From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id BBC18C433EF for ; Mon, 14 Mar 2022 01:34:45 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235747AbiCNBfo (ORCPT ); Sun, 13 Mar 2022 21:35:44 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:38366 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229932AbiCNBfm (ORCPT ); Sun, 13 Mar 2022 21:35:42 -0400 Received: from smtp.tom.com (smtprz01.163.net [106.3.154.234]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id E026725C68 for ; Sun, 13 Mar 2022 18:34:33 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by vip-app02.163.net (Postfix) with ESMTP id 54C224400C6 for ; Mon, 14 Mar 2022 09:34:32 +0800 (CST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=tom.com; s=mail; t=1647221672; bh=tCwCTuNV8vbPzPftdhc6YfcqI53uXHjpEfLiz8JtAFs=; h=Date:From:To:Cc:Subject:In-Reply-To:References:From; b=OtkTDZQB7oHyH2BgY2VnloBqnhOYt1Kp/0jAGjVujhMThs97WIUA5D+HpR21FJ6Qf wySmuuq/pld+/LxuUlW3eIdAdWGt6diH7YC7/Ngc7X6FqY7K2kWdJ3zEFKpVAgN3ml S35tZb17hrUsQmUBbsEg0egvFXxF8c5Lt622WvSg= Received: from localhost (HELO smtp.tom.com) ([127.0.0.1]) by localhost (TOM SMTP Server) with SMTP ID 1531590502 for ; Mon, 14 Mar 2022 09:34:32 +0800 (CST) X-Virus-Scanned: Debian amavisd-new at mxtest.tom.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=tom.com; s=mail; t=1647221672; bh=tCwCTuNV8vbPzPftdhc6YfcqI53uXHjpEfLiz8JtAFs=; h=Date:From:To:Cc:Subject:In-Reply-To:References:From; b=OtkTDZQB7oHyH2BgY2VnloBqnhOYt1Kp/0jAGjVujhMThs97WIUA5D+HpR21FJ6Qf wySmuuq/pld+/LxuUlW3eIdAdWGt6diH7YC7/Ngc7X6FqY7K2kWdJ3zEFKpVAgN3ml S35tZb17hrUsQmUBbsEg0egvFXxF8c5Lt622WvSg= Received: from localhost (unknown [101.93.196.13]) by antispamvip.163.net (Postfix) with ESMTPA id 96EAA154146C; Mon, 14 Mar 2022 09:34:29 +0800 (CST) Date: Mon, 14 Mar 2022 09:34:29 +0800 From: Mingbao Sun To: Sagi Grimberg Cc: Keith Busch , Jens Axboe , Christoph Hellwig , Chaitanya Kulkarni , linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org, Eric Dumazet , "David S . Miller" , Hideaki YOSHIFUJI , David Ahern , Jakub Kicinski , netdev@vger.kernel.org, tyler.sun@dell.com, ping.gan@dell.com, yanxiu.cai@dell.com, libin.zhang@dell.com, ao.sun@dell.com Subject: Re: [PATCH v2 2/3] nvme-tcp: support specifying the congestion-control Message-ID: <20220314093429.00005b95@tom.com> In-Reply-To: <7121e4be-0e25-dd5f-9d29-0fb02cdbe8de@grimberg.me> References: <20220311103414.8255-1-sunmingbao@tom.com> <20220311103414.8255-2-sunmingbao@tom.com> <7121e4be-0e25-dd5f-9d29-0fb02cdbe8de@grimberg.me> X-Mailer: Claws Mail 3.18.0 (GTK+ 2.24.33; x86_64-w64-mingw32) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Before answering the questions, I=E2=80=99d like to address the motivation behind this patchset. You know, InfiniBand/RoCE provides NVMe-oF a lossless network environment (that is zero packet loss), which is a great advantage to performance. In contrast, 'TCP/IP + ethernet' is often used as a lossy network environment (packet dropping often occurs).=20 And once packet dropping occurs, timeout-retransmission would be triggered. But once timeout-retransmission was triggered, bandwidth would drop to 0 all of a sudden. This is great damage to performance. So although NVMe/TCP may have a bandwidth competitive to that of NVMe/RDMA, but the packet dropping of the former is a flaw to its performance. However, with the combination of the following conditions, NVMe/TCP can become much more competitive to NVMe/RDMA in the data center. - Ethernet NICs supporting QoS configuration (support mapping TOS/DSCP in IP header into priority, supporting adjusting buffer size of each priority, support PFC) - Ethernet Switches supporting ECN marking, supporting adjusting buffer size of each priority. - NVMe/TCP supports specifying the tos for its TCP traffic (already implemented) - NVMe/TCP supports specifying dctcp as the congestion-control of its TCP sockets (the work of this feature) So this feature is the last item from the software aspect to form up the above combination.