1 v9fs: Plan 9 Resource Sharing for Linux
2 =======================================
7 v9fs is a Unix implementation of the Plan 9 9p remote filesystem protocol.
9 This software was originally developed by Ron Minnich <rminnich@sandia.gov>
10 and Maya Gokhale. Additional development by Greg Watson
11 <gwatson@lanl.gov> and most recently Eric Van Hensbergen
12 <ericvh@gmail.com>, Latchesar Ionkov <lucho@ionkov.net> and Russ Cox
15 The best detailed explanation of the Linux implementation and applications of
16 the 9p client is available in the form of a USENIX paper:
17 http://www.usenix.org/events/usenix05/tech/freenix/hensbergen.html
19 Other applications are described in the following papers:
21 http://xcpu.org/papers/xcpu-talk.pdf
22 * KVMFS: control file system for KVM
23 http://xcpu.org/papers/kvmfs.pdf
24 * CellFS: A New Programming Model for the Cell BE
25 http://xcpu.org/papers/cellfs-talk.pdf
26 * PROSE I/O: Using 9p to enable Application Partitions
27 http://plan9.escet.urjc.es/iwp9/cready/PROSE_iwp9_2006.pdf
28 * VirtFS: A Virtualization Aware File System pass-through
34 For remote file server:
36 mount -t 9p 10.10.1.2 /mnt/9
38 For Plan 9 From User Space applications (http://swtch.com/plan9)
40 mount -t 9p `namespace`/acme /mnt/9 -o trans=unix,uname=$USER
42 For server running on QEMU host with virtio transport:
44 mount -t 9p -o trans=virtio <mount_tag> /mnt/9
46 where mount_tag is the tag associated by the server to each of the exported
47 mount points. Each 9P export is seen by the client as a virtio device with an
48 associated "mount_tag" property. Available mount tags can be
49 seen by reading /sys/bus/virtio/drivers/9pnet_virtio/virtio<n>/mount_tag files.
54 trans=name select an alternative transport. Valid options are
56 unix - specifying a named pipe mount point
57 tcp - specifying a normal TCP/IP connection
58 fd - used passed file descriptors for connection
60 virtio - connect to the next virtio channel available
61 (from QEMU with trans_virtio module)
62 rdma - connect to a specified RDMA channel
64 uname=name user name to attempt mount as on the remote server. The
65 server may override or ignore this value. Certain user
66 names may require authentication.
68 aname=name aname specifies the file tree to access when the server is
69 offering several exported file systems.
71 cache=mode specifies a caching policy. By default, no caches are used.
72 none = default no cache policy, metadata and data
73 alike are synchronous.
74 loose = no attempts are made at consistency,
75 intended for exclusive, read-only mounts
76 fscache = use FS-Cache for a persistent, read-only
78 mmap = minimal cache that is only used for read-write
79 mmap. Northing else is cached, like cache=none
81 debug=n specifies debug level. The debug level is a bitmask.
82 0x01 = display verbose error messages
83 0x02 = developer debug (DEBUG_CURRENT)
84 0x04 = display 9p trace
85 0x08 = display VFS trace
86 0x10 = display Marshalling debug
87 0x20 = display RPC debug
88 0x40 = display transport debug
89 0x80 = display allocation debug
90 0x100 = display protocol message debug
91 0x200 = display Fid debug
92 0x400 = display packet debug
93 0x800 = display fscache tracing debug
95 rfdno=n the file descriptor for reading with trans=fd
97 wfdno=n the file descriptor for writing with trans=fd
99 msize=n the number of bytes to use for 9p packet payload
101 port=n port to connect to on the remote server
103 noextend force legacy mode (no 9p2000.u or 9p2000.L semantics)
105 version=name Select 9P protocol version. Valid options are:
106 9p2000 - Legacy mode (same as noextend)
107 9p2000.u - Use 9P2000.u protocol
108 9p2000.L - Use 9P2000.L protocol
110 dfltuid attempt to mount as a particular uid
112 dfltgid attempt to mount with a particular gid
114 afid security channel - used by Plan 9 authentication protocols
116 nodevmap do not map special files - represent them as normal files.
117 This can be used to share devices/named pipes/sockets between
118 hosts. This functionality will be expanded in later versions.
120 access there are four access modes.
121 user = if a user tries to access a file on v9fs
122 filesystem for the first time, v9fs sends an
123 attach command (Tattach) for that user.
124 This is the default mode.
125 <uid> = allows only user with uid=<uid> to access
126 the files on the mounted filesystem
127 any = v9fs does single attach and performs all
128 operations as one user
129 client = ACL based access check on the 9p client
130 side for access validation
132 cachetag cache tag to use the specified persistent cache.
133 cache tags for existing cache sessions can be listed at
134 /sys/fs/9p/caches. (applies only to cache=fscache)
139 Protocol specifications are maintained on github:
140 http://ericvh.github.com/9p-rfc/
142 9p client and server implementations are listed on
143 http://9p.cat-v.org/implementations
145 A 9p2000.L server is being developed by LLNL and can be found
146 at http://code.google.com/p/diod/
148 There are user and developer mailing lists available through the v9fs project
149 on sourceforge (http://sourceforge.net/projects/v9fs).
151 News and other information is maintained on a Wiki.
152 (http://sf.net/apps/mediawiki/v9fs/index.php).
154 Bug reports are best issued via the mailing list.
156 For more information on the Plan 9 Operating System check out
157 http://plan9.bell-labs.com/plan9
159 For information on Plan 9 from User Space (Plan 9 applications and libraries
160 ported to Linux/BSD/OSX/etc) check out http://swtch.com/plan9