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 lists.zx2c4.com (lists.zx2c4.com [165.227.139.114]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id EB1A6C433F5 for ; Mon, 21 Mar 2022 19:34:50 +0000 (UTC) Received: by lists.zx2c4.com (OpenSMTPD) with ESMTP id f5a306a2; Mon, 21 Mar 2022 19:34:49 +0000 (UTC) Received: from mxout23.cac.washington.edu (mxout23.cac.washington.edu [140.142.32.140]) by lists.zx2c4.com (OpenSMTPD) with ESMTPS id 87b48adf (TLSv1.2:ECDHE-ECDSA-AES256-GCM-SHA384:256:NO) for ; Mon, 14 Mar 2022 17:18:16 +0000 (UTC) Received: from smtp.washington.edu (smtp.washington.edu [140.142.234.157]) by mxout23.cac.washington.edu (8.14.4+UW20.07/8.14.4+UW19.10) with ESMTP id 22EHGT4d011739 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK) for ; Mon, 14 Mar 2022 10:16:30 -0700 X-Auth-Received: from mail.ipd.uw.edu (mail.ipd.uw.edu [128.95.160.136]) (authenticated authid=ipdit) by smtp.washington.edu (8.16.1+UW21.10/8.14.4+UW19.10) with ESMTPSA id 22EHGTl2014866 (version=TLSv1.2 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT) for ; Mon, 14 Mar 2022 10:16:29 -0700 X-UW-Orig-Sender: ipdit@smtp.washington.edu Received: from [192.168.2.129] (unknown [128.95.160.246]) (Authenticated sender: lugo) by mail.ipd.uw.edu (Postfix) with ESMTPSA id 3500884904 for ; Mon, 14 Mar 2022 10:16:29 -0700 (PDT) To: wireguard@lists.zx2c4.com From: Luki Goldschmidt Subject: Performance on 10G / 1500 MTU link Message-ID: <506c5c38-8451-7982-f2fe-d9a489e71752@ipd.uw.edu> Date: Mon, 14 Mar 2022 10:16:28 -0700 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.8.1 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Content-Language: en-US X-PMX-Version: 6.4.9.2830568, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2022.3.14.170619, AntiVirus-Engine: 5.89.0, AntiVirus-Data: 2022.2.25.5890002 X-PMX-Server: mxout23.cac.washington.edu X-Uwash-Spam: Gauge=IIIIIIII, Probability=8%, Report= HTML_00_01 0.05, HTML_00_10 0.05, BODYTEXTP_SIZE_3000_LESS 0, BODY_SIZE_1000_LESS 0, BODY_SIZE_2000_LESS 0, BODY_SIZE_5000_LESS 0, BODY_SIZE_7000_LESS 0, BODY_SIZE_800_899 0, DATE_TZ_NA 0, FROM_EDU_TLD 0, NO_CTA_FOUND 0, NO_CTA_URI_FOUND 0, NO_FUR_HEADER 0, NO_URI_FOUND 0, NO_URI_HTTPS 0, SENDER_NO_AUTH 0, __BODY_NO_MAILTO 0, __CT 0, __CTE 0, __CT_TEXT_PLAIN 0, __FRAUD_PARTNERSHIP 0, __FROM_DOMAIN_NOT_IN_BODY 0, __FROM_NAME_NOT_IN_ADDR 0, __FROM_NAME_NOT_IN_BODY 0, __HAS_FROM 0, __HAS_MSGID 0, __HEADER_ORDER_FROM 0, __MIME_TEXT_ONLY 0, __MIME_TEXT_P 0, __MIME_TEXT_P1 0, __MIME_VERSION 0, __MOZILLA_USER_AGENT 0, __NO_HTML_TAG_RAW 0, __PHISH_SPEAR_STRUCTURE_1 0, __SANE_MSGID 0, __SUBJ_ALPHA_END 0, __TO_MALFORMED_2 0, __TO_NO_NAME 0, __URI_NO_MAILTO 0, __USER_AGENT 0 X-Mailman-Approved-At: Mon, 21 Mar 2022 19:34:47 +0000 X-BeenThere: wireguard@lists.zx2c4.com X-Mailman-Version: 2.1.30rc1 Precedence: list List-Id: Development discussion of WireGuard List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: wireguard-bounces@lists.zx2c4.com Sender: "WireGuard" Hi, Has anyone succeeded in saturating a 10Gb link with a MTU 1500 using WireGuard? On a LAN with 10Gb or 40Gb links, I'm getting 5-6 Gbps throughput with WireGuard (tunnel MTU 1420). Without WireGuard, I have no problem pushing 9.8 and 35 Gbps, respectively. When I increase the tunnel MTU to 8920, I can easily push 9.3 Gbps through Wireguard. I'm testing testing with iperf using a single or multiple parallel transfers. I tried kernels 5.14 and 5.15, and a range of CPUs like Intel E5-2680v4, E3-1270v6, Xeon Silver 4114. The bottleneck seems to linked to packets per second. I can't use jumbo frames over the WAN connection so MTU 1500 (link) will have to be it, but I'd love to get the most out of the 10 Gbps connection. The WAN link latency is only ~1 ms so it ought to be doable. Any tuning tips are appreciated. Luki