[PATCH v3 0/9] s390/vfio-ap: Fix bugs in vfio_ap device driver callback functions

Anthony Krowiak posted 9 patches 1 month, 2 weeks ago
There is a newer version of this series
drivers/s390/crypto/vfio_ap_ops.c | 110 ++++++++++++++++++++----------
1 file changed, 75 insertions(+), 35 deletions(-)
[PATCH v3 0/9] s390/vfio-ap: Fix bugs in vfio_ap device driver callback functions
Posted by Anthony Krowiak 1 month, 2 weeks ago
During review of patches by the Sashiko AI, several pre-existing bugs were
discovered. This 9-patch series fixes those bugs

Change log v2 => v3:
~~~~~~~~~~~~~~~~~~~
Patch 3: Fix use of wrong lock in mdev probe function
Patch 7: Fix required lock not held during display of sysfs status
attribute 
* Squashed these two patches into a new patch entitled "Fix missing lock
  required to access list of ap_matrix_mdev objects"
* Added two new patches that fix pre-existing bugs:
  ~ s390/vfio-ap: Fix NULL deref in status_show() during queue probe
  ~ s390/vfio-ap: Fix memory leak when queue removed from host AP config

Anthony Krowiak (9):
  s390/vfio-ap: Fix stale do_remove flag across iterations in
    vfio_ap_mdev_cfg_remove
  s390/vfio-ap: Fix dereference matrix_mdev->kvm without checking for
    NULL
  s390/vfio-ap: Fix missing lock required to access list of
    ap_matrix_mdev objects
  s390/vfio-ap: Fix required lock not held during update of
    ap_matrix_mdev object
  s390/vfio-ap: Fix control domain removal in vfio_ap_mdev_cfg_remove
  s390/vfio-ap: fix potential use of uninitialized apm_filtered bitmap
  s390/vfio-ap: Fix hot-unplug skipped when last AP adapter or domain
    removed
  s390/vfio-ap: Fix NULL deref in status_show() during queue probe
  s390/vfio-ap: Fix memory leak when queue removed from host AP config

 drivers/s390/crypto/vfio_ap_ops.c | 110 ++++++++++++++++++++----------
 1 file changed, 75 insertions(+), 35 deletions(-)

-- 
2.53.0
Re: [PATCH v3 0/9] s390/vfio-ap: Fix bugs in vfio_ap device driver callback functions
Posted by Christian Borntraeger 1 month, 2 weeks ago
Am 12.08.26 um 13:51 schrieb Anthony Krowiak:
> During review of patches by the Sashiko AI, several pre-existing bugs were
> discovered. This 9-patch series fixes those bugs
> 
> Change log v2 => v3:
> ~~~~~~~~~~~~~~~~~~~
> Patch 3: Fix use of wrong lock in mdev probe function
> Patch 7: Fix required lock not held during display of sysfs status
> attribute
> * Squashed these two patches into a new patch entitled "Fix missing lock
>    required to access list of ap_matrix_mdev objects"
> * Added two new patches that fix pre-existing bugs:
>    ~ s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>    ~ s390/vfio-ap: Fix memory leak when queue removed from host AP config
> 
> Anthony Krowiak (9):
>    s390/vfio-ap: Fix stale do_remove flag across iterations in
>      vfio_ap_mdev_cfg_remove
>    s390/vfio-ap: Fix dereference matrix_mdev->kvm without checking for
>      NULL
>    s390/vfio-ap: Fix missing lock required to access list of
>      ap_matrix_mdev objects
>    s390/vfio-ap: Fix required lock not held during update of
>      ap_matrix_mdev object
>    s390/vfio-ap: Fix control domain removal in vfio_ap_mdev_cfg_remove
>    s390/vfio-ap: fix potential use of uninitialized apm_filtered bitmap
>    s390/vfio-ap: Fix hot-unplug skipped when last AP adapter or domain
>      removed
>    s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>    s390/vfio-ap: Fix memory leak when queue removed from host AP config
> 
>   drivers/s390/crypto/vfio_ap_ops.c | 110 ++++++++++++++++++++----------
>   1 file changed, 75 insertions(+), 35 deletions(-)
> 


Sashiko has one new finding for patch8. Everything else is unrelated.
Will you do a respin for this?
I plan to take this for the next merge window.
