From 1477b6a7edd9ffa7bba4f9779ce9a76ce92761ed Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Wed, 4 Nov 2009 16:14:16 +0900 Subject: sched: Remove unused __schedule() declaration __schedule() had been removed. Signed-off-by: Hiroshi Shimamoto Cc: Peter Zijlstra LKML-Reference: <4AF129C8.3030008@ct.jp.nec.com> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 75e6e60bf583..f18102c4d0b8 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -349,7 +349,6 @@ extern signed long schedule_timeout(signed long timeout); extern signed long schedule_timeout_interruptible(signed long timeout); extern signed long schedule_timeout_killable(signed long timeout); extern signed long schedule_timeout_uninterruptible(signed long timeout); -asmlinkage void __schedule(void); asmlinkage void schedule(void); extern int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner); -- cgit v1.2.3 From 2a2bb3142d326bb28b03875cabfc49baaac9a14a Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Wed, 4 Nov 2009 16:16:10 +0900 Subject: sched: Remove unused time_sync_thresh declaration time_sync_thresh had been removed. Signed-off-by: Hiroshi Shimamoto Cc: Peter Zijlstra LKML-Reference: <4AF12A3A.5050200@ct.jp.nec.com> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index f18102c4d0b8..754b3deed02b 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -171,8 +171,6 @@ print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) } #endif -extern unsigned long long time_sync_thresh; - /* * Task state bitmask. NOTE! These bits are also * encoded in fs/proc/array.c: get_task_state(). -- cgit v1.2.3 From 9824a2b728b63e7ff586b9fd9293c819be79f0f3 Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Wed, 4 Nov 2009 16:16:54 +0900 Subject: sched: Remove unused cpu_nr_migrations() cpu_nr_migrations() is not used, remove it. Signed-off-by: Hiroshi Shimamoto Cc: Peter Zijlstra LKML-Reference: <4AF12A66.6020609@ct.jp.nec.com> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/sched.c | 11 ----------- 2 files changed, 12 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 754b3deed02b..dfc21fb76bf1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -145,7 +145,6 @@ extern unsigned long this_cpu_load(void); extern void calc_global_load(void); -extern u64 cpu_nr_migrations(int cpu); extern unsigned long get_parent_ip(unsigned long addr); diff --git a/kernel/sched.c b/kernel/sched.c index 67be4d0dddaa..30fd0ba5f603 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -541,7 +541,6 @@ struct rq { struct load_weight load; unsigned long nr_load_updates; u64 nr_switches; - u64 nr_migrations_in; struct cfs_rq cfs; struct rt_rq rt; @@ -2049,7 +2048,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) #endif if (old_cpu != new_cpu) { p->se.nr_migrations++; - new_rq->nr_migrations_in++; #ifdef CONFIG_SCHEDSTATS if (task_hot(p, old_rq->clock, NULL)) schedstat_inc(p, se.nr_forced2_migrations); @@ -2988,15 +2986,6 @@ static void calc_load_account_active(struct rq *this_rq) } } -/* - * Externally visible per-cpu scheduler statistics: - * cpu_nr_migrations(cpu) - number of migrations into that cpu - */ -u64 cpu_nr_migrations(int cpu) -{ - return cpu_rq(cpu)->nr_migrations_in; -} - /* * Update rq->cpu_load[] statistics. This function is usually called every * scheduler tick (TICK_NSEC). -- cgit v1.2.3 From acc3f5d7cabbfd6cec71f0c1f9900621fa2d6ae7 Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Tue, 3 Nov 2009 14:53:40 +1030 Subject: cpumask: Partition_sched_domains takes array of cpumask_var_t Currently partition_sched_domains() takes a 'struct cpumask *doms_new' which is a kmalloc'ed array of cpumask_t. You can't have such an array if 'struct cpumask' is undefined, as we plan for CONFIG_CPUMASK_OFFSTACK=y. So, we make this an array of cpumask_var_t instead: this is the same for the CONFIG_CPUMASK_OFFSTACK=n case, but requires multiple allocations for the CONFIG_CPUMASK_OFFSTACK=y case. Hence we add alloc_sched_domains() and free_sched_domains() functions. Signed-off-by: Rusty Russell Cc: Peter Zijlstra LKML-Reference: <200911031453.40668.rusty@rustcorp.com.au> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 8 ++++-- kernel/cpuset.c | 19 +++++++------- kernel/sched.c | 68 ++++++++++++++++++++++++++++++++++----------------- 3 files changed, 61 insertions(+), 34 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index