I do no longer remember the details, but the goal when I went with this approach to picking the btrfs root was to be able to snapshot the current system (assuing / contain /etc/, /var/, /usr/ and everything else needed to run the system), upgrade it, boot into the new system and conclude that the upgraded system is broken, boot it again and change the boot parameter to get back to the old system by changing subvol=@ to subvol=<oldsnapshot>.
Will this use case still work with your new approach, is is the only way to change the root to have a working system to boot into to change the default subvolume? -- Happy hacking Petter Reinholdtsen
