Re: 答复: 答复: 答复: 答复: 答复: 答复: 答复: [PATCH] iommu/vt-d:Add support for ACPI device in RMRR

[Lu Baolu]

Hi,

On 2020/8/7 11:29, FelixCui-oc wrote:
> Hi baolu,
> 		I understand what you mean is that you want to put the following processing code in the acpi_device_create_direct_mappings() into the probe_acpi_namespace_devices() ,right?
> 		If you mean it , I think it's OK.

Yes. If the acpi device has companion physical devices, then probe the
companion devices; otherwise, probe the acpi device instead. Also, let's
put detecting and probing acpi devices in different patches so that they
could be easier for reviewing.

Best regards,
baolu

> 	if (pn_dev == NULL) {
> 		acpi_device->bus->iommu_ops = &intel_iommu_ops;
> 		ret = iommu_probe_device(acpi_device);
> 		if (ret) {
> 			pr_err("acpi_device probe fail! ret:%d\n", ret);
> 			return ret;
> 		}
> 		return 0;
> 	}
>
> Best regards
> Felix cui-oc
>
>
>
> 	
>
> -----邮件原件-----
> 发件人: Lu Baolu <baolu.lu@xxxxxxxxxxxxxxx>
> 发送时间: 2020年8月7日 9:08
> 收件人: FelixCui-oc <FelixCui-oc@xxxxxxxxxxx>; Joerg Roedel <joro@xxxxxxxxxx>; iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; David Woodhouse <dwmw2@xxxxxxxxxxxxx>
> 抄送: baolu.lu@xxxxxxxxxxxxxxx; RaymondPang-oc <RaymondPang-oc@xxxxxxxxxxx>; CobeChen-oc <CobeChen-oc@xxxxxxxxxxx>
> 主题: Re: 答复: 答复: 答复: 答复: 答复: 答复: [PATCH] iommu/vt-d:Add support for ACPI device in RMRR
>
> Hi Felix,
>
> On 2020/8/6 14:51, FelixCui-oc wrote:
> > Hi  baolu,
> > 		>Sure. Before that, let me sync my understanding with you. You have an acpi namespace device in ANDD table, it also shows up in the device scope of a RMRR.
> > 		>Current code doesn't enumerate that device for the RMRR, hence iommu_create_device_direct_mappings() doesn't work for this device.
> >
> > 		>At the same time, probe_acpi_namespace_devices() doesn't work for this device, hence you want to add a home-made
> > 		>acpi_device_create_direct_mappings() helper.
> >
> > 		Your understanding is right.
> > 		But there is a problem that even if the namespace device in rmrr is enumerated in the code, probe_acpi_namespace_devices() also doesn't work for this device.
> > 		This is because the dev parameter of the iommu_create_device_direct_mappings() is not the namespace device in RMRR.
> > 		The actual parameter passed in is the namespace device's physical node device.
> > 		In iommu_create_device_direct_mappings(), the physical node device passed in cannot match the namespace device in rmrr->device[],right?
> > 		We need acpi_device_create_direct_mappings() helper ?
> >
> > 		In addition, adev->physical_node_list is related to the __HID of namespace device reported by the bios.
> > 		For example, if the __HID reported by the bios belongs to acpi_pnp_device_ids[], adev->physical_node_list has no devices.
> > 		So in acpi_device_create_direct_mappings(), I added the case that adev->physical_node_list is empty.
>
> Got you. Thanks!
>
> Have you ever tried to have probe_acpi_namespace_devices() handle the case of empty adev->physical_node_list at the same time?
>
> Best regards,
> baolu
>
> > Best regards
> > Felix cui
> >
> >
> >    
> >
> > -----邮件原件-----
> > 发件人: Lu Baolu <baolu.lu@xxxxxxxxxxxxxxx>
> > 发送时间: 2020年8月6日 10:36
> > 收件人: FelixCui-oc <FelixCui-oc@xxxxxxxxxxx>; Joerg Roedel
> > <joro@xxxxxxxxxx>; iommu@xxxxxxxxxxxxxxxxxxxxxxxxxx;
> > linux-kernel@xxxxxxxxxxxxxxx; David Woodhouse <dwmw2@xxxxxxxxxxxxx>
> > 抄送: baolu.lu@xxxxxxxxxxxxxxx; RaymondPang-oc
> > <RaymondPang-oc@xxxxxxxxxxx>; CobeChen-oc <CobeChen-oc@xxxxxxxxxxx>
> > 主题: Re: 答复: 答复: 答复: 答复: 答复: [PATCH] iommu/vt-d:Add support for ACPI
> > device in RMRR
> >
> > Hi Felix,
> >
> > On 8/5/20 3:37 PM, FelixCui-oc wrote:
> > > Hi baolu,
> > > 		Let me talk about why acpi_device_create_direct_mappings() is needed and please tell me if there is an error.
> >
> > Sure. Before that, let me sync my understanding with you. You have an acpi namespace device in ANDD table, it also shows up in the device scope of a RMRR. Current code doesn't enumerate that device for the RMRR, hence iommu_create_device_direct_mappings() doesn't work for this device.
> >
> > At the same time, probe_acpi_namespace_devices() doesn't work for this
> > device, hence you want to add a home-made
> > acpi_device_create_direct_mappings() helper.
> >
> > Did I get it right?
> >
> > > 		In the probe_acpi_namespace_devices() function, only the device in the addev->physical_node_list is probed,
> > > 		but we need to establish identity mapping for the namespace device in RMRR. These are two different devices.
> >
> > The namespace device has been probed and put in one drhd's device list.
> > Hence, it should be processed by probe_acpi_namespace_devices(). So the question is why there are no devices in addev->physical_node_list?
> >
> > > 		Therefore, the namespace device in RMRR is not mapped in probe_acpi_namespace_devices().
> > > 		acpi_device_create_direct_mappings() is to create direct mappings for namespace devices in RMRR.
> >
> > Best regards,
> > baolu