Re: [net-next v4 01/12] Implementation of Virtual Bus

[Jason Gunthorpe <jgg@ziepe.ca>]

On Thu, May 21, 2020 at 07:43:58PM +0200, gregkh@linuxfoundation.org wrote:
> On Thu, May 21, 2020 at 02:57:55PM +0000, Parav Pandit wrote:
> > Hi Greg, Jason,
> > 
> > On 5/20/2020 12:32 PM, Jeff Kirsher wrote:
> > > From: Dave Ertman <david.m.ertman@intel.com>
> > > 
> > 
> > > +static const
> > > +struct virtbus_dev_id *virtbus_match_id(const struct virtbus_dev_id *id,
> > > +					struct virtbus_device *vdev)
> > > +{
> > > +	while (id->name[0]) {
> > > +		if (!strcmp(vdev->match_name, id->name))
> > > +			return id;
> > 
> > Should we have VID, DID based approach instead of _any_ string chosen by
> > vendor drivers?
> 
> No, because:
> 
> > This will required central place to define the VID, DID of the vdev in
> > vdev_ids.h to have unique ids.
> 
> That's not a good way to run things :)
> 
> Have the virtbus core create the "name", as it really doesn't matter
> what it is, just that it is unique, right?

It is being used like the compatible string in OF. Look at where
"sof-ipc-test" appears in the SOF patches.

So it has to be a compile time static, and it has to be broadly global
in some fashion since it appears in a mod alias.

I don't think the name "sof-ipc-test" is particularly good by these
metrics.

Jason