dfc21fb76bf1..78ba664474f3 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1009,9 +1009,13 @@ static inline struct cpumask *sched_domain_span(struct sched_domain *sd) return to_cpumask(sd->span); } -extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, +extern void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new); +/* Allocate an array of sched domains, for partition_sched_domains(). */ +cpumask_var_t *alloc_sched_domains(unsigned int ndoms); +void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms); + /* Test a flag in parent sched domain */ static inline int test_sd_parent(struct sched_domain *sd, int flag) { @@ -1029,7 +1033,7 @@ unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu); struct sched_domain_attr; static inline void -partition_sched_domains(int ndoms_new, struct cpumask *doms_new, +partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new) { } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index d247381e7371..3cf2183b472d 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -537,8 +537,7 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) * element of the partition (one sched domain) to be passed to * partition_sched_domains(). */ -/* FIXME: see the FIXME in partition_sched_domains() */ -static int generate_sched_domains(struct cpumask **domains, +static int generate_sched_domains(cpumask_var_t **domains, struct sched_domain_attr **attributes) { LIST_HEAD(q); /* queue of cpusets to be scanned */ @@ -546,7 +545,7 @@ static int generate_sched_domains(struct cpumask **domains, struct cpuset **csa; /* array of all cpuset ptrs */ int csn; /* how many cpuset ptrs in csa so far */ int i, j, k; /* indices for partition finding loops */ - struct cpumask *doms; /* resulting partition; i.e. sched domains */ + cpumask_var_t *doms; /* resulting partition; i.e. sched domains */ struct sched_domain_attr *dattr; /* attributes for custom domains */ int ndoms = 0; /* number of sched domains in result */ int nslot; /* next empty doms[] struct cpumask slot */ @@ -557,7 +556,8 @@ static int generate_sched_domains(struct cpumask **domains, /* Special case for the 99% of systems with one, full, sched domain */ if (is_sched_load_balance(&top_cpuset)) { - doms = kmalloc(cpumask_size(), GFP_KERNEL); + ndoms = 1; + doms = alloc_sched_domains(ndoms); if (!doms) goto done; @@ -566,9 +566,8 @@ static int generate_sched_domains(struct cpumask **domains, *dattr = SD_ATTR_INIT; update_domain_attr_tree(dattr, &top_cpuset); } - cpumask_copy(doms, top_cpuset.cpus_allowed); + cpumask_copy(doms[0], top_cpuset.cpus_allowed); - ndoms = 1; goto done; } @@ -636,7 +635,7 @@ restart: * Now we know how many domains to create. * Convert to and populate cpu masks. */ - doms = kmalloc(ndoms * cpumask_size(), GFP_KERNEL); + doms = alloc_sched_domains(ndoms); if (!doms) goto done; @@ -656,7 +655,7 @@ restart: continue; } - dp = doms + nslot; + dp = doms[nslot]; if (nslot == ndoms) { static int warnings = 10; @@ -718,7 +717,7 @@ done: static void do_rebuild_sched_domains(struct work_struct *unused) { struct sched_domain_attr *attr; - struct cpumask *doms; + cpumask_var_t *doms; int ndoms; get_online_cpus(); @@ -2052,7 +2051,7 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, unsigned long phase, void *unused_cpu) { struct sched_domain_attr *attr; - struct cpumask *doms; + cpumask_var_t *doms; int ndoms; switch (phase) { diff --git a/kernel/sched.c b/kernel/sched.c index 30fd0ba5f603..ae026aad145b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -8846,7 +8846,7 @@ static int build_sched_domains(const struct cpumask *cpu_map) return __build_sched_domains(cpu_map, NULL); } -static struct cpumask *doms_cur; /* current sched domains */ +static cpumask_var_t *doms_cur; /* current sched domains */ static int ndoms_cur; /* number of sched domains in 'doms_cur' */ static struct sched_domain_attr *dattr_cur; /* attribues of custom domains in 'doms_cur' */ @@ -8868,6 +8868,31 @@ int __attribute__((weak)) arch_update_cpu_topology(void) return 0; } +cpumask_var_t *alloc_sched_domains(unsigned int ndoms) +{ + int i; + cpumask_var_t *doms; + + doms = kmalloc(sizeof(*doms) * ndoms, GFP_KERNEL); + if (!doms) + return NULL; + for (i = 0; i < ndoms; i++) { + if (!alloc_cpumask_var(&doms[i], GFP_KERNEL)) { + free_sched_domains(doms, i); + return NULL; + } + } + return