2012/1/4 likunyun :
> If I do that , what's more should do in ovs?
>
> if NEED_CSUM_NORMALIZE is not defined and the skb could ge the flow in
> kernel. OVS does nothing about checksum. The kernel should finish the
> checksum (checksum_setup and checksum help) after skb being transmited from
> ov
On Jan 3, 2012, at 9:36 AM, likunyun wrote:
> I get it, you mean when we using the Xen's checksum , it always doesn't
> defined the csum->start and csum->offset in sk_buff.
> But in my kernel ,both are defined.
> Based on my kernel ,it will be better if I change it according under.
If you do t
2011/12/19 likunyun :
> In the kernel, Xen's checksum fields are defined in sk_buff, and CONFIG_XEN
> is defined in .config.
> They are used in many place , like in the front of dev_queue_xmit():
> ... ...
> if (!skb->proto_csum_blank)
> return 0;
>
> skb_checksum_setup()
> ...
>
> Kernel is is
t; From: je...@nicira.com
> Date: Thu, 15 Dec 2011 18:57:05 -0800
> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
> To: kunyu...@hotmail.com
> CC: discuss@openvswitch.org
>
> 2011/12/15 likunyun :
> >> The URL that you provided requires a username and password
Source address:
http://download.opensuse.org/repositories/Kernel:/SLE11-SP1/SUSE_SLE-11_SP1/src/
file: kernel-source-2.6.32.49-12.1.src.rpm
> From: je...@nicira.com
> Date: Thu, 15 Dec 2011 18:57:05 -0800
> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
&g
2011/12/15 likunyun :
>> The URL that you provided requires a username and password to login.
> The account is owned by my company, Maybe I could send the code after I was
> allowed taking the code home, it will take serveral days.
> Or you figure out the information you care, then I provide them.
xcept ovs finish the tso .
Why it set the wrong date in sk_buff in the compatibility code for older
kernels ?
> From: je...@nicira.com
> Date: Tue, 13 Dec 2011 21:52:50 -0800
> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
> To: kunyu...@hotmail.com
> CC: dis
On Wed, Dec 14, 2011 at 12:31 AM, Dave Whitla wrote:
> When you "enable" tx checksum on the vm you are actually enabling "tx
> checksum offloading". That is the vm delegates this to the switch. OVS
> doesn't so your syn is dropped by the receiver.
OVS understands checksum offloading.
__
t;
>>
>>> From: je...@nicira.com
>>> Date: Tue, 6 Dec 2011 09:52:46 -0800
>>> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
>>> To: kunyu...@hotmail.com
>>> CC: discuss@openvswitch.org
>
2011/12/7 likunyun :
>
>
>> From: je...@nicira.com
>> Date: Tue, 6 Dec 2011 09:52:46 -0800
>> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
>> To: kunyu...@hotmail.com
>> CC: discuss@openvswitch.org
>
>>
>> 2011/12/6 likunyun :
>
> From: je...@nicira.com
> Date: Tue, 6 Dec 2011 09:52:46 -0800
> Subject: Re: [ovs-discuss] discuss Digest, Vol 30, Issue 5
> To: kunyu...@hotmail.com
> CC: discuss@openvswitch.org
>
> 2011/12/6 likunyun :
> >> 1. Some questions a
2011/12/6 likunyun :
>> 1. Some questions about checksum in OVS (likunyun)
>>
>>
>> --
>>
>> Message: 1
>> Date: Sat, 3 Dec 2011 12:28:32 +
>> From: likunyun
>> Subject: [ovs-discuss] Some questions about checksum in OVS
>> To
> 1. Some questions about checksum in OVS (likunyun)
>
>
> --
>
> Message: 1
> Date: Sat, 3 Dec 2011 12:28:32 +
> From: likunyun
> Subject: [ovs-discuss] Some questions about checksum in OVS
> To:
> Message-ID:
> Content
13 matches
Mail list logo
