On Thu, 2025-09-11 at 12:39 +0800, Tiwei Bie wrote: > On Wed, 10 Sep 2025 14:10:37 +0200, Johannes Berg wrote: > > On Sun, 2025-08-10 at 13:51 +0800, Tiwei Bie wrote: > > > From: Tiwei Bie <tiwei....@antgroup.com> > > > > > > With SMP and NO_HZ enabled, the CPU may still need to sleep even > > > if the timer is disarmed. Switch to deciding whether to sleep based > > > on pending resched. This is a preparation for adding SMP support. > > > > What's the rationale for need_resched()? Is that somehow defined for > > this? Is it what other architectures use? I guess I'm just not entirely > > sure what it means. > > Here is a relevant document: > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/scheduler/sched-arch.rst?h=v6.17-rc5#n37 > > There is a similar check on x86: > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/x86/kernel/process.c?h=v6.17-rc5#n916
Ah cool, thanks for the pointers. Though "1." part there is a bit confusing, I guess I'll send a patch to reword it :) FWIW I already applied a few patches, but I'll take another look, hopefully today. johannes
