IPFire will now by default clock down its CPUs. When previously all CPU cores have been running on full clock speed by default, we were able to keep latency to a minimum as they were always ready to process any packets. As modern processors have massively improved how quickly they can clock up and down and CPUs with many cores being widely available, we have now decided to change this based on our benchmark results. Where supported, we will use Intel P-State or otherwise fall back to the new schedutil governor which has recently been introduced into the Linux kernel and has proven to not increase any packet forwarding latency in our benchmarks. When clocked down, systems will reduce their power consumption and therefore lowering the amount of emitted heat. The cpufrequtil package which used to implement this feature has been dropped as it is no longer needed.