Re: [PATCH] use NETIF_F_LLTX in bonding device

netdev

Re: [PATCH] use NETIF_F_LLTX in bonding device

To:	Arthur Kepner <akepner@xxxxxxx>
Subject:	Re: [PATCH] use NETIF_F_LLTX in bonding device
From:	"David S. Miller" <davem@xxxxxxxxxxxxx>
Date:	Tue, 22 Mar 2005 18:46:26 -0800
Cc:	netdev@xxxxxxxxxxx
In-reply-to:	<Pine.LNX.4.61.0503151016580.18478@xxxxxxxxxx>
References:	<Pine.LNX.4.61.0503151016580.18478@xxxxxxxxxx>
Sender:	netdev-bounce@xxxxxxxxxxx

On Tue, 15 Mar 2005 10:29:16 -0800 (PST)
Arthur Kepner <akepner@xxxxxxx> wrote:

> 
> Lock contention on the bonding device's xmit_lock can 
> become a bottleneck when 3 or more gige links are aggregated. 
> And it looks like it's unnecessary too, so use the 
> NETIF_F_LLTX flag to avoid grabbing this lock. 
> 
> Signed-off-by: <akepner@xxxxxxx>

Yes, bonding does all of it's own internal locking, so this
change is correct.

Applied, thanks Arthur.

[More with this subject...]

<Prev in Thread]	Current Thread	[Next in Thread>
[PATCH] use NETIF_F_LLTX in bonding device, Arthur Kepner Re: [PATCH] use NETIF_F_LLTX in bonding device, David S. Miller <=

Previous by Date:	Re: Fw: [Bug 4279] New: When I try to start vpnc the net/core/skbuff.c:91 crash, David S. Miller
Next by Date:	Re: [PATCH] net/socket.c : remove redundant NULL pointer check before kfree(), David S. Miller
Previous by Thread:	[PATCH] use NETIF_F_LLTX in bonding device, Arthur Kepner
Next by Thread:	MTU bug 2.6.8, Colin Whittaker
Indexes:	[Date] [Thread] [Top] [All Lists]