From b0432d8f162c7d5d9537b4cb749d44076b76a783 Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Thu, 7 Apr 2011 17:23:22 -0700 Subject: sched: Fix sched-domain avg_load calculation In function find_busiest_group(), the sched-domain avg_load isn't calculated at all if there is a group imbalance within the domain. This will cause erroneous imbalance calculation. The reason is that calculate_imbalance() sees sds->avg_load = 0 and it will dump entire sds->max_load into imbalance variable, which is used later on to migrate entire load from busiest CPU to the puller CPU. This has two really bad effect: 1. stampede of task migration, and they won't be able to break out of the bad state because of positive feedback loop: large load delta -> heavier load migration -> larger imbalance and the cycle goes on. 2. severe imbalance in CPU queue depth. This causes really long scheduling latency blip which affects badly on application that has tight latency requirement. The fix is to have kernel calculate domain avg_load in both cases. This will ensure that imbalance calculation is always sensible and the target is usually half way between busiest and puller CPU. Signed-off-by: Ken Chen Signed-off-by: Peter Zijlstra Cc: Link: http://lkml.kernel.org/r/20110408002322.3A0D812217F@elm.corp.google.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/sched_fair.c') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 7f00772e57c..60f9d407c5e 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -3127,6 +3127,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, if (!sds.busiest || sds.busiest_nr_running == 0) goto out_balanced; + sds.avg_load = (SCHED_LOAD_SCALE * sds.total_load) / sds.total_pwr; + /* * If the busiest group is imbalanced the below checks don't * work because they assumes all things are equal, which typically @@ -3151,7 +3153,6 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, * Don't pull any tasks if this group is already above the domain * average load. */ - sds.avg_load = (SCHED_LOAD_SCALE * sds.total_load) / sds.total_pwr; if (sds.this_load >= sds.avg_load) goto out_balanced; -- cgit v1.2.3-70-g09d2 From b30aef17f71cf9e24b10c11cbb5e5f0ebe8a85ab Mon Sep 17 00:00:00 2001 From: Ken Chen Date: Fri, 8 Apr 2011 12:20:16 -0700 Subject: sched: Fix erroneous all_pinned logic The scheduler load balancer has specific code to deal with cases of unbalanced system due to lots of unmovable tasks (for example because of hard CPU affinity). In those situation, it excludes the busiest CPU that has pinned tasks for load balance consideration such that it can perform second 2nd load balance pass on the rest of the system. This all works as designed if there is only one cgroup in the system. However, when we have multiple cgroups, this logic has false positives and triggers multiple load balance passes despite there are actually no pinned tasks at all. The reason it has false positives is that the all pinned logic is deep in the lowest function of can_migrate_task() and is too low level: load_balance_fair() iterates each task group and calls balance_tasks() to migrate target load. Along the way, balance_tasks() will also set a all_pinned variable. Given that task-groups are iterated, this all_pinned variable is essentially the status of last group in the scanning process. Task group can have number of reasons that no load being migrated, none due to cpu affinity. However, this status bit is being propagated back up to the higher level load_balance(), which incorrectly think that no tasks were moved. It kick off the all pinned logic and start multiple passes attempt to move load onto puller CPU. To fix this, move the all_pinned aggregation up at the iterator level. This ensures that the status is aggregated over all task-groups, not just last one in the list. Signed-off-by: Ken Chen Cc: stable@kernel.org Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/BANLkTi=ernzNawaR5tJZEsV_QVnfxqXmsQ@mail.gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'kernel/sched_fair.c') diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 60f9d407c5e..6fa833ab2cb 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -2104,21 +2104,20 @@ balance_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest, enum cpu_idle_type idle, int *all_pinned, int *this_best_prio, struct cfs_rq *busiest_cfs_rq) { - int loops = 0, pulled = 0, pinned = 0; + int loops = 0, pulled = 0; long rem_load_move = max_load_move; struct task_struct *p, *n; if (max_load_move == 0) goto out; - pinned = 1; - list_for_each_entry_safe(p, n, &busiest_cfs_rq->tasks, se.group_node) { if (loops++ > sysctl_sched_nr_migrate) break; if ((p->se.load.weight >> 1) > rem_load_move || - !can_migrate_task(p, busiest, this_cpu, sd, idle, &pinned)) + !can_migrate_task(p, busiest, this_cpu, sd, idle, + all_pinned)) continue; pull_task(busiest, p, this_rq, this_cpu); @@ -2153,9 +2152,6 @@ out: */ schedstat_add(sd, lb_gained[idle], pulled); - if (all_pinned) - *all_pinned = pinned; - return max_load_move - rem_load_move; } @@ -3341,6 +3337,7 @@ redo: * still unbalanced. ld_moved simply stays zero, so it is * correctly treated as an imbalance. */ + all_pinned = 1; local_irq_save(flags); double_rq_lock(this_rq, busiest); ld_moved = move_tasks(this_rq, this_cpu, busiest, -- cgit v1.2.3-70-g09d2