lkml.org 
[lkml]   [2022]   [May]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
SubjectRe: [PATCH v19 11/20] s390/vfio-ap: prepare for dynamic update of guest's APCB on queue probe/remove
From


On 5/27/22 9:50 AM, Jason J. Herne wrote:
> On 4/4/22 18:10, Tony Krowiak wrote:
>> The callback functions for probing and removing a queue device must take
>> and release the locks required to perform a dynamic update of a guest's
>> APCB in the proper order.
>>
>> The proper order for taking the locks is:
>>
>>          matrix_dev->guests_lock => kvm->lock => matrix_dev->mdevs_lock
>>
>> The proper order for releasing the locks is:
>>
>>          matrix_dev->mdevs_lock => kvm->lock => matrix_dev->guests_lock
>>
>> A new helper function is introduced to be used by the probe callback to
>> acquire the required locks. Since the probe callback only has
>> access to a queue device when it is called, the helper function will
>> find
>> the ap_matrix_mdev object to which the queue device's APQN is
>> assigned and
>> return it so the KVM guest to which the mdev is attached can be
>> dynamically
>> updated.
>>
>> Note that in order to find the ap_matrix_mdev (matrix_mdev) object,
>> it is
>> necessary to search the matrix_dev->mdev_list. This presents a
>> locking order dilemma because the matrix_dev->mdevs_lock can't be
>> taken to
>> protect against changes to the list while searching for the
>> matrix_mdev to
>> which a queue device's APQN is assigned. This is due to the fact that
>> the
>> proper locking order requires that the matrix_dev->mdevs_lock be taken
>> after both the matrix_mdev->kvm->lock and the matrix_dev->mdevs_lock.
>> Consequently, the matrix_dev->guests_lock will be used to protect
>> against
>> removal of a matrix_mdev object from the list while a queue device is
>> being probed. This necessitates changes to the mdev probe/remove
>> callback functions to take the matrix_dev->guests_lock prior to removing
>> a matrix_mdev object from the list.
>>
>> A new macro is also introduced to acquire the locks required to
>> dynamically
>> update the guest's APCB in the proper order when a queue device is
>> removed.
>>
>> Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
>> ---
>>   drivers/s390/crypto/vfio_ap_ops.c | 126 +++++++++++++++++++++---------
>>   1 file changed, 88 insertions(+), 38 deletions(-)
>>
>> diff --git a/drivers/s390/crypto/vfio_ap_ops.c
>> b/drivers/s390/crypto/vfio_ap_ops.c
>> index 2219b1069ceb..080a733f7cd2 100644
>> --- a/drivers/s390/crypto/vfio_ap_ops.c
>> +++ b/drivers/s390/crypto/vfio_ap_ops.c
>> @@ -116,6 +116,74 @@ static const struct vfio_device_ops
>> vfio_ap_matrix_dev_ops;
>>       mutex_unlock(&matrix_dev->guests_lock);        \
>>   })
>>   +/**
>> + * vfio_ap_mdev_get_update_locks_for_apqn: retrieve the matrix mdev
>> to which an
>> + *                       APQN is assigned and acquire the
>> + *                       locks required to update the APCB of
>> + *                       the KVM guest to which the mdev is
>> + *                       attached.
>> + *
>> + * @apqn: the APQN of a queue device.
>> + *
>> + * The proper locking order is:
>> + * 1. matrix_dev->guests_lock: required to use the KVM pointer to
>> update a KVM
>> + *                   guest's APCB.
>> + * 2. matrix_mdev->kvm->lock:  required to update a guest's APCB
>> + * 3. matrix_dev->mdevs_lock:  required to access data stored in a
>> matrix_mdev
>> + *
>> + * Note: If @apqn is not assigned to a matrix_mdev, the
>> matrix_mdev->kvm->lock
>> + *     will not be taken.
>> + *
>> + * Return: the ap_matrix_mdev object to which @apqn is assigned or
>> NULL if @apqn
>> + *       is not assigned to an ap_matrix_mdev.
>> + */
>> +static struct ap_matrix_mdev
>> *vfio_ap_mdev_get_update_locks_for_apqn(int apqn)
>> +{
>> +    struct ap_matrix_mdev *matrix_mdev;
>> +
>> +    mutex_lock(&matrix_dev->guests_lock);
>> +
>> +    list_for_each_entry(matrix_mdev, &matrix_dev->mdev_list, node) {
>> +        if (test_bit_inv(AP_QID_CARD(apqn), matrix_mdev->matrix.apm) &&
>> +            test_bit_inv(AP_QID_QUEUE(apqn),
>> matrix_mdev->matrix.aqm)) {
>> +            if (matrix_mdev->kvm)
>> +                mutex_lock(&matrix_mdev->kvm->lock);
>> +
>> +            mutex_lock(&matrix_dev->mdevs_lock);
>> +
>> +            return matrix_mdev;
>> +        }
>> +    }
>> +
>> +    mutex_lock(&matrix_dev->mdevs_lock);
>> +
>> +    return NULL;
>> +}
>> +
>> +/**
>> + * get_update_locks_for_queue: get the locks required to update the
>> APCB of the
>> + *                   KVM guest to which the matrix mdev linked to a
>> + *                   vfio_ap_queue object is attached.
>> + *
>> + * @queue: a pointer to a vfio_ap_queue object.
>> + *
>> + * The proper locking order is:
>> + * 1. matrix_dev->guests_lock: required to use the KVM pointer to
>> update a KVM
>> + *                guest's APCB.
>> + * 2. queue->matrix_mdev->kvm->lock: required to update a guest's APCB
>> + * 3. matrix_dev->mdevs_lock:    required to access data stored in a
>> matrix_mdev
>> + *
>> + * Note: if @queue is not linked to an ap_matrix_mdev object, the
>> KVM lock
>> + *      will not be taken.
>> + */
>> +#define get_update_locks_for_queue(queue) ({            \
>> +    struct ap_matrix_mdev *matrix_mdev = q->matrix_mdev; \
>> +    mutex_lock(&matrix_dev->guests_lock);            \
>> +    if (matrix_mdev && matrix_mdev->kvm) \
>> +        mutex_lock(&matrix_mdev->kvm->lock);        \
>> +    mutex_lock(&matrix_dev->mdevs_lock);            \
>> +})
>> +
>
>
> One more comment I forgot to include before:
> This macro is far too similar to existing macro,
> get_update_locks_for_mdev. And it is only called in one place. Let's
> remove this and replace the single invocation with:
>
> get_update_locks_for_mdev(q->matrix_mdev);