Re: [PATCH v3 0/9] s390/vfio-ap: Fix bugs in vfio_ap device driver callback functions
Posted by Matthew Rosato 1 month, 2 weeks ago
On 8/12/26 11:30 AM, Christian Borntraeger wrote:
> 
> Am 12.08.26 um 13:51 schrieb Anthony Krowiak:
>> During review of patches by the Sashiko AI, several pre-existing bugs
>> were
>> discovered. This 9-patch series fixes those bugs
>>
>> Change log v2 => v3:
>> ~~~~~~~~~~~~~~~~~~~
>> Patch 3: Fix use of wrong lock in mdev probe function
>> Patch 7: Fix required lock not held during display of sysfs status
>> attribute
>> * Squashed these two patches into a new patch entitled "Fix missing lock
>>    required to access list of ap_matrix_mdev objects"
>> * Added two new patches that fix pre-existing bugs:
>>    ~ s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>>    ~ s390/vfio-ap: Fix memory leak when queue removed from host AP config
>>
>> Anthony Krowiak (9):
>>    s390/vfio-ap: Fix stale do_remove flag across iterations in
>>      vfio_ap_mdev_cfg_remove
>>    s390/vfio-ap: Fix dereference matrix_mdev->kvm without checking for
>>      NULL
>>    s390/vfio-ap: Fix missing lock required to access list of
>>      ap_matrix_mdev objects
>>    s390/vfio-ap: Fix required lock not held during update of
>>      ap_matrix_mdev object
>>    s390/vfio-ap: Fix control domain removal in vfio_ap_mdev_cfg_remove
>>    s390/vfio-ap: fix potential use of uninitialized apm_filtered bitmap
>>    s390/vfio-ap: Fix hot-unplug skipped when last AP adapter or domain
>>      removed
>>    s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>>    s390/vfio-ap: Fix memory leak when queue removed from host AP config
>>
>>   drivers/s390/crypto/vfio_ap_ops.c | 110 ++++++++++++++++++++----------
>>   1 file changed, 75 insertions(+), 35 deletions(-)
>>
> 
> 
> Sashiko has one new finding for patch8. Everything else is unrelated.
> Will you do a respin for this?

I am going thru this version now but yes I would like a respin that
fixes patch 8.

> I plan to take this for the next merge window.

I am also going thru all of the Sashiko reports, most are already
resolved by this series.  But I will send Tony a pruned list to
investigate after this series vs adding more fixes to this one.
Re: [PATCH v3 0/9] s390/vfio-ap: Fix bugs in vfio_ap device driver callback functions
Posted by Christian Borntraeger 1 month, 2 weeks ago

Am 12.08.26 um 17:36 schrieb Matthew Rosato:
> On 8/12/26 11:30 AM, Christian Borntraeger wrote:
>>
>> Am 12.08.26 um 13:51 schrieb Anthony Krowiak:
>>> During review of patches by the Sashiko AI, several pre-existing bugs
>>> were
>>> discovered. This 9-patch series fixes those bugs
>>>
>>> Change log v2 => v3:
>>> ~~~~~~~~~~~~~~~~~~~
>>> Patch 3: Fix use of wrong lock in mdev probe function
>>> Patch 7: Fix required lock not held during display of sysfs status
>>> attribute
>>> * Squashed these two patches into a new patch entitled "Fix missing lock
>>>     required to access list of ap_matrix_mdev objects"
>>> * Added two new patches that fix pre-existing bugs:
>>>     ~ s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>>>     ~ s390/vfio-ap: Fix memory leak when queue removed from host AP config
>>>
>>> Anthony Krowiak (9):
>>>     s390/vfio-ap: Fix stale do_remove flag across iterations in
>>>       vfio_ap_mdev_cfg_remove
>>>     s390/vfio-ap: Fix dereference matrix_mdev->kvm without checking for
>>>       NULL
>>>     s390/vfio-ap: Fix missing lock required to access list of
>>>       ap_matrix_mdev objects
>>>     s390/vfio-ap: Fix required lock not held during update of
>>>       ap_matrix_mdev object
>>>     s390/vfio-ap: Fix control domain removal in vfio_ap_mdev_cfg_remove
>>>     s390/vfio-ap: fix potential use of uninitialized apm_filtered bitmap
>>>     s390/vfio-ap: Fix hot-unplug skipped when last AP adapter or domain
>>>       removed
>>>     s390/vfio-ap: Fix NULL deref in status_show() during queue probe
>>>     s390/vfio-ap: Fix memory leak when queue removed from host AP config
>>>
>>>    drivers/s390/crypto/vfio_ap_ops.c | 110 ++++++++++++++++++++----------
>>>    1 file changed, 75 insertions(+), 35 deletions(-)
>>>
>>
>>
>> Sashiko has one new finding for patch8. Everything else is unrelated.
>> Will you do a respin for this?
> 
> I am going thru this version now but yes I would like a respin that
> fixes patch 8.
> 
>> I plan to take this for the next merge window.
> 
> I am also going thru all of the Sashiko reports, most are already
> resolved by this series.  But I will send Tony a pruned list to
> investigate after this series vs adding more fixes to this one.

Let me know when this is ready. Apart from patch 8 I am happy.