Alder Lake Support Added To Intel's TCC Driver In Linux 5.15
While much of Intel's next-gen Alder Lake processor support appears to be in good shape for Linux 5.14, some remaining items are landing for the current Linux 5.15 cycle. The latest Alder Lake support hitting the kernel is for Intel's TCC cooling driver.
The Intel TCC driver is the new cooling driver merged earlier this year that allows for preemptively downclocking your CPU at a lower thermal threshold. The Intel TCC driver allows manipulating the Thermal Control Circuit offset so it's lower than the default activation temperature. The TCC driver works with Intel CPUs going back to Skylake along with all current Intel client CPUs.
With Linux 5.15 this TCC driver coverage is extended to Alder Lake for those wishing for their future Alder Lake Linux systems to run cooler albeit with restricted performance when hitting your desired offset temperature.
The Alder Lake TCC support was submitted today as part of the thermal subsystem updates for Linux 5.15. The thermal pull request also adds a long overdue NVIDIA Tegra 3 thermal sensor driver, cross-compile support for the TMON tool, and other updates.
On a similar topic, Friday saw more PM code merged. With that latest power management code is an improvement to the Intel P-State driver for improving the HWP performance level calibration to also benefit Alder Lake / Intel hybrid processors.
As mentioned at the start, much of the Intel Alder Lake support is squared away by Linux 5.14 but some remaining pieces are in place for Linux 5.15 and potentially back-ported to prior stable kernel series. In any case once Alder Lake processors begin shipping later this year stay tuned to Phoronix for my benchmarking and support guidelines about kernel requirements. The big unknown at the moment is Intel not yet having published any Linux kernel patches around Thread Director. Such kernel patches will be coming but not yet and thus given the timing won't be found mainline in a released kernel version at the point of the first Alder Lake CPUs start shipping. Thread Director is a hardware feature but remains to be seen how much OS/kernel integration is needed for it to performance efficiently.
The Intel TCC driver is the new cooling driver merged earlier this year that allows for preemptively downclocking your CPU at a lower thermal threshold. The Intel TCC driver allows manipulating the Thermal Control Circuit offset so it's lower than the default activation temperature. The TCC driver works with Intel CPUs going back to Skylake along with all current Intel client CPUs.
With Linux 5.15 this TCC driver coverage is extended to Alder Lake for those wishing for their future Alder Lake Linux systems to run cooler albeit with restricted performance when hitting your desired offset temperature.
The Alder Lake TCC support was submitted today as part of the thermal subsystem updates for Linux 5.15. The thermal pull request also adds a long overdue NVIDIA Tegra 3 thermal sensor driver, cross-compile support for the TMON tool, and other updates.
On a similar topic, Friday saw more PM code merged. With that latest power management code is an improvement to the Intel P-State driver for improving the HWP performance level calibration to also benefit Alder Lake / Intel hybrid processors.
As mentioned at the start, much of the Intel Alder Lake support is squared away by Linux 5.14 but some remaining pieces are in place for Linux 5.15 and potentially back-ported to prior stable kernel series. In any case once Alder Lake processors begin shipping later this year stay tuned to Phoronix for my benchmarking and support guidelines about kernel requirements. The big unknown at the moment is Intel not yet having published any Linux kernel patches around Thread Director. Such kernel patches will be coming but not yet and thus given the timing won't be found mainline in a released kernel version at the point of the first Alder Lake CPUs start shipping. Thread Director is a hardware feature but remains to be seen how much OS/kernel integration is needed for it to performance efficiently.
2 Comments