to build libfakechroot against old Glibc (as we know,...

> If I am not completely wrong for Bioconda and conda-forge, CentOS 7 with its glibc is currently the basis. There is even already a fakechroot package available: [anaconda.org/conda-forge/fakechroot](https://anaconda.org/conda-forge/fakechroot) Is...

> **Is your feature request related to a problem? Please describe.** Nice job. And I want to know if you can add a feature that exposes ports of container like...