Re: [PATCH 00/13] Reconcile NUMA balancing decisions with the load balancer v6

[Mel Gorman]

On Mon, Feb 24, 2020 at 04:16:41PM +0100, Ingo Molnar wrote:
> 
> * Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> wrote:
> 
> > The only differences in V6 are due to Vincent's latest patch series.
> > 
> > This is V5 which includes the latest versions of Vincent's patch
> > addressing review feedback. Patches 4-9 are Vincent's work plus one
> > important performance fix. Vincent's patches were retested and while
> > not presented in detail, it was mostly an improvement.
> > 
> > Changelog since V5:
> > o Import Vincent's latest patch set
> 
> >  include/linux/sched.h        |  31 ++-
> >  include/trace/events/sched.h |  49 ++--
> >  kernel/sched/core.c          |  13 -
> >  kernel/sched/debug.c         |  17 +-
> >  kernel/sched/fair.c          | 626 ++++++++++++++++++++++++++++---------------
> >  kernel/sched/pelt.c          |  59 ++--
> >  kernel/sched/sched.h         |  42 ++-
> >  7 files changed, 535 insertions(+), 302 deletions(-)
> 
> Applied to tip:sched/core for v5.7 inclusion, thanks Mel and Vincent!
> 

Thanks!

> Please base future iterations on top of a0f03b617c3b (current 
> sched/core).
> 

Will do.

However I noticed that "sched/fair: Fix find_idlest_group() to handle
CPU affinity" did not make it to tip/sched/core. Peter seemed to think it
was fine. Was it rejected or is it just sitting in Peter's queue somewhere?

-- 
Mel Gorman
SUSE Labs