From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [RFC/RFT PATCH 0/3] arm64: KVM: work around incoherency with uncached guest mappings Date: Wed, 04 Mar 2015 18:03:11 +0100 Message-ID: <54F73ACF.1090605@redhat.com> References: <20150220153626.GB10942@hawk.usersys.redhat.com> <20150224145529.GA5498@hawk.usersys.redhat.com> <20150302163146.GB9686@lvm> <54F51A63.4060302@samsung.com> <20150304113505.GS28951@e104818-lin.cambridge.arm.com> <20150304122956.GT28951@e104818-lin.cambridge.arm.com> <20150304141212.GA5352@hawk.usersys.redhat.com> <20150304142943.GU28951@e104818-lin.cambridge.arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from localhost (localhost [127.0.0.1]) by mm01.cs.columbia.edu (Postfix) with ESMTP id C59D947AC5 for ; Wed, 4 Mar 2015 11:57:22 -0500 (EST) Received: from mm01.cs.columbia.edu ([127.0.0.1]) by localhost (mm01.cs.columbia.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xQTIQxEudRcA for ; Wed, 4 Mar 2015 11:57:21 -0500 (EST) Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by mm01.cs.columbia.edu (Postfix) with ESMTPS id 2B25747AB6 for ; Wed, 4 Mar 2015 11:57:20 -0500 (EST) In-Reply-To: <20150304142943.GU28951@e104818-lin.cambridge.arm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: kvmarm-bounces@lists.cs.columbia.edu Sender: kvmarm-bounces@lists.cs.columbia.edu To: Catalin Marinas , Andrew Jones Cc: KVM devel mailing list , Ard Biesheuvel , Marc Zyngier , Laszlo Ersek , "kvmarm@lists.cs.columbia.edu" , "linux-arm-kernel@lists.infradead.org" List-Id: kvmarm@lists.cs.columbia.edu On 04/03/2015 15:29, Catalin Marinas wrote: > I disagree it is 100% a host-side issue. It is a host-side issue _if_ > the host tells the guest that the (virtual) device is non-coherent (or, > more precisely, it does not explicitly tell the guest that the device is > coherent). If the guest thinks the (virtual) device is non-coherent > because of information passed by the host, I fully agree that the host > needs to manage the cache coherency. > > However, the host could also pass a "dma-coherent" property in the DT > given to the guest and avoid any form of cache maintenance. If the guest > does not honour such coherency property, it's a guest problem and it > needs fixing in the guest. This isn't any different from a real physical > device behaviour. Can you add that property to the device tree for PCI devices too? Paolo