From nobody Sun Jun 20 21:29:22 2021 X-Original-To: freebsd-arm@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 3072B11E5CCC for ; Sun, 20 Jun 2021 21:31:15 +0000 (UTC) (envelope-from denis@ovsienko.info) Received: from aibo.runbox.com (aibo.runbox.com [91.220.196.211]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 4G7Qns5QQQz4vVn for ; Sun, 20 Jun 2021 21:31:13 +0000 (UTC) (envelope-from denis@ovsienko.info) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=ovsienko.info; s=selector2; h=Content-Transfer-Encoding:Content-Type: MIME-Version:References:In-Reply-To:Message-ID:Subject:To:From:Date; bh=+KNNIEIc0qxqrVQk3oOk/KB4qQaNmWifxBH/IcYO8/I=; b=hHuTTG/k0RJvQqmBrq1QMKdF6o +sRqsUKyuCcw8DqIm6RQzHy4ATWshjqOAkRK4wI3qjEOY7t0CLgN8BX653x2oRVmNHae7iv5VlhqZ ONu5Am2isyEabPCWHurlwE3nqC9+E+OkRM1JTZf7I6qMMALeHToHx+N6JnB1v7Lr5rvxQhyY4Q9Fn MXMJwb4KRWSMvM7Zil2zskFk59AWtFxOajhqx6hOEfgZHtxRWksiMzeZrZrwVP6gOJx05QcorPwTk 9J47xRzLntmJfnL0cu0sLzcgpHfz7yEdwmdYImi/Zuy6faqFIdQHrv6ol96Nkju1Fo63Z64KqFfDs JWsTQ0sQ==; Received: from [10.9.9.72] (helo=submission01.runbox) by mailtransmit02.runbox with esmtp (Exim 4.86_2) (envelope-from ) id 1lv527-00010M-HK for freebsd-arm@freebsd.org; Sun, 20 Jun 2021 23:31:11 +0200 Received: by submission01.runbox with esmtpsa [Authenticated ID (984599)] (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) id 1lv51z-0008RV-8T for freebsd-arm@freebsd.org; Sun, 20 Jun 2021 23:31:03 +0200 Date: Sun, 20 Jun 2021 22:29:22 +0100 From: Denis Ovsienko To: freebsd-arm@freebsd.org Subject: Re: Raspberry Pi 3B and pitiful network speeds Message-ID: <20210620222922.51da1818@basepc> In-Reply-To: <20210620154105.0c83bbcc@basepc> References: <20210620144513.1f91a68f@basepc> <169baf0b-3f3c-f1dc-4a6f-b8a0ef863f51@denninger.net> <20210620154105.0c83bbcc@basepc> List-Id: Porting FreeBSD to ARM processors List-Archive: https://lists.freebsd.org/archives/freebsd-arm List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-arm@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Rspamd-Queue-Id: 4G7Qns5QQQz4vVn X-Spamd-Bar: --- Authentication-Results: mx1.freebsd.org; dkim=pass header.d=ovsienko.info header.s=selector2 header.b=hHuTTG/k; dmarc=none; spf=pass (mx1.freebsd.org: domain of denis@ovsienko.info designates 91.220.196.211 as permitted sender) smtp.mailfrom=denis@ovsienko.info X-Spamd-Result: default: False [-3.10 / 15.00]; ARC_NA(0.00)[]; RBL_DBL_DONT_QUERY_IPS(0.00)[91.220.196.211:from]; R_DKIM_ALLOW(-0.20)[ovsienko.info:s=selector2]; RCVD_IN_DNSWL_LOW(-0.10)[91.220.196.211:from]; FROM_HAS_DN(0.00)[]; RWL_MAILSPIKE_GOOD(0.00)[91.220.196.211:from]; TO_MATCH_ENVRCPT_ALL(0.00)[]; R_SPF_ALLOW(-0.20)[+ip4:91.220.196.211]; MIME_GOOD(-0.10)[text/plain]; TO_DN_NONE(0.00)[]; DMARC_NA(0.00)[ovsienko.info]; RCPT_COUNT_ONE(0.00)[1]; SPAMHAUS_ZRD(0.00)[91.220.196.211:from:127.0.2.255]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; DKIM_TRACE(0.00)[ovsienko.info:+]; NEURAL_HAM_SHORT(-1.00)[-1.000]; NEURAL_HAM_LONG(-1.00)[-1.000]; FROM_EQ_ENVFROM(0.00)[]; MID_RHS_NOT_FQDN(0.50)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:50304, ipnet:91.220.196.0/24, country:NO]; RCVD_TLS_LAST(0.00)[]; MAILMAN_DEST(0.00)[freebsd-arm]; RCVD_COUNT_TWO(0.00)[2] X-ThisMailContainsUnwantedMimeParts: N On Sun, 20 Jun 2021 15:41:05 +0100 Denis Ovsienko wrote: > Notwithstanding your point, under right conditions NetBSD 9.2 can > fully utilize the RPI3B's 100Mbit/s Ethernet interface, at least in > one direction at a time. Hopefully that means it should be possible in > FreeBSD too. As it turns out, it is indeed possible with FreeBSD. In mostly-simplex tests the RPI3B managed to download at 11.4MB/s or to upload at 10.2MB/s. A simultaneous two-way test resulted in 7.6MB/s download and 7.0MB/s upload, so it looks like the 100Mb/s duplex PHY internally connects through a 150Mb/s simplex bottleneck or some such. The numbers come from a quick nc+dd test using 1000MB of pseudo-random data, but I remember observing similar results with iperf earlier (cannot remember which OS). The tests were done using FreeBSD 13.0-RELEASE [1] and RPI3B rev A22082 with proper power and cooling. The SD card was a basic one. dd was reading from /dev/urandom and writing to /dev/null. As a separate note, in my setup ifconfig tells "100baseTX ", which is expected for RPI3B. In the original message it tells "1000baseT ", which means the board is RPI3B+ or RPI4B. So another useful thing to do in addition to the method above would be to identify the model exactly and to double-check the earlier test results (Mb/s or MB/s). Hopefully this helps to troubleshoot the issue, whatever it is. 1: https://download.freebsd.org/ftp/releases/arm64/aarch64/ISO-IMAGES/13.0/FreeBSD-13.0-RELEASE-arm64-aarch64-RPI.img.xz -- Denis Ovsienko