Re: [PATCH v10 06/26] s390: vfio-ap: sysfs interfaces to configure adapters

[Tony Krowiak]

On 09/21/2018 05:40 AM, Cornelia Huck wrote:
> On Wed, 12 Sep 2018 15:42:56 -0400
> Tony Krowiak <akrowiak@xxxxxxxxxxxxxxxxxx> wrote:
>
> > From: Tony Krowiak <akrowiak@xxxxxxxxxxxxx>
> >
> > Introduces two new sysfs attributes for the VFIO mediated
> > matrix device for assigning AP adapters to and unassigning
> > AP adapters from a mediated matrix device. The IDs of the
> > AP adapters assigned to the mediated matrix device will be
> > stored in an AP mask (APM).
> >
> > The bits in the APM, from most significant to least significant
> > bit, correspond to AP adapter IDs (APID) 0 to 255. On
> > some systems, the maximum allowable adapter number may be less
> > than 255 - depending upon the host's AP configuration - and
> > assignment may be rejected if the input adapter ID exceeds the
> > limit.
> >
> > When an adapter is assigned, the bit corresponding to the APID
> > will be set in the APM. Likewise, when an adapter is
> > unassigned, the bit corresponding to the APID will be cleared
> > from the APM.
> >
> > In order to successfully assign an adapter, the APQNs derived from
> > the adapter ID being assigned and the queue indexes of all domains
> > previously assigned:
> >
> > 1. Must be bound to the vfio_ap device driver.
> >
> > 2. Must not be assigned to any other mediated matrix device
> >
> > If there are no domains assigned to the mdev, then there must
> > be an AP queue bound to the vfio_ap device driver with an
> > APQN containing the APID, otherwise all domains
> > subsequently assigned will fail because there will be no
> > AP queues bound with an APQN containing the adapter ID.
> >
> > Assigning or un-assigning an AP adapter will be rejected if
> > a guest using the mediated matrix device is running.
> >
> > The relevant sysfs structures are:
> >
> > /sys/devices/vfio_ap/matrix/
> > ...... [mdev_supported_types]
> > ......... [vfio_ap-passthrough]
> > ............ [devices]
> > ...............[$uuid]
> > .................. assign_adapter
> > .................. unassign_adapter
> >
> > To assign an adapter to the $uuid mediated matrix device's APM,
> > write the APID to the assign_adapter file. To unassign an adapter,
> > write the APID to the unassign_adapter file. The APID is specified
> > using conventional semantics: If it begins with 0x the number will
> > be parsed as a hexadecimal number; if it begins with a 0 the number
> > will be parsed as an octal number; otherwise, it will be parsed as a
> > decimal number.
> >
> > For example, to assign adapter 173 (0xad) to the mediated matrix
> > device $uuid:
> >
> > 	echo 173 > assign_adapter
> >
> > 	or
> >
> > 	echo 0xad > assign_adapter
> >
> > 	or
> >
> > 	echo 0255 > assign_adapter
> >
> > To unassign adapter 173 (0xad):
> >
> > 	echo 173 > unassign_adapter
> >
> > 	or
> >
> > 	echo 0xad > unassign_adapter
> >
> > 	or
> >
> > 	echo 0255 > unassign_adapter
> >
> > Signed-off-by: Tony Krowiak <akrowiak@xxxxxxxxxxxxx>
> > Reviewed-by: Halil Pasic <pasic@xxxxxxxxxxxxx>
> > Tested-by: Michael Mueller <mimu@xxxxxxxxxxxxx>
> > Tested-by: Farhan Ali <alifm@xxxxxxxxxxxxx>
> > Tested-by: Pierre Morel <pmorel@xxxxxxxxxxxxx>
> > Signed-off-by: Christian Borntraeger <borntraeger@xxxxxxxxxx>
> > ---
> >   drivers/s390/crypto/vfio_ap_ops.c |  295 +++++++++++++++++++++++++++++++++++++
> >   1 files changed, 295 insertions(+), 0 deletions(-)
> (...)
>
> > +/**
> > + * vfio_ap_mdev_verify_no_sharing
> > + *
> > + * Verifies that the APQNs derived from the cross product of the AP adapter IDs
> > + * and AP queue indexes comprising the AP matrix are not configured for another
> > + * mediated device. AP queue sharing is not allowed.
> > + *
> > + * @kvm: the KVM guest
> > + * @matrix: the AP matrix
> > + *
> > + * Returns 0 if the APQNs are not shared, otherwise; returns -EADDRINUSE.
> > + */
> > +static int vfio_ap_mdev_verify_no_sharing(struct ap_matrix_mdev *matrix_mdev)
> > +{
> > +	int nbits;
> > +	struct ap_matrix_mdev *lstdev;
> > +	unsigned long apm[BITS_TO_LONGS(matrix_mdev->matrix.apm_max + 1)];
> > +	unsigned long aqm[BITS_TO_LONGS(matrix_mdev->matrix.aqm_max + 1)];
> Can you please convert this to use a fixed-size array? I think
> {apm,aqm}_max has an upper bound of 255?

Sure

> (Also, this can use DECLARE_BITMAP.)

Indeed it can.

> > +
> > +	list_for_each_entry(lstdev, &matrix_dev->mdev_list, node) {
> > +		if (matrix_mdev == lstdev)
> > +			continue;
> > +
> > +		memset(apm, 0, sizeof(apm));
> > +		memset(aqm, 0, sizeof(aqm));
> > +
> > +		/*
> > +		 * We work on full longs, as we can only exclude the leftover
> > +		 * bits in non-inverse order. The leftover is all zeros.
> > +		 */
> > +		nbits = sizeof(apm) * BITS_PER_BYTE;
> > +		if (!bitmap_and(apm, matrix_mdev->matrix.apm,
> > +				lstdev->matrix.apm, nbits))
> > +			continue;
> > +
> > +		nbits = sizeof(aqm) * BITS_PER_BYTE;
> > +		if (!bitmap_and(aqm, matrix_mdev->matrix.aqm,
> > +				lstdev->matrix.aqm, nbits))
> > +			continue;
> > +
> > +		return -EADDRINUSE;
> > +	}
> > +
> > +	return 0;
> > +}