From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH net] gro: Enter slow-path if there is no tailroom Date: Tue, 10 Jan 2017 21:29:11 -0500 (EST) Message-ID: <20170110.212911.71346127320993214.davem@davemloft.net> References: <1484079841.21472.35.camel@edumazet-glaptop3.roam.corp.google.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, herbert@gondor.apana.org.au, slavash@mellanox.com, willemb@google.com, edumazet@google.com To: eric.dumazet@gmail.com Return-path: Received: from shards.monkeyblade.net ([184.105.139.130]:34972 "EHLO shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752002AbdAKC3O (ORCPT ); Tue, 10 Jan 2017 21:29:14 -0500 In-Reply-To: <1484079841.21472.35.camel@edumazet-glaptop3.roam.corp.google.com> Sender: netdev-owner@vger.kernel.org List-ID: From: Eric Dumazet Date: Tue, 10 Jan 2017 12:24:01 -0800 > From: Herbert Xu > > The GRO path has a fast-path where we avoid calling pskb_may_pull > and pskb_expand by directly accessing frag0. However, this should > only be done if we have enough tailroom in the skb as otherwise > we'll have to expand it later anyway. > > This patch adds the check by capping frag0_len with the skb tailroom. > > Fixes: cb18978cbf45 ("gro: Open-code final pskb_may_pull") > Reported-by: Slava Shwartsman > Signed-off-by: Herbert Xu > Signed-off-by: Eric Dumazet Applied and queued up for -stable.