Messages in this thread | | | From | Tom Zanussi <> | Date | Mon, 13 Oct 2003 09:53:09 -0500 | Subject | Re: [PATCH][RFC] relayfs (1/4) (Documentation) |
| |
David S. Miller writes: > On Fri, 10 Oct 2003 10:41:29 -0400 > Karim Yaghmour <karim@opersys.com> wrote: > > > The question isn't whether netlink can transfer hundreds of thousands of > > data units in one fell swoop. The question is: is it more efficient than > > relayfs at this? > > Wrong, it's the queueing model that's important for applications > like this. >
relayfs isn't trying to provide a generic queueing model - it's basically just an efficient buffering mechanism with hooks for kernel-user data transfer. It's a lower-level thing than netlink and might even be of use to netlink as a buffering layer.
In any case, applications like tracing or kernel debugging don't have a need for more of a queueing model than the in-order delivery and event buffering capabilities relayfs provides, and since applications like these either can't use netlink or would benefit from the efficiency provided by a no-frills buffering scheme, maybe there is actually a use for something like relayfs.
-- Regards,
Tom Zanussi <zanussi@us.ibm.com> IBM Linux Technology Center/RAS
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |