From owner-freebsd-virtualization@freebsd.org Sat Jan 14 05:54:40 2017 Return-Path: Delivered-To: freebsd-virtualization@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id 278E1CAFE0F for ; Sat, 14 Jan 2017 05:54:40 +0000 (UTC) (envelope-from soralx@cydem.org) Received: from smtp.triumf.ca (smtp.triumf.ca [142.90.100.188]) by mx1.freebsd.org (Postfix) with ESMTP id EB4F21CF0; Sat, 14 Jan 2017 05:54:39 +0000 (UTC) (envelope-from soralx@cydem.org) Received: from mscad14 (mscad14.triumf.ca [142.90.115.36]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.triumf.ca (Postfix) with ESMTP id D3253F805; Fri, 13 Jan 2017 21:54:38 -0800 (PST) Date: Fri, 13 Jan 2017 21:54:38 -0800 From: To: Cc: , Subject: Re: Issues with GTX960 on CentOS7 using bhyve PCI passthru (FreeBSD 11-RC2) Message-ID: <20170113215438.01c3b674@mscad14> In-Reply-To: <201701131547.v0DFlEjt054297@pdx.rh.CN85.dnsmgr.net> References: <20170113001737.5fe3001b@mscad14> <201701131547.v0DFlEjt054297@pdx.rh.CN85.dnsmgr.net> X-Mailer: Claws Mail 3.14.1 (GTK+ 2.24.29; amd64-portbld-freebsd9.3) MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-BeenThere: freebsd-virtualization@freebsd.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Discussion of various virtualization techniques FreeBSD supports." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 14 Jan 2017 05:54:40 -0000 > I suspect this is a failure to run the BIOS code that > enables the secondary power connector so you can come > out of slot only power mode. Well, that Quadro does not have a power connector, but I imagine card BIOS routines would be similar between all cards in a family, including those that require the extra power. Does bhyve not execute peripheral cards' option ROMs? I guess it doesn't. This could explain a lot of strange behaviour seen resulting from running in a VM. How does UEFI work in this regard? My guess is that cards have to explicitly support the new boot method (UEFI)? So passthrough with newer cards may be easier? This could explain why the newer RX 480 worked right away, and the older Quadro 2000 (and a lot of other nVidia cards without manufacturer's support for VMs) had no chance -- UEFI cards are somehow more "autonomous". It all is just speculation on my side, I know nothing about this UEFI stuff. Could you summarize in couple sentences what's the deal between bhyve and UEFI (if there is any), or future plans? > The general rule on other platforms is that ATI/AMD cards tend to just > work, where as the NVidia cards are very picky and unless official > listed as known to work when passed through you well fight problems. > Very few cards are listed as known to work, most populars being the > Quadro 2000, and the Quadro FX3800. Many cards are listed as known to > NOT work. Yeah, messing with nVidia for this reason (and because of the closed driver) seems to me like a huge time sink. I don't have the time, so decided to try with AMD for now. I am only interested in nVidia because it's best choice for a Windows VM I hope to run (SolidWorks, Altium, etc). OpenCL on AMD is a priority for me personally right now anyway. > GOOD WORK on getting as far as you have as quickly as you have! > Note that the https://wiki.freebsd.org/bhyve/pci_passthru has > had a small update to reflect we know that VGA passthrough is > not working at this time. Also a note about AMD IOmmu/AMD-Vi > was added, hopefully saving someone from duplicate work. Perhaps a note could be added about ATI/AMD cards partially working, to encourage others to play? -- [SorAlx] ridin' VN2000 Classic LT