| From: Chen Yu <yu.c.chen@intel.com> |
| Subject: sched/numa: add statistics of numa balance task migration |
| Date: Tue, 8 Apr 2025 18:14:44 +0800 |
| |
| On systems with NUMA balancing enabled, it is found that tracking the task |
| activities due to NUMA balancing is helpful. NUMA balancing has two |
| mechanisms for task migration: one is to migrate the task to an idle CPU |
| in its preferred node, the other is to swap tasks on different nodes if |
| they are on each other's preferred node. |
| |
| The kernel already has NUMA page migration statistics in |
| /sys/fs/cgroup/mytest/memory.stat and /proc/{PID}/sched, but does not have |
| statistics for task migration/swap. Add the task migration and swap count |
| accordingly. |
| |
| The following two new fields: |
| |
| numa_task_migrated |
| numa_task_swapped |
| |
| will be displayed in both |
| /sys/fs/cgroup/{GROUP}/memory.stat and /proc/{PID}/sched |
| |
| Introducing both pertask and permemcg NUMA balancing statistics helps to |
| quickly evaluate the performance and resource usage of the target |
| workload. For example, the user can first identify the container which |
| has high NUMA balance activity and then narrow down to a specific task |
| within that group, and tune the memory policy of that task. In summary, |
| it is plausible to iterate the /proc/$pid/sched to find the offending |
| task, but the introduction of per memcg tasks' Numa balancing aggregated |
| activity can further help users identify the task in a divide-and-conquer |
| way. |
| |
| [yu.c.chen@intel.com: v3] |
| Link: https://lkml.kernel.org/r/20250430103623.3349842-1-yu.c.chen@intel.com |
| Link: https://lkml.kernel.org/r/20250430103623.3349842-1-yu.c.chen@intel.com |
| Link: https://lkml.kernel.org/r/20250408101444.192519-1-yu.c.chen@intel.com |
| Signed-off-by: Chen Yu <yu.c.chen@intel.com> |
| Tested-by: K Prateek Nayak <kprateek.nayak@amd.com> |
| Tested-by: Madadi Vineeth Reddy <vineethr@linux.ibm.com> |
| Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> |
| Cc: Aubrey Li <aubrey.li@intel.com> |
| Cc: "Chen, Tim C" <tim.c.chen@intel.com> |
| Cc: Ingo Molnar <mingo@redhat.com> |
| Cc: Johannes Weiner <hannes@cmpxchg.org> |
| Cc: Jonathan Corbet <corbet@lwn.net> |
| Cc: Mel Gorman <mgorman <mgorman@suse.de> |
| Cc: Michal Hocko <mhocko@kernel.org> |
| Cc: Michal Koutný <mkoutny@suse.com> |
| Cc: Muchun Song <muchun.song@linux.dev> |
| Cc: Roman Gushchin <roman.gushchin@linux.dev> |
| Cc: Shakeel Butt <shakeel.butt@linux.dev> |
| Cc: Tejun Heo <tj@kernel.org> |
| Cc: Libo Chen <libo.chen@oracle.com> |
| Signed-off-by: Andrew Morton <akpm@linux-foundation.org> |
| --- |
| |
| Documentation/admin-guide/cgroup-v2.rst | 6 ++++++ |
| include/linux/sched.h | 4 ++++ |
| include/linux/vm_event_item.h | 2 ++ |
| kernel/sched/core.c | 7 +++++-- |
| kernel/sched/debug.c | 4 ++++ |
| mm/memcontrol.c | 2 ++ |
| mm/vmstat.c | 2 ++ |
| 7 files changed, 25 insertions(+), 2 deletions(-) |
| |
| --- a/Documentation/admin-guide/cgroup-v2.rst~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/Documentation/admin-guide/cgroup-v2.rst |
| @@ -1670,6 +1670,12 @@ The following nested keys are defined. |
| numa_hint_faults (npn) |
