sched/fair: Sync task util before slow-path wakeup
PATCH v2 1/1 sched/fair: Sync task util before slow-path wakeup
We use task_util in find_idlest_group via capacity_spare_wake. This task_util is updated in wake_cap. However wake_cap is not the only reason for ending up in find_idlest_group - we could have been sent there by wake_wide. So explicitly sync the task util with prev_cpu when we are about to head to find_idlest_group.
We could simply do this at the beginning of select_task_rq_fair (i.e. irrespective of whether we're heading to select_idle_sibling or find_idlest_group & co), but I didn't want to slow down the select_idle_sibling path more than necessary.
Don't do this during fork balancing, we won't need the task_util and we'd just clobber the last_update_time, which is supposed to be 0.
---
changes v1 -> v2: Just cosmetic
kernel/sched/fair.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)
kernel/sched/fair.c +10 -1
@@ -5913,8 +5913,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f new_cpu = cpu; } +if (sd && !(sd_flag &SD_BALANCE_FORK )) { +/* +* We're going to need the task's util for capacity_spare_wake +* in find_idlest_group. Sync it up to prev_cpu's +* last_update_time. +*/ +sync_entity_load_avg (&p->se); + } +if (!sd) { - pick_cpu: +pick_cpu:if (sd_flag &SD_BALANCE_WAKE )/* XXX always ? */ new_cpu =select_idle_sibling (p, prev_cpu, new_cpu);
--
2.13.0
Hi Peter,
Ping.
Log of previous discussion: https://patchwork.kernel.org/patch/9876769/
Cheers,
Brendan
Quoted 50 lines · Brendan Jackman
We use
task_utilinfind_idlest_groupviacapacity_spare_wake. Thistask_utilis updated inwake_cap. Howeverwake_capis not the only reason for ending up infind_idlest_group- we could have been sent there bywake_wide. So explicitly sync the task util withprev_cpuwhen we are about to head tofind_idlest_group.We could simply do this at the beginning of
select_task_rq_fair(i.e. irrespective of whether we're heading toselect_idle_siblingorfind_idlest_group& co), but I didn't want to slow down theselect_idle_siblingpath more than necessary.Don't do this during fork balancing, we won't need the
task_utiland we'd just clobber thelast_update_time, which is supposed to be 0.Signed-off-by Brendan Jackman <[email protected]>Cc Dietmar Eggemann <[email protected]>Cc Vincent Guittot <[email protected]>Cc Josef Bacik <[email protected]>Cc Ingo Molnar <[email protected]>Cc Morten Rasmussen <[email protected]>Cc Peter Zijlstra <[email protected]>---
changes v1 -> v2: Just cosmetic
kernel/sched/fair.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)kernel/sched/fair.c +10
-1@@ -5913,8 +5913,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f new_cpu = cpu; } +if (sd && !(sd_flag &SD_BALANCE_FORK )) { +/* +* We're going to need the task's util for capacity_spare_wake +* in find_idlest_group. Sync it up to prev_cpu's +* last_update_time. +*/ +sync_entity_load_avg (&p->se); + } +if (!sd) { - pick_cpu: +pick_cpu:if (sd_flag &SD_BALANCE_WAKE )/* XXX always ? */ new_cpu =select_idle_sibling (p, prev_cpu, new_cpu);
tip-bot for Brendan Jackman
merge botsched/fair: Sync task util before slow-path wakeup
Commit-ID: ea16f0ea6c3dc9e1aa083bd3d1792ba02860526e Gitweb: https://git.kernel.org/tip/ea16f0ea6c3dc9e1aa083bd3d1792ba02860526e Author: Brendan Jackman <[email protected]> AuthorDate: Thu, 5 Oct 2017 11:55:51 +0100 Committer: Ingo Molnar <[email protected]> CommitDate: Tue, 10 Oct 2017 11:45:31 +0200
sched/fair: Sync task util before slow-path wakeup
We use task_util() in find_idlest_group() via capacity_spare_wake(). This task_util() updated in wake_cap(). However wake_cap() is not the only reason for ending up in find_idlest_group() - we could have been sent there by wake_wide(). So explicitly sync the task util with prev_cpu when we are about to head to find_idlest_group().
We could simply do this at the beginning of select_task_rq_fair() (i.e. irrespective of whether we're heading to select_idle_sibling() or find_idlest_group() & co), but I didn't want to slow down the select_idle_sibling() path more than necessary.
Don't do this during fork balancing, we won't need the task_util and we'd just clobber the last_update_time, which is supposed to be 0.
--- kernel/sched/fair.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)
kernel/sched/fair.c +10 -1
@@ -6263,8 +6263,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f new_cpu = cpu; } +if (sd && !(sd_flag &SD_BALANCE_FORK )) { +/* +* We're going to need the task's util for capacity_spare_wake +* in find_idlest_group. Sync it up to prev_cpu's +* last_update_time. +*/ +sync_entity_load_avg (&p->se); + } +if (!sd) { - pick_cpu: +pick_cpu:if (sd_flag &SD_BALANCE_WAKE )/* XXX always ? */ new_cpu =select_idle_sibling (p, prev_cpu, new_cpu);