From mboxrd@z Thu Jan  1 00:00:00 1970
From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: Re: PCI BAR register space written with garbage in HVM
	guest.
Date: Tue, 16 Mar 2010 11:20:40 -0400
Message-ID: <20100316152040.GA28821@phenom.dumpdata.com>
References: <4779de451003151809n6cec813dp32d77fee34b1bda2@mail.gmail.com>
	<20100316014851.GE7622@phenom.dumpdata.com>
	<4779de451003151955v15863656i5f39a631a8c558ee@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <xen-devel-bounces@lists.xensource.com>
Content-Disposition: inline
In-Reply-To: <4779de451003151955v15863656i5f39a631a8c558ee@mail.gmail.com>
List-Unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xensource.com>
List-Help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-Subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
Sender: xen-devel-bounces@lists.xensource.com
Errors-To: xen-devel-bounces@lists.xensource.com
To: Dan Gora <dan.gora@gmail.com>
Cc: xen-devel@lists.xensource.com
List-Id: xen-devel@lists.xenproject.org

> <aside>
> What is pcifront/pciback's role for HVM guests exactly?  I understand

None functionaly. The purpose is to "bind" the PCI devices to pciback
(or pcistub) so that no other kernel module usurps it and starts
utilizing it.

> that you "hide" the devices from the dom0 with pciback and it
> definately loads and does *something* when the HVM guest comes up, but
> accesses from the domU don't appear to go through it at all (I
> understand that it works with qemu somehow, but that channel too is
> not at all clear how it works...)

With HVM pciback is not used. You need an virtualization aware IOMMU to
pass through PCI devices to the guest. PCIback/PCI front is for PV
guests and on machine where you don't neccessarily have this fancy
IOMMU.

> I've looked through qemu enough to kind of understand that it's
> responsible for abstracting the PCI configuration space for the domU,
> but I don't really understand how accesses get channeled through to it
> from the domU.  Does it use hypercalls somehow?  Can someone explain

The Hypervisor gets "trapped" when an outb is made (look for
emulate_privileged_op function and specifically out).

Then it somehow injects the fault in QEMU which does the rest. I don't
remember the details of how it does thought :-(

> how this whole flow is supposed to work for PCI configuration space
> accesses?
> </aside>