From a35b6466aabb051568b844e8c63f87a356d3d129 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 8 Aug 2012 21:46:40 +0200 Subject: sched, cgroup: Reduce rq->lock hold times for large cgroup hierarchies Peter Portante reported that for large cgroup hierarchies (and or on large CPU counts) we get immense lock contention on rq->lock and stuff stops working properly. His workload was a ton of processes, each in their own cgroup, everybody idling except for a sporadic wakeup once every so often. It was found that: schedule() idle_balance() load_balance() local_irq_save() double_rq_lock() update_h_load() walk_tg_tree(tg_load_down) tg_load_down() Results in an entire cgroup hierarchy walk under rq->lock for every new-idle balance and since new-idle balance isn't throttled this results in a lot of work while holding the rq->lock. This patch does two things, it removes the work from under rq->lock based on the good principle of race and pray which is widely employed in the load-balancer as a whole. And secondly it throttles the update_h_load() calculation to max once per jiffy. I considered excluding update_h_load() for new-idle balance all-together, but purely relying on regular balance passes to update this data might not work out under some rare circumstances where the new-idle busiest isn't the regular busiest for a while (unlikely, but a nightmare to debug if someone hits it and suffers). Cc: pjt@google.com Cc: Larry Woodman Cc: Mike Galbraith Reported-by: Peter Portante Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-aaarrzfpnaam7pqrekofu8a6@git.kernel.org Signed-off-by: Thomas Gleixner --- kernel/sched/fair.c | 11 +++++++++-- kernel/sched/sched.h | 6 +++++- 2 files changed, 14 insertions(+), 3 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index d0cc03b3e70b..c219bf8d704c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3387,6 +3387,14 @@ static int tg_load_down(struct task_group *tg, void *data) static void update_h_load(long cpu) { + struct rq *rq = cpu_rq(cpu); + unsigned long now = jiffies; + + if (rq->h_load_throttle == now) + return; + + rq->h_load_throttle = now; + rcu_read_lock(); walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); rcu_read_unlock(); @@ -4293,11 +4301,10 @@ redo: env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running); + update_h_load(env.src_cpu); more_balance: local_irq_save(flags); double_rq_lock(this_rq, busiest); - if (!env.loop) - update_h_load(env.src_cpu); /* * cur_ld_moved - load moved in current iteration diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index c35a1a7dd4d6..531411b1044e 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -374,7 +374,11 @@ struct rq { #ifdef CONFIG_FAIR_GROUP_SCHED /* list of leaf cfs_rq on this cpu: */ struct list_head leaf_cfs_rq_list; -#endif +#ifdef CONFIG_SMP + unsigned long h_load_throttle; +#endif /* CONFIG_SMP */ +#endif /* CONFIG_FAIR_GROUP_SCHED */ + #ifdef CONFIG_RT_GROUP_SCHED struct list_head leaf_rt_rq_list; #endif -- cgit v1.2.3 From bea6832cc8c4a0a9a65dd17da6aaa657fe27bc3e Mon Sep 17 00:00:00 2001 From: Stanislaw Gruszka Date: Wed, 8 Aug 2012 11:27:15 +0200 Subject: sched: fix divide by zero at {thread_group,task}_times On architectures where cputime_t is 64 bit type, is possible to trigger divide by zero on do_div(temp, (__force u32) total) line, if total is a non zero number but has lower 32 bit's zeroed. Removing casting is not a good solution since some do_div() implementations do cast to u32 internally. This problem can be triggered in practice on very long lived processes: PID: 2331 TASK: ffff880472814b00 CPU: 2 COMMAND: "oraagent.bin" #0 [ffff880472a51b70] machine_kexec at ffffffff8103214b #1 [ffff880472a51bd0] crash_kexec at ffffffff810b91c2 #2 [ffff880472a51ca0] oops_end at ffffffff814f0b00 #3 [ffff880472a51cd0] die at ffffffff8100f26b #4 [ffff880472a51d00] do_trap at ffffffff814f03f4 #5 [ffff880472a51d60] do_divide_error at ffffffff8100cfff #6 [ffff880472a51e00] divide_error at ffffffff8100be7b [exception RIP: thread_group_times+0x56] RIP: ffffffff81056a16 RSP: ffff880472a51eb8 RFLAGS: 00010046 RAX: bc3572c9fe12d194 RBX: ffff880874150800 RCX: 0000000110266fad RDX: 0000000000000000 RSI: ffff880472a51eb8 RDI: 001038ae7d9633dc RBP: ffff880472a51ef8 R8: 00000000b10a3a64 R9: ffff880874150800 R10: 00007fcba27ab680 R11: 0000000000000202 R12: ffff880472a51f08 R13: ffff880472a51f10 R14: 0000000000000000 R15: 0000000000000007 ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 #7 [ffff880472a51f00] do_sys_times at ffffffff8108845d #8 [ffff880472a51f40] sys_times at ffffffff81088524 #9 [ffff880472a51f80] system_call_fastpath at ffffffff8100b0f2 RIP: 0000003808caac3a RSP: 00007fcba27ab6d8 RFLAGS: 00000202 RAX: 0000000000000064 RBX: ffffffff8100b0f2 RCX: 0000000000000000 RDX: 00007fcba27ab6e0 RSI: 000000000076d58e RDI: 00007fcba27ab6e0 RBP: 00007fcba27ab700 R8: 0000000000000020 R9: 000000000000091b R10: 00007fcba27ab680 R11: 0000000000000202 R12: 00007fff9ca41940 R13: 0000000000000000 R14: 00007fcba27ac9c0 R15: 00007fff9ca41940 ORIG_RAX: 0000000000000064 CS: 0033 SS: 002b Cc: stable@vger.kernel.org Signed-off-by: Stanislaw Gruszka Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20120808092714.GA3580@redhat.com Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2cb4e7777998..e2c5841c7dfe 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3142,6 +3142,20 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) # define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) #endif +static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total) +{ + u64 temp = (__force u64) rtime; + + temp *= (__force u64) utime; + + if (sizeof(cputime_t) == 4) + temp = div_u64(temp, (__force u32) total); + else + temp = div64_u64(temp, (__force u64) total); + + return (__force cputime_t) temp; +} + void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) { cputime_t rtime, utime = p->utime, total = utime + p->stime; @@ -3151,13 +3165,9 @@ void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) */ rtime = nsecs_to_cputime(p->se.sum_exec_runtime); - if (total) { - u64 temp = (__force u64) rtime; - - temp *= (__force u64) utime; - do_div(temp, (__force u32) total); - utime = (__force cputime_t) temp; - } else + if (total) + utime = scale_utime(utime, rtime, total); + else utime = rtime; /* @@ -3184,13 +3194,9 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) total = cputime.utime + cputime.stime; rtime = nsecs_to_cputime(cputime.sum_exec_runtime); - if (total) { - u64 temp = (__force u64) rtime; - - temp *= (__force u64) cputime.utime; - do_div(temp, (__force u32) total); - utime = (__force cputime_t) temp; - } else + if (total) + utime = scale_utime(cputime.utime, rtime, total); + else utime = rtime; sig->prev_utime = max(sig->prev_utime, utime); -- cgit v1.2.3 From 35cf4e50b16331def6cfcbee11e49270b6db07f5 Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Tue, 7 Aug 2012 05:00:13 +0200 Subject: sched,cgroup: Fix up task_groups list With multiple instances of task_groups, for_each_rt_rq() is a noop, no task groups having been added to the rt.c list instance. This renders __enable/disable_runtime() and print_rt_stats() noop, the user (non) visible effect being that rt task groups are missing in /proc/sched_debug. Signed-off-by: Mike Galbraith Cc: stable@kernel.org # v3.3+ Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344308413.6846.7.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 1 + kernel/sched/sched.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e2c5841c7dfe..6aa212f3c581 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7252,6 +7252,7 @@ int in_sched_functions(unsigned long addr) #ifdef CONFIG_CGROUP_SCHED struct task_group root_task_group; +LIST_HEAD(task_groups); #endif DECLARE_PER_CPU(cpumask_var_t, load_balance_tmpmask); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 531411b1044e..f6714d009e77 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -80,7 +80,7 @@ extern struct mutex sched_domains_mutex; struct cfs_rq; struct rt_rq; -static LIST_HEAD(task_groups); +extern struct list_head task_groups; struct cfs_bandwidth { #ifdef CONFIG_CFS_BANDWIDTH -- cgit v1.2.3 From e221d028bb08b47e624c5f0a31732c642db9d19a Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Tue, 7 Aug 2012 10:02:38 +0200 Subject: sched,rt: fix isolated CPUs leaving root_task_group indefinitely throttled Root task group bandwidth replenishment must service all CPUs, regardless of where the timer was last started, and regardless of the isolation mechanism, lest 'Quoth the Raven, "Nevermore"' become rt scheduling policy. Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344326558.6968.25.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/rt.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 573e1ca01102..944cb68420e9 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -788,6 +788,19 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) const struct cpumask *span; span = sched_rt_period_mask(); +#ifdef CONFIG_RT_GROUP_SCHED + /* + * FIXME: isolated CPUs should really leave the root task group, + * whether they are isolcpus or were isolated via cpusets, lest + * the timer run on a CPU which does not service all runqueues, + * potentially leaving other CPUs indefinitely throttled. If + * isolation is really required, the user will turn the throttle + * off to kill the perturbations it causes anyway. Meanwhile, + * this maintains functionality for boot and/or troubleshooting. + */ + if (rt_b == &root_task_group.rt_bandwidth) + span = cpu_online_mask; +#endif for_each_cpu(i, span) { int enqueue = 0; struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i); -- cgit v1.2.3 From 8f6189684eb4e85e6c593cd710693f09c944450a Mon Sep 17 00:00:00 2001 From: Mike Galbraith Date: Sat, 4 Aug 2012 05:44:14 +0200 Subject: sched: Fix migration thread runtime bogosity Make stop scheduler class do the same accounting as other classes, Migration threads can be caught in the act while doing exec balancing, leading to the below due to use of unmaintained ->se.exec_start. The load that triggered this particular instance was an apparently out of control heavily threaded application that does system monitoring in what equated to an exec bomb, with one of the VERY frequently migrated tasks being ps. %CPU PID USER CMD 99.3 45 root [migration/10] 97.7 53 root [migration/12] 97.0 57 root [migration/13] 90.1 49 root [migration/11] 89.6 65 root [migration/15] 88.7 17 root [migration/3] 80.4 37 root [migration/8] 78.1 41 root [migration/9] 44.2 13 root [migration/2] Signed-off-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344051854.6739.19.camel@marge.simpson.net Signed-off-by: Thomas Gleixner --- kernel/sched/stop_task.