[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <568EB5B7.9090304@solarflare.com>
Date: Thu, 7 Jan 2016 19:00:07 +0000
From: Edward Cree <ecree@...arflare.com>
To: Tom Herbert <tom@...bertland.com>
CC: David Miller <davem@...emloft.net>,
Linux Kernel Network Developers <netdev@...r.kernel.org>,
<linux-net-drivers@...arflare.com>
Subject: Re: [PATCH v2 net-next 4/5] fou: enable LCO in FOU and GUE
On 07/01/16 18:51, Tom Herbert wrote:
> On Thu, Jan 7, 2016 at 9:13 AM, Edward Cree <ecree@...arflare.com> wrote:
>> @@ -787,7 +786,7 @@ int fou_build_header(struct sk_buff *skb, struct ip_tunnel_encap *e,
>> int type = csum ? SKB_GSO_UDP_TUNNEL_CSUM : SKB_GSO_UDP_TUNNEL;
>> __be16 sport;
>>
>> - skb = iptunnel_handle_offloads(skb, csum, type);
>> + skb = iptunnel_handle_offloads(skb, false, type);
> It might good to have another iptunnel_handle_offloads (lifke
> iptunnel_handle_offloads_nocsum) function that doesn't have csum
> parameter for readability.
I hope to follow up soon with a patch to use LCO for GRE, at which point I
believe that all callers will be passing false anyway, so the csum_help
parameter can be removed from the function entirely.
-ed
Powered by blists - more mailing lists