From f6db8347993256b58bd4746b0c4c5b935c32210d Mon Sep 17 00:00:00 2001 From: "Naveen N. Rao" Date: Thu, 25 Jun 2015 23:53:37 +0530 Subject: [PATCH 1/5] sched/stat: Simplify the sched_info accounting dependency Both CONFIG_SCHEDSTATS=y and CONFIG_TASK_DELAY_ACCT=y track task sched_info, which results in ugly #if clauses. Simplify the code by introducing a synthethic CONFIG_SCHED_INFO switch, selected by both. Signed-off-by: Naveen N. Rao Cc: Balbir Singh Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Thomas Gleixner Cc: a.p.zijlstra@chello.nl Cc: ricklind@us.ibm.com Link: http://lkml.kernel.org/r/8d19eef800811a94b0f91bcbeb27430a884d7433.1435255405.git.naveen.n.rao@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 6 +++--- init/Kconfig | 1 + kernel/sched/core.c | 2 +- kernel/sched/stats.h | 4 ++-- lib/Kconfig.debug | 5 +++++ 5 files changed, 12 insertions(+), 6 deletions(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index 6633e83e608a..9bf4bc0e3b8b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -849,7 +849,7 @@ extern struct user_struct root_user; struct backing_dev_info; struct reclaim_state; -#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) +#ifdef CONFIG_SCHED_INFO struct sched_info { /* cumulative counters */ unsigned long pcount; /* # of times run on this cpu */ @@ -859,7 +859,7 @@ struct sched_info { unsigned long long last_arrival,/* when we last ran on a cpu */ last_queued; /* when we were last queued to run */ }; -#endif /* defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) */ +#endif /* CONFIG_SCHED_INFO */ #ifdef CONFIG_TASK_DELAY_ACCT struct task_delay_info { @@ -1408,7 +1408,7 @@ struct task_struct { int rcu_tasks_idle_cpu; #endif /* #ifdef CONFIG_TASKS_RCU */ -#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) +#ifdef CONFIG_SCHED_INFO struct sched_info sched_info; #endif diff --git a/init/Kconfig b/init/Kconfig index b999fa381bf9..12cb556ad160 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -435,6 +435,7 @@ config TASKSTATS config TASK_DELAY_ACCT bool "Enable per-task delay accounting" depends on TASKSTATS + select SCHED_INFO help Collect information on time spent by a task waiting for system resources like cpu, synchronous block I/O completion and swapping diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c86935a7f1f8..abb8785ed1ba 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1975,7 +1975,7 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p) set_task_cpu(p, cpu); raw_spin_unlock_irqrestore(&p->pi_lock, flags); -#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) +#ifdef CONFIG_SCHED_INFO if (likely(sched_info_on())) memset(&p->sched_info, 0, sizeof(p->sched_info)); #endif diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h index 077ebbd5e10f..b0fbc7632de5 100644 --- a/kernel/sched/stats.h +++ b/kernel/sched/stats.h @@ -47,7 +47,7 @@ rq_sched_info_depart(struct rq *rq, unsigned long long delta) # define schedstat_set(var, val) do { } while (0) #endif -#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) +#ifdef CONFIG_SCHED_INFO static inline void sched_info_reset_dequeued(struct task_struct *t) { t->sched_info.last_queued = 0; @@ -156,7 +156,7 @@ sched_info_switch(struct rq *rq, #define sched_info_depart(rq, t) do { } while (0) #define sched_info_arrive(rq, next) do { } while (0) #define sched_info_switch(rq, t, next) do { } while (0) -#endif /* CONFIG_SCHEDSTATS || CONFIG_TASK_DELAY_ACCT */ +#endif /* CONFIG_SCHED_INFO */ /* * The following are functions that support scheduler-internal time accounting. diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index b908048f8d6a..e2894b23efb6 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -841,9 +841,14 @@ config SCHED_DEBUG that can help debug the scheduler. The runtime overhead of this option is minimal. +config SCHED_INFO + bool + default n + config SCHEDSTATS bool "Collect scheduler statistics" depends on DEBUG_KERNEL && PROC_FS + select SCHED_INFO help If you say Y here, additional code will be inserted into the scheduler and related routines to collect statistics about From 5968cecedd7a09f23e9fcb5f9fb4e893712f35ba Mon Sep 17 00:00:00 2001 From: "Naveen N. Rao" Date: Tue, 30 Jun 2015 14:36:03 +0530 Subject: [PATCH 2/5] sched/stat: Expose /proc/pid/schedstat if CONFIG_SCHED_INFO=y Expand /proc/pid/schedstat output: - enable it on CONFIG_TASK_DELAY_ACCT=y && !CONFIG_SCHEDSTATS kernels. - dump all zeroes on kernels that are booted with the 'nodelayacct' option, which boot option disables delay accounting on CONFIG_TASK_DELAY_ACCT=y kernels. Signed-off-by: Naveen N. Rao Cc: Balbir Singh Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Srikar Dronamraju Cc: Thomas Gleixner Cc: a.p.zijlstra@chello.nl Cc: ricklind@us.ibm.com Link: http://lkml.kernel.org/r/5ccbef17d4bc841084ea6e6421d4e4a23b7b806f.1435654789.git.naveen.n.rao@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- fs/proc/base.