From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH net-next 3/9] IB/ipoib: Add support for acting as VIF Date: Wed, 11 Jul 2012 11:05:12 +0300 Message-ID: <4FFD33B8.5010604@mellanox.com> References: <1341922569-4118-1-git-send-email-ogerlitz@mellanox.com> <1341922569-4118-4-git-send-email-ogerlitz@mellanox.com> <1341923180.3265.5005.camel@edumazet-glaptop> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8"; format=flowed Content-Transfer-Encoding: 7bit Cc: , , , , , Erez Shitrit To: Eric Dumazet Return-path: Received: from eu1sys200aog105.obsmtp.com ([207.126.144.119]:57607 "HELO eu1sys200aog105.obsmtp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1752631Ab2GKIGc (ORCPT ); Wed, 11 Jul 2012 04:06:32 -0400 In-Reply-To: <1341923180.3265.5005.camel@edumazet-glaptop> Sender: netdev-owner@vger.kernel.org List-ID: On 7/10/2012 3:26 PM, Eric Dumazet wrote: > On Tue, 2012-07-10 at 15:16 +0300, Or Gerlitz wrote: >> The IPoIB code checks the VIF private flag on the RX path, and accoriding >> to the value of the flag prepares the skb CB data, etc. >> > > skb->cb[] can be destroyed in netif_receive_skb() / __netif_receive_skb() > > #ifdef CONFIG_NET_CLS_ACT > skb = handle_ing(skb, &pt_prev, &ret, orig_dev); > ... > #endif Eric, Can you elaborate a little further? we're not sure to see where skb->cb[] is touched on the flow you pointed to. Or.