On my smp platform which is made of 5 cores in 2 clusters, I have the
nr_busy_cpu field of sched_group_power struct that is not null when the
platform is fully idle. The root cause is:
During the boot sequence, some CPUs reach the idle loop and set their
NOHZ_IDLE flag while waiting for others CPUs to boot. But the nr_busy_cpus
field is initialized later with the assumption that all CPUs are in the busy
state whereas some CPUs have already set their NOHZ_IDLE flag.
More generally, the NOHZ_IDLE flag must be initialized when new sched_domains
are created in order to ensure that NOHZ_IDLE and nr_busy_cpus are aligned.
This condition can be ensured by adding a synchronize_rcu between the
destruction of old sched_domains and the creation of new ones so the NOHZ_IDLE
flag will not be updated with old sched_domain once it has been initialized.
But this solution introduces a additionnal latency in the rebuild sequence
that is called during cpu hotplug.
As suggested by Frederic Weisbecker, another solution is to have the same
rcu lifecycle for both NOHZ_IDLE and sched_domain struct.
A new nohz_idle field is added to sched_domain so both status and sched_domain
will share the same RCU lifecycle and will be always synchronized.
In addition, there is no more need to protect nohz_idle against concurrent
access as it is only modified by 2 exclusive functions called by local cpu.
This solution has been prefered to the creation of a new struct with an extra
pointer indirection for sched_domain.
The synchronization is done at the cost of :
- An additional indirection and a rcu_dereference for accessing nohz_idle.
- We use only the nohz_idle field of the top sched_domain.
Change since v7:
- remove atomic access which is useless now.
- refactor the sequence that update nohz_idle status and nr_busy_cpus.
Change since v6:
- Add the flags in struct sched_domain instead of creating a sched_domain_rq.
Change since v5:
- minor variable and function name change.
- remove a useless null check before kfree
- fix a compilation error when NO_HZ is not set.
Change since v4:
- link both sched_domain and NOHZ_IDLE flag in one RCU object so
their states are always synchronized.
Change since V3;
- NOHZ flag is not cleared if a NULL domain is attached to the CPU
- Remove patch 2/2 which becomes useless with latest modifications
Change since V2:
- change the initialization to idle state instead of busy state so a CPU that
enters idle during the build of the sched_domain will not corrupt the
initialization state
Change since V1:
- remove the patch for SCHED softirq on an idle core use case as it was
a side effect of the other use cases.
Signed-off-by: Vincent Guittot <[email protected]>
---
include/linux/sched.h | 3 +++
kernel/sched/fair.c | 26 ++++++++++++++++----------
kernel/sched/sched.h | 1 -
3 files changed, 19 insertions(+), 11 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index d35d2b6..22bcbe8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -899,6 +899,9 @@ struct sched_domain {
unsigned int wake_idx;
unsigned int forkexec_idx;
unsigned int smt_gain;
+#ifdef CONFIG_NO_HZ
+ int nohz_idle; /* NOHZ IDLE status */
+#endif
int flags; /* See SD_* */
int level;
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 7a33e59..5db1817 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5395,13 +5395,16 @@ static inline void set_cpu_sd_state_busy(void)
struct sched_domain *sd;
int cpu = smp_processor_id();
- if (!test_bit(NOHZ_IDLE, nohz_flags(cpu)))
- return;
- clear_bit(NOHZ_IDLE, nohz_flags(cpu));
-
rcu_read_lock();
- for_each_domain(cpu, sd)
+ sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd);
+
+ if (!sd || !sd->nohz_idle)
+ goto unlock;
+ sd->nohz_idle = 0;
+
+ for (; sd; sd = sd->parent)
atomic_inc(&sd->groups->sgp->nr_busy_cpus);
+unlock:
rcu_read_unlock();
}
@@ -5410,13 +5413,16 @@ void set_cpu_sd_state_idle(void)
struct sched_domain *sd;
int cpu = smp_processor_id();
- if (test_bit(NOHZ_IDLE, nohz_flags(cpu)))
- return;
- set_bit(NOHZ_IDLE, nohz_flags(cpu));
-
rcu_read_lock();
- for_each_domain(cpu, sd)
+ sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd);
+
+ if (!sd || sd->nohz_idle)
+ goto unlock;
+ sd->nohz_idle = 1;
+
+ for (; sd; sd = sd->parent)
atomic_dec(&sd->groups->sgp->nr_busy_cpus);
+unlock:
rcu_read_unlock();
}
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index cc03cfd..03b13c8 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1187,7 +1187,6 @@ extern void account_cfs_bandwidth_used(int enabled, int was_enabled);
enum rq_nohz_flag_bits {
NOHZ_TICK_STOPPED,
NOHZ_BALANCE_KICK,
- NOHZ_IDLE,
};
#define nohz_flags(cpu) (&cpu_rq(cpu)->nohz_flags)
--
1.7.9.5
On Tue, Apr 23, 2013 at 04:59:02PM +0200, Vincent Guittot wrote:
> On my smp platform which is made of 5 cores in 2 clusters, I have the
> nr_busy_cpu field of sched_group_power struct that is not null when the
> platform is fully idle. The root cause is:
> During the boot sequence, some CPUs reach the idle loop and set their
> NOHZ_IDLE flag while waiting for others CPUs to boot. But the nr_busy_cpus
> field is initialized later with the assumption that all CPUs are in the busy
> state whereas some CPUs have already set their NOHZ_IDLE flag.
>
> More generally, the NOHZ_IDLE flag must be initialized when new sched_domains
> are created in order to ensure that NOHZ_IDLE and nr_busy_cpus are aligned.