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c index 7b386e86fd23..da5eb5bed84a 100644 --- a/kernel/sched/stop_task.c +++ b/kernel/sched/stop_task.c @@ -27,8 +27,10 @@ static struct task_struct *pick_next_task_stop(struct rq *rq) { struct task_struct *stop = rq->stop; - if (stop && stop->on_rq) + if (stop && stop->on_rq) { + stop->se.exec_start = rq->clock_task; return stop; + } return NULL; } @@ -52,6 +54,21 @@ static void yield_task_stop(struct rq *rq) static void put_prev_task_stop(struct rq *rq, struct task_struct *prev) { + struct task_struct *curr = rq->curr; + u64 delta_exec; + + delta_exec = rq->clock_task - curr->se.exec_start; + if (unlikely((s64)delta_exec < 0)) + delta_exec = 0; + + schedstat_set(curr->se.statistics.exec_max, + max(curr->se.statistics.exec_max, delta_exec)); + + curr->se.sum_exec_runtime += delta_exec; + account_group_exec_runtime(curr, delta_exec); + + curr->se.exec_start = rq->clock_task; + cpuacct_charge(curr, delta_exec); } static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) @@ -60,6 +77,9 @@ static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) static void set_curr_task_stop(struct rq *rq) { + struct task_struct *stop = rq->stop; + + stop->se.exec_start = rq->clock_task; } static void switched_to_stop(struct rq *rq, struct task_struct *p) -- cgit v1.2.3 From 532b1858c5241bedfff5ab863d7cf012e8b81a6b Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 8 Aug 2012 16:16:04 +0200 Subject: sched: Fix __sched_period comment It should be sched_nr_latency so fix it before it annoys me more. Signed-off-by: Borislav Petkov Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344435364-18632-1-git-send-email-bp@amd64.org Signed-off-by: Thomas Gleixner --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c219bf8d704c..99285a85e210 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -597,7 +597,7 @@ calc_delta_fair(unsigned long delta, struct sched_entity *se) /* * The idea is to set a period in which each task runs once. * - * When there are too many tasks (sysctl_sched_nr_latency) we have to stretch + * When there are too many tasks (sched_nr_latency) we have to stretch * this period because otherwise the slices get too small. * * p = (nr <= nl) ? l : l*nr/nl -- cgit v1.2.3 From edde96eafc91a510f404e7b82cfc0ecb608505ee Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Sat, 4 Aug 2012 11:49:47 +0300 Subject: sched: Document schedule() entry points This patch adds a comment on top of the schedule() function to explain to scheduler newbies how the main scheduler function is entered. Acked-by: Randy Dunlap Explained-by: Ingo Molnar Explained-by: Peter Zijlstra Signed-off-by: Pekka Enberg Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1344070187-2420-1-git-send-email-penberg@kernel.org Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 34 ++++++++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index fbf1fd098dc6..c9a3655e572d 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3367,6 +3367,40 @@ pick_next_task(struct rq *rq) /* * __schedule() is the main scheduler function. + * + * The main means of driving the scheduler and thus entering this function are: + * + * 1. Explicit blocking: mutex, semaphore, waitqueue, etc. + * + * 2. TIF_NEED_RESCHED flag is checked on interrupt and userspace return + * paths. For example, see arch/x86/entry_64.S. + * + * To drive preemption between tasks, the scheduler sets the flag in timer + * interrupt handler scheduler_tick(). + * + * 3. Wakeups don't really cause entry into schedule(). They add a + * task to the run-queue and that's it. + * + * Now, if the new task added to the run-queue preempts the current + * task, then the wakeup sets TIF_NEED_RESCHED and schedule() gets + * called on the nearest possible occasion: + * + * - If the kernel is preemptible (CONFIG_PREEMPT=y): + * + * - in syscall or exception context, at the next outmost + * preempt_enable(). (this might be as soon as the wake_up()'s + * spin_unlock()!) + * + * - in IRQ context, return from interrupt-handler to + * preemptible context + * + * - If the kernel is not preemptible (CONFIG_PREEMPT is not set) + * then at the next: + * + * - cond_resched() call + * - explicit schedule() call + * - return from syscall or exception to user-space + * - return from interrupt-handler to user-space */ static void __sched __schedule(void) { -- cgit v1.2.3 From 78feefc512a09165627dd534111f651b6c8e605f Mon Sep 17 00:00:00 2001 From: Michael Wang Date: Mon, 6 Aug 2012 16:41:59 +0800 Subject: sched: using dst_rq instead of this_rq during load balance As we already have dst_rq in lb_env, using or changing "this_rq" do not make sense. This patch will replace "this_rq" with dst_rq in load_balance, and we don't need to change "this_rq" while process LBF_SOME_PINNED any more. Signed-off-by: Michael Wang Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/501F8357.3070102@linux.vnet.ibm.com Signed-off-by: Thomas Gleixner --- kernel/sched/fair.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 99285a85e210..287bfaca6420 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4283,7 +4283,7 @@ redo: goto out_balanced; } - BUG_ON(busiest == this_rq); + BUG_ON(busiest == env.dst_rq); schedstat_add(sd, lb_imbalance[idle], env.imbalance); @@ -4304,7 +4304,7 @@ redo: update_h_load(env.src_cpu); more_balance: local_irq_save(flags); - double_rq_lock(this_rq, busiest); + double_rq_lock(env.dst_rq, busiest); /* * cur_ld_moved - load moved in current iteration @@ -4312,7 +4312,7 @@ more_balance: */ cur_ld_moved = move_tasks(&env); ld_moved += cur_ld_moved; - double_rq_unlock(this_rq, busiest); + double_rq_unlock(env.dst_rq, busiest); local_irq_restore(flags); if (env.flags & LBF_NEED_BREAK) { @@ -4348,8 +4348,7 @@ more_balance: if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0 && lb_iterations++ < max_lb_iterations) { - this_rq = cpu_rq(env.new_dst_cpu); - env.dst_rq = this_rq; + env.dst_rq = cpu_rq(env.new_dst_cpu); env.dst_cpu = env.new_dst_cpu; env.flags &= ~LBF_SOME_PINNED; env.loop = 0; -- cgit v1.2.3 From f03542a7019c600163ac4441d8a826c92c1bd510 Mon Sep 17 00:00:00 2001 From: Alex Shi Date: Thu, 26 Jul 2012 08:55:34 +0800 Subject: sched: recover SD_WAKE_AFFINE in select_task_rq_fair and code clean up Since power saving code was removed from sched now, the implement code is out of service in this function, and even pollute other logical. like, 'want_sd' never has chance to be set '0', that remove the effect of SD_WAKE_AFFINE here. So, clean up the obsolete code, includes SD_PREFER_LOCAL. Signed-off-by: Alex Shi Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/5028F431.6000306@intel.com Signed-off-by: Thomas Gleixner --- kernel/sched/core.c | 1 - kernel/sched/fair.c | 34 +++------------------------------- 2 files changed, 3 insertions(+), 32 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c9a3655e572d..4376c9f34790 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6622,7 +6622,6 @@ sd_numa_init(struct sched_domain_topology_level *tl, int cpu) | 0*SD_BALANCE_FORK | 0*SD_BALANCE_WAKE | 0*SD_WAKE_AFFINE - | 0*SD_PREFER_LOCAL | 0*SD_SHARE_CPUPOWER | 0*SD_SHARE_PKG_RESOURCES | 1*SD_SERIALIZE diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 287bfaca6420..01d3eda6b7f9 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2686,7 +2686,6 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) int prev_cpu = task_cpu(p); int new_cpu = cpu; int want_affine = 0; - int want_sd = 1; int sync = wake_flags & WF_SYNC; if (p->nr_cpus_allowed == 1) @@ -2703,27 +2702,6 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) if (!(tmp->flags & SD_LOAD_BALANCE)) continue; - /* - * If power savings logic is enabled for a domain, see if we - * are not overloaded, if so, don't balance wider. - */ - if (tmp->flags & (SD_PREFER_LOCAL)) { - unsigned long power = 0; - unsigned long nr_running = 0; - unsigned long capacity; - int i; - - for_each_cpu(i, sched_domain_span(tmp)) { - power += power_of(i); - nr_running += cpu_rq(i)->cfs.nr_running; - } - - capacity = DIV_ROUND_CLOSEST(power, SCHED_POWER_SCALE); - - if (nr_running < capacity) - want_sd = 0; - } - /* * If both cpu and prev_cpu are part of this domain, * cpu is a valid SD_WAKE_AFFINE target. @@ -2731,21 +2709,15 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) if (want_affine && (tmp->flags & SD_WAKE_AFFINE) && cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) { affine_sd = tmp; - want_affine = 0; - } - - if (!want_sd && !want_affine) break; + } - if (!(tmp->flags & sd_flag)) - continue; - - if (want_sd) + if (tmp->flags & sd_flag) sd = tmp; } if (affine_sd) { - if (cpu == prev_cpu || wake_affine(affine_sd, p, sync)) + if (cpu != prev_cpu && wake_affine(affine_sd, p, sync)) prev_cpu = cpu; new_cpu = select_idle_sibling(p, prev_cpu); -- cgit v1.2.3 From 73fbec604432e1fbfeb1bc59a110dac1f98160f6 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 16 Jun 2012 15:57:37 +0200 Subject: sched: Move cputime code to its own file Extract cputime code from the giant sched/core.c and put it in its own file. This make it easier to deal with this particular area and de-bloat a bit more core.c Signed-off-by: Frederic Weisbecker Acked-by: Martin Schwidefsky Cc: Tony Luck Cc: Fenghua Yu Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Heiko Carstens Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Peter Zijlstra --- kernel/sched/Makefile | 2 +- kernel/sched/core.c | 557 +------------------------------------------------ kernel/sched/cputime.c | 504 ++++++++++++++++++++++++++++++++++++++++++++ kernel/sched/sched.h | 63 ++++++ 4 files changed, 570 insertions(+), 556 deletions(-) create mode 100644 kernel/sched/cputime.c (limited to 'kernel/sched') diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile index 173ea52f3af0..f06d249e103b 100644 --- a/kernel/sched/Makefile +++ b/kernel/sched/Makefile @@ -11,7 +11,7 @@ ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer endif -obj-y += core.o clock.o idle_task.o fair.o rt.o stop_task.o +obj-y += core.o clock.o cputime.o idle_task.o fair.o rt.o stop_task.o obj-$(CONFIG_SMP) += cpupri.o obj-$(CONFIG_SCHED_AUTOGROUP) += auto_group.o obj-$(CONFIG_SCHEDSTATS) += stats.o diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4376c9f34790..ae3bcaa3afbf 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -740,126 +740,6 @@ void deactivate_task(struct rq *rq, struct task_struct *p, int flags) dequeue_task(rq, p, flags); } -#ifdef CONFIG_IRQ_TIME_ACCOUNTING - -/* - * There are no locks covering percpu hardirq/softirq time. - * They are only modified in account_system_vtime, on corresponding CPU - * with interrupts disabled. So, writes are safe. - * They are read and saved off onto struct rq in update_rq_clock(). - * This may result in other CPU reading this CPU's irq time and can - * race with irq/account_system_vtime on this CPU. We would either get old - * or new value with a side effect of accounting a slice of irq time to wrong - * task when irq is in progress while we read rq->clock. That is a worthy - * compromise in place of having locks on each irq in account_system_time. - */ -static DEFINE_PER_CPU(u64, cpu_hardirq_time); -static DEFINE_PER_CPU(u64, cpu_softirq_time); - -static DEFINE_PER_CPU(u64, irq_start_time); -static int sched_clock_irqtime; - -void enable_sched_clock_irqtime(void) -{ - sched_clock_irqtime = 1; -} - -void disable_sched_clock_irqtime(void) -{ - sched_clock_irqtime = 0; -} - -#ifndef CONFIG_64BIT -static DEFINE_PER_CPU(seqcount_t, irq_time_seq); - -static inline void irq_time_write_begin(void) -{ - __this_cpu_inc(irq_time_seq.sequence); - smp_wmb(); -} - -static inline void irq_time_write_end(void) -{ - smp_wmb(); - __this_cpu_inc(irq_time_seq.sequence); -} - -static inline u64 irq_time_read(int cpu) -{ - u64 irq_time; - unsigned seq; - - do { - seq = read_seqcount_begin(&per_cpu(irq_time_seq, cpu)); - irq_time = per_cpu(cpu_softirq_time, cpu) + - per_cpu(cpu_hardirq_time, cpu); - } while (read_seqcount_retry(&per_cpu(irq_time_seq, cpu), seq)); - - return irq_time; -} -#else /* CONFIG_64BIT */ -static inline void irq_time_write_begin(void) -{ -} - -static inline void irq_time_write_end(void) -{ -} - -static inline u64 irq_time_read(int cpu) -{ - return per_cpu(cpu_softirq_time, cpu) + per_cpu(cpu_hardirq_time, cpu); -} -#endif /* CONFIG_64BIT */ - -/* - * Called before incrementing preempt_count on {soft,}irq_enter - * and before decrementing preempt_count on {soft,}irq_exit. - */ -void account_system_vtime(struct task_struct *curr) -{ - unsigned long flags; - s64 delta; - int cpu; - - if (!sched_clock_irqtime) - return; - - local_irq_save(flags); - - cpu = smp_processor_id(); - delta = sched_clock_cpu(cpu) - __this_cpu_read(irq_start_time); - __this_cpu_add(irq_start_time, delta); - - irq_time_write_begin(); - /* - * We do not account for softirq time from ksoftirqd here. - * We want to continue accounting softirq time to ksoftirqd thread - * in that case, so as not to confuse scheduler with a special task - * that do not consume any time, but still wants to run. - */ - if (hardirq_count()) - __this_cpu_add(cpu_hardirq_time, delta); - else if (in_serving_softirq() && curr != this_cpu_ksoftirqd()) - __this_cpu_add(cpu_softirq_time, delta); - - irq_time_write_end(); - local_irq_restore(flags); -} -EXPORT_SYMBOL_GPL(account_system_vtime); - -#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ - -#ifdef CONFIG_PARAVIRT -static inline u64 steal_ticks(u64 steal) -{ - if (unlikely(steal > NSEC_PER_SEC)) - return div_u64(steal, TICK_NSEC); - - return __iter_div_u64_rem(steal, TICK_NSEC, &steal); -} -#endif - static void update_rq_clock_task(struct rq *rq, s64 delta) { /* @@ -920,43 +800,6 @@ static void update_rq_clock_task(struct rq *rq, s64 delta) #endif } -#ifdef CONFIG_IRQ_TIME_ACCOUNTING -static int irqtime_account_hi_update(void) -{ - u64 *cpustat = kcpustat_this_cpu->cpustat; - unsigned long flags; - u64 latest_ns; - int ret = 0; - - local_irq_save(flags); - latest_ns = this_cpu_read(cpu_hardirq_time); - if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_IRQ]) - ret = 1; - local_irq_restore(flags); - return ret; -} - -static int irqtime_account_si_update(void) -{ - u64 *cpustat = kcpustat_this_cpu->cpustat; - unsigned long flags; - u64 latest_ns; - int ret = 0; - - local_irq_save(flags); - latest_ns = this_cpu_read(cpu_softirq_time); - if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_SOFTIRQ]) - ret = 1; - local_irq_restore(flags); - return ret; -} - -#else /* CONFIG_IRQ_TIME_ACCOUNTING */ - -#define sched_clock_irqtime (0) - -#endif - void sched_set_stop_task(int cpu, struct task_struct *stop) { struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; @@ -2809,404 +2652,6 @@ unsigned long long task_sched_runtime(struct task_struct *p) return ns; } -#ifdef CONFIG_CGROUP_CPUACCT -struct cgroup_subsys cpuacct_subsys; -struct cpuacct root_cpuacct; -#endif - -static inline void task_group_account_field(struct task_struct *p, int index, - u64 tmp) -{ -#ifdef CONFIG_CGROUP_CPUACCT - struct kernel_cpustat *kcpustat; - struct cpuacct *ca; -#endif - /* - * Since all updates are sure to touch the root cgroup, we - * get ourselves ahead and touch it first. If the root cgroup - * is the only cgroup, then nothing else should be necessary. - * - */ - __get_cpu_var(kernel_cpustat).cpustat[index] += tmp; - -#ifdef CONFIG_CGROUP_CPUACCT - if (unlikely(!cpuacct_subsys.active)) - return; - - rcu_read_lock(); - ca = task_ca(p); - while (ca && (ca != &root_cpuacct)) { - kcpustat = this_cpu_ptr(ca->cpustat); - kcpustat->cpustat[index] += tmp; - ca = parent_ca(ca); - } - rcu_read_unlock(); -#endif -} - - -/* - * Account user cpu time to a process. - * @p: the process that the cpu time gets accounted to - * @cputime: the cpu time spent in user space since the last update - * @cputime_scaled: cputime scaled by cpu frequency - */ -void account_user_time(struct task_struct *p, cputime_t cputime, - cputime_t cputime_scaled) -{ - int index; - - /* Add user time to process. */ - p->utime += cputime; - p->utimescaled += cputime_scaled; - account_group_user_time(p, cputime); - - index = (TASK_NICE(p) > 0) ? CPUTIME_NICE : CPUTIME_USER; - - /* Add user time to cpustat. */ - task_group_account_field(p, index, (__force u64) cputime); - - /* Account for user time used */ - acct_update_integrals(p); -} - -/* - * Account guest cpu time to a process. - * @p: the process that the cpu time gets accounted to - * @cputime: the cpu time spent in virtual machine since the last update - * @cputime_scaled: cputime scaled by cpu frequency - */ -static void account_guest_time(struct task_struct *p, cputime_t cputime, - cputime_t cputime_scaled) -{ - u64 *cpustat = kcpustat_this_cpu->cpustat; - - /* Add guest time to process. */ - p->utime += cputime; - p->utimescaled += cputime_scaled; - account_group_user_time(p, cputime); - p->gtime += cputime; - - /* Add guest time to cpustat. */ - if (TASK_NICE(p) > 0) { - cpustat[CPUTIME_NICE] += (__force u64) cputime; - cpustat[CPUTIME_GUEST_NICE] += (__force u64) cputime; - } else { - cpustat[CPUTIME_USER] += (__force u64) cputime; - cpustat[CPUTIME_GUEST] += (__force u64) cputime; - } -} - -/* - * Account system cpu time to a process and desired cpustat field - * @p: the process that the cpu time gets accounted to - * @cputime: the cpu time spent in kernel space since the last update - * @cputime_scaled: cputime scaled by cpu frequency - * @target_cputime64: pointer to cpustat field that has to be updated - */ -static inline -void __account_system_time(struct task_struct *p, cputime_t cputime, - cputime_t cputime_scaled, int index) -{ - /* Add system time to process. */ - p->stime += cputime; - p->stimescaled += cputime_scaled; - account_group_system_time(p, cputime); - - /* Add system time to cpustat. */ - task_group_account_field(p, index, (__force u64) cputime); - - /* Account for system time used */ - acct_update_integrals(p); -} - -/* - * Account system cpu time to a process. - * @p: the process that the cpu time gets accounted to - * @hardirq_offset: the offset to subtract from hardirq_count() - * @cputime: the cpu time spent in kernel space since the last update - * @cputime_scaled: cputime scaled by cpu frequency - */ -void account_system_time(struct task_struct *p, int hardirq_offset, - cputime_t cputime, cputime_t cputime_scaled) -{ - int index; - - if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) { - account_guest_time(p, cputime, cputime_scaled); - return; - } - - if (hardirq_count() - hardirq_offset) - index = CPUTIME_IRQ; - else if (in_serving_softirq()) - index = CPUTIME_SOFTIRQ; - else - index = CPUTIME_SYSTEM; - - __account_system_time(p, cputime, cputime_scaled, index); -} - -/* - * Account for involuntary wait time. - * @cputime: the cpu time spent in involuntary wait - */ -void account_steal_time(cputime_t cputime) -{ - u64 *cpustat = kcpustat_this_cpu->cpustat; - - cpustat[CPUTIME_STEAL] += (__force u64) cputime; -} - -/* - * Account for idle time. - * @cputime: the cpu time spent in idle wait - */ -void account_idle_time(cputime_t cputime) -{ - u64 *cpustat = kcpustat_this_cpu->cpustat; - struct rq *rq = this_rq(); - - if (atomic_read(&rq->nr_iowait) > 0) - cpustat[CPUTIME_IOWAIT] += (__force u64) cputime; - else - cpustat[CPUTIME_IDLE] += (__force u64) cputime; -} - -static __always_inline bool steal_account_process_tick(void) -{ -#ifdef CONFIG_PARAVIRT - if (static_key_false(¶virt_steal_enabled)) { - u64 steal, st = 0; - - steal = paravirt_steal_clock(smp_processor_id()); - steal -= this_rq()->prev_steal_time; - - st = steal_ticks(steal); - this_rq()->prev_steal_time += st * TICK_NSEC; - - account_steal_time(st); - return st; - } -#endif - return false; -} - -#ifndef CONFIG_VIRT_CPU_ACCOUNTING - -#ifdef CONFIG_IRQ_TIME_ACCOUNTING -/* - * Account a tick to a process and cpustat - * @p: the process that the cpu time gets accounted to - * @user_tick: is the tick from userspace - * @rq: the pointer to rq - * - * Tick demultiplexing follows the order - * - pending hardirq update - * - pending softirq update - * - user_time - * - idle_time - * - system time - * - check for guest_time - * - else account as system_time - * - * Check for hardirq is done both for system and user time as there is - * no timer going off while we are on hardirq and hence we may never get an - * opportunity to update it solely in system time. - * p->stime and friends are only updated on system time and not on irq - * softirq as those do not count in task exec_runtime any more. - */ -static void irqtime_account_process_tick(struct task_struct *p, int user_tick, - struct rq *rq) -{ - cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); - u64 *cpustat = kcpustat_this_cpu->cpustat; - - if (steal_account_process_tick()) - return; - - if (irqtime_account_hi_update()) { - cpustat[CPUTIME_IRQ] += (__force u64) cputime_one_jiffy; - } else if (irqtime_account_si_update()) { - cpustat[CPUTIME_SOFTIRQ] += (__force u64) cputime_one_jiffy; - } else if (this_cpu_ksoftirqd() == p) { - /* - * ksoftirqd time do not get accounted in cpu_softirq_time. - * So, we have to handle it separately here. - * Also, p->stime needs to be updated for ksoftirqd. - */ - __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, - CPUTIME_SOFTIRQ); - } else if (user_tick) { - account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); - } else if (p == rq->idle) { - account_idle_time(cputime_one_jiffy); - } else if (p->flags & PF_VCPU) { /* System time or guest time */ - account_guest_time(p, cputime_one_jiffy, one_jiffy_scaled); - } else { - __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, - CPUTIME_SYSTEM); - } -} - -static void irqtime_account_idle_ticks(int ticks) -{ - int i; - struct rq *rq = this_rq(); - - for (i = 0; i < ticks; i++) - irqtime_account_process_tick(current, 0, rq); -} -#else /* CONFIG_IRQ_TIME_ACCOUNTING */ -static void irqtime_account_idle_ticks(int ticks) {} -static void irqtime_account_process_tick(struct task_struct *p, int user_tick, - struct rq *rq) {} -#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ - -/* - * Account a single tick of cpu time. - * @p: the process that the cpu time gets accounted to - * @user_tick: indicates if the tick is a user or a system tick - */ -void account_process_tick(struct task_struct *p, int user_tick) -{ - cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); - struct rq *rq = this_rq(); - - if (sched_clock_irqtime) { - irqtime_account_process_tick(p, user_tick, rq); - return; - } - - if (steal_account_process_tick()) - return; - - if (user_tick) - account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); - else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET)) - account_system_time(p, HARDIRQ_OFFSET, cputime_one_jiffy, - one_jiffy_scaled); - else - account_idle_time(cputime_one_jiffy); -} - -/* - * Account multiple ticks of steal time. - * @p: the process from which the cpu time has been stolen - * @ticks: number of stolen ticks - */ -void account_steal_ticks(unsigned long ticks) -{ - account_steal_time(jiffies_to_cputime(ticks)); -} - -/* - * Account multiple ticks of idle time. - * @ticks: number of stolen ticks - */ -void account_idle_ticks(unsigned long ticks) -{ - - if (sched_clock_irqtime) { - irqtime_account_idle_ticks(ticks); - return; - } - - account_idle_time(jiffies_to_cputime(ticks)); -} - -#endif - -/* - * Use precise platform statistics if available: - */ -#ifdef CONFIG_VIRT_CPU_ACCOUNTING -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 -# define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) -#endif - -static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total) -{ - u64 temp = (__force u64) rtime; - - temp *= (__force u64) utime; - - if (sizeof(cputime_t) == 4) - temp = div_u64(temp, (__force u32) total); - else - temp = div64_u64(temp, (__force u64) total); - - return (__force cputime_t) temp; -} - -void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) -{ - cputime_t rtime, utime = p->utime, total = utime + p->stime; - - /* - * Use CFS's precise accounting: - */ - rtime = nsecs_to_cputime(p->se.sum_exec_runtime); - - if (total) - utime = scale_utime(utime, rtime, total); - else - utime = rtime; - - /* - * 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); - - *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.utime + cputime.stime; - rtime = nsecs_to_cputime(cputime.sum_exec_runtime); - - if (total) - utime = scale_utime(cputime.utime, rtime, total); - else - utime = rtime; - - sig->prev_utime = max(sig->prev_utime, utime); - sig->prev_stime = max(sig->prev_stime, rtime - sig->prev_utime); - - *ut = sig->prev_utime; - *st = sig->prev_stime; -} -#endif - /* * This function gets called by the timer code, with HZ frequency. * We call it with interrupts disabled. @@ -8419,6 +7864,8 @@ struct cgroup_subsys cpu_cgroup_subsys = { * (balbir@in.ibm.com). */ +struct cpuacct root_cpuacct; + /* create a new cpu accounting group */ static struct cgroup_subsys_state *cpuacct_create(struct cgroup *cgrp) { diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c new file mode 100644 index 000000000000..372692bd5376 --- /dev/null +++ b/kernel/sched/cputime.c @@ -0,0 +1,504 @@ +#include +#include +#include +#include +#include +#include "sched.h" + + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + +/* + * There are no locks covering percpu hardirq/softirq time. + * They are only modified in account_system_vtime, on corresponding CPU + * with interrupts disabled. So, writes are safe. + * They are read and saved off onto struct rq in update_rq_clock(). + * This may result in other CPU reading this CPU's irq time and can + * race with irq/account_system_vtime on this CPU. We would either get old + * or new value with a side effect of accounting a slice of irq time to wrong + * task when irq is in progress while we read rq->clock. That is a worthy + * compromise in place of having locks on each irq in account_system_time. + */ +DEFINE_PER_CPU(u64, cpu_hardirq_time); +DEFINE_PER_CPU(u64, cpu_softirq_time); + +static DEFINE_PER_CPU(u64, irq_start_time); +static int sched_clock_irqtime; + +void enable_sched_clock_irqtime(void) +{ + sched_clock_irqtime = 1; +} + +void disable_sched_clock_irqtime(void) +{ + sched_clock_irqtime = 0; +} + +#ifndef CONFIG_64BIT +DEFINE_PER_CPU(seqcount_t, irq_time_seq); +#endif /* CONFIG_64BIT */ + +/* + * Called before incrementing preempt_count on {soft,}irq_enter + * and before decrementing preempt_count on {soft,}irq_exit. + */ +void account_system_vtime(struct task_struct *curr) +{ + unsigned long flags; + s64 delta; + int cpu; + + if (!sched_clock_irqtime) + return; + + local_irq_save(flags); + + cpu = smp_processor_id(); + delta = sched_clock_cpu(cpu) - __this_cpu_read(irq_start_time); + __this_cpu_add(irq_start_time, delta); + + irq_time_write_begin(); + /* + * We do not account for softirq time from ksoftirqd here. + * We want to continue accounting softirq time to ksoftirqd thread + * in that case, so as not to confuse scheduler with a special task + * that do not consume any time, but still wants to run. + */ + if (hardirq_count()) + __this_cpu_add(cpu_hardirq_time, delta); + else if (in_serving_softirq() && curr != this_cpu_ksoftirqd()) + __this_cpu_add(cpu_softirq_time, delta); + + irq_time_write_end(); + local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(account_system_vtime); + +static int irqtime_account_hi_update(void) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + unsigned long flags; + u64 latest_ns; + int ret = 0; + + local_irq_save(flags); + latest_ns = this_cpu_read(cpu_hardirq_time); + if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_IRQ]) + ret = 1; + local_irq_restore(flags); + return ret; +} + +static int irqtime_account_si_update(void) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + unsigned long flags; + u64 latest_ns; + int ret = 0; + + local_irq_save(flags); + latest_ns = this_cpu_read(cpu_softirq_time); + if (nsecs_to_cputime64(latest_ns) > cpustat[CPUTIME_SOFTIRQ]) + ret = 1; + local_irq_restore(flags); + return ret; +} + +#else /* CONFIG_IRQ_TIME_ACCOUNTING */ + +#define sched_clock_irqtime (0) + +#endif /* !CONFIG_IRQ_TIME_ACCOUNTING */ + +static inline void task_group_account_field(struct task_struct *p, int index, + u64 tmp) +{ +#ifdef CONFIG_CGROUP_CPUACCT + struct kernel_cpustat *kcpustat; + struct cpuacct *ca; +#endif + /* + * Since all updates are sure to touch the root cgroup, we + * get ourselves ahead and touch it first. If the root cgroup + * is the only cgroup, then nothing else should be necessary. + * + */ + __get_cpu_var(kernel_cpustat).cpustat[index] += tmp; + +#ifdef CONFIG_CGROUP_CPUACCT + if (unlikely(!cpuacct_subsys.active)) + return; + + rcu_read_lock(); + ca = task_ca(p); + while (ca && (ca != &root_cpuacct)) { + kcpustat = this_cpu_ptr(ca->cpustat); + kcpustat->cpustat[index] += tmp; + ca = parent_ca(ca); + } + rcu_read_unlock(); +#endif +} + +/* + * Account user cpu time to a process. + * @p: the process that the cpu time gets accounted to + * @cputime: the cpu time spent in user space since the last update + * @cputime_scaled: cputime scaled by cpu frequency + */ +void account_user_time(struct task_struct *p, cputime_t cputime, + cputime_t cputime_scaled) +{ + int index; + + /* Add user time to process. */ + p->utime += cputime; + p->utimescaled += cputime_scaled; + account_group_user_time(p, cputime); + + index = (TASK_NICE(p) > 0) ? CPUTIME_NICE : CPUTIME_USER; + + /* Add user time to cpustat. */ + task_group_account_field(p, index, (__force u64) cputime); + + /* Account for user time used */ + acct_update_integrals(p); +} + +/* + * Account guest cpu time to a process. + * @p: the process that the cpu time gets accounted to + * @cputime: the cpu time spent in virtual machine since the last update + * @cputime_scaled: cputime scaled by cpu frequency + */ +static void account_guest_time(struct task_struct *p, cputime_t cputime, + cputime_t cputime_scaled) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + + /* Add guest time to process. */ + p->utime += cputime; + p->utimescaled += cputime_scaled; + account_group_user_time(p, cputime); + p->gtime += cputime; + + /* Add guest time to cpustat. */ + if (TASK_NICE(p) > 0) { + cpustat[CPUTIME_NICE] += (__force u64) cputime; + cpustat[CPUTIME_GUEST_NICE] += (__force u64) cputime; + } else { + cpustat[CPUTIME_USER] += (__force u64) cputime; + cpustat[CPUTIME_GUEST] += (__force u64) cputime; + } +} + +/* + * Account system cpu time to a process and desired cpustat field + * @p: the process that the cpu time gets accounted to + * @cputime: the cpu time spent in kernel space since the last update + * @cputime_scaled: cputime scaled by cpu frequency + * @target_cputime64: pointer to cpustat field that has to be updated + */ +static inline +void __account_system_time(struct task_struct *p, cputime_t cputime, + cputime_t cputime_scaled, int index) +{ + /* Add system time to process. */ + p->stime += cputime; + p->stimescaled += cputime_scaled; + account_group_system_time(p, cputime); + + /* Add system time to cpustat. */ + task_group_account_field(p, index, (__force u64) cputime); + + /* Account for system time used */ + acct_update_integrals(p); +} + +/* + * Account system cpu time to a process. + * @p: the process that the cpu time gets accounted to + * @hardirq_offset: the offset to subtract from hardirq_count() + * @cputime: the cpu time spent in kernel space since the last update + * @cputime_scaled: cputime scaled by cpu frequency + */ +void account_system_time(struct task_struct *p, int hardirq_offset, + cputime_t cputime, cputime_t cputime_scaled) +{ + int index; + + if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) { + account_guest_time(p, cputime, cputime_scaled); + return; + } + + if (hardirq_count() - hardirq_offset) + index = CPUTIME_IRQ; + else if (in_serving_softirq()) + index = CPUTIME_SOFTIRQ; + else + index = CPUTIME_SYSTEM; + + __account_system_time(p, cputime, cputime_scaled, index); +} + +/* + * Account for involuntary wait time. + * @cputime: the cpu time spent in involuntary wait + */ +void account_steal_time(cputime_t cputime) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + + cpustat[CPUTIME_STEAL] += (__force u64) cputime; +} + +/* + * Account for idle time. + * @cputime: the cpu time spent in idle wait + */ +void account_idle_time(cputime_t cputime) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + struct rq *rq = this_rq(); + + if (atomic_read(&rq->nr_iowait) > 0) + cpustat[CPUTIME_IOWAIT] += (__force u64) cputime; + else + cpustat[CPUTIME_IDLE] += (__force u64) cputime; +} + +static __always_inline bool steal_account_process_tick(void) +{ +#ifdef CONFIG_PARAVIRT + if (static_key_false(¶virt_steal_enabled)) { + u64 steal, st = 0; + + steal = paravirt_steal_clock(smp_processor_id()); + steal -= this_rq()->prev_steal_time; + + st = steal_ticks(steal); + this_rq()->prev_steal_time += st * TICK_NSEC; + + account_steal_time(st); + return st; + } +#endif + return false; +} + +#ifndef CONFIG_VIRT_CPU_ACCOUNTING + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING +/* + * Account a tick to a process and cpustat + * @p: the process that the cpu time gets accounted to + * @user_tick: is the tick from userspace + * @rq: the pointer to rq + * + * Tick demultiplexing follows the order + * - pending hardirq update + * - pending softirq update + * - user_time + * - idle_time + * - system time + * - check for guest_time + * - else account as system_time + * + * Check for hardirq is done both for system and user time as there is + * no timer going off while we are on hardirq and hence we may never get an + * opportunity to update it solely in system time. + * p->stime and friends are only updated on system time and not on irq + * softirq as those do not count in task exec_runtime any more. + */ +static void irqtime_account_process_tick(struct task_struct *p, int user_tick, + struct rq *rq) +{ + cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); + u64 *cpustat = kcpustat_this_cpu->cpustat; + + if (steal_account_process_tick()) + return; + + if (irqtime_account_hi_update()) { + cpustat[CPUTIME_IRQ] += (__force u64) cputime_one_jiffy; + } else if (irqtime_account_si_update()) { + cpustat[CPUTIME_SOFTIRQ] += (__force u64) cputime_one_jiffy; + } else if (this_cpu_ksoftirqd() == p) { + /* + * ksoftirqd time do not get accounted in cpu_softirq_time. + * So, we have to handle it separately here. + * Also, p->stime needs to be updated for ksoftirqd. + */ + __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, + CPUTIME_SOFTIRQ); + } else if (user_tick) { + account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); + } else if (p == rq->idle) { + account_idle_time(cputime_one_jiffy); + } else if (p->flags & PF_VCPU) { /* System time or guest time */ + account_guest_time(p, cputime_one_jiffy, one_jiffy_scaled); + } else { + __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, + CPUTIME_SYSTEM); + } +} + +static void irqtime_account_idle_ticks(int ticks) +{ + int i; + struct rq *rq = this_rq(); + + for (i = 0; i < ticks; i++) + irqtime_account_process_tick(current, 0, rq); +} +#else /* CONFIG_IRQ_TIME_ACCOUNTING */ +static void irqtime_account_idle_ticks(int ticks) {} +static void irqtime_account_process_tick(struct task_struct *p, int user_tick, + struct rq *rq) {} +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ + +/* + * Account a single tick of cpu time. + * @p: the process that the cpu time gets accounted to + * @user_tick: indicates if the tick is a user or a system tick + */ +void account_process_tick(struct task_struct *p, int user_tick) +{ + cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); + struct rq *rq = this_rq(); + + if (sched_clock_irqtime) { + irqtime_account_process_tick(p, user_tick, rq); + return; + } + + if (steal_account_process_tick()) + return; + + if (user_tick) + account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); + else if ((p != rq->idle) || (irq_count() != HARDIRQ_OFFSET)) + account_system_time(p, HARDIRQ_OFFSET, cputime_one_jiffy, + one_jiffy_scaled); + else + account_idle_time(cputime_one_jiffy); +} + +/* + * Account multiple ticks of steal time. + * @p: the process from which the cpu time has been stolen + * @ticks: number of stolen ticks + */ +void account_steal_ticks(unsigned long ticks) +{ + account_steal_time(jiffies_to_cputime(ticks)); +} + +/* + * Account multiple ticks of idle time. + * @ticks: number of stolen ticks + */ +void account_idle_ticks(unsigned long ticks) +{ + + if (sched_clock_irqtime) { + irqtime_account_idle_ticks(ticks); + return; + } + + account_idle_time(jiffies_to_cputime(ticks)); +} + +#endif + +/* + * Use precise platform statistics if available: + */ +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +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 +# define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) +#endif + +static cputime_t scale_utime(cputime_t utime, cputime_t rtime, cputime_t total) +{ + u64 temp = (__force u64) rtime; + + temp *= (__force u64) utime; + + if (sizeof(cputime_t) == 4) + temp = div_u64(temp, (__force u32) total); + else + temp = div64_u64(temp, (__force u64) total); + + return (__force cputime_t) temp; +} + +void task_times(struct task_struct *p, cputime_t *ut, cputime_t *st) +{ + cputime_t rtime, utime = p->utime, total = utime + p->stime; + + /* + * Use CFS's precise accounting: + */ + rtime = nsecs_to_cputime(p->se.sum_exec_runtime); + + if (total) + utime = scale_utime(utime, rtime, total); + else + utime = rtime; + + /* + * 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); + + *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.utime + cputime.stime; + rtime = nsecs_to_cputime(cputime.sum_exec_runtime); + + if (total) + utime = scale_utime(cputime.utime, rtime, total); + else + utime = rtime; + + sig->prev_utime = max(sig->prev_utime, utime); + sig->prev_stime = max(sig->prev_stime, rtime - sig->prev_utime); + + *ut = sig->prev_utime; + *st = sig->prev_stime; +} +#endif diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index f6714d009e77..804c2e5e7872 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -891,6 +891,9 @@ struct cpuacct { struct kernel_cpustat __percpu *cpustat; }; +extern struct cgroup_subsys cpuacct_subsys; +extern struct cpuacct root_cpuacct; + /* return cpu accounting group corresponding to this container */ static inline struct cpuacct *cgroup_ca(struct cgroup *cgrp) { @@ -917,6 +920,16 @@ extern void cpuacct_charge(struct task_struct *tsk, u64 cputime); static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {} #endif +#ifdef CONFIG_PARAVIRT +static inline u64 steal_ticks(u64 steal) +{ + if (unlikely(steal > NSEC_PER_SEC)) + return div_u64(steal, TICK_NSEC); + + return __iter_div_u64_rem(steal, TICK_NSEC, &steal); +} +#endif + static inline void inc_nr_running(struct rq *rq) { rq->nr_running++; @@ -1157,3 +1170,53 @@ enum rq_nohz_flag_bits { #define nohz_flags(cpu) (&cpu_rq(cpu)->nohz_flags) #endif + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + +DECLARE_PER_CPU(u64, cpu_hardirq_time); +DECLARE_PER_CPU(u64, cpu_softirq_time); + +#ifndef CONFIG_64BIT +DECLARE_PER_CPU(seqcount_t, irq_time_seq); + +static inline void irq_time_write_begin(void) +{ + __this_cpu_inc(irq_time_seq.sequence); + smp_wmb(); +} + +static inline void irq_time_write_end(void) +{ + smp_wmb(); + __this_cpu_inc(irq_time_seq.sequence); +} + +static inline u64 irq_time_read(int cpu) +{ + u64 irq_time; + unsigned seq; + + do { + seq = read_seqcount_begin(&per_cpu(irq_time_seq, cpu)); + irq_time = per_cpu(cpu_softirq_time, cpu) + + per_cpu(cpu_hardirq_time, cpu); + } while (read_seqcount_retry(&per_cpu(irq_time_seq, cpu), seq)); + + return irq_time; +} +#else /* CONFIG_64BIT */ +static inline void irq_time_write_begin(void) +{ +} + +static inline void irq_time_write_end(void) +{ +} + +static inline u64 irq_time_read(int cpu) +{ + return per_cpu(cpu_softirq_time, cpu) + per_cpu(cpu_hardirq_time, cpu); +} +#endif /* CONFIG_64BIT */ +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ + -- cgit v1.2.3 From baa36046d09ea6dbc122c795566992318663d9eb Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 18 Jun 2012 17:54:14 +0200 Subject: cputime: Consolidate vtime handling on context switch The archs that implement virtual cputime accounting all flush the cputime of a task when it gets descheduled and sometimes set up some ground initialization for the next task to account its cputime. These archs all put their own hooks in their context switch callbacks and handle the off-case themselves. Consolidate this by creating a new account_switch_vtime() callback called in generic code right after a context switch and that these archs must implement to flush the prev task cputime and initialize the next task cputime related state. Signed-off-by: Frederic Weisbecker Acked-by: Martin Schwidefsky Cc: Tony Luck Cc: Fenghua Yu Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Heiko Carstens Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Peter Zijlstra --- kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ae3bcaa3afbf..78d9c965433a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1796,6 +1796,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) * Manfred Spraul */ prev_state = prev->state; + account_switch_vtime(prev); finish_arch_switch(prev); #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_disable(); -- cgit v1.2.3 From f319da0c6894fcf55e21320e40506418a2aad629 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 20 Aug 2012 11:26:57 +0200 Subject: sched: Fix load avg vs cpu-hotplug Rabik and Paul reported two different issues related to the same few lines of code. Rabik's issue is that the nr_uninterruptible migration code is wrong in that he sees artifacts due to this (Rabik please do expand in more detail). Paul's issue is that this code as it stands relies on us using stop_machine() for unplug, we all would like to remove this assumption so that eventually we can remove this stop_machine() usage altogether. The only reason we'd have to migrate nr_uninterruptible is so that we could use for_each_online_cpu() loops in favour of for_each_possible_cpu() loops, however since nr_uninterruptible() is the only such loop and its using possible lets not bother at all. The problem Rabik sees is (probably) caused by the fact that by migrating nr_uninterruptible we screw rq->calc_load_active for both rqs involved. So don't bother with fancy migration schemes (meaning we now have to keep using for_each_possible_cpu()) and instead fold any nr_active delta after we migrate all tasks away to make sure we don't have any skewed nr_active accounting. Reported-by: Rakib Mullick Reported-by: Paul E. McKenney Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1345454817.23018.27.camel@twins Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 31 ++++++++++--------------------- 1 file changed, 10 insertions(+), 21 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index fbf1fd098dc6..207a81c769d4 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5304,27 +5304,17 @@ void idle_task_exit(void) } /* - * While a dead CPU has no uninterruptible tasks queued at this point, - * it might still have a nonzero ->nr_uninterruptible counter, because - * for performance reasons the counter is not stricly tracking tasks to - * their home CPUs. So we just add the counter to another CPU's counter, - * to keep the global sum constant after CPU-down: - */ -static void migrate_nr_uninterruptible(struct rq *rq_src) -{ - struct rq *rq_dest = cpu_rq(cpumask_any(cpu_active_mask)); - - rq_dest->nr_uninterruptible += rq_src->nr_uninterruptible; - rq_src->nr_uninterruptible = 0; -} - -/* - * remove the tasks which were accounted by rq from calc_load_tasks. + * Since this CPU is going 'away' for a while, fold any nr_active delta + * we might have. Assumes we're called after migrate_tasks() so that the + * nr_active count is stable. + * + * Also see the comment "Global load-average calculations". */ -static void calc_global_load_remove(struct rq *rq) +static void calc_load_migrate(struct rq *rq) { - atomic_long_sub(rq->calc_load_active, &calc_load_tasks); - rq->calc_load_active = 0; + long delta = calc_load_fold_active(rq); + if (delta) + atomic_long_add(delta, &calc_load_tasks); } /* @@ -5618,8 +5608,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) BUG_ON(rq->nr_running != 1); /* the migration thread */ raw_spin_unlock_irqrestore(&rq->lock, flags); - migrate_nr_uninterruptible(rq); - calc_global_load_remove(rq); + calc_load_migrate(rq); break; #endif } -- cgit v1.2.3 From a4c96ae319b8047f62dedbe1eac79e321c185749 Mon Sep 17 00:00:00 2001 From: Peter Boonstoppel Date: Thu, 9 Aug 2012 15:34:47 -0700 Subject: sched: Unthrottle rt runqueues in __disable_runtime() migrate_tasks() uses _pick_next_task_rt() to get tasks from the real-time runqueues to be migrated. When rt_rq is throttled _pick_next_task_rt() won't return anything, in which case migrate_tasks() can't move all threads over and gets stuck in an infinite loop. Instead unthrottle rt runqueues before migrating tasks. Additionally: move unthrottle_offline_cfs_rqs() to rq_offline_fair() Signed-off-by: Peter Boonstoppel Signed-off-by: Peter Zijlstra Cc: Paul Turner Link: http://lkml.kernel.org/r/5FBF8E85CA34454794F0F7ECBA79798F379D3648B7@HQMAIL04.nvidia.com Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 3 --- kernel/sched/fair.c | 7 +++++-- kernel/sched/rt.c | 1 + kernel/sched/sched.h | 1 - 4 files changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 207a81c769d4..a4ea245f3d85 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5342,9 +5342,6 @@ static void migrate_tasks(unsigned int dead_cpu) */ rq->stop = NULL; - /* Ensure any throttled groups are reachable by pick_next_task */ - unthrottle_offline_cfs_rqs(rq); - for ( ; ; ) { /* * There's this thread running, bail when that's the only diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c219bf8d704c..86ad83c45dae 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2052,7 +2052,7 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) hrtimer_cancel(&cfs_b->slack_timer); } -void unthrottle_offline_cfs_rqs(struct rq *rq) +static void unthrottle_offline_cfs_rqs(struct rq *rq) { struct cfs_rq *cfs_rq; @@ -2106,7 +2106,7 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) return NULL; } static inline void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} -void unthrottle_offline_cfs_rqs(struct rq *rq) {} +static inline void unthrottle_offline_cfs_rqs(struct rq *rq) {} #endif /* CONFIG_CFS_BANDWIDTH */ @@ -4956,6 +4956,9 @@ static void rq_online_fair(struct rq *rq) static void rq_offline_fair(struct rq *rq) { update_sysctl(); + + /* Ensure any throttled groups are reachable by pick_next_task */ + unthrottle_offline_cfs_rqs(rq); } #endif /* CONFIG_SMP */ diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 944cb68420e9..e0b7ba9c040f 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -691,6 +691,7 @@ balanced: * runtime - in which case borrowing doesn't make sense. */ rt_rq->rt_runtime = RUNTIME_INF; + rt_rq->rt_throttled = 0; raw_spin_unlock(&rt_rq->rt_runtime_lock); raw_spin_unlock(&rt_b->rt_runtime_lock); } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index f6714d009e77..0848fa36c383 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1144,7 +1144,6 @@ extern void print_rt_stats(struct seq_file *m, int cpu); extern void init_cfs_rq(struct cfs_rq *cfs_rq); extern void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq); -extern void unthrottle_offline_cfs_rqs(struct rq *rq); extern void account_cfs_bandwidth_used(int enabled, int