[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Tue, 11 Nov 2008 00:37:37 -0800
From: "Waskiewicz Jr, Peter P" <peter.p.waskiewicz.jr@...el.com>
To: Jeff Garzik <jeff@...zik.org>,
"Kirsher, Jeffrey T" <jeffrey.t.kirsher@...el.com>
CC: "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"davem@...emloft.net" <davem@...emloft.net>
Subject: RE: [PATCH] ixgbe: Fix usage of netif_*_all_queues() with
netif_carrier_{off|on}()
> Jeff Kirsher wrote:
>> From: PJ Waskiewicz <peter.p.waskiewicz.jr@...el.com>
>>
>> netif_carrier_off() is sufficient to stop Tx into the driver. Stopping the Tx
>> queues is redundant and unnecessary. By the same token, netif_carrier_on()
>> will be sufficient to re-enable Tx, so waking the queues is unnecessary.
> Is this a fix as the subject says (2.6.28), or an optimization as the
> description implies (2.6.29)?
> AFAICS from a quick glance, things are not actually _broken_, are they?
> Jeff
This can be applied to 2.6.29; it doesn't need 2.6.28 inclusion. Nothing is broken here, it's just poor usage. DaveM pointed this out to us awhile ago, and I've finally gotten around to getting the simple patch out to correct the unneeded overhead.
Cheers,
-PJ Waskiewicz
Powered by blists - more mailing lists