From patchwork Mon Feb 11 16:04:17 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tariq Toukan X-Patchwork-Id: 1039923 X-Patchwork-Delegate: davem@davemloft.net Return-Path: X-Original-To: patchwork-incoming-netdev@ozlabs.org Delivered-To: patchwork-incoming-netdev@ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=netdev-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=fail (p=none dis=none) header.from=mellanox.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 43yrKg5zBNz9sMM for ; Tue, 12 Feb 2019 03:07:51 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727442AbfBKQHt (ORCPT ); Mon, 11 Feb 2019 11:07:49 -0500 Received: from mail-il-dmz.mellanox.com ([193.47.165.129]:51525 "EHLO mellanox.co.il" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1730064AbfBKQHs (ORCPT ); Mon, 11 Feb 2019 11:07:48 -0500 Received: from Internal Mail-Server by MTLPINE1 (envelope-from tariqt@mellanox.com) with ESMTPS (AES256-SHA encrypted); 11 Feb 2019 18:04:25 +0200 Received: from dev-l-vrt-206-006.mtl.labs.mlnx (dev-l-vrt-206-006.mtl.labs.mlnx [10.134.206.6]) by labmailer.mlnx (8.13.8/8.13.8) with ESMTP id x1BG4M7w015878; Mon, 11 Feb 2019 18:04:22 +0200 From: Tariq Toukan To: "David S. Miller" Cc: netdev@vger.kernel.org, Eran Ben Elisha , Saeed Mahameed , Eric Dumazet , Tariq Toukan Subject: [PATCH net V2] net/mlx4_en: Force CHECKSUM_NONE for short ethernet frames Date: Mon, 11 Feb 2019 18:04:17 +0200 Message-Id: <1549901057-2614-1-git-send-email-tariqt@mellanox.com> X-Mailer: git-send-email 1.8.3.1 Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org From: Saeed Mahameed When an ethernet frame is padded to meet the minimum ethernet frame size, the padding octets are not covered by the hardware checksum. Fortunately the padding octets are usually zero's, which don't affect checksum. However, it is not guaranteed. For example, switches might choose to make other use of these octets. This repeatedly causes kernel hardware checksum fault. Prior to the cited commit below, skb checksum was forced to be CHECKSUM_NONE when padding is detected. After it, we need to keep skb->csum updated. However, fixing up CHECKSUM_COMPLETE requires to verify and parse IP headers, it does not worth the effort as the packets are so small that CHECKSUM_COMPLETE has no significant advantage. Future work: when reporting checksum complete is not an option for IP non-TCP/UDP packets, we can actually fallback to report checksum unnecessary, by looking at cqe IPOK bit. Fixes: 88078d98d1bb ("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends") Cc: Eric Dumazet Signed-off-by: Saeed Mahameed Signed-off-by: Tariq Toukan Reviewed-by: Eric Dumazet --- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) Please queue for -stable >= v4.18. v2: Changed the comment per the mailing list discussion. diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index 9a0881cb7f51..6c01314e87b0 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -617,6 +617,8 @@ static int get_fixed_ipv6_csum(__wsum hw_checksum, struct sk_buff *skb, } #endif +#define short_frame(size) ((size) <= ETH_ZLEN + ETH_FCS_LEN) + /* We reach this function only after checking that any of * the (IPv4 | IPv6) bits are set in cqe->status. */ @@ -624,9 +626,20 @@ static int check_csum(struct mlx4_cqe *cqe, struct sk_buff *skb, void *va, netdev_features_t dev_features) { __wsum hw_checksum = 0; + void *hdr; + + /* CQE csum doesn't cover padding octets in short ethernet + * frames. And the pad field is appended prior to calculating + * and appending the FCS field. + * + * Detecting these padded frames requires to verify and parse + * IP headers, so we simply force all those small frames to skip + * checksum complete. + */ + if (short_frame(skb->len)) + return -EINVAL; - void *hdr = (u8 *)va + sizeof(struct ethhdr); - + hdr = (u8 *)va + sizeof(struct ethhdr); hw_checksum = csum_unfold((__force __sum16)cqe->checksum); if (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_CVLAN_PRESENT_MASK) && @@ -819,6 +832,11 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud skb_record_rx_queue(skb, cq_ring); if (likely(dev->features & NETIF_F_RXCSUM)) { + /* TODO: For IP non TCP/UDP packets when csum complete is + * not an option (not supported or any other reason) we can + * actually check cqe IPOK status bit and report + * CHECKSUM_UNNECESSARY rather than CHECKSUM_NONE + */ if ((cqe->status & cpu_to_be16(MLX4_CQE_STATUS_TCP | MLX4_CQE_STATUS_UDP)) && (cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPOK)) &&