was_enabled); -- cgit v1.2.3 From 9450d57eab5cad36774c297da123062744472588 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 18 Aug 2012 17:45:08 -0700 Subject: sched: Fix kernel-doc warnings in kernel/sched/fair.c Fix two kernel-doc warnings in kernel/sched/fair.c: Warning(kernel/sched/fair.c:3660): Excess function parameter 'cpus' description in 'update_sg_lb_stats' Warning(kernel/sched/fair.c:3806): Excess function parameter 'cpus' description in 'update_sd_lb_stats' Signed-off-by: Randy Dunlap Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/50303714.3090204@xenotime.net Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 86ad83c45dae..42d9df6a5ca4 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3658,7 +3658,6 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) * @group: sched_group whose statistics are to be updated. * @load_idx: Load index of sched_domain of this_cpu for load calc. * @local_group: Does group contain this_cpu. - * @cpus: Set of cpus considered for load balancing. * @balance: Should we balance. * @sgs: variable to hold the statistics for this group. */ @@ -3805,7 +3804,6 @@ static bool update_sd_pick_busiest(struct lb_env *env, /** * update_sd_lb_stats - Update sched_domain's statistics for load balancing. * @env: The load balancing environment. - * @cpus: Set of cpus considered for load balancing. * @balance: Should we balance. * @sds: variable to hold the statistics for this sched_domain. */ -- cgit v1.2.3 From c751134ef8b070070d5f06348286b29d86424677 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Thu, 16 Aug 2012 13:21:05 +0900 Subject: sched: Remove AFFINE_WAKEUPS feature flag Commit beac4c7e4a1c ("sched: Remove AFFINE_WAKEUPS feature") removed use of the flag but left the definition. Get rid of it. Signed-off-by: Namhyung Kim Signed-off-by: Peter Zijlstra Cc: Mike Galbraith Link: http://lkml.kernel.org/r/1345090865-20851-1-git-send-email-namhyung@kernel.org Signed-off-by: Ingo Molnar --- kernel/sched/features.h | 8 -------- 1 file changed, 8 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/features.h b/kernel/sched/features.h index de00a486c5c6..c38f52ea53dd 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -11,14 +11,6 @@ SCHED_FEAT(GENTLE_FAIR_SLEEPERS, true) */ SCHED_FEAT(START_DEBIT, true) -/* - * Based on load and program behaviour, see if it makes sense to place - * a newly woken task on the same cpu as the task that woke it -- - * improve cache locality. Typically used with SYNC wakeups as - * generated by pipes and the like, see also SYNC_WAKEUPS. - */ -SCHED_FEAT(AFFINE_WAKEUPS, true) - /* * Prefer to schedule the task we woke last (assuming it failed * wakeup-preemption), since its likely going to consume data we -- cgit v1.2.3 From 201c373e8e4823700d3160d5c28e1ab18fd1193e Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Thu, 16 Aug 2012 17:03:24 +0900 Subject: sched/debug: Limit sd->*_idx range on sysctl Various sd->*_idx's are used for refering the rq's load average table when selecting a cpu to run. However they can be set to any number with sysctl knobs so that it can crash the kernel if something bad is given. Fix it by limiting them into the actual range. Signed-off-by: Namhyung Kim Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1345104204-8317-1-git-send-email-namhyung@kernel.org Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 35 ++++++++++++++++++++++------------- 1 file changed, 22 insertions(+), 13 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ae66229238a0..ec0f2b81b81c 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4896,16 +4896,25 @@ static void sd_free_ctl_entry(struct ctl_table **tablep) *tablep = NULL; } +static int min_load_idx = 0; +static int max_load_idx = CPU_LOAD_IDX_MAX; + static void set_table_entry(struct ctl_table *entry, const char *procname, void *data, int maxlen, - umode_t mode, proc_handler *proc_handler) + umode_t mode, proc_handler *proc_handler, + bool load_idx) { entry->procname = procname; entry->data = data; entry->maxlen = maxlen; entry->mode = mode; entry->proc_handler = proc_handler; + + if (load_idx) { + entry->extra1 = &min_load_idx; + entry->extra2 = &max_load_idx; + } } static struct ctl_table * @@ -4917,30 +4926,30 @@ sd_alloc_ctl_domain_table(struct sched_domain *sd) return NULL; set_table_entry(&table[0], "min_interval", &sd->min_interval, - sizeof(long), 0644, proc_doulongvec_minmax); + sizeof(long), 0644, proc_doulongvec_minmax, false); set_table_entry(&table[1], "max_interval", &sd->max_interval, - sizeof(long), 0644, proc_doulongvec_minmax); + sizeof(long), 0644, proc_doulongvec_minmax, false); set_table_entry(&table[2], "busy_idx", &sd->busy_idx, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, true); set_table_entry(&table[3], "idle_idx", &sd->idle_idx, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, true); set_table_entry(&table[4], "newidle_idx", &sd->newidle_idx, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, true); set_table_entry(&table[5], "wake_idx", &sd->wake_idx, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, true); set_table_entry(&table[6], "forkexec_idx", &sd->forkexec_idx, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, true); set_table_entry(&table[7], "busy_factor", &sd->busy_factor, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, false); set_table_entry(&table[8], "imbalance_pct", &sd->imbalance_pct, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, false); set_table_entry(&table[9], "cache_nice_tries", &sd->cache_nice_tries, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, false); set_table_entry(&table[10], "flags", &sd->flags, - sizeof(int), 0644, proc_dointvec_minmax); + sizeof(int), 0644, proc_dointvec_minmax, false); set_table_entry(&table[11], "name", sd->name, - CORENAME_MAX_SIZE, 0444, proc_dostring); + CORENAME_MAX_SIZE, 0444, proc_dostring, false); /* &table[12] is terminator */ return table; -- cgit v1.2.3 From 38b8dd6f87398524d02c21ff614c507ba8c9d295 Mon Sep 17 00:00:00 2001 From: Michael Wang Date: Tue, 3 Jul 2012 14:34:02 +0800 Subject: sched: Remove useless code in yield_to() It's impossible to enter the else branch if we have set skip_clock_update in task_yield_fair(), as yield_to_task_fair() will directly return true after invoke task_yield_fair(). Signed-off-by: Michael Wang Acked-by: Mike Galbraith Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/4FF2925A.9060005@linux.vnet.ibm.com Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 7 ------- 1 file changed, 7 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ec0f2b81b81c..c46a011ce5db 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4348,13 +4348,6 @@ again: */ if (preempt && rq != p_rq) resched_task(p_rq->curr); - } else { - /* - * We might have set it in task_yield_fair(), but are - * not going to schedule(), so don't want to skip - * the next update. - */ - rq->skip_clock_update = 0; } out: -- cgit v1.2.3 From 5ed4f1d96deee82ee92cd1ac1e0108c27e80e9b0 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Thu, 13 Sep 2012 06:11:26 +0200 Subject: sched: Fix nohz_idle_balance() On tickless systems, one CPU runs load balance for all idle CPUs. The cpu_load of this CPU is updated before starting the load balance of each other idle CPUs. We should instead update the cpu_load of the balance_cpu. Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra Cc: Venkatesh Pallipadi Cc: Suresh Siddha Link: http://lkml.kernel.org/r/1347509486-8688-1-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 1ca4fe423528..9ae3a5b68ba4 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4794,14 +4794,15 @@ static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) if (need_resched()) break; - raw_spin_lock_irq(&this_rq->lock); - update_rq_clock(this_rq); - update_idle_cpu_load(this_rq); - raw_spin_unlock_irq(&this_rq->lock); + rq = cpu_rq(balance_cpu); + + raw_spin_lock_irq(&rq->lock); + update_rq_clock(rq); + update_idle_cpu_load(rq); + raw_spin_unlock_irq(&rq->lock); rebalance_domains(balance_cpu, CPU_IDLE); - rq = cpu_rq(balance_cpu); if (time_after(this_rq->next_balance, rq->next_balance)) this_rq->next_balance = rq->next_balance; } -- cgit v1.2.3 From f3e947867478af9a12b9956bcd000ac7613a8a95 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 12 Sep 2012 11:22:00 +0200 Subject: sched: Remove __ARCH_WANT_INTERRUPTS_ON_CTXSW Now that the last architecture to use this has stopped doing so (ARM, thanks Catalin!) we can remove this complexity from the scheduler core. Signed-off-by: Peter Zijlstra Cc: Oleg Nesterov Cc: Catalin Marinas Link: http://lkml.kernel.org/n/tip-g9p2a1w81xxbrze25v9zpzbf@git.kernel.org Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 40 +--------------------------------------- kernel/sched/rt.c | 5 ----- kernel/sched/sched.h | 6 ------ 3 files changed, 1 insertion(+), 50 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c46a011ce5db..8b51b2d9b1fd 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1361,25 +1361,6 @@ static void ttwu_queue_remote(struct task_struct *p, int cpu) smp_send_reschedule(cpu); } -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW -static int ttwu_activate_remote(struct task_struct *p, int wake_flags) -{ - struct rq *rq; - int ret = 0; - - rq = __task_rq_lock(p); - if (p->on_cpu) { - ttwu_activate(rq, p, ENQUEUE_WAKEUP); - ttwu_do_wakeup(rq, p, wake_flags); - ret = 1; - } - __task_rq_unlock(rq); - - return ret; - -} -#endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ - bool cpus_share_cache(int this_cpu, int that_cpu) { return per_cpu(sd_llc_id, this_cpu) == per_cpu(sd_llc_id, that_cpu); @@ -1440,21 +1421,8 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) * If the owning (remote) cpu is still in the middle of schedule() with * this task as prev, wait until its done referencing the task. */ - while (p->on_cpu) { -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - /* - * In case the architecture enables interrupts in - * context_switch(), we cannot busy wait, since that - * would lead to deadlocks when an interrupt hits and - * tries to wake up @prev. So bail and do a complete - * remote wakeup. - */ - if (ttwu_activate_remote(p, wake_flags)) - goto stat; -#else + while (p->on_cpu) cpu_relax(); -#endif - } /* * Pairs with the smp_wmb() in finish_lock_switch(). */ @@ -1798,13 +1766,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) prev_state = prev->state; account_switch_vtime(prev); finish_arch_switch(prev); -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - local_irq_disable(); -#endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ perf_event_task_sched_in(prev, current); -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - local_irq_enable(); -#endif /* __ARCH_WANT_INTERRUPTS_ON_CTXSW */ finish_lock_switch(rq, prev); finish_arch_post_lock_switch(); diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index e0b7ba9c040f..418feb01344e 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1632,11 +1632,6 @@ static int push_rt_task(struct rq *rq) if (!next_task) return 0; -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - if (unlikely(task_running(rq, next_task))) - return 0; -#endif - retry: if (unlikely(next_task == rq->curr)) { WARN_ON(1); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 09871698e80c..7a7db09cfabc 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -737,11 +737,7 @@ static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next) */ next->on_cpu = 1; #endif -#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - raw_spin_unlock_irq(&rq->lock); -#else raw_spin_unlock(&rq->lock); -#endif } static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) @@ -755,9 +751,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) smp_wmb(); prev->on_cpu = 0; #endif -#ifndef __ARCH_WANT_INTERRUPTS_ON_CTXSW local_irq_enable(); -#endif } #endif /* __ARCH_WANT_UNLOCKED_CTXSW */ -- cgit v1.2.3 From 08bedae1d0acd8c9baf514fb69fa199d0c8345f6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 6 Sep 2012 00:03:50 +0200 Subject: sched: Fix load avg vs. cpu-hotplug Commit f319da0c68 ("sched: Fix load avg vs cpu-hotplug") was an incomplete fix: In particular, the problem is that at the point it calls calc_load_migrate() nr_running := 1 (the stopper thread), so move the call to CPU_DEAD where we're sure that nr_running := 0. Also note that we can call calc_load_migrate() without serialization, we know the state of rq is stable since its cpu is dead, and we modify the global state using appropriate atomic ops. Suggested-by: Paul E. McKenney Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1346882630.2600.59.camel@twins Signed-off-by: Ingo Molnar --- kernel/sched/core.