From: Tony Krowiak <akrowiak@linux.ibm.com>
To: Halil Pasic <pasic@linux.ibm.com>
Cc: linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org,
kvm@vger.kernel.org, freude@linux.ibm.com,
borntraeger@de.ibm.com, cohuck@redhat.com,
mjrosato@linux.ibm.com, alex.williamson@redhat.com,
kwankhede@nvidia.com, fiuczy@linux.ibm.com,
frankja@linux.ibm.com, david@redhat.com, hca@linux.ibm.com,
gor@linux.ibm.com
Subject: Re: [PATCH v13 05/15] s390/vfio-ap: manage link between queue struct and matrix mdev
Date: Thu, 14 Jan 2021 16:10:37 -0500 [thread overview]
Message-ID: <e32ea832-8d4c-1069-9bd8-d92ae210a55a@linux.ibm.com> (raw)
In-Reply-To: <20210114035009.375c5496.pasic@linux.ibm.com>
On 1/13/21 9:50 PM, Halil Pasic wrote:
> On Wed, 13 Jan 2021 16:41:27 -0500
> Tony Krowiak <akrowiak@linux.ibm.com> wrote:
>
>> On 1/11/21 2:17 PM, Halil Pasic wrote:
>>> On Tue, 22 Dec 2020 20:15:56 -0500
>>> Tony Krowiak <akrowiak@linux.ibm.com> wrote:
>>>
>>>> Let's create links between each queue device bound to the vfio_ap device
>>>> driver and the matrix mdev to which the queue's APQN is assigned. The idea
>>>> is to facilitate efficient retrieval of the objects representing the queue
>>>> devices and matrix mdevs as well as to verify that a queue assigned to
>>>> a matrix mdev is bound to the driver.
>>>>
>>>> The links will be created as follows:
>>>>
>>>> * When the queue device is probed, if its APQN is assigned to a matrix
>>>> mdev, the structures representing the queue device and the matrix mdev
>>>> will be linked.
>>>>
>>>> * When an adapter or domain is assigned to a matrix mdev, for each new
>>>> APQN assigned that references a queue device bound to the vfio_ap
>>>> device driver, the structures representing the queue device and the
>>>> matrix mdev will be linked.
>>>>
>>>> The links will be removed as follows:
>>>>
>>>> * When the queue device is removed, if its APQN is assigned to a matrix
>>>> mdev, the structures representing the queue device and the matrix mdev
>>>> will be unlinked.
>>>>
>>>> * When an adapter or domain is unassigned from a matrix mdev, for each
>>>> APQN unassigned that references a queue device bound to the vfio_ap
>>>> device driver, the structures representing the queue device and the
>>>> matrix mdev will be unlinked.
>>>>
>>>> Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
>>> Reviewed-by: Halil Pasic <pasic@linux.ibm.com>
>>>
> [..]
>
>>>> +
>>>> int vfio_ap_mdev_probe_queue(struct ap_device *apdev)
>>>> {
>>>> struct vfio_ap_queue *q;
>>>> @@ -1324,9 +1404,13 @@ int vfio_ap_mdev_probe_queue(struct ap_device *apdev)
>>>> q = kzalloc(sizeof(*q), GFP_KERNEL);
>>>> if (!q)
>>>> return -ENOMEM;
>>>> + mutex_lock(&matrix_dev->lock);
>>>> dev_set_drvdata(&apdev->device, q);
>>>> q->apqn = to_ap_queue(&apdev->device)->qid;
>>>> q->saved_isc = VFIO_AP_ISC_INVALID;
>>>> + vfio_ap_queue_link_mdev(q);
>>>> + mutex_unlock(&matrix_dev->lock);
>>>> +
>>> Does the critical section have to include more than just
>>> vfio_ap_queue_link_mdev()? Did we need the critical section
>>> before this patch?
>> We did not need the critical section before this patch because
>> the only function that retrieved the vfio_ap_queue via the queue
>> device's drvdata was the remove callback. I included the initialization
>> of the vfio_ap_queue object under lock because the
>> vfio_ap_find_queue() function retrieves the vfio_ap_queue object from
>> the queue device's drvdata so it might be advantageous to initialize
>> it under the mdev lock. On the other hand, I can't come up with a good
>> argument to change this.
>>
>>
> I was asking out of curiosity, not because I want it changed. I was
> also wondering if somebody could see a partially initialized device:
> we even first call dev_set_drvdata() and only then finish the
> initialization. Before 's390/vfio-ap: use new AP bus interface to search
> for queue devices', which is the previous patch, we had the klist code
> in between, which uses spinlocks, which I think ensure, that all
> effects of probe are seen when we get the queue from
> vfio_ap_find_queue(). But with patch 4 in place that is not the case any
> more. Or am I wrong?
