NeilBrown <ne...@suse.de> wrote: > I'm not in favour of any filesystem depending on this for correct > functionality today. As long as the filesystem isn't so large that > inum+volnum simply cannot fit in 64 bits, we should make a reasonable > effort to present them both in 64 bits.
The Auristor version of AFS (which is supported by the in-kernel afs filesystem) has a file handle (FID) that consists of a 64-bit volume ID (which is arguably a superblock-level thing), a 96-bit vnode ID (equivalent to the inode number) and a 32-bit uniquifier. I don't think the capacity of these values is fully utilised by the server... yet, but I can't support them correctly with the UAPI that we have. Allowing some expansion of stx_ino and/or replacing that with an stx_fid would be helpful in that regard. That said, getting userspace to be able to handle this is I think going to be a large undertaking with lots of auditing required. David
