53 messages in com.xensource.lists.xen-develRe: [Xen-devel] [PATCH] Network Check...
FromSent OnAttachments
Jon Mason20 May 2005 16:29 
Keir Fraser21 May 2005 07:52 
Keir Fraser21 May 2005 12:15 
Jon Mason21 May 2005 14:49 
Andrew Theurer23 May 2005 08:28 
Bin Ren23 May 2005 08:31 
Andrew Theurer23 May 2005 08:46 
Bin Ren23 May 2005 08:56 
Bin Ren23 May 2005 09:05 
Jon Mason23 May 2005 09:15 
Bin Ren23 May 2005 09:35 
Keir Fraser23 May 2005 10:54 
Bin Ren23 May 2005 11:07 
Jon Mason23 May 2005 11:17 
Keir Fraser23 May 2005 11:42 
Bin Ren23 May 2005 11:53 
Bin Ren23 May 2005 12:55 
Keir Fraser23 May 2005 13:12 
Jon Mason23 May 2005 13:20 
Ian Pratt23 May 2005 13:22 
Ian Pratt23 May 2005 13:26 
Keir Fraser23 May 2005 13:37 
Jon Mason23 May 2005 13:44 
Bin Ren23 May 2005 14:01 
Andrew Theurer23 May 2005 14:09 
Nivedita Singhvi23 May 2005 14:12 
Bin Ren23 May 2005 14:48 
Bin Ren23 May 2005 14:52 
Jon Mason23 May 2005 14:58 
Bin Ren23 May 2005 15:05 
Jon Mason23 May 2005 15:41 
Rolf Neugebauer23 May 2005 16:55 
Ian Pratt23 May 2005 16:58 
Bin Ren23 May 2005 17:38 
Ian Pratt23 May 2005 18:21 
Bin Ren23 May 2005 18:35 
Jon Mason24 May 2005 09:11 
Andrew Theurer24 May 2005 13:45 
Bin Ren24 May 2005 15:54 
Andrew Theurer25 May 2005 07:38 
Ian Pratt25 May 2005 09:48 
Jon Mason25 May 2005 10:12 
Nivedita Singhvi25 May 2005 11:19 
Ian Pratt25 May 2005 13:06 
Keir Fraser25 May 2005 14:13 
Jon Mason25 May 2005 14:35 
Cédric Schieli25 May 2005 14:37 
Keir Fraser25 May 2005 14:40 
Keir Fraser25 May 2005 14:47 
Keir Fraser25 May 2005 14:53 
Jon Mason25 May 2005 16:41 
Keir Fraser26 May 2005 01:07 
Jon Mason26 May 2005 06:36 
Subject:Re: [Xen-devel] [PATCH] Network Checksum Removal
From:Jon Mason (jdma@us.ibm.com)
Date:05/25/2005 02:35:14 PM
List:com.xensource.lists.xen-devel

On Wednesday 25 May 2005 04:14 pm, Keir Fraser wrote:

On 25 May 2005, at 21:06, Ian Pratt wrote:

There is a patch on netdev which can decrease the CPU load of bridging. specifically, it allows the bridge device to take advantage of the network device features (like hardware checksum offload). Stephen Hemminger says it should go in the 2.6.13 kernel.

Please can you post it as a patch so that we can include it in our 2.6.11 patches directory.

With the patch, csum offload will be much more interesting in the rx case

The code we already have offloads rx csums both for dom0 and domU's (the dom0 traffic has to be received through veth0 though, not the bridge device itself).

The problem with the bridge device is that all traffic generated in dom0 will be software checksummed, regardless of whether it needs to be or not. In Xen's case, it will software checksum all traffic to domU, even though the vif device is advertising NETIF_F_NO_CSUM.

This is because the stack doesn't see the features of the children devices of the bridge, only the features of the bridge device itself. I created a quick hack to work around this, and started the discussion on the Linux netdev mailing list about how to fix the problem. From this discussion, a patch was created which does most of what we want, but needs to be slightly modified to be optimal for Xen. I will post the Xen optimized patch as soon as I have it done.

Thanks, Jon

_______________________________________________ Xen-devel mailing list Xen-@lists.xensource.com http://lists.xensource.com/xen-devel