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 8b51b2d9b1fd..ba144b121f3d 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5048,7 +5048,9 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) migrate_tasks(cpu); BUG_ON(rq->nr_running != 1); /* the migration thread */ raw_spin_unlock_irqrestore(&rq->lock, flags); + break; + case CPU_DEAD: calc_load_migrate(rq); break; #endif -- cgit v1.2.3 From c1cc017c59c44d9ede7003631c43adc0cfdce2f9 Mon Sep 17 00:00:00 2001 From: Alex Shi Date: Mon, 10 Sep 2012 15:10:58 +0800 Subject: sched/nohz: Clean up select_nohz_load_balancer() There is no load_balancer to be selected now. It just sets the state of the nohz tick to stop. So rename the function, pass the 'cpu' as a parameter and then remove the useless call from tick_nohz_restart_sched_tick(). [ s/set_nohz_tick_stopped/nohz_balance_enter_idle/g s/clear_nohz_tick_stopped/nohz_balance_exit_idle/g ] Signed-off-by: Alex Shi Acked-by: Suresh Siddha Cc: Venkatesh Pallipadi Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1347261059-24747-1-git-send-email-alex.shi@intel.com Signed-off-by: Ingo Molnar --- kernel/sched/fair.c | 25 ++++++++++--------------- 1 file changed, 10 insertions(+), 15 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 9ae3a5b68ba4..de596a2f626c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4603,7 +4603,7 @@ static void nohz_balancer_kick(int cpu) return; } -static inline void clear_nohz_tick_stopped(int cpu) +static inline void nohz_balance_exit_idle(int cpu) { if (unlikely(test_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu)))) { cpumask_clear_cpu(cpu, nohz.idle_cpus_mask); @@ -4643,28 +4643,23 @@ void set_cpu_sd_state_idle(void) } /* - * This routine will record that this cpu is going idle with tick stopped. + * This routine will record that the cpu is going idle with tick stopped. * This info will be used in performing idle load balancing in the future. */ -void select_nohz_load_balancer(int stop_tick) +void nohz_balance_enter_idle(int cpu) { - int cpu = smp_processor_id(); - /* * If this cpu is going down, then nothing needs to be done. */ if (!cpu_active(cpu)) return; - if (stop_tick) { - if (test_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu))) - return; + if (test_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu))) + return; - cpumask_set_cpu(cpu, nohz.idle_cpus_mask); - atomic_inc(&nohz.nr_cpus); - set_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu)); - } - return; + cpumask_set_cpu(cpu, nohz.idle_cpus_mask); + atomic_inc(&nohz.nr_cpus); + set_bit(NOHZ_TICK_STOPPED, nohz_flags(cpu)); } static int __cpuinit sched_ilb_notifier(struct notifier_block *nfb, @@ -4672,7 +4667,7 @@ static int __cpuinit sched_ilb_notifier(struct notifier_block *nfb, { switch (action & ~CPU_TASKS_FROZEN) { case CPU_DYING: - clear_nohz_tick_stopped(smp_processor_id()); + nohz_balance_exit_idle(smp_processor_id()); return NOTIFY_OK; default: return NOTIFY_DONE; @@ -4833,7 +4828,7 @@ static inline int nohz_kick_needed(struct rq *rq, int cpu) * busy tick after returning from idle, we will update the busy stats. */ set_cpu_sd_state_busy(); - clear_nohz_tick_stopped(cpu); + nohz_balance_exit_idle(cpu); /* * None are in tickless mode and hence no need for NOHZ idle load -- cgit v1.2.3 From bc2a27cd27271c5257989a57f511be86b26f5e54 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Mon, 9 Jul 2012 11:27:06 +0200 Subject: sched: cpu_power: enable ARCH_POWER Heteregeneous ARM platform uses arch_scale_freq_power function to reflect the relative capacity of each core Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1341826026-6504-6-git-send-email-vincent.guittot@linaro.org Signed-off-by: Ingo Molnar --- kernel/sched/features.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/features.h b/kernel/sched/features.h index c38f52ea53dd..eebefcad7027 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -34,7 +34,7 @@ SCHED_FEAT(CACHE_HOT_BUDDY, true) /* * Use arch dependent cpu power functions */ -SCHED_FEAT(ARCH_POWER, false) +SCHED_FEAT(ARCH_POWER, true) SCHED_FEAT(HRTICK, false) SCHED_FEAT(DOUBLE_TICK, false) -- cgit v1.2.3 From 37407ea7f93864c2cfc03edf8f37872ec539ea2b Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 16 Sep 2012 12:29:43 -0700 Subject: Revert "sched: Improve scalability via 'CPU buddies', which withstand random perturbations" This reverts commit 970e178985cadbca660feb02f4d2ee3a09f7fdda. Nikolay Ulyanitsky reported thatthe 3.6-rc5 kernel has a 15-20% performance drop on PostgreSQL 9.2 on his machine (running "pgbench"). Borislav Petkov was able to reproduce this, and bisected it to this commit 970e178985ca ("sched: Improve scalability via 'CPU buddies' ...") apparently because the new single-idle-buddy model simply doesn't find idle CPU's to reschedule on aggressively enough. Mike Galbraith suspects that it is likely due to the user-mode spinlocks in PostgreSQL not reacting well to preemption, but we don't really know the details - I'll just revert the commit for now. There are hopefully other approaches to improve scheduler scalability without it causing these kinds of downsides. Reported-by: Nikolay Ulyanitsky Bisected-by: Borislav Petkov Acked-by: Mike Galbraith Cc: Andrew Morton Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Linus Torvalds --- kernel/sched/core.c | 39 +-------------------------------------- kernel/sched/fair.c | 28 +++++++++++++++++++++------- 2 files changed, 22 insertions(+), 45 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index a4ea245f3d85..649c9f876cb1 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6014,11 +6014,6 @@ static void destroy_sched_domains(struct sched_domain *sd, int cpu) * SD_SHARE_PKG_RESOURCE set (Last Level Cache Domain) for this * allows us to avoid some pointer chasing select_idle_sibling(). * - * Iterate domains and sched_groups downward, assigning CPUs to be - * select_idle_sibling() hw buddy. Cross-wiring hw makes bouncing - * due to random perturbation self canceling, ie sw buddies pull - * their counterpart to their CPU's hw counterpart. - * * Also keep a unique ID per domain (we use the first cpu number in * the cpumask of the domain), this allows us to quickly tell if * two cpus are in the same cache domain, see cpus_share_cache(). @@ -6032,40 +6027,8 @@ static void update_top_cache_domain(int cpu) int id = cpu; sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES); - if (sd) { - struct sched_domain *tmp = sd; - struct sched_group *sg, *prev; - bool right; - - /* - * Traverse to first CPU in group, and count hops - * to cpu from there, switching direction on each - * hop, never ever pointing the last CPU rightward. - */ - do { - id = cpumask_first(sched_domain_span(tmp)); - prev = sg = tmp->groups; - right = 1; - - while (cpumask_first(sched_group_cpus(sg)) != id) - sg = sg->next; - - while (!cpumask_test_cpu(cpu, sched_group_cpus(sg))) { - prev = sg; - sg = sg->next; - right = !right; - } - - /* A CPU went down, never point back to domain start. */ - if (right && cpumask_first(sched_group_cpus(sg->next)) == id) - right = false; - - sg = right ? sg->next : prev; - tmp->idle_buddy = cpumask_first(sched_group_cpus(sg)); - } while ((tmp = tmp->child)); - + if (sd) id = cpumask_first(sched_domain_span(sd)); - } rcu_assign_pointer(per_cpu(sd_llc, cpu), sd); per_cpu(sd_llc_id, cpu) = id; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 42d9df6a5ca4..96e2b18b6283 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2637,6 +2637,8 @@ static int select_idle_sibling(struct task_struct *p, int target) int cpu = smp_processor_id(); int prev_cpu = task_cpu(p); struct sched_domain *sd; + struct sched_group *sg; + int i; /* * If the task is going to be woken-up on this cpu and if it is @@ -2653,17 +2655,29 @@ static int select_idle_sibling(struct task_struct *p, int target) return prev_cpu; /* - * Otherwise, check assigned siblings to find an elegible idle cpu. + * Otherwise, iterate the domains and find an elegible idle cpu. */ sd = rcu_dereference(per_cpu(sd_llc, target)); - for_each_lower_domain(sd) { - if (!cpumask_test_cpu(sd->idle_buddy, tsk_cpus_allowed(p))) - continue; - if (idle_cpu(sd->idle_buddy)) - return sd->idle_buddy; - } + sg = sd->groups; + do { + if (!cpumask_intersects(sched_group_cpus(sg), + tsk_cpus_allowed(p))) + goto next; + for_each_cpu(i, sched_group_cpus(sg)) { + if (!idle_cpu(i)) + goto next; + } + + target = cpumask_first_and(sched_group_cpus(sg), + tsk_cpus_allowed(p)); + goto done; +next: + sg = sg->next; + } while (sg != sd->groups); + } +done: return target; } -- cgit v1.2.3 From 5d18023294abc22984886bd7185344e0c2be0daf Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 20 Aug 2012 11:26:57 +0200 Subject: sched: Fix load avg vs cpu-hotplug Rabik and Paul reported two different issues related to the same few lines of code. Rabik's issue is that the nr_uninterruptible migration code is wrong in that he sees artifacts due to this (Rabik please do expand in more detail). Paul's issue is that this code as it stands relies on us using stop_machine() for unplug, we all would like to remove this assumption so that eventually we can remove this stop_machine() usage altogether. The only reason we'd have to migrate nr_uninterruptible is so that we could use for_each_online_cpu() loops in favour of for_each_possible_cpu() loops, however since nr_uninterruptible() is the only such loop and its using possible lets not bother at all. The problem Rabik sees is (probably) caused by the fact that by migrating nr_uninterruptible we screw rq->calc_load_active for both rqs involved. So don't bother with fancy migration schemes (meaning we now have to keep using for_each_possible_cpu()) and instead fold any nr_active delta after we migrate all tasks away to make sure we don't have any skewed nr_active accounting. [ paulmck: Move call to calc_load_migration to CPU_DEAD to avoid miscounting noted by Rakib. ] Reported-by: Rakib Mullick Reported-by: Paul E. McKenney Signed-off-by: Peter Zijlstra Signed-off-by: Paul E. McKenney --- kernel/sched/core.c | 41 ++++++++++++++++++++--------------------- 1 file changed, 20 insertions(+), 21 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index fbf1fd098dc6..8c38b5e7ce47 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5304,27 +5304,17 @@ void idle_task_exit(void) } /* - * While a dead CPU has no uninterruptible tasks queued at this point, - * it might still have a nonzero ->nr_uninterruptible counter, because - * for performance reasons the counter is not stricly tracking tasks to - * their home CPUs. So we just add the counter to another CPU's counter, - * to keep the global sum constant after CPU-down: - */ -static void migrate_nr_uninterruptible(struct