Linux's Turbostat Adds Arrow Lake & Lunar Lake Support, Table-Based Feature Enumeration
The Intel-developed Turbostat Linux CLI utility for reporting processor frequency and idle statistics is seeing a number of feature updates for Linux 6.7 as well as new hardware support.
Turbostat is useful for providing insight into modern x86_64 CPU power-state statistics, turbo frequency boosting details, and related bits along with being able to show specific MSR values and other CPU details.
With Linux 6.7, initial support is added to Turbostat for working on Intel's upcoming Arrow Lake and Lunar Lake processor platforms. Additionally, over on the server side is also initial support for Granite Rapids, Sierra Forest, and Grand Ridge. This goes along with all of the other Intel driver work we've been seeing in recent months for these upcoming Intel client and server platforms.
In addition to adding support for the next two generations of Intel Core processors and the upcoming Intel Xeon processors, the Turbostat utility now supports the notion of table-based feature enumeration for improving platform handling moving forward. The existing Turbostat code relies on various CPU model checks throughout the code-base for figuring out the paths to take. This table-driven feature enumeration is intended to ultimately replace all of the different CPU model checks and global variables throughout the code. Reworking Turbostat to use this table-defined feature information represents the bulk of the Turbostat code changes for Linux 6.7.
The Turbostat CLI utility continues to live within the Linux kernel source tree and this pull request lays out all of the changes submitted for the nearly-closed Linux 6.7 merge window.
Turbostat is useful for providing insight into modern x86_64 CPU power-state statistics, turbo frequency boosting details, and related bits along with being able to show specific MSR values and other CPU details.
With Linux 6.7, initial support is added to Turbostat for working on Intel's upcoming Arrow Lake and Lunar Lake processor platforms. Additionally, over on the server side is also initial support for Granite Rapids, Sierra Forest, and Grand Ridge. This goes along with all of the other Intel driver work we've been seeing in recent months for these upcoming Intel client and server platforms.
In addition to adding support for the next two generations of Intel Core processors and the upcoming Intel Xeon processors, the Turbostat utility now supports the notion of table-based feature enumeration for improving platform handling moving forward. The existing Turbostat code relies on various CPU model checks throughout the code-base for figuring out the paths to take. This table-driven feature enumeration is intended to ultimately replace all of the different CPU model checks and global variables throughout the code. Reworking Turbostat to use this table-defined feature information represents the bulk of the Turbostat code changes for Linux 6.7.
The Turbostat CLI utility continues to live within the Linux kernel source tree and this pull request lays out all of the changes submitted for the nearly-closed Linux 6.7 merge window.
1 Comment