AMD Turin processors with PCI Multi-segment enabled in the BIOS might cause ESXi 8.0U3 host failures

Products

• VMware vSphere ESXi 8.0

Issue/Introduction

In AMD 5th generation EPYC 9005 series with code name Turin,

the PCI Multi-segment support is active in the server's BIOS settings by default.

When the ESXi8.0U3 was installed or upgraded on the server.

Following symptoms may be encountered:

• The ESXi host fails with a purple diagnostic screen.
• The ESXi host becomes unresponsive.
• Some I/O devices do not work.

Environment

VMware ESXi 8.0U3

Cause

The issue occurs because AMD Turin has preset PCIe multi-segment support,

but ESXi does not provide PCIe multi-segment support for CPUs that have an input–output memory management unit (IOMMU).

This issue is specific for AMD Turin CPUs.

Resolution

This issue is resolved in the VMware ESXi 8.0 Update 3e (Build 24674464)

Workaround:

Disable PCI Multi-Segment (Set it to One Segment) in the server BIOS.

Additional Information

The procedure to update the BIOS setting for the server vendor.

PowerEdge: 17G AMD - PSOD related to PCI Multi-Segment during ESXi 8.0 U3 installation
https://www.dell.com/support/kbdoc/en-us/000241608/

VMware ESXi 8.0 Update 3e Release Notes
https://techdocs.broadcom.com/us/en/vmware-cis/vsphere/vsphere/8-0/release-notes/esxi-update-and-patch-release-notes/vsphere-esxi-80u3e-release-notes.html