Also remember that systemd isn’t generally doing this in series, waiting for each unit before starting the next. It’s firing off a bunch of units and then continuing what it does. If it were measuring the actual time that a unit takes without including the fact that it’s waiting for resources that other units are using, it’s highly unlikely that bare, which is basically empty, would take longer than massive snaps like Firefox and the GNOME content snaps.
Theoretically with a huge number of snaps and slow enough storage media this could have a noticeable effect, but in practice that case is highly unlikely.
Also remember that systemd isn’t generally doing this in series, waiting for each unit before starting the next. It’s firing off a bunch of units and then continuing what it does. If it were measuring the actual time that a unit takes without including the fact that it’s waiting for resources that other units are using, it’s highly unlikely that
bare
, which is basically empty, would take longer than massive snaps like Firefox and the GNOME content snaps.Theoretically with a huge number of snaps and slow enough storage media this could have a noticeable effect, but in practice that case is highly unlikely.