We can't do that, but your comment does point out a flaw in this macro;
namely, we must take the matrix_dev->guests_lock before attempting to
access q->matrix_mdev.

An ap_matrix_mdev can be unlinked from a vfio_ap_queue (and vice versa)
when the queue is removed, but it also can be unlinked when an adapter
or domain is unassigned from an ap_matrix_mdev. In order to ensure that
the q->matrix_mdev is not in the process of being nullified (unlinked),
we must be holding the guests_lock which is also held when an adapter or
domain is unassigned.

>
>
>>   /**
>>    * vfio_ap_mdev_get_queue - retrieve a queue with a specific APQN
>> from a
>>    *                hash table of queues assigned to a matrix mdev
>> @@ -615,21 +683,18 @@ static int vfio_ap_mdev_probe(struct
>> mdev_device *mdev)
>>       matrix_mdev->pqap_hook = handle_pqap;
>>       vfio_ap_matrix_init(&matrix_dev->info, &matrix_mdev->shadow_apcb);
>>       hash_init(matrix_mdev->qtable.queues);
>> -    mdev_set_drvdata(mdev, matrix_mdev);
>> -    mutex_lock(&matrix_dev->mdevs_lock);
>> -    list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
>> -    mutex_unlock(&matrix_dev->mdevs_lock);
>>         ret = vfio_register_emulated_iommu_dev(&matrix_mdev->vdev);
>>       if (ret)
>>           goto err_list;
>> +    mdev_set_drvdata(mdev, matrix_mdev);
>> +    mutex_lock(&matrix_dev->mdevs_lock);
>> +    list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
>> +    mutex_unlock(&matrix_dev->mdevs_lock);
>>       dev_set_drvdata(&mdev->dev, matrix_mdev);
>>       return 0;
>>     err_list:
>> -    mutex_lock(&matrix_dev->mdevs_lock);
>> -    list_del(&matrix_mdev->node);
>> -    mutex_unlock(&matrix_dev->mdevs_lock);
>>       vfio_uninit_group_dev(&matrix_mdev->vdev);
>>       kfree(matrix_mdev);
>>   err_dec_available:
>> @@ -692,11 +757,13 @@ static void vfio_ap_mdev_remove(struct
>> mdev_device *mdev)
>>         vfio_unregister_group_dev(&matrix_mdev->vdev);
>>   +    mutex_lock(&matrix_dev->guests_lock);
>>       mutex_lock(&matrix_dev->mdevs_lock);
>>       vfio_ap_mdev_reset_queues(matrix_mdev);
>>       vfio_ap_mdev_unlink_fr_queues(matrix_mdev);
>>       list_del(&matrix_mdev->node);
>>       mutex_unlock(&matrix_dev->mdevs_lock);
>> +    mutex_unlock(&matrix_dev->guests_lock);
>>       vfio_uninit_group_dev(&matrix_mdev->vdev);
>>       kfree(matrix_mdev);
>>       atomic_inc(&matrix_dev->available_instances);
>> @@ -1665,49 +1732,30 @@ void vfio_ap_mdev_unregister(void)
>>       mdev_unregister_driver(&vfio_ap_matrix_driver);
>>   }
>>   -/*
>> - * vfio_ap_queue_link_mdev
>> - *
>> - * @q: The queue to link with the matrix mdev.
>> - *
>> - * Links @q with the matrix mdev to which the queue's APQN is assigned.