>
> This condition can be ensured by adding a synchronize_rcu between the
> destruction of old sched_domains and the creation of new ones so the NOHZ_IDLE
> flag will not be updated with old sched_domain once it has been initialized.
> But this solution introduces a additionnal latency in the rebuild sequence
> that is called during cpu hotplug.
>
> As suggested by Frederic Weisbecker, another solution is to have the same
> rcu lifecycle for both NOHZ_IDLE and sched_domain struct.
> A new nohz_idle field is added to sched_domain so both status and sched_domain
> will share the same RCU lifecycle and will be always synchronized.
> In addition, there is no more need to protect nohz_idle against concurrent
> access as it is only modified by 2 exclusive functions called by local cpu.
>
> This solution has been prefered to the creation of a new struct with an extra
> pointer indirection for sched_domain.
> Signed-off-by: Vincent Guittot <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Thanks for persisting!
Commit-ID: 25f55d9d01ad7a7ad248fd5af1d22675ffd202c5
Gitweb: http://git.kernel.org/tip/25f55d9d01ad7a7ad248fd5af1d22675ffd202c5
Author: Vincent Guittot <[email protected]>
AuthorDate: Tue, 23 Apr 2013 16:59:02 +0200
Committer: Ingo Molnar <[email protected]>
CommitDate: Fri, 26 Apr 2013 12:13:44 +0200
sched: Fix init NOHZ_IDLE flag
On my SMP platform which is made of 5 cores in 2 clusters, I
have the nr_busy_cpu field of sched_group_power struct that is
not null when the platform is fully idle - which makes the
scheduler unhappy.
The root cause is:
During the boot sequence, some CPUs reach the idle loop and set
their NOHZ_IDLE flag while waiting for others CPUs to boot. But
the nr_busy_cpus field is initialized later with the assumption
that all CPUs are in the busy state whereas some CPUs have
already set their NOHZ_IDLE flag.
More generally, the NOHZ_IDLE flag must be initialized when new
sched_domains are created in order to ensure that NOHZ_IDLE and
nr_busy_cpus are aligned.
This condition can be ensured by adding a synchronize_rcu()
between the destruction of old sched_domains and the creation of
new ones so the NOHZ_IDLE flag will not be updated with old
sched_domain once it has been initialized. But this solution
introduces a additionnal latency in the rebuild sequence that is
called during cpu hotplug.
As suggested by Frederic Weisbecker, another solution is to have
the same rcu lifecycle for both NOHZ_IDLE and sched_domain
struct. A new nohz_idle field is added to sched_domain so both
status and sched_domain will share the same RCU lifecycle and
will be always synchronized. In addition, there is no more need
to protect nohz_idle against concurrent access as it is only
modified by 2 exclusive functions called by local cpu.
This solution has been prefered to the creation of a new struct
with an extra pointer indirection for sched_domain.
The synchronization is done at the cost of :
- An additional indirection and a rcu_dereference for accessing nohz_idle.
- We use only the nohz_idle field of the top sched_domain.
Signed-off-by: Vincent Guittot <[email protected]>
Acked-by: Peter Zijlstra <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
[ Fixed !NO_HZ build bug. ]
Signed-off-by: Ingo Molnar <[email protected]>
---
include/linux/sched.h | 2 ++
kernel/sched/fair.c | 26 ++++++++++++++++----------
kernel/sched/sched.h | 1 -
3 files changed, 18 insertions(+), 11 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6bdaa73..a25168f 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -808,6 +808,8 @@ struct sched_domain {
unsigned int wake_idx;
unsigned int forkexec_idx;
unsigned int smt_gain;
+
+ int nohz_idle; /* NOHZ IDLE status */
int flags; /* See SD_* */
int level;
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index acaf567..8bf7081 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5420,13 +5420,16 @@ static inline void set_cpu_sd_state_busy(void)
struct sched_domain *sd;
int cpu = smp_processor_id();
- if (!test_bit(NOHZ_IDLE, nohz_flags(cpu)))
- return;
- clear_bit(NOHZ_IDLE, nohz_flags(cpu));
-
rcu_read_lock();
- for_each_domain(cpu, sd)
+ sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd);
+
+ if (!sd || !sd->nohz_idle)
+ goto unlock;
+ sd->nohz_idle = 0;
+
+ for (; sd; sd = sd->parent)
atomic_inc(&sd->groups->sgp->nr_busy_cpus);
+unlock:
rcu_read_unlock();
}
@@ -5435,13 +5438,16 @@ void set_cpu_sd_state_idle(void)
struct sched_domain *sd;
int cpu = smp_processor_id();
- if (test_bit(NOHZ_IDLE, nohz_flags(cpu)))
- return;
- set_bit(NOHZ_IDLE, nohz_flags(cpu));
-
rcu_read_lock();
- for_each_domain(cpu, sd)
+ sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd);
+
+ if (!sd || sd->nohz_idle)
+ goto unlock;
+ sd->nohz_idle = 1;
+
+ for (; sd; sd = sd->parent)
atomic_dec(&sd->groups->sgp->nr_busy_cpus);
+unlock:
rcu_read_unlock();
}
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 605426a..4c225c4 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1303,7 +1303,6 @@ extern void account_cfs_bandwidth_used(int enabled, int was_enabled);
enum rq_nohz_flag_bits {
NOHZ_TICK_STOPPED,
NOHZ_BALANCE_KICK,
- NOHZ_IDLE,
};
#define nohz_flags(cpu) (&cpu_rq(cpu)->nohz_flags)