Re: [PATCH v2 4/9] vfio: Introduce vfio_get_vfio_device()

["Cédric Le Goater" <clg@redhat.com>]

On 2/10/25 15:32, Philippe Mathieu-Daudé wrote:
> On 30/1/25 14:43, Cédric Le Goater wrote:
>> This helper will be useful in the listener handlers to extract the
>> VFIO device from a memory region using memory_region_owner(). At the
>> moment, we only care for PCI passthrough devices. If the need arises,
>> we will add more.
>>
>> Signed-off-by: Cédric Le Goater <clg@redhat.com>
>> ---
>>   include/hw/vfio/vfio-common.h |  1 +
>>   hw/vfio/helpers.c             | 10 ++++++++++
>>   2 files changed, 11 insertions(+)
>>
>> diff --git a/include/hw/vfio/vfio-common.h b/include/hw/vfio/vfio-common.h
>> index 0c60be5b15c70168f4f94ad7054d9bd750a162d3..ac35136a11051b079cd9d04e6becd344a0e0f7e7 100644
>> --- a/include/hw/vfio/vfio-common.h
>> +++ b/include/hw/vfio/vfio-common.h
>> @@ -252,6 +252,7 @@ bool vfio_device_hiod_realize(VFIODevice *vbasedev, Error **errp);
>>   bool vfio_attach_device(char *name, VFIODevice *vbasedev,
>>                           AddressSpace *as, Error **errp);
>>   void vfio_detach_device(VFIODevice *vbasedev);
>> +VFIODevice *vfio_get_vfio_device(Object *obj);
>>   int vfio_kvm_device_add_fd(int fd, Error **errp);
>>   int vfio_kvm_device_del_fd(int fd, Error **errp);
>> diff --git a/hw/vfio/helpers.c b/hw/vfio/helpers.c
>> index 913796f437f84eece8711cb4b4b654a44040d17c..4b255d4f3a9e81f55df00c68fc71da769fd5bd04 100644
>> --- a/hw/vfio/helpers.c
>> +++ b/hw/vfio/helpers.c
>> @@ -23,6 +23,7 @@
>>   #include <sys/ioctl.h>
>>   #include "hw/vfio/vfio-common.h"
>> +#include "hw/vfio/pci.h"
>>   #include "hw/hw.h"
>>   #include "trace.h"
>>   #include "qapi/error.h"
>> @@ -728,3 +729,12 @@ bool vfio_device_hiod_realize(VFIODevice *vbasedev, Error **errp)
>>       return HOST_IOMMU_DEVICE_GET_CLASS(hiod)->realize(hiod, vbasedev, errp);
>>   }
>> +
>> +VFIODevice *vfio_get_vfio_device(Object *obj)
> 
> Can't we take a VFIOPCIDevice argument?

It could be some other VFIO (AP, Plateform) device type and we wouldn't
need to do the cast below in that case.


Thanks,

C.



> 
>> +{
>> +    if (object_dynamic_cast(obj, TYPE_VFIO_PCI)) {
>> +        return &VFIO_PCI(obj)->vbasedev;
>> +    } else {
>> +        return NULL;
>> +    }
>> +}
>