RE: [PATCH v2 5/6] hw/arm/virt: Add support for smmuv3 device

[Shameerali Kolothum Thodi via]

> -----Original Message-----
> From: Nicolin Chen <nicol...@nvidia.com>
> Sent: Friday, May 2, 2025 6:55 PM
> To: Shameerali Kolothum Thodi <shameerali.kolothum.th...@huawei.com>
> Cc: qemu-...@nongnu.org; qemu-devel@nongnu.org;
> eric.au...@redhat.com; peter.mayd...@linaro.org; j...@nvidia.com;
> ddut...@redhat.com; berra...@redhat.com; nath...@nvidia.com;
> mo...@nvidia.com; smost...@google.com; Linuxarm
> <linux...@huawei.com>; Wangzhou (B) <wangzh...@hisilicon.com>;
> jiangkunkun <jiangkun...@huawei.com>; Jonathan Cameron
> <jonathan.came...@huawei.com>; zhangfei....@linaro.org
> Subject: Re: [PATCH v2 5/6] hw/arm/virt: Add support for smmuv3 device
> 
> On Fri, May 02, 2025 at 11:27:06AM +0100, Shameer Kolothum wrote:
> > @@ -2972,6 +3004,21 @@ static void
> virt_machine_device_plug_cb(HotplugHandler *hotplug_dev,
> >          virtio_md_pci_plug(VIRTIO_MD_PCI(dev), MACHINE(hotplug_dev),
> errp);
> >      }
> >
> > +    if (object_dynamic_cast(OBJECT(dev), TYPE_ARM_SMMUV3)) {
> 
> This seems to be on the path of a hotplug function? Mind elaborating
> why, while PATCH-1 sets hotpluggable to false?

Not really. Qemu has this path for both cold plug case(which is before the Guest
is booted, ie, -device arm-smmuv3, case) and for hot plug case (where we try to
add the device after Guest is booted, ie,  device_add arm-smmuv3, case).

We are preventing only the hot plug case.

> 
> > +        if (!vms->legacy_smmuv3_present && vms->platform_bus_dev) {
> > +            VirtMachineClass *vmc = VIRT_MACHINE_GET_CLASS(vms);
> > +
> > +            create_smmuv3_dev_dtb(vms, dev);
> > +            if (vms->iommu != VIRT_IOMMU_SMMUV3) {
> 
> Should this be VIRT_IOMMU_NONE only as the other cases are rejected?

The check is basically to handle cases where we will have multiple SMMUv3 
devices
and to set this only once.

Thanks,
Shameer