From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754498AbaIVRig (ORCPT ); Mon, 22 Sep 2014 13:38:36 -0400 Received: from shards.monkeyblade.net ([149.20.54.216]:52418 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753525AbaIVRif (ORCPT ); Mon, 22 Sep 2014 13:38:35 -0400 Date: Mon, 22 Sep 2014 13:38:31 -0400 (EDT) Message-Id: <20140922.133831.220769383344675250.davem@davemloft.net> To: clem@clem.clem-digital.net Cc: xiyou.wangcong@gmail.com, linux-kernel@vger.kernel.org, netdev@vger.kernel.org Subject: Re: Corrupt Network Packets -- 3.17-rc5/rc6 From: David Miller In-Reply-To: <201409221724.s8MHOxe5002845@clem.clem-digital.net> References: <201409221724.s8MHOxe5002845@clem.clem-digital.net> X-Mailer: Mew version 6.5 on Emacs 24.1 / Mule 6.0 (HANACHIRUSATO) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.5.7 (shards.monkeyblade.net [149.20.54.216]); Mon, 22 Sep 2014 10:38:34 -0700 (PDT) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Pete Clements Date: Mon, 22 Sep 2014 13:24:59 -0400 (EDT) > Quoting Cong Wang > > wrote: > > > Heads up fyi: system is i386 UP. > > > > > > Started getting some pop3 reception errors from the above system. > > > Investigating the problem discovered scp of the mailbox failed > > > with "Corrupted MAC on input" > > > > > > Problem started with rc5, still in rc6. Running rc4 with no > > > problem. > > > > > > Sorry for lack of better definition. > > > > > > Which driver are you using? > > > Here's the output from boot log: > > PCI: setting IRQ 11 as level-triggered > 3c59x 0000:00:0d.0: found PCI INT A -> IRQ 11 > 3c59x: Donald Becker and others. > 0000:00:0d.0: 3Com PCI 3c900 Cyclone 10Mbps TPC at d8812000. Should be fixed by: commit 8400dd029e764e30361a51a7ae35900ae0032b43 Author: Neil Horman Date: Wed Sep 17 09:04:45 2014 -0400 3c59x: Fix bad offset spec in skb_frag_dma_map Recently aded the use of skb_frag_dma_map to 3c59x, but didn't realize it automatically included the frag_offset internally, as well as provided an option to specify an extra offset in the parameter list. We need to specify an offset of 0 in the parameter list to avoid skb corruption that results in lost connections. Signed-off-by: Neil Horman CC: Linux Kernel list CC: "David S. Miller" CC: Meelis Roos Tested-by: Meelis Roos diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c index 8ab87ff..8ca49f04 100644 --- a/drivers/net/ethernet/3com/3c59x.c +++ b/drivers/net/ethernet/3com/3c59x.c @@ -2186,7 +2186,7 @@ boomerang_start_xmit(struct sk_buff *skb, struct net_device *dev) skb_frag_t *frag = &skb_shinfo(skb)->frags[i]; dma_addr = skb_frag_dma_map(&VORTEX_PCI(vp)->dev, frag, - frag->page_offset, + 0, frag->size, DMA_TO_DEVICE); if (dma_mapping_error(&VORTEX_PCI(vp)->dev, dma_addr)) {