Re: [Qemu-devel] Re: [PATCH] pci: initialize header type register.

["Michael S. Tsirkin" <mst@redhat.com>]

On Mon, Feb 08, 2010 at 07:56:27PM +0200, Blue Swirl wrote:
> On Mon, Feb 8, 2010 at 7:37 PM, Michael S. Tsirkin <mst@redhat.com> wrote:
> > On Mon, Feb 08, 2010 at 06:37:41PM +0100, Gerd Hoffmann wrote:
> >> On 02/08/10 18:32, Michael S. Tsirkin wrote:
> >>> On Mon, Feb 08, 2010 at 06:24:57PM +0100, Gerd Hoffmann wrote:
> >>>> On 02/08/10 17:27, Michael S. Tsirkin wrote:
> >>>>> On Mon, Feb 08, 2010 at 12:14:11PM +0100, Gerd Hoffmann wrote:
> >>>>>> On 02/08/10 11:17, Michael S. Tsirkin wrote:
> >>>>>>> On Mon, Feb 08, 2010 at 03:41:47PM +0900, Isaku Yamahata wrote:
> >>>>>>>> initialize header type register in pci generic code.
> >>>>>>>>
> >>>>>>>> Cc: Blue Swirl<blauwirbel@gmail.com>
> >>>>>>>> Cc: "Michael S. Tsirkin"<mst@redhat.com>
> >>>>>>>> Signed-off-by: Isaku Yamahata<yamahata@valinux.co.jp>
> >>>>>>>
> >>>>>>> No objections here, I am assuming this will be followed
> >>>>>>> by patches removing header type init from bridges?
> >>>>>>>    From qdev perspective, it is probably cleaner to make
> >>>>>>> multifunction bit a separate qdev property though, right?
> >>>>>>
> >>>>>>    From a qdev perspective it would make *alot* of sense to move a bunch of
> >>>>>> pci config stuff (including, but not limited to header type) into
> >>>>>> PCIDeviceInfo.
> >>>>>>
> >>>>>> cheers,
> >>>>>>     Gerd
> >>>>>
> >>>>> Actually - won't this make it possible to create broken configurations
> >>>>> by tweaking properties from command-line?
> >>>>
> >>>> Not as property, as struct element in PCIDeviceInfo.  i.e.
> >>>>
> >>>> static PCIDeviceInfo e1000_info = {
> >>>>      [ stuff which is here right now ]
> >>>>      .vendor_id = PCI_VENDOR_ID_INTEL,
> >>>>      .device_id = E1000_DEVID,
> >>>>      .class     = PCI_CLASS_NETWORK_ETHERNET,
> >>>>      [ probably more stuff which makes sense ]
> >>>> }
> >>>>
> >>>> Then setup this in generic pci code instead of having each driver doing
> >>>> a bunch of pci_config_set_*() calls.
> >>>>
> >>>> cheers,
> >>>>    Gerd
> >>>
> >>> We still end up with class, vendor etc duplicated in 2 places.
> >>
> >> No.  The info should be *only* in PCIDeviceInfo then.
> >
> > That would put a lot of code in pci config cycle path.  A single array
> > mirroring the whole config space is much cleaner.
> 
> I'd suppose the arrays would remain as they are now, they just would
> be initialized (using the pci functions) based on PCIDeviceInfo
> structure.

This still means we have two copies of same data
and need to maintain code that keeps them in sync,
even if that is called just at init time.

> This would simplify the device code a lot.

Well, I think

pci_set_class(pci_dev, PCI_CLASS_NETWORK_ETHERNET)

is simpler than

	.class = PCI_CLASS_NETWORK_ETHERNET

and some magic that copies that to pci config.


> >>>  Why do
> >>> we want stuff like vendor id in PCIDeviceInfo at all?  Why can't
> >>> everyone just use pci_config_set/get calls?
> >>
> >> You can do nice stuff like printing vendor/device IDs in the '-device ?'
> >> list then.
> >
> > That should use pci functions as well.
> >
> >> cheers,
> >>   Gerd
> >