[clug] 'Rump' filesystem. Old news or not?

Daniel Pittman daniel at rimspace.net
Sun Aug 16 21:19:34 MDT 2009

steve jenkin <sjenkin at canb.auug.org.au> writes:

> I thought this was an interesting idea - running kernel (filesystem) modules
> in Userspace.

F(ilesystems in)USE(rspace), just sayin'  Not that the hurd, and minix, and
plan9, and other systems didn't get there first.

> "Rump File Systems: Kernel Code Reborn" Antti Kantee


> When kernel functionality is desired in userspace, the common approach is to
> reimplement it for userspace interfaces. We show that use of existing kernel
> file systems in userspace programs is possible without modifying the kernel
> file system code base.

The plan9 system, and NFS, are both built on more or less this idea: plan9
used the same message passing interface for in-kernel, out-of-kernel and
out-of-machine filesystems.  NFS, of course, just exported the VFS of the time
over the network.


> The prototype of a similar framework for Linux was also implemented and
> portability was verified: Linux file systems work on NetBSD and NetBSD file
> systems work on Linux.

So, in essence, the same sort of plan9 structure.  More or less.

Not that their performance numbers are anything to sneeze at; if they can
reduce the overheads so far then they might well do to replace FUSE.


I wonder if they can manage MMAP and MMAP-vs-write coherence, though.
✣ Daniel Pittman            ✉ daniel at rimspace.net            ☎ +61 401 155 707
               ♽ made with 100 percent post-consumer electrons
   Looking for work?  Love Perl?  In Melbourne, Australia?  We are hiring.

More information about the linux mailing list