From owner-freebsd-xen@freebsd.org Fri Oct 11 21:53:30 2019 Return-Path: Delivered-To: freebsd-xen@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id A845113374B for ; Fri, 11 Oct 2019 21:53:30 +0000 (UTC) (envelope-from buhrow@lothlorien.nfbcal.org) Received: from lothlorien.nfbcal.org (ns.NFBCAL.ORG [157.22.230.125]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "nfbcal.org", Issuer "Let's Encrypt Authority X3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 46qhXn35V7z44qv for ; Fri, 11 Oct 2019 21:53:29 +0000 (UTC) (envelope-from buhrow@lothlorien.nfbcal.org) Received: from lothlorien.nfbcal.org (localhost [127.0.0.1]) by lothlorien.nfbcal.org (8.15.2/8.14.1-NFBNETBSD) with ESMTPS id x9BLrQwd009994 (version=TLSv1.2 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Fri, 11 Oct 2019 14:53:26 -0700 (PDT) X-Virus-Status: Clean X-Virus-Scanned: clamav-milter 0.97.3 at lothlorien.nfbcal.org Received: (from buhrow@localhost) by lothlorien.nfbcal.org (8.15.2/8.12.11) id x9BLrQSI028763; Fri, 11 Oct 2019 14:53:26 -0700 (PDT) Message-Id: <201910112153.x9BLrQSI028763@lothlorien.nfbcal.org> From: Brian Buhrow Date: Fri, 11 Oct 2019 14:53:26 -0700 In-Reply-To: <201910102323.x9ANNACe003999@lothlorien.nfbcal.org> X-Mailer: Mail User's Shell (7.2.6 beta(4.pl1)+dynamic 20000103) To: freebsd-xen@freebsd.org Subject: Re: i/o is very slow on FreeBSD dom0 with Xen-4.12 and Freebsd-12 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.4.3 (lothlorien.nfbcal.org [127.0.0.1]); Fri, 11 Oct 2019 14:53:26 -0700 (PDT) X-Rspamd-Queue-Id: 46qhXn35V7z44qv X-Spamd-Bar: -- Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=pass (mx1.freebsd.org: domain of buhrow@lothlorien.nfbcal.org designates 157.22.230.125 as permitted sender) smtp.mailfrom=buhrow@lothlorien.nfbcal.org X-Spamd-Result: default: False [-2.00 / 15.00]; ARC_NA(0.00)[]; NEURAL_HAM_MEDIUM(-0.99)[-0.993,0]; FROM_HAS_DN(0.00)[]; R_SPF_ALLOW(-0.20)[+a:ns.nfbcal.org:c]; NEURAL_HAM_LONG(-1.00)[-0.999,0]; MIME_GOOD(-0.10)[text/plain]; TO_DN_NONE(0.00)[]; DMARC_NA(0.00)[nfbcal.org]; TO_MATCH_ENVRCPT_SOME(0.00)[]; RCPT_COUNT_TWO(0.00)[2]; IP_SCORE(-0.01)[country: US(-0.05)]; FORGED_SENDER(0.30)[buhrow@nfbcal.org,buhrow@lothlorien.nfbcal.org]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:7091, ipnet:157.22.0.0/16, country:US]; FROM_NEQ_ENVFROM(0.00)[buhrow@nfbcal.org,buhrow@lothlorien.nfbcal.org]; RCVD_TLS_ALL(0.00)[]; RCVD_COUNT_TWO(0.00)[2] X-BeenThere: freebsd-xen@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Discussion of the freebsd port to xen - implementation and usage List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Oct 2019 21:53:30 -0000 hello. Another data point on this issue. It appears the problem may be with the onboard IDE controllers, or, possibly, the way interrupts are handled. If I switch the BIOS to enable ahcisata mode instead of IDE mode for the SATA controllers, things boot more quickly, but the FreeBSD kernel is unable to talk to the hard drives. The ahci channels time out and we get stuck in an endless loop of CAM system retries. When the BIOS is set to Enhanced IDE mode, the FreeBSD kernel can talk to the drives, but the data rate is so slow as to be unusable. When FreeBSD boots on bare metal it can talk to the ahci or IDE drives without a problem. This makes me think Xen is messing with the interrupt handling of the IDE controller somehow. Is it possible the BIOS sets things up in a way that FreeBSD likes, but Xen does not? Perhaps the ioapic is programmed in a way that FreeBSD doesn't like? -thanks -Brian