Messages in this thread | | | From | Tom Zanussi <> | Date | Thu, 9 Oct 2003 10:25:49 -0500 | Subject | Re: [PATCH][RFC] relayfs (1/4) (Documentation) |
| |
James Morris writes: > On Tue, 7 Oct 2003, Tom Zanussi wrote: > > > This 4-part patch contains code for an interim version of relayfs (see > > Documentation below for a description of relayfs). > > What is wrong with using Netlink sockets instead of this?
Nothing, if they meet your needs. One thing you can do with relayfs files is mmap() them. That combined with the kernel-side API, designed to make writing data into buffers and transferring it as large blocks to user-space efficient and flexible, allows for high-speed, high-volume applications which I'm not sure Netlink was designed for.
relayfs can also be used in 'packet' mode, using read(2) to read data as it becomes available, so it can be used for low-speed, low-volume applications as well. Also, some people might find the file-based approach more natural to deal with. Personal preference, I suppose.
Tom
> > > - James > -- > James Morris > <jmorris@redhat.com> > > >
-- 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/
| |