RE: [PATCH v3 18/37] vfio/container: Implement attach/detach_device

[Duan, Zhenzhong]

>-----Original Message-----
>From: Cédric Le Goater <clg@redhat.com>
>Sent: Saturday, October 28, 2023 12:07 AM
>Subject: Re: [PATCH v3 18/37] vfio/container: Implement attach/detach_device
>
>Sorry, previous email was empty. Friday effect !
>
>On 10/26/23 12:30, Zhenzhong Duan wrote:
>> From: Eric Auger <eric.auger@redhat.com>
>>
>> No fucntional change intended.
>>
>> Signed-off-by: Eric Auger <eric.auger@redhat.com>
>> Signed-off-by: Yi Liu <yi.l.liu@intel.com>
>> Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
>> Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
>> Signed-off-by: Cédric Le Goater <clg@redhat.com>
>> ---
>>   hw/vfio/common.c    | 16 ++++++++++++++++
>>   hw/vfio/container.c | 12 +++++-------
>>   2 files changed, 21 insertions(+), 7 deletions(-)
>>
>> diff --git a/hw/vfio/common.c b/hw/vfio/common.c
>> index d62c815d7f..64565b4ae9 100644
>> --- a/hw/vfio/common.c
>> +++ b/hw/vfio/common.c
>> @@ -1500,3 +1500,19 @@ retry:
>>
>>       return info;
>>   }
>> +
>> +int vfio_attach_device(char *name, VFIODevice *vbasedev,
>> +                       AddressSpace *as, Error **errp)
>> +{
>> +    const VFIOIOMMUOps *ops = &vfio_legacy_ops;
>
>hmm, this looks wrong. please explain.

The final shape will be:

int vfio_attach_device(char *name, VFIODevice *vbasedev,
                       AddressSpace *as, Error **errp)
{
    const VFIOIOMMUOps *ops;

#ifdef CONFIG_IOMMUFD
    if (vbasedev->iommufd) {
        ops = &vfio_iommufd_ops;
    } else
#endif
    {
        ops = &vfio_legacy_ops;
    }
    return ops->attach_device(name, vbasedev, as, errp);
}

Depending on if iommufd is selected, different ops will be chosen.
Then corresponding attach_device callback is called to attach to
iommufd or legacy container.

Thanks
Zhenzhong