From patchwork Mon Jan 25 18:40:07 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Greg Kroah-Hartman X-Patchwork-Id: 372472 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-18.8 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER, INCLUDES_PATCH, MAILING_LIST_MULTI, SPF_HELO_NONE, SPF_PASS, URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id CE04FC433DB for ; Tue, 26 Jan 2021 04:52:02 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 9E7F522B3F for ; Tue, 26 Jan 2021 04:52:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728747AbhAZEvl (ORCPT ); Mon, 25 Jan 2021 23:51:41 -0500 Received: from mail.kernel.org ([198.145.29.99]:33670 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1730533AbhAYSrD (ORCPT ); Mon, 25 Jan 2021 13:47:03 -0500 Received: by mail.kernel.org (Postfix) with ESMTPSA id 502E72310E; Mon, 25 Jan 2021 18:46:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1611600393; bh=MML0RhjYqf4iDB6sL55kedkhNv5IKT2gLQybQyxIeXA=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=08AsxYBN3OchZcyWwuVqgnKeCtoL5+VPbZC1pW+LLnKc2ketv+XfzixjB4m6WgXTc 8GhjAqiUb8bfa/WxOJBRhy0Aa9OZwV4fLgUd2+A4SFwteNrSRlLf0pvsuNUd4bT60V PZjz5oa753DqsCKTFzVoelpfy3S2Jv//Kswf+3x0= From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, Eric Dumazet , Juerg Haefliger , Heiner Kallweit , Jakub Kicinski Subject: [PATCH 5.4 85/86] tcp: do not mess with cloned skbs in tcp_add_backlog() Date: Mon, 25 Jan 2021 19:40:07 +0100 Message-Id: <20210125183204.641595910@linuxfoundation.org> X-Mailer: git-send-email 2.30.0 In-Reply-To: <20210125183201.024962206@linuxfoundation.org> References: <20210125183201.024962206@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org From: Eric Dumazet commit b160c28548bc0a87cbd16d5af6d3edcfd70b8c9a upstream. Heiner Kallweit reported that some skbs were sent with the following invalid GSO properties : - gso_size > 0 - gso_type == 0 This was triggerring a WARN_ON_ONCE() in rtl8169_tso_csum_v2. Juerg Haefliger was able to reproduce a similar issue using a lan78xx NIC and a workload mixing TCP incoming traffic and forwarded packets. The problem is that tcp_add_backlog() is writing over gso_segs and gso_size even if the incoming packet will not be coalesced to the backlog tail packet. While skb_try_coalesce() would bail out if tail packet is cloned, this overwriting would lead to corruptions of other packets cooked by lan78xx, sharing a common super-packet. The strategy used by lan78xx is to use a big skb, and split it into all received packets using skb_clone() to avoid copies. The drawback of this strategy is that all the small skb share a common struct skb_shared_info. This patch rewrites TCP gso_size/gso_segs handling to only happen on the tail skb, since skb_try_coalesce() made sure it was not cloned. Fixes: 4f693b55c3d2 ("tcp: implement coalescing on backlog queue") Signed-off-by: Eric Dumazet Bisected-by: Juerg Haefliger Tested-by: Juerg Haefliger Reported-by: Heiner Kallweit Link: https://bugzilla.kernel.org/show_bug.cgi?id=209423 Link: https://lore.kernel.org/r/20210119164900.766957-1-eric.dumazet@gmail.com Signed-off-by: Jakub Kicinski Signed-off-by: Greg Kroah-Hartman --- net/ipv4/tcp_ipv4.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1657,6 +1657,7 @@ int tcp_v4_early_demux(struct sk_buff *s bool tcp_add_backlog(struct sock *sk, struct sk_buff *skb) { u32 limit = READ_ONCE(sk->sk_rcvbuf) + READ_ONCE(sk->sk_sndbuf); + u32 tail_gso_size, tail_gso_segs; struct skb_shared_info *shinfo; const struct tcphdr *th; struct tcphdr *thtail; @@ -1664,6 +1665,7 @@ bool tcp_add_backlog(struct sock *sk, st unsigned int hdrlen; bool fragstolen; u32 gso_segs; + u32 gso_size; int delta; /* In case all data was pulled from skb frags (in __pskb_pull_tail()), @@ -1689,13 +1691,6 @@ bool tcp_add_backlog(struct sock *sk, st */ th = (const struct tcphdr *)skb->data; hdrlen = th->doff * 4; - shinfo = skb_shinfo(skb); - - if (!shinfo->gso_size) - shinfo->gso_size = skb->len - hdrlen; - - if (!shinfo->gso_segs) - shinfo->gso_segs = 1; tail = sk->sk_backlog.tail; if (!tail) @@ -1718,6 +1713,15 @@ bool tcp_add_backlog(struct sock *sk, st goto no_coalesce; __skb_pull(skb, hdrlen); + + shinfo = skb_shinfo(skb); + gso_size = shinfo->gso_size ?: skb->len; + gso_segs = shinfo->gso_segs ?: 1; + + shinfo = skb_shinfo(tail); + tail_gso_size = shinfo->gso_size ?: (tail->len - hdrlen); + tail_gso_segs = shinfo->gso_segs ?: 1; + if (skb_try_coalesce(tail, skb, &fragstolen, &delta)) { TCP_SKB_CB(tail)->end_seq = TCP_SKB_CB(skb)->end_seq; @@ -1744,11 +1748,8 @@ bool tcp_add_backlog(struct sock *sk, st } /* Not as strict as GRO. We only need to carry mss max value */ - skb_shinfo(tail)->gso_size = max(shinfo->gso_size, - skb_shinfo(tail)->gso_size); - - gso_segs = skb_shinfo(tail)->gso_segs + shinfo->gso_segs; - skb_shinfo(tail)->gso_segs = min_t(u32, gso_segs, 0xFFFF); + shinfo->gso_size = max(gso_size, tail_gso_size); + shinfo->gso_segs = min_t(u32, gso_segs + tail_gso_segs, 0xFFFF); sk->sk_backlog.len += delta; __NET_INC_STATS(sock_net(sk),