netdev
[Top] [All Lists]

Re: odd behavior with r8169 and pcap

To: Francois Romieu <romieu@xxxxxxxxxxxxx>
Subject: Re: odd behavior with r8169 and pcap
From: Terry Griffin <terryg@xxxxxxxxx>
Date: Mon, 29 Nov 2004 16:01:09 -0800
Cc: linux-kernel@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <20041129231800.GB3880@xxxxxxxxxxxxxxxxxxxxxxxxxx>
References: <1101751909.2291.21.camel@xxxxxxxxxxxxxxxx> <20041129210213.GA3880@xxxxxxxxxxxxxxxxxxxxxxxxxx> <1101766059.3382.57.camel@xxxxxxxxxxxxxxxx> <20041129231800.GB3880@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
On Mon, 2004-11-29 at 15:18, Francois Romieu wrote:
> Terry Griffin <terryg@xxxxxxxxx> :
> [...]
> 
> Ok, thanks for the info. It may sound like voodoo, but...
> 
> Could you pass acpi=off to the kernel and disable whatever acpi
> or USB option appearing in the bios ?
> 
> --
> Ueimor

It was already voodoo with pcap in the causal loop. So more
voodoo doesn't have the thrill it might have otherwise.

Passing acpi=off did the trick. Throughput is at the higher rate
with or without pcap monitoring. I did not have to change any BIOS
settings.

More info previously promised:

- Removing one of the RealTek adapters did not help.
- The problem still exists in 2.6.10-rc2 and 2.6.10-rc2-bk13.

Terry



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