netdev
[Top] [All Lists]

Re: PATCH Re: [Bonding-devel] Re: Bonding driver unreliable under high C

To: Jeff Garzik <jgarzik@xxxxxxxxxxxxxxxx>
Subject: Re: PATCH Re: [Bonding-devel] Re: Bonding driver unreliable under high CPUload
From: "Jay Vosburgh" <fubar@xxxxxxxxxx>
Date: Wed, 18 Sep 2002 17:56:15 -0700
Cc: Andrew Morton <akpm@xxxxxxxxx>, "Cureington, Tony" <tony.cureington@xxxxxx>, Pascal Brisset <pascal.brisset-ml@xxxxxxxxxx>, bonding-devel@xxxxxxxxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx
Importance: Normal
Sender: netdev-bounce@xxxxxxxxxxx
Sensitivity:

      No, it doesn't.  This is a mere band aid until somebody can get time
to do it right.  It's "less bad."

      -J


Jeff Garzik <jgarzik@xxxxxxxxxxxxxxxx> on 09/18/2002 05:48:24 PM

To:    Jay Vosburgh/Beaverton/IBM@IBMUS
cc:    Andrew Morton <akpm@xxxxxxxxx>, "Cureington, Tony"
       <tony.cureington@xxxxxx>, Pascal Brisset
       <pascal.brisset-ml@xxxxxxxxxx>, bonding-devel@xxxxxxxxxxxxxxxxxxxxx,
       netdev@xxxxxxxxxxx
Subject:    Re: PATCH Re: [Bonding-devel] Re: Bonding driver unreliable
       under high CPUload



Jay Vosburgh wrote:
>
>       Ok, here is a patch for bonding against 2.4.20-pre7.  This gets rid
> of the MII_whatever macros, wraps ioctl() to make it less bad, and
replaces
> a couple of magic numbers with friendly labels.
>
>       Comments?


The patch looks ok but doesn't address the core problem that the link
state is checked in interrupt context.

 Jeff








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