netdev
[Top] [All Lists]
<prev [Date] next>
[Advanced]
 <prev [Thread] next>

Re: 3c59x (was Route cache performance under stress)

from [Robert Olsson] [Permanent Link][Original]
To: Andi Kleen <ak@xxxxxxx>
Subject: Re: 3c59x (was Route cache performance under stress)
From: Robert Olsson <Robert.Olsson@xxxxxxxxxxx>
Date: Wed, 11 Jun 2003 11:54:34 +0200
Cc: Bogdan Costescu <bogdan.costescu@xxxxxxxxxxxxxxxxxxxxx>, "David S. Miller" <davem@xxxxxxxxxx>, sim@xxxxxxxxxxxxx, ralph+d@xxxxxxxxx, hadi@xxxxxxxxxxxxxxxx, xerox@xxxxxxxxxx, fw@xxxxxxxxxxxxx, netdev@xxxxxxxxxxx, linux-net@xxxxxxxxxxxxxxx
In-reply-to: <20030610164949.GB13246@wotan.suse.de>
References: <20030610.085600.71109220.davem@redhat.com> <Pine.LNX.4.44.0306101815550.26879-100000@kenzo.iwr.uni-heidelberg.de> <20030610164949.GB13246@wotan.suse.de>
Sender: netdev-bounce@xxxxxxxxxxx 

Andi Kleen writes:

 > You can play some tricks with the driver to make eth_type_trans disappear
 > from the profiles. This usually helps a lot because it avoids one
 > full "fetch from cache cold memory" roundtrip per packet, which is slow on
 > any CPU.


 Andi!
 Interesting. Can we get into details?

 Cheers.

                                                --ro
[More with this subject...]
<Prev in Thread] Current Thread [Next in Thread>
Previous by Date:  Re: [patch]: CONFIG_IPV6_SUBTREES fix for MIPv6, Henrik Petander
Next by Date:  Re: 3c59x (was Route cache performance under stress), Andi Kleen
Previous by Thread:  Re: 3c59x (was Route cache performance under stress), Andi Kleen
Next by Thread:  Re: 3c59x (was Route cache performance under stress), Andi Kleen
Indexes:  [Date] [Thread] [Top] [All Lists]