>> - */
>> -static void vfio_ap_queue_link_mdev(struct vfio_ap_queue *q)
>> -{
>> -    unsigned long apid = AP_QID_CARD(q->apqn);
>> -    unsigned long apqi = AP_QID_QUEUE(q->apqn);
>> -    struct ap_matrix_mdev *matrix_mdev;
>> -
>> -    list_for_each_entry(matrix_mdev, &matrix_dev->mdev_list, node) {
>> -        if (test_bit_inv(apid, matrix_mdev->matrix.apm) &&
>> -            test_bit_inv(apqi, matrix_mdev->matrix.aqm)) {
>> -            vfio_ap_mdev_link_queue(matrix_mdev, q);
>> -            break;
>> -        }
>> -    }
>> -}
>> -
>>   int vfio_ap_mdev_probe_queue(struct ap_device *apdev)
>>   {
>>       struct vfio_ap_queue *q;
>> +    struct ap_matrix_mdev *matrix_mdev;
>>       DECLARE_BITMAP(apm_delta, AP_DEVICES);
>>         q = kzalloc(sizeof(*q), GFP_KERNEL);
>>       if (!q)
>>           return -ENOMEM;
>> -    mutex_lock(&matrix_dev->mdevs_lock);
>>       q->apqn = to_ap_queue(&apdev->device)->qid;
>>       q->saved_isc = VFIO_AP_ISC_INVALID;
>> -    vfio_ap_queue_link_mdev(q);
>> -    if (q->matrix_mdev) {
>> +
>> +    matrix_mdev = vfio_ap_mdev_get_update_locks_for_apqn(q->apqn);
>> +
>> +    if (matrix_mdev) {
>> +        vfio_ap_mdev_link_queue(matrix_mdev, q);
>>           memset(apm_delta, 0, sizeof(apm_delta));
>>           set_bit_inv(AP_QID_CARD(q->apqn), apm_delta);
>>           vfio_ap_mdev_filter_matrix(apm_delta,
>> -                       q->matrix_mdev->matrix.aqm,
>> -                       q->matrix_mdev);
>> +                       matrix_mdev->matrix.aqm,
>> +                       matrix_mdev);
>>       }
>>       dev_set_drvdata(&apdev->device, q);
>> -    mutex_unlock(&matrix_dev->mdevs_lock);
>> +    release_update_locks_for_mdev(matrix_mdev);
>>         return 0;
>>   }
>> @@ -1716,11 +1764,13 @@ void vfio_ap_mdev_remove_queue(struct
>> ap_device *apdev)
>>   {
>>       unsigned long apid;
>>       struct vfio_ap_queue *q;
>> +    struct ap_matrix_mdev *matrix_mdev;
>>   -    mutex_lock(&matrix_dev->mdevs_lock);
>>       q = dev_get_drvdata(&apdev->device);
>> +    get_update_locks_for_queue(q);
>> +    matrix_mdev = q->matrix_mdev;
>>   -    if (q->matrix_mdev) {
>> +    if (matrix_mdev) {
>>           vfio_ap_unlink_queue_fr_mdev(q);
>>             apid = AP_QID_CARD(q->apqn);
>> @@ -1731,5 +1781,5 @@ void vfio_ap_mdev_remove_queue(struct ap_device
>> *apdev)
>>       vfio_ap_mdev_reset_queue(q, 1);
>>       dev_set_drvdata(&apdev->device, NULL);
>>       kfree(q);
>> -    mutex_unlock(&matrix_dev->mdevs_lock);
>> +    release_update_locks_for_mdev(matrix_mdev);
>>   }
>
>

\
 
 \ /
  Last update: 2022-05-31 13:58    [W:0.425 / U:0.352 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site