netdev
[Top] [All Lists]

Re: Intel and TOE in the news

To: Andi Kleen <ak@xxxxxx>
Subject: Re: Intel and TOE in the news
From: Eugene Surovegin <ebs@xxxxxxxxxxx>
Date: Sun, 20 Feb 2005 08:46:07 -0800
Cc: "David S. Miller" <davem@xxxxxxxxxxxxx>, jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <m1psywb8i4.fsf@xxxxxx>
Mail-followup-to: Andi Kleen <ak@xxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx>, jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx
References: <4216B62D.6000502@xxxxxxxxx> <20050219041007.GA17896@xxxxxxxxxxxxxxxxx> <20050219114624.373af63f.davem@xxxxxxxxxxxxx> <m1psywb8i4.fsf@xxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.5.1i
On Sat, Feb 19, 2005 at 09:27:31PM +0100, Andi Kleen wrote:
> <speculating freely>
> 
> It would be nice if the NIC could asynchronously trigger prefetches in
> the CPU. Currently a lot of the packet processing cost goes
> to waiting for read cache misses.

Just FYI :), Freescale 85xx TSECs can prefetch buffers into L2 cache. 
IIRC they call it buffer "stashing" and gianfar driver has a config 
option to enable such behavior.

But in embedded world you usually don't see flashy PR releases for 
such features :).

--
Eugene.

<Prev in Thread] Current Thread [Next in Thread>