You are correct insofar as patch 4 replaces the driver_find_device()
function call with a call to AP bus's ap_get_qdev() function which
does not use spinlocks. Without digging deeply into the probe call
chain I do not know whether or not the use of spinlocks by the klist
code ensures all effects of the probe are seen when we get the
queue from vfio_ap_find_queue(). What I'm sure about is that since
both vfio_ap_find_queue() and the setting of the drvdata in the
probe function are always done under the mdev lock, consistency
should be maintained. What I did decide when thinking about your
previous review comment is that we should probably initialize the
vfio_ap_queue object before setting the drvdata, so I made that change.
>
> Regards,
> Halil
next prev parent reply other threads:[~2021-01-14 21:11 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-23 1:15 [PATCH v13 00/15] s390/vfio-ap: dynamic configuration support Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 01/15] s390/vfio-ap: clean up vfio_ap resources when KVM pointer invalidated Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 02/15] s390/vfio-ap: No need to disable IRQ after queue reset Tony Krowiak
2021-01-11 16:32 ` Halil Pasic
2021-01-13 17:06 ` Tony Krowiak
2021-01-13 21:21 ` Halil Pasic
2021-01-14 0:46 ` Tony Krowiak
2021-01-14 3:13 ` Halil Pasic
2020-12-23 1:15 ` [PATCH v13 03/15] s390/vfio-ap: move probe and remove callbacks to vfio_ap_ops.c Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 04/15] s390/vfio-ap: use new AP bus interface to search for queue devices Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 05/15] s390/vfio-ap: manage link between queue struct and matrix mdev Tony Krowiak
2021-01-11 19:17 ` Halil Pasic
2021-01-13 21:41 ` Tony Krowiak
2021-01-14 2:50 ` Halil Pasic
2021-01-14 21:10 ` Tony Krowiak [this message]
2020-12-23 1:15 ` [PATCH v13 06/15] s390/vfio-ap: allow assignment of unavailable AP queues to mdev device Tony Krowiak
2021-01-11 20:40 ` Halil Pasic
2021-01-14 17:54 ` Tony Krowiak
2021-01-15 1:08 ` Halil Pasic
2021-01-15 1:44 ` Halil Pasic
2021-03-31 14:36 ` Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 07/15] s390/vfio-ap: introduce shadow APCB Tony Krowiak
2021-01-11 22:50 ` Halil Pasic
2021-01-14 21:35 ` Tony Krowiak
2020-12-23 1:15 ` [PATCH v13 08/15] s390/vfio-ap: sysfs attribute to display the guest's matrix Tony Krowiak
2021-01-11 22:58 ` Halil Pasic
2021-01-28 21:29 ` Tony Krowiak
2020-12-23 1:16 ` [PATCH v13 09/15] s390/vfio-ap: allow hot plug/unplug of AP resources using mdev device Tony Krowiak
2021-01-12 1:12 ` Halil Pasic
2021-01-12 17:55 ` Halil Pasic
2021-02-01 14:41 ` Tony Krowiak
2021-02-03 23:13 ` Tony Krowiak
2021-02-04 0:21 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 10/15] s390/zcrypt: driver callback to indicate resource in use Tony Krowiak
2021-01-12 16:50 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 11/15] s390/vfio-ap: implement in-use callback for vfio_ap driver Tony Krowiak
2021-01-12 1:20 ` Halil Pasic
2021-01-12 14:14 ` Matthew Rosato
2021-01-12 16:49 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 12/15] s390/zcrypt: Notify driver on config changed and scan complete callbacks Tony Krowiak
2021-01-12 16:58 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 13/15] s390/vfio-ap: handle host AP config change notification Tony Krowiak
2021-01-12 18:39 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 14/15] s390/vfio-ap: handle AP bus scan completed notification Tony Krowiak
2021-01-12 18:44 ` Halil Pasic
2020-12-23 1:16 ` [PATCH v13 15/15] s390/vfio-ap: update docs to include dynamic config support Tony Krowiak
2021-01-06 15:16 ` [PATCH v13 00/15] s390/vfio-ap: dynamic configuration support Tony Krowiak
2021-01-07 14:41 ` Halil Pasic
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e32ea832-8d4c-1069-9bd8-d92ae210a55a@linux.ibm.com \
--to=akrowiak@linux.ibm.com \
--cc=alex.williamson@redhat.com \
--cc=borntraeger@de.ibm.com \
--cc=cohuck@redhat.com \
--cc=david@redhat.com \
--cc=fiuczy@linux.ibm.com \
--cc=frankja@linux.ibm.com \
--cc=freude@linux.ibm.com \
--cc=gor@linux.ibm.com \
--cc=hca@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=kwankhede@nvidia.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=mjrosato@linux.ibm.com \
--cc=pasic@linux.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).