doms; +} + +void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms) +{ + unsigned int i; + for (i = 0; i < ndoms; i++) + free_cpumask_var(doms[i]); + kfree(doms); +} + /* * Set up scheduler domains and groups. Callers must hold the hotplug lock. * For now this just excludes isolated cpus, but could be used to @@ -8879,12 +8904,12 @@ static int arch_init_sched_domains(const struct cpumask *cpu_map) arch_update_cpu_topology(); ndoms_cur = 1; - doms_cur = kmalloc(cpumask_size(), GFP_KERNEL); + doms_cur = alloc_sched_domains(ndoms_cur); if (!doms_cur) - doms_cur = fallback_doms; - cpumask_andnot(doms_cur, cpu_map, cpu_isolated_map); + doms_cur = &fallback_doms; + cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); dattr_cur = NULL; - err = build_sched_domains(doms_cur); + err = build_sched_domains(doms_cur[0]); register_sched_domain_sysctl(); return err; @@ -8934,19 +8959,19 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * doms_new[] to the current sched domain partitioning, doms_cur[]. * It destroys each deleted domain and builds each new domain. * - * 'doms_new' is an array of cpumask's of length 'ndoms_new'. + * 'doms_new' is an array of cpumask_var_t's of length 'ndoms_new'. * The masks don't intersect (don't overlap.) We should setup one * sched domain for each mask. CPUs not in any of the cpumasks will * not be load balanced. If the same cpumask appears both in the * current 'doms_cur' domains and in the new 'doms_new', we can leave * it as it is. * - * The passed in 'doms_new' should be kmalloc'd. This routine takes - * ownership of it and will kfree it when done with it. If the caller - * failed the kmalloc call, then it can pass in doms_new == NULL && - * ndoms_new == 1, and partition_sched_domains() will fallback to - * the single partition 'fallback_doms', it also forces the domains - * to be rebuilt. + * The passed in 'doms_new' should be allocated using + * alloc_sched_domains. This routine takes ownership of it and will + * free_sched_domains it when done with it. If the caller failed the + * alloc call, then it can pass in doms_new == NULL && ndoms_new == 1, + * and partition_sched_domains() will fallback to the single partition + * 'fallback_doms', it also forces the domains to be rebuilt. * * If doms_new == NULL it will be replaced with cpu_online_mask. * ndoms_new == 0 is a special case for destroying existing domains, @@ -8954,8 +8979,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * * Call with hotplug lock held */ -/* FIXME: Change to struct cpumask *doms_new[] */ -void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, +void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new) { int i, j, n; @@ -8974,40 +8998,40 @@ void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, /* Destroy deleted domains */ for (i = 0; i < ndoms_cur; i++) { for (j = 0; j < n && !new_topology; j++) { - if (cpumask_equal(&doms_cur[i], &doms_new[j]) + if (cpumask_equal(doms_cur[i], doms_new[j]) && dattrs_equal(dattr_cur, i, dattr_new, j)) goto match1; } /* no match - a current sched domain not in new doms_new[] */ - detach_destroy_domains(doms_cur + i); + detach_destroy_domains(doms_cur[i]); match1: ; } if (doms_new == NULL) { ndoms_cur = 0; - doms_new = fallback_doms; - cpumask_andnot(&doms_new[0], cpu_online_mask, cpu_isolated_map); + doms_new = &fallback_doms; + cpumask_andnot(doms_new[0], cpu_online_mask, cpu_isolated_map); WARN_ON_ONCE(dattr_new); } /* Build new domains */ for (i = 0; i < ndoms_new; i++) { for (j = 0; j < ndoms_cur && !new_topology; j++) { - if (cpumask_equal(&doms_new[i], &doms_cur[j]) + if (cpumask_equal(doms_new[i], doms_cur[j]) && dattrs_equal(dattr_new, i, dattr_cur, j)) goto match2; } /* no match - add a new doms_new */ - __build_sched_domains(doms_new + i, + __build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL); match2: ; } /* Remember the new sched domains */ - if (doms_cur != fallback_doms) - kfree(doms_cur); + if (doms_cur != &fallback_doms) + free_sched_domains(doms_cur, ndoms_cur); kfree(dattr_cur); /* kfree(NULL) is safe */ doms_cur = doms_new; dattr_cur = dattr_new; -- cgit v1.2.3 From d180c5bccec02612256fd8076ff3c1fac3429553 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Thu, 26 Nov 2009 14:48:30 +0900 Subject: sched: Introduce task_times() to replace task_{u,s}time() pair Functions task_{u,s}time() are called in pair