While macOS uses DAS-CTS to schedule hundreds of background activities, third-parties normally use launchd. Comes with a full diagram explaining DAS-CTS.
How running a background task takes a tiny fraction of second, although the task itself takes seconds or minutes, and why it’s run on E cores.
Some threads are set to run in the background, and get allocated to the E cores. Could you run them in a VM, and effectively promote them to run on P cores instead?
Virtual CPU cores are of one type, and QoS has no effect in virtualised macOS. This has consequences for both the host and guest macOS.
Many apps could benefit users of Apple silicon Macs by giving them controls over core use by their threads. Here’s how that can be done simply and effectively.
How you can use the taskpolicy command to confine all the threads of a process to the E cores, as a brake, but there’s no accelerator in macOS.
How can the two E cores in an M1 Pro or Max equal performance of the four in the original M1? Why does running two threads complete in half the time taken to run one?
Threads, GCD and core allocation in Apple silicon explained. How thread priority is baked into code, and how important it is to performance.
It’s a strange coincidence that Intel and Microsoft came up with similar hardware of P and E core types in a SoC, and identical terminology for thread allocation using QoS.
Both P and E cores are run at different frequencies according to the load on M1 chips. This explores how macOS manages their frequencies and why.