From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexander Duyck Subject: Re: eth_get_headlen() and unaligned accesses... Date: Thu, 09 Oct 2014 20:10:01 -0700 Message-ID: <54374E09.1020907@redhat.com> References: <20141009.201248.1210454965155680255.davem@davemloft.net> Reply-To: alexander.h.duyck@redhat.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit To: David Miller , netdev@vger.kernel.org Return-path: Received: from mx1.redhat.com ([209.132.183.28]:31784 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751081AbaJJDKE (ORCPT ); Thu, 9 Oct 2014 23:10:04 -0400 In-Reply-To: <20141009.201248.1210454965155680255.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On 10/09/2014 05:12 PM, David Miller wrote: > So, we have a bit of a problem, this is on sparc64: > > [423475.740836] Kernel unaligned access at TPC[81d330] __skb_flow_get_ports+0x70/0xe0 > [423475.755756] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.17.0+ #2 > [423475.767854] Call Trace: > [423475.772877] [0000000000433288] kernel_unaligned_trap+0x368/0x5c0 > [423475.785203] [000000000042a824] sun4v_do_mna+0x84/0xa0 > [423475.795624] [0000000000406cd0] sun4v_mna+0x5c/0x68 > [423475.805521] [000000000081d330] __skb_flow_get_ports+0x70/0xe0 > [423475.817323] [000000000081d6ac] __skb_flow_dissect+0x1ac/0x460 > [423475.829128] [0000000000843c98] eth_get_headlen+0x38/0xa0 > [423475.840083] [0000000010064d54] igb_poll+0x8d4/0xf60 [igb] > [423475.851184] [00000000008243c8] net_rx_action+0xa8/0x1c0 > > The chip DMA's to the beginning of a frag page and (unless timestamps > are enabled) that's where the ethernet header begins. > > So any larger than 16-bit access to the IP and later headers will be > unaligned. > > We have various ways we can deal with this based upon the capabilities > of the chips involved. Can we configure the IGB to put 2 "don't care" > bytes at the beginning of the packet? The problem is the igb part expects to be able to use 2K buffers which means it will always try to use the full half of a page. I had forgotten that the function this replaced had worked with unaligned accesses as all of the fields I was pulling were only 16b in width. I think I assumed that this function was already setup to handle that. Actually the fix should be pretty simple. Just do what we already appear to be doing for the iph_to_flow_copy_addrs. We can use memcpy to copy the 4 bytes for the port data instead of doing the direct assignment. I'll try to submit a patch, just need to see if I have a tree setup as it has been a couple weeks. Thanks, Alex