Re: [PATCH v2 00/14] Introducing TIF_NOTIFY_IPI flag

Peter Zijlstra <[email protected]> Sat, 15 Jun 2024 03:45:21 +0200

Newsgroups	gmane.linux.ports.parisc,gmane.linux.kernel,gmane.linux.ports.alpha,gmane.linux.ports.arm.kernel,gmane.linux.ports.ppc64.devel,gmane.linux.ports.sh.devel,gmane.linux.ports.sparc,gmane.linux.power-management.general
Message-ID	<[email protected]>

On Sat, Jun 15, 2024 at 03:28:14AM +0200, Peter Zijlstra wrote:
> On Fri, Jun 14, 2024 at 12:48:37PM +0200, Vincent Guittot wrote:

> > The main problem is that need_resched becomes somewhat meaningless
> > because it doesn't  only mean "I need to resched a task" and we have
> > to add more tests around even for those not using polling
> 
> True, however we already had some of that by having the wakeup list,
> that made nr_running less 'reliable'.

Doesn't using !idle_cpu() instead of need_resched() in those balance
paths already do the right thing?

Checking need_resched() as an indicator of it getting work is already a
bit an assumption.

Also, Ingo, idle_cpu() and friends don't really belong in syscalls.c...