Re: [PATCH 2/2] sched: Fix next_balance logic in rebalance_domains() and idle_balance()

From: Jason Low
Date: Thu May 08 2014 - 20:49:30 EST


On Thu, 2014-05-08 at 15:14 -0700, Jason Low wrote:
> Hi Ingo, Peter,
>
> Were there NULL domains on the test system? If so, I think we can
> address the problem by doing update_next_balance() only if the below
> rcu_dereference_check_sched_domain() returns a non-null domain.

Also, below is a patch which has the change:

---

Subject: [PATCH v2] sched: Fix next_balance logic in rebalance_domains()

v1->v2:
- Address a potential null pointer dereference.

Currently, in idle_balance(), we update rq->next_balance when we pull_tasks.
However, it is also important to update this in the !pulled_tasks case too.

When the CPU is "busy" (the CPU isn't idle), rq->next_balance gets computed
using sd->busy_factor (so we increase the balance interval when the CPU is
busy). However, when the CPU goes idle, rq->next_balance could still be set
to a large value that was computed with the sd->busy_factor.

Thus, we need to also update rq->next_balance in idle_balance() in the cases
where !pulled_tasks too, so that rq->next_balance gets updated without taking
the busy_factor into account when the CPU is about to go idle.

This patch makes rq->next_balance get updated independently of whether or
not we pulled_task. Also, we add logic to ensure that we always traverse
at least 1 of the sched domains to get a proper next_balance value for
updating rq->next_balance.

Additionally, since load_balance() modifies the sd->balance_interval, we
need to re-obtain the sched domain's interval after the call to
load_balance() in rebalance_domains() before we update rq->next_balance.

This patch adds and uses 2 new helper functions, update_next_balance() and
get_sd_balance_interval() to update next_balance and obtain the sched
domain's balance_interval.

Cc: daniel.lezcano@xxxxxxxxxx
Cc: alex.shi@xxxxxxxxxx
Cc: preeti@xxxxxxxxxxxxxxxxxx
Cc: efault@xxxxxx
Cc: vincent.guittot@xxxxxxxxxx
Cc: morten.rasmussen@xxxxxxx
Cc: aswin@xxxxxx
Cc: mingo@xxxxxxxxxx
Reviewed-by: Preeti U Murthy <preeti@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Jason Low <jason.low2@xxxxxx>
---
kernel/sched/fair.c | 69 ++++++++++++++++++++++++++++++++++-----------------
1 files changed, 46 insertions(+), 23 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 28ccf50..57b7390 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6669,17 +6669,44 @@ out:
return ld_moved;
}

+static inline unsigned long
+get_sd_balance_interval(struct sched_domain *sd, int cpu_busy)
+{
+ unsigned long interval = sd->balance_interval;
+
+ if (cpu_busy)
+ interval *= sd->busy_factor;
+
+ /* scale ms to jiffies */
+ interval = msecs_to_jiffies(interval);
+ interval = clamp(interval, 1UL, max_load_balance_interval);
+
+ return interval;
+}
+
+static inline void
+update_next_balance(struct sched_domain *sd, int cpu_busy, unsigned long *next_balance)
+{
+ unsigned long interval, next;
+
+ interval = get_sd_balance_interval(sd, cpu_busy);
+ next = sd->last_balance + interval;
+
+ if (time_after(*next_balance, next))
+ *next_balance = next;
+}
+
/*
* idle_balance is called by schedule() if this_cpu is about to become
* idle. Attempts to pull tasks from other CPUs.
*/
static int idle_balance(struct rq *this_rq)
{
+ unsigned long next_balance = jiffies + HZ;
+ int this_cpu = this_rq->cpu;
struct sched_domain *sd;
int pulled_task = 0;
- unsigned long next_balance = jiffies + HZ;
u64 curr_cost = 0;
- int this_cpu = this_rq->cpu;

idle_enter_fair(this_rq);

@@ -6689,8 +6716,15 @@ static int idle_balance(struct rq *this_rq)
*/
this_rq->idle_stamp = rq_clock(this_rq);

- if (this_rq->avg_idle < sysctl_sched_migration_cost)
+ if (this_rq->avg_idle < sysctl_sched_migration_cost) {
+ rcu_read_lock();
+ sd = rcu_dereference_check_sched_domain(this_rq->sd);
+ if (sd)
+ update_next_balance(sd, 0, &next_balance);
+ rcu_read_unlock();
+
goto out;
+ }

/*
* Drop the rq->lock, but keep IRQ/preempt disabled.
@@ -6700,15 +6734,16 @@ static int idle_balance(struct rq *this_rq)
update_blocked_averages(this_cpu);
rcu_read_lock();
for_each_domain(this_cpu, sd) {
- unsigned long interval;
int continue_balancing = 1;
u64 t0, domain_cost;

if (!(sd->flags & SD_LOAD_BALANCE))
continue;

- if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost)
+ if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) {
+ update_next_balance(sd, 0, &next_balance);
break;
+ }

if (sd->flags & SD_BALANCE_NEWIDLE) {
t0 = sched_clock_cpu(this_cpu);
@@ -6724,9 +6759,7 @@ static int idle_balance(struct rq *this_rq)
curr_cost += domain_cost;
}

- interval = msecs_to_jiffies(sd->balance_interval);
- if (time_after(next_balance, sd->last_balance + interval))
- next_balance = sd->last_balance + interval;
+ update_next_balance(sd, 0, &next_balance);

/*
* Stop searching for tasks to pull if there are
@@ -6750,15 +6783,11 @@ static int idle_balance(struct rq *this_rq)
if (this_rq->cfs.h_nr_running && !pulled_task)
pulled_task = 1;

- if (pulled_task || time_after(jiffies, this_rq->next_balance)) {
- /*
- * We are going idle. next_balance may be set based on
- * a busy processor. So reset next_balance.
- */
+out:
+ /* Move the next balance forward */
+ if (time_after(this_rq->next_balance, next_balance))
this_rq->next_balance = next_balance;
- }

-out:
/* Is there a task of a high priority class? */
if (this_rq->nr_running != this_rq->cfs.h_nr_running)
pulled_task = -1;
@@ -7041,16 +7070,9 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle)
break;
}

- interval = sd->balance_interval;
- if (idle != CPU_IDLE)
- interval *= sd->busy_factor;
-
- /* scale ms to jiffies */
- interval = msecs_to_jiffies(interval);
- interval = clamp(interval, 1UL, max_load_balance_interval);
+ interval = get_sd_balance_interval(sd, idle != CPU_IDLE);

need_serialize = sd->flags & SD_SERIALIZE;
-
if (need_serialize) {
if (!spin_trylock(&balancing))
goto out;
@@ -7066,6 +7088,7 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle)
idle = idle_cpu(cpu) ? CPU_IDLE : CPU_NOT_IDLE;
}
sd->last_balance = jiffies;
+ interval = get_sd_balance_interval(sd, idle != CPU_IDLE);
}
if (need_serialize)
spin_unlock(&balancing);
--
1.7.1



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/