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/fs/proc/base.c b/fs/proc/base.c index 286a422f440e..ad63fa3306df 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -304,14 +304,17 @@ static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns, } #endif -#ifdef CONFIG_SCHEDSTATS +#ifdef CONFIG_SCHED_INFO /* * Provides /proc/PID/schedstat */ static int proc_pid_schedstat(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task) { - seq_printf(m, "%llu %llu %lu\n", + if (unlikely(!sched_info_on())) + seq_printf(m, "0 0 0\n"); + else + seq_printf(m, "%llu %llu %lu\n", (unsigned long long)task->se.sum_exec_runtime, (unsigned long long)task->sched_info.run_delay, task->sched_info.pcount); @@ -2600,7 +2603,7 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_STACKTRACE ONE("stack", S_IRUSR, proc_pid_stack), #endif -#ifdef CONFIG_SCHEDSTATS +#ifdef CONFIG_SCHED_INFO ONE("schedstat", S_IRUGO, proc_pid_schedstat), #endif #ifdef CONFIG_LATENCYTOP @@ -2948,7 +2951,7 @@ static const struct pid_entry tid_base_stuff[] = { #ifdef CONFIG_STACKTRACE ONE("stack", S_IRUSR, proc_pid_stack), #endif -#ifdef CONFIG_SCHEDSTATS +#ifdef CONFIG_SCHED_INFO ONE("schedstat", S_IRUGO, proc_pid_schedstat), #endif #ifdef CONFIG_LATENCYTOP From 6b55c9654fccf69ae7ace23ca101dc37b903181b Mon Sep 17 00:00:00 2001 From: Srikar Dronamraju Date: Thu, 25 Jun 2015 22:51:41 +0530 Subject: [PATCH 3/5] sched/debug: Move print_cfs_rq() declaration to kernel/sched/sched.h Currently print_cfs_rq() is declared in include/linux/sched.h. However it's not used outside kernel/sched. Hence move the declaration to kernel/sched/sched.h Also some functions are only available for CONFIG_SCHED_DEBUG=y. Hence move the declarations to within the #ifdef. Signed-off-by: Srikar Dronamraju Acked-by: Rik van Riel Cc: Iulia Manda Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1435252903-1081-2-git-send-email-srikar@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 -- kernel/sched/sched.h | 5 +++++ 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index 9bf4bc0e3b8b..3505352dd296 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -191,8 +191,6 @@ struct task_group; #ifdef CONFIG_SCHED_DEBUG extern void proc_sched_show_task(struct task_struct *p, struct seq_file *m); extern void proc_sched_set_task(struct task_struct *p); -extern void -print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq); #endif /* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index aea7c1f393cb..7d5895258fe3 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1668,9 +1668,14 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) extern struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq); extern struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq); + +#ifdef CONFIG_SCHED_DEBUG extern void print_cfs_stats(struct seq_file *m, int cpu); extern void print_rt_stats(struct seq_file *m, int cpu); extern void print_dl_stats(struct seq_file *m, int cpu); +extern void +print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq); +#endif extern void init_cfs_rq(struct cfs_rq *cfs_rq); extern void init_rt_rq(struct rt_rq *rt_rq); From e3d24d0a6048a826de5562d75dedb664d3a2a1b2 Mon Sep 17 00:00:00 2001 From: Srikar Dronamraju Date: Thu, 25 Jun 2015 22:51:42 +0530 Subject: [PATCH 4/5] sched/numa: Show numa_group ID in /proc/sched_debug task listings Having the numa group ID in /proc/sched_debug helps to see how the numa groups have spread across the system. Signed-off-by: Srikar Dronamraju Acked-by: Rik van Riel Cc: Iulia Manda Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1435252903-1081-3-git-send-email-srikar@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 315c68e015d9..f1dcd1d390c1 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -142,7 +142,7 @@ print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) 0LL, 0L); #endif #ifdef CONFIG_NUMA_BALANCING - SEQ_printf(m, " %d", task_node(p)); + SEQ_printf(m, " %d %d", task_node(p), task_numa_group_id(p)); #endif #ifdef CONFIG_CGROUP_SCHED SEQ_printf(m, " %s", task_group_path(task_group(p))); From 397f2378f136128623fc237746157aa2564d1082 Mon Sep 17 00:00:00 2001 From: Srikar Dronamraju Date: Thu, 25 Jun 2015 22:51:43 +0530 Subject: [PATCH 5/5] sched/numa: Fix numa balancing stats in /proc/pid/sched Commit 44dba3d5d6a1 ("sched: Refactor task_struct to use numa_faults instead of numa_* pointers") modified the way tsk->numa_faults stats are accounted. However that commit never touched show_numa_stats() that is displayed in /proc/pid/sched and thus the numbers displayed in /proc/pid/sched don't match the actual numbers. Fix it by making sure that /proc/pid/sched reflects the task fault numbers. Also add group fault stats too. Also couple of more modifications are added here: 1. Format changes: - Previously we would list two entries per node, one for private and one for shared. Also the home node info was listed in each entry. - Now preferred node, total_faults and current node are displayed separately. - Now there is one entry per node, that lists private,shared task and group faults. 