rq *rq_src) -{ - struct rq *rq_dest = cpu_rq(cpumask_any(cpu_active_mask)); - - rq_dest->nr_uninterruptible += rq_src->nr_uninterruptible; - rq_src->nr_uninterruptible = 0; -} - -/* - * remove the tasks which were accounted by rq from calc_load_tasks. + * Since this CPU is going 'away' for a while, fold any nr_active delta + * we might have. Assumes we're called after migrate_tasks() so that the + * nr_active count is stable. + * + * Also see the comment "Global load-average calculations". */ -static void calc_global_load_remove(struct rq *rq) +static void calc_load_migrate(struct rq *rq) { - atomic_long_sub(rq->calc_load_active, &calc_load_tasks); - rq->calc_load_active = 0; + long delta = calc_load_fold_active(rq); + if (delta) + atomic_long_add(delta, &calc_load_tasks); } /* @@ -5617,9 +5607,18 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) migrate_tasks(cpu); BUG_ON(rq->nr_running != 1); /* the migration thread */ raw_spin_unlock_irqrestore(&rq->lock, flags); + break; - migrate_nr_uninterruptible(rq); - calc_global_load_remove(rq); + case CPU_DEAD: + { + struct rq *dest_rq; + + local_irq_save(flags); + dest_rq = cpu_rq(smp_processor_id()); + raw_spin_lock(&dest_rq->lock); + calc_load_migrate(rq); + raw_spin_unlock_irqrestore(&dest_rq->lock, flags); + } break; #endif } -- cgit v1.2.3 From bf9fae9f5e4ca8dce4708812f9ad6281e61df109 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 8 Sep 2012 15:23:11 +0200 Subject: cputime: Use a proper subsystem naming for vtime related APIs Use a naming based on vtime as a prefix for virtual based cputime accounting APIs: - account_system_vtime() -> vtime_account() - account_switch_vtime() -> vtime_task_switch() It makes it easier to allow for further declension such as vtime_account_system(), vtime_account_idle(), ... if we want to find out the context we account to from generic code. This also make it better to know on which subsystem these APIs refer to. Signed-off-by: Frederic Weisbecker Cc: Tony Luck Cc: Fenghua Yu Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Heiko Carstens Cc: Martin Schwidefsky Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Peter Zijlstra --- kernel/sched/core.c | 2 +- kernel/sched/cputime.c | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ba144b121f3d..21e4dcff18f3 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1764,7 +1764,7 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) * Manfred Spraul */ prev_state = prev->state; - account_switch_vtime(prev); + vtime_task_switch(prev); finish_arch_switch(prev); perf_event_task_sched_in(prev, current); finish_lock_switch(rq, prev); diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 372692bd5376..53f5b12f2821 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -10,11 +10,11 @@ /* * There are no locks covering percpu hardirq/softirq time. - * They are only modified in account_system_vtime, on corresponding CPU + * They are only modified in vtime_account, on corresponding CPU * with interrupts disabled. So, writes are safe. * They are read and saved off onto struct rq in update_rq_clock(). * This may result in other CPU reading this CPU's irq time and can - * race with irq/account_system_vtime on this CPU. We would either get old + * race with irq/vtime_account on this CPU. We would either get old * or new value with a side effect of accounting a slice of irq time to wrong * task when irq is in progress while we read rq->clock. That is a worthy * compromise in place of having locks on each irq in account_system_time. @@ -43,7 +43,7 @@ DEFINE_PER_CPU(seqcount_t, irq_time_seq); * Called before incrementing preempt_count on {soft,}irq_enter * and before decrementing preempt_count on {soft,}irq_exit. */ -void account_system_vtime(struct task_struct *curr) +void vtime_account(struct task_struct *curr) { unsigned long flags; s64 delta; @@ -73,7 +73,7 @@ void account_system_vtime(struct task_struct *curr) irq_time_write_end(); local_irq_restore(flags); } -EXPORT_SYMBOL_GPL(account_system_vtime); +EXPORT_SYMBOL_GPL(vtime_account); static int irqtime_account_hi_update(void) { -- cgit v1.2.3 From a7e1a9e3af71b45ecae2dae35851f238117b317d Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 8 Sep 2012 16:14:02 +0200 Subject: vtime: Consolidate system/idle context detection Move the code that finds out to which context we account the cputime into generic layer. Archs that consider the whole time spent in the idle task as idle time (ia64, powerpc) can rely on the generic vtime_account() and implement vtime_account_system() and vtime_account_idle(), letting the generic code to decide when to call which API. Archs that have their own meaning of idle time, such as s390 that only considers the time spent in CPU low power mode as idle time, can just override vtime_account(). Signed-off-by: Frederic Weisbecker Cc: Tony Luck Cc: Fenghua Yu Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: Ingo Molnar Cc: Thomas Gleixner Cc: Peter Zijlstra --- kernel/sched/cputime.c | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 53f5b12f2821..81b763ba58a6 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -432,6 +432,32 @@ void thread_group_times(struct task_struct *p, cputime_t *ut, cputime_t *st) *ut = cputime.utime; *st = cputime.stime; } + +/* + * Archs that account the whole time spent in the idle task + * (outside irq) as idle time can rely on this and just implement + * vtime_account_system() and vtime_account_idle(). Archs that + * have other meaning of the idle time (s390 only includes the + * time spent by the CPU when it's in low power mode) must override + * vtime_account(). + */ +#ifndef __ARCH_HAS_VTIME_ACCOUNT +void vtime_account(struct task_struct *tsk) +{ + unsigned long flags; + + local_irq_save(flags); + + if (in_interrupt() || !is_idle_task(tsk)) + vtime_account_system(tsk); + else + vtime_account_idle(tsk); + + local_irq_restore(flags); +} +EXPORT_SYMBOL_GPL(vtime_account); +#endif /* __ARCH_HAS_VTIME_ACCOUNT */ + #else #ifndef nsecs_to_cputime -- cgit v1.2.3 From 04e7e951532b390b16feb070be9972b8fad2fc57 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Mon, 16 Jul 2012 15:06:40 -0700 Subject: rcu: Switch task's syscall hooks on context switch Clear the syscalls hook of a task when it's scheduled out so that if the task migrates, it doesn't run the syscall slow path on a CPU that might not need it. Also set the syscalls hook on the next task if needed. Signed-off-by: Frederic Weisbecker Cc: Alessio Igor Bogani Cc: Andrew Morton Cc: Avi Kivity Cc: Chris Metcalf Cc: Christoph Lameter Cc: Geoff Levand Cc: Gilad Ben Yossef Cc: Hakan Akkan Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Josh Triplett Cc: Kevin Hilman Cc: Max Krasnyansky Cc: Peter Zijlstra Cc: Stephen Hemminger Cc: Steven Rostedt Cc: Sven-Thorsten Dietrich Cc: Thomas Gleixner Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 1a48cdbc8631..ea2213b07d9d 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2081,6 +2081,7 @@ context_switch(struct rq *rq, struct task_struct *prev, #endif /* Here we just switch the register state and the stack. */ + rcu_switch(prev, next); switch_to(prev, next, prev); barrier(); -- cgit v1.2.3 From 90a340ed53f0f3bcc3fdf1b2cff56c0e4e911d01 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 11 Jul 2012 20:26:36 +0200 Subject: rcu: Exit RCU extended QS on kernel preemption after irq/exception When an exception or an irq exits, and we are going to resume into interrupted kernel code, the low level architecture code calls preempt_schedule_irq() if there is a need to reschedule. If the interrupt/exception occured between a call to rcu_user_enter() (from syscall exit, exception exit, do_notify_resume exit, ...) and a real resume to userspace (iret,...), preempt_schedule_irq() can be called whereas RCU thinks we are in userspace. But preempt_schedule_irq() is going to run kernel code and may be some RCU read side critical section. We must exit the userspace extended quiescent state before we call it. To solve this, just call rcu_user_exit() in the beginning of preempt_schedule_irq(). Signed-off-by: Frederic Weisbecker Cc: Alessio Igor Bogani Cc: Andrew Morton Cc: Avi Kivity Cc: Chris Metcalf Cc: Christoph Lameter Cc: Geoff Levand Cc: Gilad Ben Yossef Cc: Hakan Akkan Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Josh Triplett Cc: Kevin Hilman Cc: Max Krasnyansky Cc: Peter Zijlstra Cc: Stephen Hemminger Cc: Steven Rostedt Cc: Sven-Thorsten Dietrich Cc: Thomas Gleixner Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ea2213b07d9d..4adcd237c545 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3570,6 +3570,7 @@ asmlinkage void __sched preempt_schedule_irq(void) /* Catch callers which need to be fixed */ BUG_ON(ti->preempt_count || !irqs_disabled()); + rcu_user_exit(); do { add_preempt_count(PREEMPT_ACTIVE); local_irq_enable(); -- cgit v1.2.3 From 20ab65e33f469c35f3dabde3445b668aa9c943ee Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 11 Jul 2012 20:26:37 +0200 Subject: rcu: Exit RCU extended QS on user preemption When exceptions or irq are about to resume userspace, if the task needs to be rescheduled, the arch low level code calls schedule() directly. If we call it, it is because we have the TIF_RESCHED flag: - It can be set after random local calls to set_need_resched() (RCU, drm, ...) - A wake up happened and the CPU needs preemption. This can happen in several ways: * Remotely: the remote waking CPU has set TIF_RESCHED and send the wakee an IPI to schedule the new task. * Remotely enqueued: the remote waking CPU sends an IPI to the target and the wake up is made by the target. * Locally: waking CPU == wakee CPU and the wakeup is done locally. set_need_resched() is called without IPI. In the case of local and remotely enqueued wake ups, the tick can be restarted when we enqueue the new task and RCU can exit the extended quiescent state at the same time. Then by the time we reach irq exit path and we call schedule, we are not in RCU user mode. But if we call schedule() only because something called set_need_resched(), RCU may still be in user mode when we reach schedule. Also if a wake up is done remotely, the CPU might see the TIF_RESCHED flag and call schedule while the IPI has not yet happen to restart the tick and exit RCU user mode. We need to manually protect against these corner cases. Create a new API schedule_user() that calls schedule() inside rcu_user_exit()-rcu_user_enter() in order to protect it. Archs will need to rely on it now to implement user preemption safely. Signed-off-by: Frederic Weisbecker Cc: Alessio Igor Bogani Cc: Andrew Morton Cc: Avi Kivity Cc: Chris Metcalf Cc: Christoph Lameter Cc: Geoff Levand Cc: Gilad Ben Yossef Cc: Hakan Akkan Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Josh Triplett Cc: Kevin Hilman Cc: Max Krasnyansky Cc: Peter Zijlstra Cc: Stephen Hemminger Cc: Steven Rostedt Cc: Sven-Thorsten Dietrich Cc: Thomas Gleixner Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- kernel/sched/core.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4adcd237c545..3c4dec0594d6 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3469,6 +3469,21 @@ asmlinkage void __sched schedule(void) } EXPORT_SYMBOL(schedule); +#ifdef CONFIG_RCU_USER_QS +asmlinkage void __sched schedule_user(void) +{ + /* + * If we come here after a random call to set_need_resched(), + * or we have been woken up remotely but the IPI has not yet arrived, + * we haven't yet exited the RCU idle mode. Do it here manually until + * we find a better solution. + */ + rcu_user_exit(); + schedule(); + rcu_user_enter(); +} +#endif + /** * schedule_preempt_disabled - called with preemption disabled * -- cgit v1.2.3