RE: [RFC v2 2/4] hisi_acc_vfio_pci: Override ioctl method to limit BAR2 region size

[Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>]

> -----Original Message-----
> From: Alex Williamson [mailto:alex.williamson@redhat.com]
> Sent: 02 July 2021 21:30
> To: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
> Cc: kvm@vger.kernel.org; linux-kernel@vger.kernel.org;
> linux-crypto@vger.kernel.org; jgg@nvidia.com; mgurtovoy@nvidia.com;
> Linuxarm <linuxarm@huawei.com>; liulongfang <liulongfang@huawei.com>;
> Zengtao (B) <prime.zeng@hisilicon.com>; yuzenghui
> <yuzenghui@huawei.com>; Jonathan Cameron
> <jonathan.cameron@huawei.com>; Wangzhou (B) <wangzhou1@hisilicon.com>
> Subject: Re: [RFC v2 2/4] hisi_acc_vfio_pci: Override ioctl method to limit
> BAR2 region size
> 
> On Fri, 2 Jul 2021 10:58:47 +0100
> Shameer Kolothum <shameerali.kolothum.thodi@huawei.com> wrote:
> 
> > HiSilicon ACC VF device BAR2 region consists of both functional register
> > space and migration control register space. From a security point of
> > view, it's not advisable to export the migration control region to Guest.
> >
> > Hence, hide the migration region and report only the functional register
> > space.
> >
> > Signed-off-by: Shameer Kolothum
> <shameerali.kolothum.thodi@huawei.com>
> > ---
> >  drivers/vfio/pci/hisi_acc_vfio_pci.c | 42 +++++++++++++++++++++++++++-
> >  1 file changed, 41 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/vfio/pci/hisi_acc_vfio_pci.c
> b/drivers/vfio/pci/hisi_acc_vfio_pci.c
> > index a9e173098ab5..d57cf6d7adaf 100644
> > --- a/drivers/vfio/pci/hisi_acc_vfio_pci.c
> > +++ b/drivers/vfio/pci/hisi_acc_vfio_pci.c
> > @@ -12,6 +12,46 @@
> >  #include <linux/vfio.h>
> >  #include <linux/vfio_pci_core.h>
> >
> > +static long hisi_acc_vfio_pci_ioctl(struct vfio_device *core_vdev, unsigned
> int cmd,
> > +				    unsigned long arg)
> > +{
> > +	struct vfio_pci_core_device *vdev =
> > +		container_of(core_vdev, struct vfio_pci_core_device, vdev);
> > +
> > +	if (cmd == VFIO_DEVICE_GET_REGION_INFO) {
> > +		struct pci_dev *pdev = vdev->pdev;
> > +		struct vfio_region_info info;
> > +		unsigned long minsz;
> > +
> > +		minsz = offsetofend(struct vfio_region_info, offset);
> > +
> > +		if (copy_from_user(&info, (void __user *)arg, minsz))
> > +			return -EFAULT;
> > +
> > +		if (info.argsz < minsz)
> > +			return -EINVAL;
> > +
> > +		if (info.index == VFIO_PCI_BAR2_REGION_INDEX) {
> > +			info.offset = VFIO_PCI_INDEX_TO_OFFSET(info.index);
> > +
> > +			/*
> > +			 * ACC VF dev BAR2 region(64K) consists of both functional
> > +			 * register space and migration control register space.
> > +			 * Report only the first 32K(functional region) to Guest.
> > +			 */
> > +			info.size = pci_resource_len(pdev, info.index) / 2;
> > +
> 
> Great, but what actually prevents the user from accessing the full
> extent of the BAR since you're re-using core code for read/write/mmap,
> which are all basing access on pci_resource_len()?  Thanks,

Ah..true. Missed that. I will add overrides for read/write/mmap and limit
the access.

Thanks,
Shameer

> 
> Alex
> 
> > +			info.flags = VFIO_REGION_INFO_FLAG_READ |
> > +					VFIO_REGION_INFO_FLAG_WRITE |
> > +					VFIO_REGION_INFO_FLAG_MMAP;
> > +
> > +			return copy_to_user((void __user *)arg, &info, minsz) ?
> > +					    -EFAULT : 0;
> > +		}
> > +	}
> > +	return vfio_pci_core_ioctl(core_vdev, cmd, arg);
> > +}
> > +
> >  static int hisi_acc_vfio_pci_open(struct vfio_device *core_vdev)
> >  {
> >  	struct vfio_pci_core_device *vdev =
> > @@ -33,7 +73,7 @@ static const struct vfio_device_ops
> hisi_acc_vfio_pci_ops = {
> >  	.name		= "hisi-acc-vfio-pci",
> >  	.open		= hisi_acc_vfio_pci_open,
> >  	.release	= vfio_pci_core_release,
> > -	.ioctl		= vfio_pci_core_ioctl,
> > +	.ioctl		= hisi_acc_vfio_pci_ioctl,
> >  	.read		= vfio_pci_core_read,
> >  	.write		= vfio_pci_core_write,
> >  	.mmap		= vfio_pci_core_mmap,