[RFC 7/8] drivers: introduce rpmsg, a remote-processor messaging bus

Pavel Machek pavel at ucw.cz
Mon Jul 18 18:07:29 EDT 2011


Hi!

> @@ -0,0 +1,75 @@
> +What:		/sys/bus/rpmsg/devices/.../name
> +Date:		June 2011
> +KernelVersion:	3.2
> +Contact:	Ohad Ben-Cohen <ohad at wizery.com>
> +Description:
> +		Every rpmsg device is a communication channel with a remote
> +		processor. Channels are identified with a (textual) name,
> +		which is maximum 32 bytes long (defined as RPMSG_NAME_SIZE in
> +		rpmsg.h).
> +
> +		This sysfs entry contains the name of this channel.
> +
> +What:		/sys/bus/rpmsg/devices/.../src
> +Date:		June 2011
> +KernelVersion:	3.2
> +Contact:	Ohad Ben-Cohen <ohad at wizery.com>
> +Description:
> +		Every rpmsg device is a communication channel with a remote
> +		processor. Channels have a local ("source") rpmsg address,
> +		and remote ("destination") rpmsg address. When an entity
> +		starts listening on one end of a channel, it assigns it with
> +		a unique rpmsg address (a 32 bits integer). This way when
> +		inbound messages arrive to this address, the rpmsg core
> +		dispatches them to the listening entity (a kernel driver).
> +
> +		This sysfs entry contains the src (local) rpmsg address
> +		of this channel. If it contains 0xffffffff, then an address
> +		wasn't assigned (can happen if no driver exists for this
> +		channel).

So this is basically networking... right? Why not implement it as
sockets? (accept, connect, read, write)?

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html



More information about the linux-arm-kernel mailing list