Earlier this month [1]AMD Fast CPPC support was queued into the Linux power management subsystem's "-next" codebase ahead of the Linux 6.11 cycle. Additional AMD P-State driver enhancements are now deemed ready and submitted for staging ahead of next month's Linux 6.11 cycle kicking off.

AMD Linux engineer Mario Limonciello has sent out a second set of AMD P-State CPU frequency scaling driver improvements for Linux 6.11.

These latest "amd_pstate" driver changes include enabling amd-pstate by default in "shared memory" designs without a

dedicated MSR, extra infrastructure for debugging purposes, and various bug fixes. The driver also now honors a "default" Energy Performance Preference (EPP) value for those wanting to reset their EPP preference to the platform default.

As for now enabling AMD P-State driver use on shared memory type systems by default, that patch message explains:

"The amd-pstate-epp driver has been implemented and resolves the performance drop issue seen in passive mode. Users who enable the active mode driver will not experience a performance drop compared to the passive mode driver. Therefore, the EPP driver should be loaded by default at system boot."

Previously some CPU performance regressions were noted by SUSE and in turn preferred ACPI CPUFreq instead but those issues should now be resolved.

More details on these latest AMD P-State driver improvements via [2]this pull request . Still being worked on for the AMD P-State driver meanwhile is the [3]Core Performance Boost and [4]better heterogeneous CPU handling .



[1] https://www.phoronix.com/news/AMD-Fast-CPPC-For-Linux-6.11

[2] https://lore.kernel.org/linux-pm/111b995d-b99e-416e-85a9-39c144c8b23c@amd.com/

[3] https://www.phoronix.com/news/AMD-Core-Perf-Boost-Per-CPU

[4] https://www.phoronix.com/news/AMD-P-State-Hetero-v3