2019-06-20 11:38:10

by Wanpeng Li

Subject: [PATCH] sched/isolation: Prefer housekeeping cpu in local node

From: Wanpeng Li <[email protected]>

In real product setup, there will be houseeking cpus in each nodes, it
is prefer to do housekeeping from local node, fallback to global online
cpumask if failed to find houseeking cpu from local node.

Cc: Ingo Molnar <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Frederic Weisbecker <[email protected]>
Signed-off-by: Wanpeng Li <[email protected]>
---
kernel/sched/isolation.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c
index 123ea07..9eb6805 100644
--- a/kernel/sched/isolation.c
+++ b/kernel/sched/isolation.c
@@ -16,9 +16,16 @@ static unsigned int housekeeping_flags;

int housekeeping_any_cpu(enum hk_flags flags)
{
+ int cpu;
+
if (static_branch_unlikely(&housekeeping_overridden))
- if (housekeeping_flags & flags)
- return cpumask_any_and(housekeeping_mask, cpu_online_mask);
+ if (housekeeping_flags & flags) {
+ cpu = cpumask_any_and(housekeeping_mask, cpu_cpu_mask(smp_processor_id()));
+ if (cpu < nr_cpu_ids)
+ return cpu;
+ else
+ return cpumask_any_and(housekeeping_mask, cpu_online_mask);
+ }
return smp_processor_id();
}
EXPORT_SYMBOL_GPL(housekeeping_any_cpu);
--
2.7.4

2019-06-20 12:38:38

by Peter Zijlstra

Subject: Re: [PATCH] sched/isolation: Prefer housekeeping cpu in local node

On Thu, Jun 20, 2019 at 07:36:54PM +0800, Wanpeng Li wrote:
> From: Wanpeng Li <[email protected]>
>
> In real product setup, there will be houseeking cpus in each nodes, it
> is prefer to do housekeeping from local node, fallback to global online
> cpumask if failed to find houseeking cpu from local node.
>
> Cc: Ingo Molnar <[email protected]>
> Cc: Peter Zijlstra <[email protected]>
> Cc: Frederic Weisbecker <[email protected]>
> Signed-off-by: Wanpeng Li <[email protected]>
> ---
> kernel/sched/isolation.c | 11 +++++++++--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c
> index 123ea07..9eb6805 100644
> --- a/kernel/sched/isolation.c
> +++ b/kernel/sched/isolation.c
> @@ -16,9 +16,16 @@ static unsigned int housekeeping_flags;
>
> int housekeeping_any_cpu(enum hk_flags flags)
> {
> + int cpu;
> +
> if (static_branch_unlikely(&housekeeping_overridden))
> - if (housekeeping_flags & flags)
> - return cpumask_any_and(housekeeping_mask, cpu_online_mask);
> + if (housekeeping_flags & flags) {
> + cpu = cpumask_any_and(housekeeping_mask, cpu_cpu_mask(smp_processor_id()));
> + if (cpu < nr_cpu_ids)
> + return cpu;
> + else
> + return cpumask_any_and(housekeeping_mask, cpu_online_mask);
> + }
> return smp_processor_id();
> }
> EXPORT_SYMBOL_GPL(housekeeping_any_cpu);

Why not something like so? IIRC there's more places that want this, but
I can't seem to remember quite where.

diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c
index 123ea07a3f3b..1cceab5f094c 100644
--- a/kernel/sched/isolation.c
+++ b/kernel/sched/isolation.c
@@ -16,9 +16,15 @@ static unsigned int housekeeping_flags;