in almost all cases. However task_stime() is implemented to call task_utime() from its inside, so such paired calls run task_utime() twice. It means we do heavy divisions (div_u64 + do_div) twice to get utime and stime which can be obtained at same time by one set of divisions. This patch introduces a function task_times(*tsk, *utime, *stime) to retrieve utime and stime at once in better, optimized way. Signed-off-by: Hidetoshi Seto Acked-by: Peter Zijlstra Cc: Stanislaw Gruszka Cc: Spencer Candland Cc: Oleg Nesterov Cc: Balbir Singh Cc: Americo Wang LKML-Reference: <4B0E16AE.906@jp.fujitsu.com> Signed-off-by: Ingo Molnar --- fs/proc/array.c | 3 +-- include/linux/sched.h | 1 + kernel/exit.c | 7 +++++-- kernel/sched.c | 55 ++++++++++++++++++++++++++++++++------------------- kernel/sys.c | 3 +-- 5 files changed, 43 insertions(+), 26 deletions(-) (limited to 'include/linux/sched.h') diff --git a/fs/proc/array.c b/fs/proc/array.c index e209f64ab27b..330deda70d08 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -535,8 +535,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, if (!whole) { min_flt = task->min_flt; maj_flt = task->maj_flt; - utime = task_utime(task); - stime = task_stime(task); + task_times(task, &utime, &stime); gtime = task_gtime(task); } diff --git a/include/linux/sched.h b/include/linux/sched.h index 78ba664474f3..fe6ae1516640 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1723,6 +1723,7 @@ static inline void put_task_struct(struct task_struct *t) extern cputime_t task_utime(struct task_struct *p); extern cputime_t task_stime(struct task_struct *p); extern cputime_t task_gtime(struct task_struct *p); +extern void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st); /* * Per process flags diff --git a/kernel/exit.c b/kernel/exit.c index f7864ac2ecc1..29068ab2670a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -91,6 +91,8 @@ static void __exit_signal(struct task_struct *tsk) if (atomic_dec_and_test(&sig->count)) posix_cpu_timers_exit_group(tsk); else { + cputime_t utime, stime; + /* * If there is any task waiting for the group exit * then notify it: @@ -110,8 +112,9 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ - sig->utime = cputime_add(sig->utime, task_utime(tsk)); - sig->stime = cputime_add(sig->stime, task_stime(tsk)); + task_times(tsk, &utime, &stime); + sig->utime = cputime_add(sig->utime, utime); + sig->stime = cputime_add(sig->stime, stime); sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; diff --git a/kernel/sched.c b/kernel/sched.c index 315ba4059f93..475a6f2b7158 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5191,6 +5191,14 @@ cputime_t task_stime(struct task_struct *p) { return p->stime; } + +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) +{ + if (ut) + *ut = task_utime(p); + if (st) + *st = task_stime(p); +} #else #ifndef nsecs_to_cputime @@ -5198,41 +5206,48 @@ cputime_t task_stime(struct task_struct *p) msecs_to_cputime(div_u64((__nsecs), NSEC_PER_MSEC)) #endif -cputime_t task_utime(struct task_struct *p) +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - cputime_t utime = p->utime, total = utime + p->stime; - u64 temp; + cputime_t rtime, utime = p->utime, total = utime + p->stime; /* * Use CFS's precise accounting: */ - temp = (u64)nsecs_to_cputime(p->se.sum_exec_runtime); + rtime = nsecs_to_cputime(p->se.sum_exec_runtime); if (total) { - temp *= utime; + u64 temp; + + temp = (u64)(rtime * utime); do_div(temp, total); - } - utime = (cputime_t)temp; + utime = (cputime_t)temp; + } else + utime = rtime; + /* + * Compare with previous values, to keep monotonicity: + */ p->prev_utime = max(p->prev_utime, utime); - return p->prev_utime; + p->prev_stime = max(p->prev_stime, rtime - p->prev_utime); + + if (ut) + *ut = p->prev_utime; + if (st) + *st = p->prev_stime; +} + +cputime_t task_utime(struct task_struct *p) +{ + cputime_t utime; + task_times(p, &utime, NULL); + return utime; } cputime_t task_stime(struct task_struct *p) { cputime_t stime; - - /* - * Use CFS's precise accounting. (we subtract utime from - * the total, to make sure the total observed by userspace - * grows monotonically - apps rely on that): - */ - stime = nsecs_to_cputime(p->se.sum_exec_runtime) - task_utime(p); - - if (stime >= 0) - p->prev_stime = max(p->prev_stime, stime); - - return p->prev_stime; + task_times(p, NULL, &stime); + return