| Number of NUMA hinting faults. |
| |
| + numa_task_migrated (npn) |
| + Number of task migration by NUMA balancing. |
| + |
| + numa_task_swapped (npn) |
| + Number of task swap by NUMA balancing. |
| + |
| pgdemote_kswapd |
| Number of pages demoted by kswapd. |
| |
| --- a/include/linux/sched.h~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/include/linux/sched.h |
| @@ -549,6 +549,10 @@ struct sched_statistics { |
| u64 nr_failed_migrations_running; |
| u64 nr_failed_migrations_hot; |
| u64 nr_forced_migrations; |
| +#ifdef CONFIG_NUMA_BALANCING |
| + u64 numa_task_migrated; |
| + u64 numa_task_swapped; |
| +#endif |
| |
| u64 nr_wakeups; |
| u64 nr_wakeups_sync; |
| --- a/include/linux/vm_event_item.h~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/include/linux/vm_event_item.h |
| @@ -66,6 +66,8 @@ enum vm_event_item { PGPGIN, PGPGOUT, PS |
| NUMA_HINT_FAULTS, |
| NUMA_HINT_FAULTS_LOCAL, |
| NUMA_PAGE_MIGRATE, |
| + NUMA_TASK_MIGRATE, |
| + NUMA_TASK_SWAP, |
| #endif |
| #ifdef CONFIG_MIGRATION |
| PGMIGRATE_SUCCESS, PGMIGRATE_FAIL, |
| --- a/kernel/sched/core.c~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/kernel/sched/core.c |
| @@ -3352,6 +3352,9 @@ void set_task_cpu(struct task_struct *p, |
| #ifdef CONFIG_NUMA_BALANCING |
| static void __migrate_swap_task(struct task_struct *p, int cpu) |
| { |
| + __schedstat_inc(p->stats.numa_task_swapped); |
| + count_memcg_events_mm(p->mm, NUMA_TASK_SWAP, 1); |
| + |
| if (task_on_rq_queued(p)) { |
| struct rq *src_rq, *dst_rq; |
| struct rq_flags srf, drf; |
| @@ -7953,8 +7956,8 @@ int migrate_task_to(struct task_struct * |
| if (!cpumask_test_cpu(target_cpu, p->cpus_ptr)) |
| return -EINVAL; |
| |
| - /* TODO: This is not properly updating schedstats */ |
| - |
| + __schedstat_inc(p->stats.numa_task_migrated); |
| + count_memcg_events_mm(p->mm, NUMA_TASK_MIGRATE, 1); |
| trace_sched_move_numa(p, curr_cpu, target_cpu); |
| return stop_one_cpu(curr_cpu, migration_cpu_stop, &arg); |
| } |
| --- a/kernel/sched/debug.c~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/kernel/sched/debug.c |
| @@ -1206,6 +1206,10 @@ void proc_sched_show_task(struct task_st |
| P_SCHEDSTAT(nr_failed_migrations_running); |
| P_SCHEDSTAT(nr_failed_migrations_hot); |
| P_SCHEDSTAT(nr_forced_migrations); |
| +#ifdef CONFIG_NUMA_BALANCING |
| + P_SCHEDSTAT(numa_task_migrated); |
| + P_SCHEDSTAT(numa_task_swapped); |
| +#endif |
| P_SCHEDSTAT(nr_wakeups); |
| P_SCHEDSTAT(nr_wakeups_sync); |
| P_SCHEDSTAT(nr_wakeups_migrate); |
| --- a/mm/memcontrol.c~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/mm/memcontrol.c |
| @@ -470,6 +470,8 @@ static const unsigned int memcg_vm_event |
| NUMA_PAGE_MIGRATE, |
| NUMA_PTE_UPDATES, |
| NUMA_HINT_FAULTS, |
| + NUMA_TASK_MIGRATE, |
| + NUMA_TASK_SWAP, |
| #endif |
| }; |
| |
| --- a/mm/vmstat.c~sched-numa-add-statistics-of-numa-balance-task-migration-and-swap |
| +++ a/mm/vmstat.c |
| @@ -1347,6 +1347,8 @@ const char * const vmstat_text[] = { |
| "numa_hint_faults", |
| "numa_hint_faults_local", |
| "numa_pages_migrated", |
| + "numa_task_migrated", |
| + "numa_task_swapped", |
| #endif |
| #ifdef CONFIG_MIGRATION |
| "pgmigrate_success", |
| _ |
