Re: [PATCH] sched/fair: Add SMT4 group_smt_balance handling

[Tim Chen <tim.c.chen@linux.intel.com>]

On Thu, 2023-09-07 at 14:28 +0530, Shrikanth Hegde wrote:
> > 
> > You mean simplify the patch as below?  I think that should be fine.  Can you
> > make sure it works for SMT4?  And I can update the patch once you confirm it
> > works properly.
> > 
> > Tim
> > 
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 6e7ee2efc1ba..48e9ab7f8a87 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -9764,16 +9764,9 @@ static bool update_sd_pick_busiest(struct lb_env *env,
> >  
> >         case group_smt_balance:
> >                 /* no idle cpus on both groups handled by group_fully_busy below */
> > -               if (sgs->idle_cpus != 0 || busiest->idle_cpus != 0) {
> > -                       if (sgs->idle_cpus > busiest->idle_cpus)
> > -                               return false;
> > -                       if (sgs->idle_cpus < busiest->idle_cpus)
> > -                               return true;
> > -                       if (sgs->sum_nr_running <= busiest->sum_nr_running)
> > -                               return false;
> > -                       else
> > -                               return true;
> > -               }
> > +               if (sgs->idle_cpus != 0 || busiest->idle_cpus != 0)
> > +                       goto has_spare;
> > +
> >                 goto fully_busy;
> >  
> >         case group_fully_busy:
> > @@ -9809,6 +9802,7 @@ static bool update_sd_pick_busiest(struct lb_env *env,
> >                  * as we do not want to pull task off SMT core with one task
> >                  * and make the core idle.
> >                  */
> > +has_spare:
> >                 if (smt_vs_nonsmt_groups(sds->busiest, sg)) {
> >                         if (sg->flags & SD_SHARE_CPUCAPACITY && sgs->sum_h_nr_running <= 1)
> >                                 return false;
> > 
> > 
> > 
> 
> Hi Tim,
> 
> In case you were waiting for my reply as inferred from other email. 
> The above change looks fine as well. This would avoid duplication of
> code for group_smt_balance. 
> 
> Acked-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>

Peter,

Here's the updated patch.  Please consider it for inclusion.

Thanks.

Tim

From 979e261fed6e3765316a4de794f595f93c02cef0 Mon Sep 17 00:00:00 2001
From: Tim Chen <tim.c.chen@linux.intel.com>
Subject: [PATCH] sched/fair: Fix SMT4 group_smt_balance handling
To: Peter Zijlstra <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@redhat.com>, Vincent Guittot <vincent.guittot@linaro.org>, Ricardo Neri <ricardo.neri@intel.com>, Ravi V. Shankar <ravi.v.shankar@intel.com>, Ben Segall
<bsegall@google.com>, Daniel Bristot de Oliveira <bristot@redhat.com>, Dietmar Eggemann <dietmar.eggemann@arm.com>, Len Brown <len.brown@intel.com>, Mel Gorman <mgorman@suse.de>, Rafael J. Wysocki
<rafael.j.wysocki@intel.com>, Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>, Steven Rostedt <rostedt@goodmis.org>, Tim Chen <tim.c.chen@linux.intel.com>, Valentin Schneider
<vschneid@redhat.com>, Ionela Voinescu <ionela.voinescu@arm.com>, x86@kernel.org, linux-kernel@vger.kernel.org, Shrikanth Hegde <sshegde@linux.vnet.ibm.com>, Srikar Dronamraju
<srikar@linux.vnet.ibm.com>, naveen.n.rao@linux.vnet.ibm.com, Yicong Yang <yangyicong@hisilicon.com>, Barry Song <v-songbaohua@oppo.com>, Chen Yu <yu.c.chen@intel.com>, Hillf Danton <hdanton@sina.com>

For SMT4, any group with more than 2 tasks will be marked as
group_smt_balance. Retain the behaviour of group_has_spare by marking
the busiest group as the group which has the least number of idle_cpus.

Also, handle rounding effect of adding (ncores_local + ncores_busy) when
the local is fully idle and busy group imbalance is less than 2 tasks.
Local group should try to pull at least 1 task in this case so imbalance
should be set to 2 instead.

Fixes: fee1759e4f04 ("sched/fair: Determine active load balance for SMT sched groups")
Acked-by: Shrikanth Hegde <sshegde@linux.vnet.ibm.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
---
 kernel/sched/fair.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 0b7445cd5af9..fd9e594b5623 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9575,7 +9575,7 @@ static inline long sibling_imbalance(struct lb_env *env,
 	imbalance /= ncores_local + ncores_busiest;
 
 	/* Take advantage of resource in an empty sched group */
-	if (imbalance == 0 && local->sum_nr_running == 0 &&
+	if (imbalance <= 1 && local->sum_nr_running == 0 &&
 	    busiest->sum_nr_running > 1)
 		imbalance = 2;
 
@@ -9763,6 +9763,15 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 		break;
 
 	case group_smt_balance:
+		/*
+		 * Check if we have spare CPUs on either SMT group to
+		 * choose has spare or fully busy handling.
+		 */
+		if (sgs->idle_cpus != 0 || busiest->idle_cpus != 0)
+			goto has_spare;
+
+		fallthrough;
+
 	case group_fully_busy:
 		/*
 		 * Select the fully busy group with highest avg_load. In
@@ -9802,6 +9811,7 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 			else
 				return true;
 		}
+has_spare:
 
 		/*
 		 * Select not overloaded group with lowest number of idle cpus
-- 
2.32.0