stime; } #endif diff --git a/kernel/sys.c b/kernel/sys.c index ce17760d9c51..bbdfce0d4347 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1346,8 +1346,7 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) utime = stime = cputime_zero; if (who == RUSAGE_THREAD) { - utime = task_utime(current); - stime = task_stime(current); + task_times(current, &utime, &stime); accumulate_thread_rusage(p, r); maxrss = p->signal->maxrss; goto out; -- cgit v1.2.3 From d5b7c78e975302a1bab28263266c39ecb71acad4 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Thu, 26 Nov 2009 14:49:05 +0900 Subject: sched: Remove task_{u,s,g}time() Now all task_{u,s}time() pairs are replaced by task_times(). And task_gtime() is too simple to be an inline function. Cleanup them all. Signed-off-by: Hidetoshi Seto Acked-by: Peter Zijlstra Cc: Stanislaw Gruszka Cc: Spencer Candland Cc: Oleg Nesterov Cc: Balbir Singh Cc: Americo Wang LKML-Reference: <4B0E16D1.70902@jp.fujitsu.com> Signed-off-by: Ingo Molnar --- fs/proc/array.c | 4 ++-- include/linux/sched.h | 3 --- kernel/exit.c | 2 +- kernel/sched.c | 33 ++------------------------------- 4 files changed, 5 insertions(+), 37 deletions(-) (limited to 'include/linux/sched.h') diff --git a/fs/proc/array.c b/fs/proc/array.c index 330deda70d08..ca61a88aed66 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -511,7 +511,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, do { min_flt += t->min_flt; maj_flt += t->maj_flt; - gtime = cputime_add(gtime, task_gtime(t)); + gtime = cputime_add(gtime, t->gtime); t = next_thread(t); } while (t != task); @@ -536,7 +536,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, min_flt = task->min_flt; maj_flt = task->maj_flt; task_times(task, &utime, &stime); - gtime = task_gtime(task); + gtime = task->gtime; } /* scale priority and nice values from timeslices to -20..20 */ diff --git a/include/linux/sched.h b/include/linux/sched.h index fe6ae1516640..0395b0f4df3a 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1720,9 +1720,6 @@ static inline void put_task_struct(struct task_struct *t) __put_task_struct(t); } -extern cputime_t task_utime(struct task_struct *p); -extern cputime_t task_stime(struct task_struct *p); -extern cputime_t task_gtime(struct task_struct *p); extern void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st); /* diff --git a/kernel/exit.c b/kernel/exit.c index 29068ab2670a..2eaf68b634e3 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -115,7 +115,7 @@ static void __exit_signal(struct task_struct *tsk) task_times(tsk, &utime, &stime); sig->utime = cputime_add(sig->utime, utime); sig->stime = cputime_add(sig->stime, stime); - sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); + sig->gtime = cputime_add(sig->gtime, tsk->gtime); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; sig->nvcsw += tsk->nvcsw; diff --git a/kernel/sched.c b/kernel/sched.c index 475a6f2b7158..82251c21f785 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5182,22 +5182,12 @@ void account_idle_ticks(unsigned long ticks) * Use precise platform statistics if available: */ #ifdef CONFIG_VIRT_CPU_ACCOUNTING -cputime_t task_utime(struct task_struct *p) -{ - return p->utime; -} - -cputime_t task_stime(struct task_struct *p) -{ - return p->stime; -} - void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { if (ut) - *ut = task_utime(p); + *ut = p->utime; if (st) - *st = task_stime(p); + *st = p->stime; } #else @@ -5235,27 +5225,8 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) if (st) *st = p->prev_stime; } - -cputime_t task_utime(struct task_struct *p) -{ - cputime_t utime; - task_times(p, &utime, NULL); - return utime; -} - -cputime_t task_stime(struct task_struct *p) -{ - cputime_t stime; - task_times(p, NULL, &stime); - return stime; -} #endif -inline cputime_t task_gtime(struct task_struct *p) -{ - return p->gtime; -} - /* * This function gets called by the timer code, with HZ frequency. * We call it with interrupts disabled. -- cgit v1.2.3 From d99ca3b977fc5a93141304f571475c2af9e6c1c5 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Wed, 2 Dec 2009 17:26:47 +0900 Subject: sched, cputime: Cleanups related to task_times() - Remove if({u,s}t)s because no one call it with NULL now. - Use cputime_{add,sub}(). - Add ifndef-endif for prev_{u,s}time since they are used only when !VIRT_CPU_ACCOUNTING. Signed-off-by: Hidetoshi Seto Cc: Peter Zijlstra Cc: Spencer Candland Cc: Americo Wang Cc: Oleg Nesterov Cc: Balbir Singh Cc: Stanislaw Gruszka LKML-Reference: <4B1624C7.7040302@jp.fujitsu.com> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 2 ++ kernel/fork.c | 2 ++ kernel/sched.c | 16 ++++++---------- 3 files changed, 10 insertions(+), 10 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 0395b0f4df3a..dff85e58264e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1331,7 +1331,9 @@ struct task_struct { cputime_t utime, stime, utimescaled, stimescaled; cputime_t gtime; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING cputime_t prev_utime, prev_stime; +#endif unsigned long nvcsw, nivcsw; /* context switch counts */ struct timespec start_time; /* monotonic time */ struct timespec real_start_time; /* boot based time */ diff --git a/kernel/fork.c b/kernel/fork.c index 166b8c49257c..ad7cb6d1193c 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1066,8 +1066,10 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->gtime = cputime_zero; p->utimescaled = cputime_zero; p->stimescaled = cputime_zero; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING p->prev_utime = cputime_zero; p->prev_stime = cputime_zero; +#endif p->default_timer_slack_ns = current->timer_slack_ns; diff --git a/kernel/sched.c b/kernel/sched.c index 4883fee99314..17e2c1db2bde 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5184,10 +5184,8 @@ void account_idle_ticks(unsigned long ticks) #ifdef CONFIG_VIRT_CPU_ACCOUNTING void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - if (ut) - *ut = p->utime; - if (st) - *st = p->stime; + *ut = p->utime; + *st = p->stime; } #else @@ -5197,7 +5195,7 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { - cputime_t rtime, utime = p->utime, total = utime + p->stime; + cputime_t rtime, utime = p->utime, total = cputime_add(utime, p->stime); /* * Use CFS's precise accounting: @@ -5217,12 +5215,10 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) * Compare with previous values, to keep monotonicity: */ p->prev_utime = max(p->prev_utime, utime); - p->prev_stime = max(p->prev_stime, rtime - p->prev_utime); + p->prev_stime = max(p->prev_stime, cputime_sub(rtime, p->prev_utime)); - if (ut) - *ut = p->prev_utime; - if (st) - *st = p->prev_stime; + *ut = p->prev_utime; + *st = p->prev_stime; } #endif -- cgit v1.2.3 From 0cf55e1ec08bb5a22e068309e2d8ba1180ab4239 Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Wed, 2 Dec 2009 17:28:07 +0900 Subject: sched, cputime: Introduce thread_group_times() This is a real fix for problem of utime/stime values decreasing described in the thread: http://lkml.org/lkml/2009/11/3/522 Now cputime is accounted in the following way: - {u,s}time in task_struct are increased every time when the thread is interrupted by a tick (timer interrupt). - When a thread exits, its {u,s}time are added to signal->{u,s}time, after adjusted by task_times(). - When all threads in a thread_group exits, accumulated {u,s}time (and also c{u,s}time) in signal struct are added to c{u,s}time in signal struct of the group's parent. So {u,s}time in task struct are "raw" tick count, while {u,s}time and c{u,s}time in signal struct are "adjusted" values. And accounted values are used by: - task_times(), to get cputime of a thread: This function returns adjusted values that originates from raw {u,s}time and scaled by sum_exec_runtime that accounted by CFS. - thread_group_cputime(), to get cputime of a thread group: This function returns sum of all {u,s}time of living threads in the group, plus {u,s}time in the signal struct that is sum of adjusted cputimes of all exited threads belonged to the group. The problem is the return value of thread_group_cputime(), because it is mixed sum of "raw" value and "adjusted" value: group's {u,s}time = foreach(thread){{u,s}time} + exited({u,s}time) This misbehavior can break {u,s}time monotonicity. Assume that if there is a thread that have raw values greater than adjusted values (e.g. interrupted by 1000Hz ticks 50 times but only runs 45ms) and if it exits, cputime will decrease (e.g. -5ms). To fix this, we could do: group's {u,s}time = foreach(t){task_times(t)} + exited({u,s}time) But task_times() contains hard divisions, so applying it for every thread should be avoided. This patch fixes the