CPU Configuration

Press <Enter> to view the information related to CPU 1.

Can be used to disable symmetric multithreading. To re-enable SMT, a POWER CYCLE is needed after selecting Enable.

The default option is Auto.

Disable CPB.

The default option is Auto.

CPPC (cooperative processor performance control) is a way for the OS to influence the performance of a CPU on a contiguous and abstract scale without knowledge of power budgets or discrete processor frequencies.

Maximum boost frequency. Auto set the boost frequency to the fused value for the installed CPU. When a manual value is entered, the value entered is a 4 digit number representing the maximum boost frequency in MHz. The value entered applies to all cores.

Specify the boost Fmax frequency, the value entered is a 4 digit number representing the maximum boost frequency in MHz. The value entered applies to all cores.

Total activated cores = # Cores Per Die * Dies Per CPU * Number of Installed CPUs. Dies Per CPU is in the CPU1 Information page. The total number of activated cores is limited by the maximum number of cores of the installed CPU SKU.

The default option is Auto.

Enable or disable the L1 Stream HW Prefetcher.

The default option is Auto.

Enable or disable L1 stride prefetcher.

The default option is Auto.

Use memory access history to fetch additional data lines into L1 cache when each access is a constant distance from the previous. Some workloads may benefit from having it disabled.

Enable or disable L1 region prefetcher.

The default option is Auto.

Fetch additional data lines into L1 cache when the data access for a given instruction tends to be followed by a consistent pattern of subsequent accesses. Some workloads may benefit from having it disabled.

Enable or disable the L2 Stream HW Prefetcher.

The default option is Auto.

Enable or disable L2 Up/Down prefetcher.

The default option is Auto.

Use memory access history to determine whether to fetch the next or previous line for all memory accesses. Some workloads may benefit from having it disabled.

The default option is No Priority.

When No Priority is selected, there is no preferred IO bus. When a specific bus is selected for higher IO priority, the format of the field is XX, where XX is the bus number in hex (For Redfish, the bus number is in decimal).

When the item Preferred IO Bus set to the preferred I/O device bus number, you will see this item.

Enabling the Enhanced Preferred I/O mode assures an LCLK valuate for best performance. (Note: Setting 'LCLK Freq Control' on the same Root Complex which the Preferred IO Bus belongs to, to anything other than 'Auto' will override the Enhanced Preferred IO Mode.)

Enable Secured-Core support. When Secured-core is 'Enable', the 4 related settings are 'Enable' and locked. When Secured-core is 'Custom', the related settings can be changed independently as needed. If all 4 related settings are 'Enable', it is effectively equivalent to Secured-core being 'Enable'.

The default option is Custom.

Enable DMAr system protection during POST.

The default option is Disable.

Enable DRTM ACPI virtual device.

The default option is Disable.

Enable or Disable Transparent Secure Memory Encryption. TSME provides hardware memory encryption for all data stored in system memory. Disabling TSME will result in a small decrease to memory latency.

Enables or disables the IO based C-state generation and DF C-states control.

The default option is Enable.

Enable or disable CPU Virtualization.

The default option is Enable.

Enable or disable IOMMU.

The default option is Auto.

Auto: Use default performance determinism settings.

The default option is Power.

Sets the maximum power consumption for the CPU. Auto sets cTDP =TDP for the installed CPU SKU. Maximum sets the maximum allowed cTDP value for the installed CPU SKU. Usually, maximum is greater than TDP. If a manual value is entered that is larger than the max value allowed, the value will be internally limited to the maximum allowable value. cTDP is only configurable before OS boot.

The default option is Auto.

Enable AMD SEV-ES support

This fields specifies the maximum valid ASID, which affects the maximum system physical address space. 16TB of physical address space is available for systems that support 253 ASIDs, while 8TB of physical address space is available for systems that support 509 ASIDs.

SEV VMs using ASIDs below the SEV-ES ASID Space Limit must enable the SEV-ES feature. ASIDs from SEV-ES ASID Space Limit to (SEV ASID Count + 1) can only be used with SEV VMs. If this field is set to (SEV ASID Count + 1), all ASIDs are forced to be SEV-ES ASIDs. Hence, the valid values for this field is 1 - (SEV ASID Count + 1)

Enable or Disable SEV-SNP Support