[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <1203531242.15017.20.camel@nimitz.home.sr71.net>
Date: Wed, 20 Feb 2008 10:14:02 -0800
From: Dave Hansen <haveblue@...ibm.com>
To: Jan-Bernd Themann <ossthema@...ibm.com>
Cc: linuxppc-dev@...abs.org, Christoph Raisch <RAISCH@...ibm.com>,
Thomas Q Klein <TKLEIN@...ibm.com>,
ossthema@...ux.vnet.ibm.com,
Jan-Bernd Themann <THEMANN@...ibm.com>,
Greg KH <greg@...ah.com>, apw <apw@...ibm.com>,
linux-kernel <linux-kernel@...r.kernel.org>,
Badari Pulavarty <pbadari@...ibm.com>,
netdev <netdev@...r.kernel.org>, tklein@...ux.ibm.com
Subject: Re: [PATCH] drivers/base: export gpl (un)register_memory_notifier
On Mon, 2008-02-18 at 11:00 +0100, Jan-Bernd Themann wrote:
> Dave Hansen <haveblue@...ibm.com> wrote on 15.02.2008 17:55:38:
>
> > I've been thinking about that, and I don't think you really *need* to
> > keep a comprehensive map like that.
> >
> > When the memory is in a particular configuration (range of memory
> > present along with unique set of holes) you get a unique ehea_bmap
> > configuration. That layout is completely predictable.
> >
> > So, if at any time you want to figure out what the ehea_bmap address for
> > a particular *Linux* virtual address is, you just need to pretend that
> > you're creating the entire ehea_bmap, use the same algorithm and figure
> > out host you would have placed things, and use that result.
> >
> > Now, that's going to be a slow, crappy linear search (but maybe not as
> > slow as recreating the silly thing). So, you might eventually run into
> > some scalability problems with a lot of packets going around. But, I'd
> > be curious if you do in practice.
>
> Up to 14 addresses translation per packet (sg_list) might be required on
> the transmit side. On receive side it is only 1. Most packets require only
> very few translations (1 or sometimes more) translations. However,
> with more then 700.000 packets per second this approach does not seem
> reasonable from performance perspective when memory is fragmented as you
> described.
OK, but let's see the data. *SHOW* me that it's slow. If the algorithm
works, then perhaps we can simply speed it up with a little caching and
*MUCH* less memory overhead.
-- Dave
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists