Re: [PATCH V4 7/7] vDPA/ifcvf: deduce VIRTIO device ID from pdev ids

From: Jason Wang <jasowang@redhat.com>
To: Zhu Lingshan <lingshan.zhu@intel.com>,
	mst@redhat.com, lulu@redhat.com, leonro@nvidia.com
Cc: virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH V4 7/7] vDPA/ifcvf: deduce VIRTIO device ID from pdev ids
Date: Wed, 17 Mar 2021 12:07:53 +0800	[thread overview]
Message-ID: <a873deef-81e9-9177-b297-7c7063077ff6@redhat.com> (raw)
In-Reply-To: <20210315074501.15868-8-lingshan.zhu@intel.com>

在 2021/3/15 下午3:45, Zhu Lingshan 写道:
>   static u32 ifcvf_vdpa_get_device_id(struct vdpa_device *vdpa_dev)
>   {
> -	return VIRTIO_ID_NET;
> +	struct ifcvf_hw *vf = vdpa_to_vf(vdpa_dev);
> +	u32 ret = -EOPNOTSUPP;
> +
> +	if (ifcvf_probed_virtio_net(vf))
> +		ret = VIRTIO_ID_NET;

So the point is to simplify the future extension.

How about simply?

if (device_id>0x1040)
     return devce_id - 0x1040;
else
     return device_id;

Since I don't think you plan to introduce device whose vendor id is not 
1AF4 and the subsys vendor/device id is not interesting to vDPA bus.

Thanks

> +
> +	return ret;
>   }