Oops, I forgot to CC Ian.
It's weird the 2/3 patch was not sent out successfully. I will resend out it
From: Keir Fraser [mailto:keir.fraser@xxxxxxxxxxxxx]
Sent: Thursday, February 04, 2010 4:25 PM
To: Han, Weidong; xen-devel@xxxxxxxxxxxxxxxxxxx
Cc: Kay, Allen M; Ian Jackson
Subject: Re: [Xen-devel][PATCH 0/3] VT-d: support Intel IGD passthrough
Cc'ing Ian Jackson for applying patch 1/3. And I didn't receive your patch
2/3. I've received and will apply your patch 3/3 however.
On 04/02/2010 08:12, "Han, Weidong" <weidong.han@xxxxxxxxx> wrote:
> The patchset supports Intel IGD passthrough in Xen. There are some
> assumptions/dependencies on chipset, VBIOS, and even drivers for IGD. But
> virtual platform is not aware of it. This patchset handles it to make IGD work
> in guest. What's more, FLR capability of IGD is not exposed on some platforms,
> this patchset uses vendor specific FLR to reset those IGDs.
> Also add an option into hvm config file to enable graphics passthrough.
> The patchset was tested in Intel McCreary platform, guest WinXP, Win7 and
> Linux all worked.
> Patch 1: qemu-igd.patch:
> Some registers of Intel IGD are mapped in host bridge, so it needs to
> passthrough these registers of physical host bridge to guest because emulated
> host bridge in guest doesn't have these mappings.
> Some VBIOSs and drivers assume the IGD BDF (bus:device:function) is always
> 00:02.0, so this patch reserves 00:02.0 for assigned IGD in guest.
> Patch 2: igd-flr.patch:
> Due to FLR capability of IGD is not exposed on some platforms, this patch uses
> vendor specific FLR to reset those IGDs.
> Patch 3: gfx-conf.patch:
> The patch adds an option into hvm config file to enable graphics passthrough
> including discrete and IGD. To passthrough graphics to guest, need to set
> gfx_passthru=1, and also specify graphics device BDF in pci passthrough
> option, like pci=['xx:xx.x'] in hvm config file.
> Signed-off-by: Weidong Han <weidong.han@xxxxxxxxx>
Xen-devel mailing list