Recent ARM chips, like the Qualcomm Snapdragon 810, seem to be using a hybrid architecture called "big.LITTLE" with cores of different architectures and speeds combined into one CPU.

On a Linux (or other unixoid) system, how does the kernel distribute processes to those cores, and how does the userland representation of this architecture look? Does the niceness of processes play a role in core allocation?

I was curious to check this in the AOSP code, but could not find the relevant information. Hence, pointers are very much appreciated, too.

  • Start by reading the Wikipedia article. It doesn't get into the details of how it's done in Linux, but it explains that different SoC families use different strategies (corollary: look in SoC-specific code, not just in generic ARM code). – Gilles Dec 23 '15 at 12:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.