Re: [PATCH v4 00/14] Add kdbus implementation

Hi

On Thu, Mar 19, 2015 at 4:48 PM, Andy Lutomirski [off-list ref] wrote:

On Thu, Mar 19, 2015 at 4:26 AM, David Herrmann [off-list ref] wrote:

quoted

metadata handling is local to the connection that sends the message.
It does not affect the overall performance of other bus operations in
parallel.

Sure it does if it writes to shared cachelines.  Given that you're
incrementing refcounts, I'm reasonable sure that you're touching lots
of shared cachelines.

Ok, sure, but it's still mostly local to the sending task. We take
locks and ref-counts on the task-struct and mm, which is for most
parts local to the CPU the task runs on. But this is inherent to
accessing this kind of data, which is the fundamental difference in
our views here, as seen below..

quoted

Furthermore, it's way faster than collecting the "same" data
via /proc, so I don't think it slows down the overall transaction at
all. If a receiver doesn't want metadata, it should not request it (by
setting the receiver-metadata-mask). If a sender doesn't like the
overhead, it should not send the metadata (by setting the
sender-metadata-mask). Only if both peers set the metadata mask, it
will be transmitted.

But you're comparing to the wrong thing, IMO.  Of course it's much
faster than /proc hackery, but it's probably much slower to do the
metadata operation once per message than to do it when you connect to
the endpoint.  (Gah!  It's a "bus" that could easily have tons of
users but a single "endpoint".  I'm still not used to it.)

Yes, of course your assumption is right if you compare against
per-connection caches, instead of per-message metadata. But we do
support _both_ use-cases, so we don't impose any policy.
We still believe "live"-metadata is a crucial feature of kdbus,
despite the known performance penalties.

Thanks
David