lkml.org 
[lkml]   [2010]   [Mar]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
Subject[patch repost] sched: Fix group_capacity for sched_smt_powersavings=1
Hi Peter,

This is repost of the same patch http://lkml.org/lkml/2010/3/2/216

After applying Suresh's fixes from discussion thread
http://lkml.org/lkml/2010/2/12/352, we still need the attached patch
to restore sched_smt_powersavings=1 functionality where tasks prefer
sibling threads and keep more cores idle.

Please apply to sched-tip, the patch is rebased and tested on today's
sched-tip master.

The attached patch will run 4 while(1) loops in two cores when
sched_smt_power_savings=1. Tested on two socket, quad core, hyper
threaded system.
Additional testing was done on POWER platform where
sched_smt_powersavings was able to consolidate tasks on sibling
threads leaving more idle cores.

Thanks,
Vaidy
---
sched: Fix group_capacity for sched_smt_powersavings=1
sched_smt_powersavings for threaded systems need this fix for
consolidation to sibling threads to work. Since threads have
fractional capacity, group_capacity will turn out to be one
always and not accommodate another task in the sibling thread.

This fix makes group_capacity a function of cpumask_weight that
will enable the power saving load balancer to pack tasks among
sibling threads and keep more cores idle.

Signed-off-by: Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 5a5ea2c..7c0a29a 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -2538,6 +2538,21 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu,
*/
if (prefer_sibling)
sgs.group_capacity = min(sgs.group_capacity, 1UL);
+ /*
+ * If power savings balance is set at this domain, then
+ * make capacity equal to number of hardware threads to
+ * accommodate more tasks until capacity is reached.
+ */
+ else if (sd->flags & SD_POWERSAVINGS_BALANCE)
+ sgs.group_capacity =
+ cpumask_weight(sched_group_cpus(group));
+
+ /*
+ * The default group_capacity is rounded from sum of
+ * fractional cpu_powers of sibling hardware threads
+ * in order to enable fair use of available hardware
+ * resources.
+ */

if (local_group) {
sds->this_load = sgs.avg_load;
@@ -2863,7 +2878,8 @@ static int need_active_balance(struct sched_domain *sd, int sd_idle, int idle)
!test_sd_parent(sd, SD_POWERSAVINGS_BALANCE))
return 0;

- if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
+ if (sched_mc_power_savings < POWERSAVINGS_BALANCE_WAKEUP &&
+ sched_smt_power_savings < POWERSAVINGS_BALANCE_WAKEUP)
return 0;
}


\
 
 \ /
  Last update: 2010-03-22 17:37    [from the cache]
©2003-2014 Jasper Spaans. Advertise on this site