Re: [GIT PULL] sched_ext: Initial pull request for v6.11

[Peter Zijlstra <peterz@infradead.org>]

On Tue, Aug 06, 2024 at 09:56:36AM -1000, Tejun Heo wrote:
> Hello, Peter.
> 
> On Wed, Jul 24, 2024 at 10:52:21AM +0200, Peter Zijlstra wrote:
> ...
> > So pick_task() came from the SCHED_CORE crud, which does a remote pick
> > and as such isn't able to do a put -- remote is still running its
> > current etc.
> > 
> > So pick_task() *SHOULD* already be considering its current and pick
> > that if it is a better candidate than whatever is on the queue.
> > 
> > If we have a pick_task() that doesn't do that, it's a pre-existing bug
> > and needs fixing anyhow.
> 
> I haven't applied the patch to make balance_fair() test
> sched_fair_runnable() instead of rq->nr_running yet but after that I think
> all review points that you raised should be addressed except for the above
> pick_task() conversion. If you want to go ahead with this change, please do
> so. I'll pull in tip/sched/core and update accordingly. AFAICS, it doesn't
> make any significant difference to SCX one way or the other and I think
> updating shouldn't be too painful.

OK. So my plan was to finish reading the for 6.11 pull diff, and then
merge that eevdf patch-set I send out. Post those patches I had in
sched/prep that re-arrange the put_prev thing. Then merge those, and
then ask you to rebase the whole lot on top of that, after which I'll
pull your branch.

And while I'm behind my own schedule, I think we can still get all that
sorted.

Anyway, I'll go stick this patch somewhere so it don't get lost.