So the realtime patch set currently puts a lock around per-CPU data accesses, eliminating the preemption problem but wrecking scalability. Here, too, a real solution has not yet been found.
It sounds like what is needed is a way to direct the kernel that a task can be pre-empted but not manipulated in a way that is unsafe for pre-CPU state.
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds