Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 7 Oct 2022 13:24:58 -0700 (PDT)
From:   Stefano Stabellini <sstabellini@kernel.org>
To:     Juergen Gross <jgross@suse.com>
cc:     Oleksandr Tyshchenko <olekstysh@gmail.com>,
        xen-devel@lists.xenproject.org, linux-kernel@vger.kernel.org,
        Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>,
        Stefano Stabellini <sstabellini@kernel.org>
Subject: Re: [PATCH] xen/virtio: Handle PCI devices which Host controller is
 described in DT
In-Reply-To: <014e7036-5daf-6465-663c-b690d6f719ee@suse.com>
Message-ID: <alpine.DEB.2.22.394.2210071324010.3690179@ubuntu-linux-20-04-desktop>
References: <20221006174804.2003029-1-olekstysh@gmail.com> <014e7036-5daf-6465-663c-b690d6f719ee@suse.com>
User-Agent: Alpine 2.22 (DEB 394 2020-01-19)
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Precedence: bulk

On Fri, 7 Oct 2022, Juergen Gross wrote:
> On 06.10.22 19:48, Oleksandr Tyshchenko wrote:
> > From: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > 
> > Use the same "xen-grant-dma" device concept (based on generic IOMMU
> > device-tree bindings) for the PCI devices behind device-tree based
> > PCI Host controller.
> > 
> > Signed-off-by: Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>
> > ---
> > Slightly RFC. This is needed to support Xen grant mappings for virtio-pci
> > devices
> > on Arm at some point in the future. The Xen toolstack side is not published
> > yet.
> > Here, for PCI devices we use the same way to pass backend domid to the guest
> > as for
> > platform devices.
> 
> I should mention we decided at the Xen Summit, that I will start a try to
> modify the virtio spec to include the backend id (domid in the Xen case)
> in the device independent config part.

Good idea


> As this will take some time to be accepted (if ever), other means to
> specify the backend domid are needed until then. DT is one possibility
> (at least on Arm), while Xenstore is the way to go for setups with a
> Xen toolstack.

What do you think of my idea of using PCI config registers on the *root
complex* (not the device) to retrieve the information?