2. Unit changes: - p->numa_pages_migrated was getting reset after every read of /proc/pid/sched. It's more useful to have absolute numbers since differential migrations between two accesses can be more easily calculated. Signed-off-by: Srikar Dronamraju Acked-by: Rik van Riel Cc: Iulia Manda Cc: Linus Torvalds Cc: Mel Gorman Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1435252903-1081-4-git-send-email-srikar@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/debug.c | 38 +++++++++++++++++--------------------- kernel/sched/fair.c | 22 +++++++++++++++++++++- kernel/sched/sched.h | 10 +++++++++- 3 files changed, 47 insertions(+), 23 deletions(-) diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index f1dcd1d390c1..4222ec50ab88 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -517,11 +517,21 @@ __initcall(init_sched_debug_procfs); SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)p->F)) +#ifdef CONFIG_NUMA_BALANCING +void print_numa_stats(struct seq_file *m, int node, unsigned long tsf, + unsigned long tpf, unsigned long gsf, unsigned long gpf) +{ + SEQ_printf(m, "numa_faults node=%d ", node); + SEQ_printf(m, "task_private=%lu task_shared=%lu ", tsf, tpf); + SEQ_printf(m, "group_private=%lu group_shared=%lu\n", gsf, gpf); +} +#endif + + static void sched_show_numa(struct task_struct *p, struct seq_file *m) { #ifdef CONFIG_NUMA_BALANCING struct mempolicy *pol; - int node, i; if (p->mm) P(mm->numa_scan_seq); @@ -533,26 +543,12 @@ static void sched_show_numa(struct task_struct *p, struct seq_file *m) mpol_get(pol); task_unlock(p); - SEQ_printf(m, "numa_migrations, %ld\n", xchg(&p->numa_pages_migrated, 0)); - - for_each_online_node(node) { - for (i = 0; i < 2; i++) { - unsigned long nr_faults = -1; - int cpu_current, home_node; - - if (p->numa_faults) - nr_faults = p->numa_faults[2*node + i]; - - cpu_current = !i ? (task_node(p) == node) : - (pol && node_isset(node, pol->v.nodes)); - - home_node = (p->numa_preferred_nid == node); - - SEQ_printf(m, "numa_faults_memory, %d, %d, %d, %d, %ld\n", - i, node, cpu_current, home_node, nr_faults); - } - } - + P(numa_pages_migrated); + P(numa_preferred_nid); + P(total_numa_faults); + SEQ_printf(m, "current_node=%d, numa_group_id=%d\n", + task_node(p), task_numa_group_id(p)); + show_numa_stats(p, m); mpol_put(pol); #endif } diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 40a7fcbf491e..7245039fc67b 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8468,7 +8468,27 @@ void print_cfs_stats(struct seq_file *m, int cpu) print_cfs_rq(m, cpu, cfs_rq); rcu_read_unlock(); } -#endif + +#ifdef CONFIG_NUMA_BALANCING +void show_numa_stats(struct task_struct *p, struct seq_file *m) +{ + int node; + unsigned long tsf = 0, tpf = 0, gsf = 0, gpf = 0; + + for_each_online_node(node) { + if (p->numa_faults) { + tsf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 0)]; + tpf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 1)]; + } + if (p->numa_group) { + gsf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 0)], + gpf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 1)]; + } + print_numa_stats(m, node, tsf, tpf, gsf, gpf); + } +} +#endif /* CONFIG_NUMA_BALANCING */ +#endif /* CONFIG_SCHED_DEBUG */ __init void init_sched_fair_class(void) { diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 7d5895258fe3..7ef596837dac 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1675,7 +1675,15 @@ extern void print_rt_stats(struct seq_file *m, int cpu); extern void print_dl_stats(struct seq_file *m, int cpu); extern void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq); -#endif + +#ifdef CONFIG_NUMA_BALANCING +extern void +show_numa_stats(struct task_struct *p, struct seq_file *m); +extern void +print_numa_stats(struct seq_file *m, int node, unsigned long tsf, + unsigned long tpf, unsigned long gsf, unsigned long gpf); +#endif /* CONFIG_NUMA_BALANCING */ +#endif /* CONFIG_SCHED_DEBUG */ extern void init_cfs_rq(struct cfs_rq *cfs_rq); extern void init_rt_rq(struct rt_rq *rt_rq);