On Wed, Mar 25, 2009 at 04:53:07PM +0800, Cui, Dexuan wrote:
> Hi Simon,
> Did you actually test your patch and verify it works? -- It doesn't work at
> least in my side.
>
> For a statically assigned device ("a pass-through device that is inserted
> at boot time"), needn't we modify ioemu to handle the vslot info
> properly?
There has been an additional patch posted[1] as well as some ioemu
changes[2, 3]. I think you have subsequently found these. Sorry for the
confusion, I did make some miss-steps along the way.
> BTW, in the python codes inside tools/python/xen, somewhere we use the
> stirng 'vslot', somewhere we use 'vslt'. We may as well use the same
> string.
Agreed. Though I am working on allowing multi-function devices to be
passed-through as multi-function devices. As part of this change it seems
to be convenient to change vslot to vdevfn. I wonder if at that time
changing thins around to vdevfn would make sense.
I was planing to post multi-function these patches after 3.4 has been
released. Partly to avoid any confusion about what I am submitting
for 3.4. And partly because they aren't quite finished yet.
If it weren't for these multi-function changes that I am working on,
I would be in favour on standardising on vslot throughout the code.
And if you want to change it anyway, I have no objections.
> In your patch, tools/python/xen/xend/server/pciif.py: +
> vslot = parse_hex(pci_config.get('vslot', 0)) ===> in the newly-added
> line, vslot is not used later??
This was addressed in subsequent patches[2,3].
> And in your patch, you only allow a vslot 0x0~0xf. I think we should
> support 0x0~0x1f.
Thanks, I'll get that fixed.
> Thanks,
Thans for your review :-)
--
Simon Horman
VA Linux Systems Japan K.K., Sydney, Australia Satellite Office
H: www.vergenet.net/~horms/ W: www.valinux.co.jp/en
[1] [patch] xend: Supply pass-through vslot to qemu-dm
[2] [patch 0/4 v2] ioemu: specify the slot for pass-through devices
[3] [rfc 0/4] ioemu: specify the slot for pass-through devices
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|