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

Re: [e1000 2.6 10/11] TxDescriptors -> 1024 default

from [Ben Greear] [Permanent Link][Original]
Subject: Re: [e1000 2.6 10/11] TxDescriptors -> 1024 default
From: Ben Greear <greearb@xxxxxxxxxxxxxxx>
Date: Thu, 11 Sep 2003 16:22:35 -0700
Cc: jgarzik@xxxxxxxxx, scott.feldman@xxxxxxxxx, netdev@xxxxxxxxxxx, ricardoz@xxxxxxxxxx
In-reply-to: <20030911160252.6cd6c07d.davem@redhat.com>
Organization: Candela Technologies
References: <Pine.LNX.4.44.0309081953510.1261-100000@localhost.localdomain> <3F60CA6D.9090503@pobox.com> <3F60D0F3.8080006@candelatech.com> <20030911131219.0ab8dfdd.davem@redhat.com> <3F60DDCC.5020906@candelatech.com> <20030911140746.4f0384a1.davem@redhat.com> <3F60E947.4090005@candelatech.com> <20030911142906.74d9dfe5.davem@redhat.com> <3F60F3F7.6090203@candelatech.com> <20030911160252.6cd6c07d.davem@redhat.com>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5b) Gecko/20030827
David S. Miller wrote: 
On Thu, 11 Sep 2003 15:15:19 -0700
Ben Greear <greearb@xxxxxxxxxxxxxxx> wrote:


And the original poster shows how a similar problem slows down TCP
as well due to local dropped packets.


So, again, dampen the per-socket send queue sizes. 

That's just a band-aid to cover up the flaw with the lack
of queue-pressure feedback to the higher stacks, as would be increasing the
TxDescriptors for that matter.

--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc  http://www.candelatech.com



[More with this subject...]
<Prev in Thread] Current Thread [Next in Thread>
Previous by Date:  Re: [PATCH] ipv4 -- proc_net_fops_create, David S. Miller
Next by Date:  Re: [e1000 2.6 10/11] TxDescriptors -> 1024 default, David S. Miller
Previous by Thread:  Re: [e1000 2.6 10/11] TxDescriptors -> 1024 default, David S. Miller
Next by Thread:  Re: [e1000 2.6 10/11] TxDescriptors -> 1024 default, David S. Miller
Indexes:  [Date] [Thread] [Top] [All Lists]