above problem in the following way: - Modify thread's exit (= __exit_signal()) not to use task_times(). It means {u,s}time in signal struct accumulates raw values instead of adjusted values. As the result it makes thread_group_cputime() to return pure sum of "raw" values. - Introduce a new function thread_group_times(*task, *utime, *stime) that converts "raw" values of thread_group_cputime() to "adjusted" values, in same calculation procedure as task_times(). - Modify group's exit (= wait_task_zombie()) to use this introduced thread_group_times(). It make c{u,s}time in signal struct to have adjusted values like before this patch. - Replace some thread_group_cputime() by thread_group_times(). This replacements are only applied where conveys the "adjusted" cputime to users, and where already uses task_times() near by it. (i.e. sys_times(), getrusage(), and /proc//stat.) This patch have a positive side effect: - Before this patch, if a group contains many short-life threads (e.g. runs 0.9ms and not interrupted by ticks), the group's cputime could be invisible since thread's cputime was accumulated after adjusted: imagine adjustment function as adj(ticks, runtime), {adj(0, 0.9) + adj(0, 0.9) + ....} = {0 + 0 + ....} = 0. After this patch it will not happen because the adjustment is applied after accumulated. v2: - remove if()s, put new variables into signal_struct. Signed-off-by: Hidetoshi Seto Acked-by: Peter Zijlstra Cc: Spencer Candland Cc: Americo Wang Cc: Oleg Nesterov Cc: Balbir Singh Cc: Stanislaw Gruszka LKML-Reference: <4B162517.8040909@jp.fujitsu.com> Signed-off-by: Ingo Molnar --- fs/proc/array.c | 5 +---- include/linux/sched.h | 4 ++++ kernel/exit.c | 23 ++++++++++++----------- kernel/fork.c | 3 +++ kernel/sched.c | 41 +++++++++++++++++++++++++++++++++++++++++ kernel/sys.c | 18 ++++++++---------- 6 files changed, 69 insertions(+), 25 deletions(-) (limited to 'include/linux/sched.h') diff --git a/fs/proc/array.c b/fs/proc/array.c index ca61a88aed66..2571da43c736 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -506,7 +506,6 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, /* add up live thread stats at the group level */ if (whole) { - struct task_cputime cputime; struct task_struct *t = task; do { min_flt += t->min_flt; @@ -517,9 +516,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, min_flt += sig->min_flt; maj_flt += sig->maj_flt; - thread_group_cputime(task, &cputime); - utime = cputime.utime; - stime = cputime.stime; + thread_group_times(task, &utime, &stime); gtime = cputime_add(gtime, sig->gtime); } diff --git a/include/linux/sched.h b/include/linux/sched.h index dff85e58264e..34238bd10ebf 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -624,6 +624,9 @@ struct signal_struct { cputime_t utime, stime, cutime, cstime; cputime_t gtime; cputime_t cgtime; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING + cputime_t prev_utime, prev_stime; +#endif unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw; unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt; unsigned long inblock, oublock, cinblock, coublock; @@ -1723,6 +1726,7 @@ static inline void put_task_struct(struct task_struct *t) } extern void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st); +extern void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st); /* * Per process flags diff --git a/kernel/exit.c b/kernel/exit.c index 2eaf68b634e3..b221ad65fd20 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -91,8 +91,6 @@ static void __exit_signal(struct task_struct *tsk) if (atomic_dec_and_test(&sig->count)) posix_cpu_timers_exit_group(tsk); else { - cputime_t utime, stime; - /* * If there is any task waiting for the group exit * then notify it: @@ -112,9 +110,8 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ - task_times(tsk, &utime, &stime); - sig->utime = cputime_add(sig->utime, utime); - sig->stime = cputime_add(sig->stime, stime); + sig->utime = cputime_add(sig->utime, tsk->utime); + sig->stime = cputime_add(sig->stime, tsk->stime); sig->gtime = cputime_add(sig->gtime, tsk->gtime); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; @@ -1208,6 +1205,7 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) struct signal_struct *psig; struct signal_struct *sig; unsigned long maxrss; + cputime_t tgutime, tgstime; /* * The resource counters for the group leader are