From owner-soc-status@FreeBSD.ORG Tue Apr 16 14:27:47 2013 Return-Path: Delivered-To: soc-status@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id A33E254E; Tue, 16 Apr 2013 14:27:47 +0000 (UTC) (envelope-from gibbs@FreeBSD.org) Received: from aslan.scsiguy.com (www.scsiguy.com [70.89.174.89]) by mx1.freebsd.org (Postfix) with ESMTP id 603C6ED6; Tue, 16 Apr 2013 14:27:46 +0000 (UTC) Received: from macbook.scsiguy.com (macbook.scsiguy.com [192.168.0.99]) (authenticated bits=0) by aslan.scsiguy.com (8.14.5/8.14.5) with ESMTP id r3GERjOG026943 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 16 Apr 2013 08:27:46 -0600 (MDT) (envelope-from gibbs@FreeBSD.org) Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 6.3 \(1503\)) Subject: =?utf-8?Q?Re=3A_Mentor_of_the_Project_=22PCI_frontend_XEN_PV_dri?= =?utf-8?Q?ver=22=EF=BC=9F?= From: "Justin T. Gibbs" In-Reply-To: <5BDCBCBB-E65D-472E-B990-F1382A4BADF0@FreeBSD.org> Date: Tue, 16 Apr 2013 08:27:47 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: References: <920A79FE-8C56-4D26-8611-3CD48602FC3B@freebsd.org> <5BDCBCBB-E65D-472E-B990-F1382A4BADF0@FreeBSD.org> To: David Chisnall X-Mailer: Apple Mail (2.1503) X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.7 (aslan.scsiguy.com [70.89.174.89]); Tue, 16 Apr 2013 08:27:46 -0600 (MDT) Cc: soc-status X-BeenThere: soc-status@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Summer of Code Status Reports and Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 Apr 2013 14:27:47 -0000 On Apr 12, 2013, at 1:21 AM, David Chisnall = wrote: > On 12 Apr 2013, at 05:13, "Justin T. Gibbs" wrote: >=20 >> This particular project was proposed for last year's GSoC when both = attilio@FreeBSD.org and I were actively working on full Xen PV support = for FreeBSD. That project still continues, but at a slower pace due to = changing needs at our respective employers. Given the lack of maturity = of our Xen PV port, I think this would be a difficult project to attempt = at this time. >>=20 >> If you are interested in Xen and have access to x86 hardware with HVM = capabilities, writing a SCSI or USB PV frontend for FreeBSD is a better = project to select. FreeBSD's Xen HVM support is very robust and, unlike = the PCI frontend driver, these drivers can be developed using an HVM = guest. The resulting drivers should work equally well in full PV mode = (with perhaps a few minor tweaks) when it is better supported. >>=20 >> I will update the wiki sometime tomorrow with more information on Xen = projects that are appropriate for this year's GSoC. >=20 > Do we support PHV mode in Xen yet? If not, then this might be a good = GSoC project. Most of the effort - it uses the PV boot mechanism, but = then the HVM pmap and PV drivers (including event channels, no APIC = emulation). Unlike the PV mode, it won't require us to maintain two = i386 pmap implementations for the increasingly rare case of trying to = run FreeBSD VMs on CPUs without virtualisation support, and it doesn't = have the QEMU overhead in dom0 that HVM does (even when you're using = fully PV devices, you still need it to boot), and which causes hosting = providers to charge more for HVM. >=20 > David The last time I was really active in the Xen community (over a year = ago), PHV was still in heavy development. However, it is certainly = coming. Once Cherry has full PV working for us, we should certainly = start the process of supporting it. -- Justin=