From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willem de Bruijn Subject: Re: [PATCH net-next v4 5/6] net-timestamp: TCP timestamping Date: Wed, 6 Aug 2014 10:23:25 -0400 Message-ID: References: <1407204710-21778-1-git-send-email-willemb@google.com> <1407204710-21778-6-git-send-email-willemb@google.com> <1407308749.14097.13.camel@edumazet-glaptop2.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Network Development , David Miller , Richard Cochran To: Eric Dumazet Return-path: Received: from mail-qg0-f54.google.com ([209.85.192.54]:39752 "EHLO mail-qg0-f54.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754087AbaHFOX4 (ORCPT ); Wed, 6 Aug 2014 10:23:56 -0400 Received: by mail-qg0-f54.google.com with SMTP id z60so2801834qgd.27 for ; Wed, 06 Aug 2014 07:23:55 -0700 (PDT) In-Reply-To: <1407308749.14097.13.camel@edumazet-glaptop2.roam.corp.google.com> Sender: netdev-owner@vger.kernel.org List-ID: > I believe you missed fact that a GSO packet can be split into 2 or many > skbs, when tcp_write_xmit() has to send a lower amount. > > This is done in tso_fragment() Thanks, I missed that case. I'll look into it. > Also, I am a bit worried about retransmits ? For the same reason, i.e., possibly calling tcp_fragment?