From: Paolo Bonzini <pbonzini@redhat.com>
To: Catalin Marinas <catalin.marinas@arm.com>,
Andrew Jones <drjones@redhat.com>
Cc: KVM devel mailing list <kvm@vger.kernel.org>,
Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Marc Zyngier <marc.zyngier@arm.com>,
Laszlo Ersek <lersek@redhat.com>,
"kvmarm@lists.cs.columbia.edu" <kvmarm@lists.cs.columbia.edu>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>
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 [thread overview]
Message-ID: <54F73ACF.1090605@redhat.com> (raw)
In-Reply-To: <20150304142943.GU28951@e104818-lin.cambridge.arm.com>
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
WARNING: multiple messages have this Message-ID (diff)
From: pbonzini@redhat.com (Paolo Bonzini)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFC/RFT PATCH 0/3] arm64: KVM: work around incoherency with uncached guest mappings
Date: Wed, 04 Mar 2015 18:03:11 +0100 [thread overview]
Message-ID: <54F73ACF.1090605@redhat.com> (raw)
In-Reply-To: <20150304142943.GU28951@e104818-lin.cambridge.arm.com>
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
next prev parent reply other threads:[~2015-03-04 16:57 UTC|newest]
Thread overview: 110+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-19 10:54 [RFC/RFT PATCH 0/3] arm64: KVM: work around incoherency with uncached guest mappings Ard Biesheuvel
2015-02-19 10:54 ` Ard Biesheuvel
2015-02-19 10:54 ` [RFC/RFT PATCH 1/3] arm64: KVM: handle some sysreg writes in EL2 Ard Biesheuvel
2015-02-19 10:54 ` Ard Biesheuvel
2015-03-03 17:59 ` Mario Smarduch
2015-03-03 17:59 ` Mario Smarduch
2015-02-19 10:54 ` [RFC/RFT PATCH 2/3] arm64: KVM: mangle MAIR register to prevent uncached guest mappings Ard Biesheuvel
2015-02-19 10:54 ` Ard Biesheuvel
2015-02-19 10:54 ` [RFC/RFT PATCH 3/3] arm64: KVM: keep trapping of VM sysreg writes enabled Ard Biesheuvel
2015-02-19 10:54 ` Ard Biesheuvel
2015-02-19 13:40 ` Marc Zyngier
2015-02-19 13:40 ` Marc Zyngier
2015-02-19 13:44 ` Ard Biesheuvel
2015-02-19 13:44 ` Ard Biesheuvel
2015-02-19 15:19 ` Marc Zyngier
2015-02-19 15:19 ` Marc Zyngier
2015-02-19 15:22 ` Ard Biesheuvel
2015-02-19 15:22 ` Ard Biesheuvel
2015-02-19 14:50 ` [RFC/RFT PATCH 0/3] arm64: KVM: work around incoherency with uncached guest mappings Alexander Graf
2015-02-19 14:50 ` Alexander Graf
2015-02-19 14:56 ` Ard Biesheuvel
2015-02-19 14:56 ` Ard Biesheuvel
2015-02-19 15:27 ` Alexander Graf
2015-02-19 15:27 ` Alexander Graf
2015-02-19 15:31 ` Ard Biesheuvel
2015-02-19 15:31 ` Ard Biesheuvel
2015-02-19 16:57 ` Andrew Jones
2015-02-19 16:57 ` Andrew Jones
2015-02-19 17:19 ` Ard Biesheuvel
2015-02-19 17:19 ` Ard Biesheuvel
2015-02-19 17:55 ` Andrew Jones
2015-02-19 17:55 ` Andrew Jones
2015-02-19 17:57 ` Paolo Bonzini
2015-02-19 17:57 ` Paolo Bonzini
2015-02-20 14:29 ` Andrew Jones
2015-02-20 14:29 ` Andrew Jones
