why PCI devices have 0x1000x domain address

[GuangyaoJiang]

Hi Dears, 

when i passthough one PCI device to my VM，The qemu-kvm got an error: 

2024-01-05T07:40:51.123478Z qemu-kvm: -device vfio-pci,host=10001:02:00.0,id=hostdev0,bus=pci.0,addr=0xc,rombar=0: Property 'vfio-pci.host' doesn't take value '10001:02:00.0'

I use lspci -D to check the pci info, it really has 0x1000x domain address. 

#lspci -D 

10000:80:01.0 PCI bridge: Intel Corporation Device 352a (rev 04)

10000:80:03.0 PCI bridge: Intel Corporation Device 352b (rev 04)

10000:80:05.0 PCI bridge: Intel Corporation Device 352c (rev 04)

10000:80:07.0 PCI bridge: Intel Corporation Device 352d (rev 04)

10001:80:01.0 PCI bridge: Intel Corporation Device 352a (rev 04)

10001:80:03.0 PCI bridge: Intel Corporation Device 352b (rev 04)

10001:80:05.0 PCI bridge: Intel Corporation Device 352c (rev 04)

10001:80:07.0 PCI bridge: Intel Corporation Device 352d (rev 04)

10001:01:00.0 Non-Volatile memory controller: Device 1f24:1058 (rev 01)

10001:02:00.0 Non-Volatile memory controller: Device 1f24:1058 (rev 01)

as you know, the default domain address normally is 0x0000.  so i need your help and want to know why the PCI devices domain that the address with 0x1000x. 

my server with 2* Intel(R) Xeon(R) Gold 6454S processers.

Thanks. 

Guangyao Jiang 

[Meghak]

Hi GuangyaoJiang,

Thank you for posting on Intel Community.

Please note the PCI address domain consists of three distinct address spaces: configuration, memory, and I/O space.

We have identified the following link that may be helpful.

https://docs.oracle.com/cd/E19455-01/805-7378/hwovr-25/index.html

Please note the query reported is more generic and we request to confirm if you're looking for something which is more Xeon processor specific.

Regards,

Megha K

[GuangyaoJiang]

Hi Megha，

          there is an new feedback,  i check the bios setting and disable the VMD function.  And after disable the VMD, the domain address back to  0x0000.

so can you find some KB about why the VMD can impact PCIe domain?

Regards,

Guangyao Jiang

[Meghak]

Hi GuangyaoJiang,

Thank you for posting on Intel Community.

We request to please confirm the server board and system details that is being used to check the issue reported.

Regards,

Megha K

[GuangyaoJiang]

Hi Megha，

The Server details：

        2* Intel(R) Xeon(R) Gold 6454S processers

        12 * DDR 5 32GB Memory

        2 * PM9A3（Sumsang）1.92TB NVMe

        Board is OEM by xFusion

        RHEL 9.0 (kernel version:  5.14.0-70.13.1.el9_0

[Meghak]

Hi GuangyaoJiang,

Thank you for confirming the details.

Please allow us some time to review the query reported and we shall get back with an update at the earliest.

Regards,

Megha K

[Meghak]

Hi GuangyaoJiang,

Thank you for posting on Intel Community.

Please note we have reviewed the issue reported and this seems to be an issue with board-BIOS configuration.

We request to please confirm if the query is related to Intel VMD/Intel® Volume Management Device.

If so, please be informed the issue is related to the board and we request to please check with OEM for more details.

We request to please check the below link for processor integrity.

https://www.intel.com/content/www/us/en/support/articles/000058107/processors/intel-xeon-processors.html

Regards,

Megha k

[Meghak]

Hi GuangyaoJiang,

Thank you for posting on Intel Community.

Please let us know if the query reported is resolved and if we can proceed with the case closure.

Regards,

Megha K

[GuangyaoJiang]

Hi Megha，

Sorry for my delay.

 1st, You are right, it is related to Intel VMD/Intel® Volume Management Device.

2nd,  I have used DCDIAG tools to check the CPU status, No error apperred.

[Meghak]

Hi GuangyaoJiang,

Thank you for confirming the details.

As mentioned on the previous post, if the query is related to Intel® Volume Management Device, the issue is related to the board.

Hence we request to please check with OEM for more details.

Please let us know if we can proceed with the case closure.

Regards,

Megha k

[Meghak]

Hi GuangyaoJiang,

Thank you for posting on Intel Community.

Please let us know if the query reported is resolved and if we can proceed with the case closure.

Regards,

Megha K

[Meghak]

Hi GuangyaoJiang,,

Thank you for posting on Intel community.

We have not received a reply from you , and as such, we will be closing your case.

If you want to continue support, please reply to this post and we will reopen the case or create a new one so that we can continue to support you.

Best Regards,

Megha k

[GuangyaoJiang]

Hi Megha，

Sorry for my Delay, due to Chinese New Year Holiday.  You can close this case.

Thanks for your supporting.