Re: [PATCH v7 01/15] s390/vfio-ap: store queue struct in hash table for quick access

From: Tony Krowiak <akrowiak@linux.ibm.com>
To: Cornelia Huck <cohuck@redhat.com>
Cc: linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org,
	kvm@vger.kernel.org, freude@linux.ibm.com,
	borntraeger@de.ibm.com, mjrosato@linux.ibm.com,
	pmorel@linux.ibm.com, pasic@linux.ibm.com,
	alex.williamson@redhat.com, kwankhede@nvidia.com,
	jjherne@linux.ibm.com, fiuczy@linux.ibm.com
Subject: Re: [PATCH v7 01/15] s390/vfio-ap: store queue struct in hash table for quick access
Date: Wed, 8 Apr 2020 11:38:07 -0400	[thread overview]
Message-ID: <e0d56b61-749a-3646-18e7-47bb5c8ca862@linux.ibm.com> (raw)
In-Reply-To: <20200408124801.2d61bc5b.cohuck@redhat.com>

On 4/8/20 6:48 AM, Cornelia Huck wrote:
> On Tue,  7 Apr 2020 15:20:01 -0400
> Tony Krowiak <akrowiak@linux.ibm.com> wrote:
>
>> Rather than looping over potentially 65535 objects, let's store the
>> structures for caching information about queue devices bound to the
>> vfio_ap device driver in a hash table keyed by APQN.
> This also looks like a nice code simplification.
>
>> Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
>> ---
>>   drivers/s390/crypto/vfio_ap_drv.c     | 28 +++------
>>   drivers/s390/crypto/vfio_ap_ops.c     | 90 ++++++++++++++-------------
>>   drivers/s390/crypto/vfio_ap_private.h | 10 ++-
>>   3 files changed, 60 insertions(+), 68 deletions(-)
>>
> (...)
>
>> diff --git a/drivers/s390/crypto/vfio_ap_ops.c b/drivers/s390/crypto/vfio_ap_ops.c
>> index 5c0f53c6dde7..134860934fe7 100644
>> --- a/drivers/s390/crypto/vfio_ap_ops.c
>> +++ b/drivers/s390/crypto/vfio_ap_ops.c
>> @@ -26,45 +26,16 @@
>>   
>>   static int vfio_ap_mdev_reset_queues(struct mdev_device *mdev);
>>   
>> -static int match_apqn(struct device *dev, const void *data)
>> -{
>> -	struct vfio_ap_queue *q = dev_get_drvdata(dev);
>> -
>> -	return (q->apqn == *(int *)(data)) ? 1 : 0;
>> -}
>> -
>> -/**
>> - * vfio_ap_get_queue: Retrieve a queue with a specific APQN from a list
>> - * @matrix_mdev: the associated mediated matrix
>> - * @apqn: The queue APQN
>> - *
>> - * Retrieve a queue with a specific APQN from the list of the
>> - * devices of the vfio_ap_drv.
>> - * Verify that the APID and the APQI are set in the matrix.
>> - *
>> - * Returns the pointer to the associated vfio_ap_queue
> Any reason you're killing this comment, instead of adapting it? The
> function is even no longer static...

I can update and restore the comment and the function should be static.

>
>> - */
>> -static struct vfio_ap_queue *vfio_ap_get_queue(
>> -					struct ap_matrix_mdev *matrix_mdev,
>> -					int apqn)
>> +struct vfio_ap_queue *vfio_ap_get_queue(unsigned long apqn)
>>   {
>>   	struct vfio_ap_queue *q;
>> -	struct device *dev;
>> -
>> -	if (!test_bit_inv(AP_QID_CARD(apqn), matrix_mdev->matrix.apm))
>> -		return NULL;
>> -	if (!test_bit_inv(AP_QID_QUEUE(apqn), matrix_mdev->matrix.aqm))
>> -		return NULL;
> These were just optimizations and therefore can be dropped now?

The purpose of this function has changed from its previous incarnation.
This function was originally called from the handle_pqap() function and
served two purposes: It retrieved the struct vfio_ap_queue as driver data
and linked the matrix_mdev to the  vfio_ap_queue. The linking of the
matrix_mdev and the vfio_ap_queue are now done when queue devices
are probed and when adapters and domains are assigned; so now, the
handle_pqap() function calls this function to retrieve both the
vfio_ap_queue as well as the matrix_mdev to which it is linked. 
Consequently,
the above code is no longer needed.

>
>> -
>> -	dev = driver_find_device(&matrix_dev->vfio_ap_drv->driver, NULL,
>> -				 &apqn, match_apqn);
>> -	if (!dev)
>> -		return NULL;
>> -	q = dev_get_drvdata(dev);
>> -	q->matrix_mdev = matrix_mdev;
>> -	put_device(dev);
>>   
>> -	return q;
>> +	hash_for_each_possible(matrix_dev->qtable, q, qnode, apqn) {
>> +		if (q && (apqn == q->apqn))
>> +			return q;
>> +	}
> Do we need any serialization here? Previously, the driver core made
> sure we could get a reference only if the device was still registered;
> not sure if we need any further guarantees now.

The vfio_ap_queue structs are created when the queue device is
probed and removed when the queue device is removed.

>
>> +
>> +	return NULL;
>>   }
>>   
>>   /**
> (...)
>