2015-02-20 14:37 ` Ard Biesheuvel
2015-02-20 14:37 ` Ard Biesheuvel
2015-02-20 15:36 ` Andrew Jones
2015-02-20 15:36 ` Andrew Jones
2015-02-24 14:55 ` Andrew Jones
2015-02-24 14:55 ` Andrew Jones
2015-02-24 17:47 ` Ard Biesheuvel
2015-02-24 17:47 ` Ard Biesheuvel
2015-02-24 19:12 ` Andrew Jones
2015-02-24 19:12 ` Andrew Jones
2015-03-02 16:31 ` Christoffer Dall
2015-03-02 16:31 ` Christoffer Dall
2015-03-02 16:47 ` Paolo Bonzini
2015-03-02 16:47 ` Paolo Bonzini
2015-03-02 16:55 ` Laszlo Ersek
2015-03-02 16:55 ` Laszlo Ersek
2015-03-02 17:05 ` Andrew Jones
2015-03-02 17:05 ` Andrew Jones
2015-03-02 16:48 ` Andrew Jones
2015-03-02 16:48 ` Andrew Jones
2015-03-03 2:20 ` Mario Smarduch
2015-03-03 2:20 ` Mario Smarduch
2015-03-04 11:35 ` Catalin Marinas
2015-03-04 11:35 ` Catalin Marinas
2015-03-04 11:50 ` Ard Biesheuvel
2015-03-04 11:50 ` Ard Biesheuvel
2015-03-04 12:29 ` Catalin Marinas
2015-03-04 12:29 ` Catalin Marinas
2015-03-04 12:43 ` Ard Biesheuvel
2015-03-04 12:43 ` Ard Biesheuvel
2015-03-04 14:12 ` Andrew Jones
2015-03-04 14:12 ` Andrew Jones
2015-03-04 14:29 ` Catalin Marinas
2015-03-04 14:29 ` Catalin Marinas
2015-03-04 14:34 ` Peter Maydell
2015-03-04 14:34 ` Peter Maydell
2015-03-04 17:03 ` Paolo Bonzini [this message]
2015-03-04 17:03 ` Paolo Bonzini
2015-03-04 17:28 ` Catalin Marinas
2015-03-04 17:28 ` Catalin Marinas
2015-03-05 10:12 ` Paolo Bonzini
2015-03-05 10:12 ` Paolo Bonzini
2015-03-05 11:04 ` Catalin Marinas
2015-03-05 11:04 ` Catalin Marinas
2015-03-05 11:52 ` Peter Maydell
2015-03-05 11:52 ` Peter Maydell
2015-03-05 12:03 ` Catalin Marinas
2015-03-05 12:03 ` Catalin Marinas
2015-03-05 12:26 ` Paolo Bonzini
2015-03-05 12:26 ` Paolo Bonzini
2015-03-05 14:58 ` Catalin Marinas
2015-03-05 14:58 ` Catalin Marinas
2015-03-05 17:43 ` Paolo Bonzini
2015-03-05 17:43 ` Paolo Bonzini
2015-03-06 21:08 ` Mario Smarduch
2015-03-06 21:08 ` Mario Smarduch
2015-03-09 14:26 ` Andrew Jones
2015-03-09 14:26 ` Andrew Jones
2015-03-09 15:33 ` Mario Smarduch
2015-03-09 15:33 ` Mario Smarduch
2015-03-05 19:13 ` Ard Biesheuvel
2015-03-05 19:13 ` Ard Biesheuvel
2015-03-06 20:33 ` Mario Smarduch
2015-03-06 20:33 ` Mario Smarduch
2015-02-19 18:44 ` Ard Biesheuvel
2015-02-19 18:44 ` Ard Biesheuvel
2015-03-03 17:34 ` Alexander Graf
2015-03-03 17:34 ` Alexander Graf
2015-03-03 18:13 ` Laszlo Ersek
2015-03-03 18:13 ` Laszlo Ersek
2015-03-03 20:58 ` Andrew Jones
2015-03-03 20:58 ` Andrew Jones
2015-03-03 18:32 ` Catalin Marinas
2015-03-03 18:32 ` Catalin Marinas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=54F73ACF.1090605@redhat.com \
--to=pbonzini@redhat.com \
--cc=ard.biesheuvel@linaro.org \
--cc=catalin.marinas@arm.com \
--cc=drjones@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=lersek@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=marc.zyngier@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.