in its @@ -1223,20 +1221,23 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) * need to protect the access to parent->signal fields, * as other threads in the parent group can be right * here reaping other children at the same time. + * + * We use thread_group_times() to get times for the thread + * group, which consolidates times for all threads in the + * group including the group leader. */ + thread_group_times(p, &tgutime, &tgstime); spin_lock_irq(&p->real_parent->sighand->siglock); psig = p->real_parent->signal; sig = p->signal; psig->cutime = cputime_add(psig->cutime, - cputime_add(p->utime, - cputime_add(sig->utime, - sig->cutime))); + cputime_add(tgutime, + sig->cutime)); psig->cstime = cputime_add(psig->cstime, - cputime_add(p->stime, - cputime_add(sig->stime, - sig->cstime))); + cputime_add(tgstime, + sig->cstime)); psig->cgtime = cputime_add(psig->cgtime, cputime_add(p->gtime, diff --git a/kernel/fork.c b/kernel/fork.c index ad7cb6d1193c..3d6f121bbe8a 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -884,6 +884,9 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->utime = sig->stime = sig->cutime = sig->cstime = cputime_zero; sig->gtime = cputime_zero; sig->cgtime = cputime_zero; +#ifndef CONFIG_VIRT_CPU_ACCOUNTING + sig->prev_utime = sig->prev_stime = cputime_zero; +#endif sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0; sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0; sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0; diff --git a/kernel/sched.c b/kernel/sched.c index 17e2c1db2bde..e6ba726941ae 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5187,6 +5187,16 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) *ut = p->utime; *st = p->stime; } + +void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) +{ + struct task_cputime cputime; + + thread_group_cputime(p, &cputime); + + *ut = cputime.utime; + *st = cputime.stime; +} #else #ifndef nsecs_to_cputime @@ -5220,6 +5230,37 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) *ut = p->prev_utime; *st = p->prev_stime; } + +/* + * Must be called with siglock held. + */ +void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) +{ + struct signal_struct *sig = p->signal; + struct task_cputime cputime; + cputime_t rtime, utime, total; + + thread_group_cputime(p, &cputime); + + total = cputime_add(cputime.utime, cputime.stime); + rtime = nsecs_to_cputime(cputime.sum_exec_runtime); + + if (total) { + u64 temp; + + temp = (u64)(rtime * cputime.utime); + do_div(temp, total); + utime = (cputime_t)temp; + } else + utime = rtime; + + sig->prev_utime = max(sig->prev_utime, utime); + sig->prev_stime = max(sig->prev_stime, + cputime_sub(rtime, sig->prev_utime)); + + *ut = sig->prev_utime; + *st = sig->prev_stime; +} #endif /* diff --git a/kernel/sys.c b/kernel/sys.c index bbdfce0d4347..9968c5fb55b9 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -911,16 +911,15 @@ change_okay: void do_sys_times(struct tms *tms) { - struct task_cputime cputime; - cputime_t cutime, cstime; + cputime_t tgutime, tgstime, cutime, cstime; - thread_group_cputime(current, &cputime); spin_lock_irq(¤t->sighand->siglock); + thread_group_times(current, &tgutime, &tgstime); cutime = current->signal->cutime; cstime = current->signal->cstime; spin_unlock_irq(¤t->sighand->siglock); - tms->tms_utime = cputime_to_clock_t(cputime.utime); - tms->tms_stime = cputime_to_clock_t(cputime.stime); + tms->tms_utime = cputime_to_clock_t(tgutime); + tms->tms_stime = cputime_to_clock_t(tgstime); tms->tms_cutime = cputime_to_clock_t(cutime); tms->tms_cstime = cputime_to_clock_t(cstime); } @@ -1338,8 +1337,7 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) { struct task_struct *t; unsigned long flags; - cputime_t utime, stime; - struct task_cputime cputime; + cputime_t tgutime, tgstime, utime, stime; unsigned long maxrss = 0; memset((char *) r, 0, sizeof *r); @@ -1372,9 +1370,9 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r) break; case RUSAGE_SELF: - thread_group_cputime(p, &cputime); - utime = cputime_add(utime, cputime.utime); - stime = cputime_add(stime, cputime.stime); + thread_group_times(p, &tgutime, &tgstime); + utime = cputime_add(utime, tgutime); + stime = cputime_add(stime, tgstime); r->ru_nvcsw += p->signal->nvcsw; r->ru_nivcsw += p->signal->nivcsw; r->ru_minflt += p->signal->min_flt; -- cgit v1.2.3