From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hannes Frederic Sowa Subject: Re: WARN trace - skb_warn_bad_offload - vxlan - large udp packet - udp checksum disabled Date: Thu, 7 Jan 2016 03:36:27 +0100 Message-ID: <568DCF2B.2090502@stressinduktion.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit To: "Nelson, Shannon" , Tom Herbert , Jesse Gross , Jiri Benc , David Miller , netdev Return-path: Received: from out5-smtp.messagingengine.com ([66.111.4.29]:45588 "EHLO out5-smtp.messagingengine.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751924AbcAGCgc (ORCPT ); Wed, 6 Jan 2016 21:36:32 -0500 Received: from compute5.internal (compute5.nyi.internal [10.202.2.45]) by mailout.nyi.internal (Postfix) with ESMTP id 7AE5320813 for ; Wed, 6 Jan 2016 21:36:31 -0500 (EST) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On 15.12.2015 02:35, Nelson, Shannon wrote: > Using a slightly modified version of udpspam (see diff below - hopefully not mangled by corporate email servers), where I set the SO_NO_CHECK socket option and can specify a large buffer size, I can reliably get the following WARN trace. I have reproduced this on both ixgbe and i40e drivers using "udpspam-no-check 6000". > > It looks to me like this is in the Tx path before we get to the actual NIC drivers, but I may be wrong. Does UFO offloading on the tunnel interface fix this error? ethtool -K vxlan ufo off If yes we can't feed SO_NO_CHECK udp packets into UFO as gso/ufo requires CHECKSUM_PARTIAL (or add some more logic into the skb or query socket status from skb_gso_segment). Thanks, Hannes