From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sowmini Varadhan Subject: Re: [Xen-devel] xennet_start_xmit assumptions Date: Wed, 25 Jan 2017 10:45:46 -0500 Message-ID: <20170125154546.GF2212@oracle.com> References: <20170118153132.GB9258@oracle.com> <20170118192528.GA6847@char.us.oracle.com> <20170119111426.GA22018@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Konrad Rzeszutek Wilk , Wei Liu , "netdev@vger.kernel.org" , "xen-devel@lists.xenproject.org" To: Paul Durrant Return-path: Received: from aserp1040.oracle.com ([141.146.126.69]:45562 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750768AbdAYPqA (ORCPT ); Wed, 25 Jan 2017 10:46:00 -0500 Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On (01/25/17 15:06), Paul Durrant wrote: > > Making netfront cope with a fully non-linear skb looks like it would > be quite intrusive and probably not worth it so I opted for just doing > the ETH_HLEN pull-tail if necessary. Can you check it works for you? I tested it, and it works fine, but note that DaveM's comments in this thread: the DKI is that we *must* provide at least the hard_header_len in the non-paged part of the skb. So might not even be necessary to handle the fully non-linear skb (though it's probably prudent to check and bail for this, as your patch does) I just posted an RFC patch for fixing the pf_packet layer, just in case other drivers like xen_netfront dont explicitly check for this http://patchwork.ozlabs.org/patch/719236/