int housekeeping_any_cpu(enum hk_flags flags)
{
- if (static_branch_unlikely(&housekeeping_overridden))
- if (housekeeping_flags & flags)
+ if (static_branch_unlikely(&housekeeping_overridden)) {
+ if (housekeeping_flags & flags) {
+ cpu = sched_numa_find_closest(housekeeping_mask, smp_processor_id());
+ if (cpu < nr_cpu_ids)
+ return cpu;
+
return cpumask_any_and(housekeeping_mask, cpu_online_mask);
+ }
+ }
return smp_processor_id();
}
EXPORT_SYMBOL_GPL(housekeeping_any_cpu);
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index b08dee29ef5e..0db7431c7207 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1212,9 +1212,6 @@ enum numa_topology_type {
extern enum numa_topology_type sched_numa_topology_type;
extern int sched_max_numa_distance;
extern bool find_numa_distance(int distance);
-#endif
-
-#ifdef CONFIG_NUMA
extern void sched_init_numa(void);
extern void sched_domains_numa_masks_set(unsigned int cpu);
extern void sched_domains_numa_masks_clear(unsigned int cpu);
@@ -1224,6 +1221,8 @@ static inline void sched_domains_numa_masks_set(unsigned int cpu) { }
static inline void sched_domains_numa_masks_clear(unsigned int cpu) { }
#endif

+extern int sched_numa_find_closest(const struct cpumask *cpus, int cpu);
+
#ifdef CONFIG_NUMA_BALANCING
/* The regions in numa_faults array from task_struct */
enum numa_faults_stats {
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index 63184cf0d0d7..408e94a6637c 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -1726,6 +1726,20 @@ void sched_domains_numa_masks_clear(unsigned int cpu)

#endif /* CONFIG_NUMA */

+int sched_numa_find_closest(const struct cpumask *cpus, int cpu)
+{
+#ifdef CONFIG_NUMA
+ int i, j = cpu_to_node(cpu);
+
+ for (i = 0; i < sched_domains_numa_levels; ++) {
+ cpu = cpumask_any_and(cpus, sched_domains_numa_mask[i][j]);
+ if (cpu < nr_cpu_ids)
+ return cpu;
+ }
+#endif
+ return nr_cpu_ids;
+}
+
static int __sdt_alloc(const struct cpumask *cpu_map)
{
struct sched_domain_topology_level *tl;

2019-06-21 01:41:48

by Wanpeng Li

Subject: Re: [PATCH] sched/isolation: Prefer housekeeping cpu in local node

On Thu, 20 Jun 2019 at 20:38, Peter Zijlstra <[email protected]> wrote:
>
> On Thu, Jun 20, 2019 at 07:36:54PM +0800, Wanpeng Li wrote:
> > From: Wanpeng Li <[email protected]>
> >
> > In real product setup, there will be houseeking cpus in each nodes, it
> > is prefer to do housekeeping from local node, fallback to global online
> > cpumask if failed to find houseeking cpu from local node.
> >
> > Cc: Ingo Molnar <[email protected]>
> > Cc: Peter Zijlstra <[email protected]>
> > Cc: Frederic Weisbecker <[email protected]>
> > Signed-off-by: Wanpeng Li <[email protected]>
> > ---
> > kernel/sched/isolation.c | 11 +++++++++--
> > 1 file changed, 9 insertions(+), 2 deletions(-)
> >
> > diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c
> > index 123ea07..9eb6805 100644
> > --- a/kernel/sched/isolation.c
> > +++ b/kernel/sched/isolation.c
> > @@ -16,9 +16,16 @@ static unsigned int housekeeping_flags;
> >
> > int housekeeping_any_cpu(enum hk_flags flags)
> > {
> > + int cpu;
> > +
> > if (static_branch_unlikely(&housekeeping_overridden))
> > - if (housekeeping_flags & flags)
> > - return cpumask_any_and(housekeeping_mask, cpu_online_mask);
> > + if (housekeeping_flags & flags) {
> > + cpu = cpumask_any_and(housekeeping_mask, cpu_cpu_mask(smp_processor_id()));
> > + if (cpu < nr_cpu_ids)
> > + return cpu;
> > + else
> > + return cpumask_any_and(housekeeping_mask, cpu_online_mask);
> > + }
> > return smp_processor_id();
> > }
> > EXPORT_SYMBOL_GPL(housekeeping_any_cpu);
>
> Why not something like so? IIRC there's more places that want this, but
> I can't seem to remember quite where.

Good point, do it in v2. Btw, could you have a look this patch?
https://lkml.org/lkml/2019/6/17/1723

Regards,
Wanpeng Li