Greg KH wrote:
On Sat, May 22, 2010 at 04:17:13PM +0200, Sven Eckelmann wrote:
From: Marek Lindner lindner_marek@yahoo.de
Add documents about new sysfs entries in
- /sys/class/net/<iface>/batman-adv/mesh_iface
- /sys/class/net/<mesh_iface>/mesh/vis_mode
Signed-off-by: Marek Lindner lindner_marek@yahoo.de Signed-off-by: Sven Eckelmann sven.eckelmann@gmx.de
.../ABI/testing/sysfs-class-net-batman-adv | 14 ++++++++++ Documentation/ABI/testing/sysfs-class-net-mesh | 27 ++++++++++++++++++++
For now, can you keep this within the drivers/staging/batman/ directory, just so everything is nice and self-contained? When it moves to the main part of the kernel tree, then it can move into the Documentation/ABI/testing/ directory, ok?
+What: /sys/class/net/<mesh_iface>/mesh/vis_mode +Date: May 2010 +Contact: Marek Lindner lindner_marek@yahoo.de +Description:
Each batman node only maintains information about itsown local neighborhood, therefore generating graphsshowing the topology of the entire mesh is not easilyfeasible without having a central instance to collectthe local topologies from all nodes. This file allowsto activate the collecting (server) mode. Onceactivated the topology can be retrieved by readingthe vis_data file.Isn't the vis_data file in debugfs now? If so, shouldn't this file move there as well? Or at least don't point to a file that doesn't exist in sysfs :)
No, vis_data is the output. vis_mode says if a node is a client or a server. A server can generate that output... but as vis_mode and vis_data are two different things, I think that not both must be moved to debugfs. For example a client should be able to move into client mode or read his current status without having debugfs compiled inside his kernel - the client will never get the output vis_data, so it isn't important for him to have debugfs. But I will remove that sentence.
Best regards, Sven