From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756033AbaKTH5O (ORCPT ); Thu, 20 Nov 2014 02:57:14 -0500 Received: from mail-wg0-f54.google.com ([74.125.82.54]:34703 "EHLO mail-wg0-f54.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751727AbaKTH5M (ORCPT ); Thu, 20 Nov 2014 02:57:12 -0500 Date: Thu, 20 Nov 2014 08:57:06 +0100 From: Ingo Molnar To: Linus Torvalds Cc: linux-kernel@vger.kernel.org, Peter Zijlstra , Thomas Gleixner , Andrew Morton Subject: [GIT PULL] scheduler fixes Message-ID: <20141120075706.GA4424@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Linus, Please pull the latest sched-urgent-for-linus git tree from: git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched-urgent-for-linus # HEAD: 6e998916dfe327e785e7c2447959b2c1a3ea4930 sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency Misc fixes: two NUMA fixes, two cputime fixes and an RCU/lockdep fix. Thanks, Ingo ------------------> Andrey Ryabinin (1): sched/numa: Fix out of bounds read in sched_init_numa() Kirill Tkhai (1): sched: Remove lockdep check in sched_move_task() Peter Zijlstra (2): sched/numa: Avoid selecting oneself as swap target sched/cputime: Fix cpu_timer_sample_group() double accounting Stanislaw Gruszka (1): sched/cputime: Fix clock_nanosleep()/clock_gettime() inconsistency include/linux/kernel_stat.h | 5 ---- kernel/sched/core.c | 63 ++++++++++++++---------------------------- kernel/sched/deadline.c | 2 ++ kernel/sched/fair.c | 14 ++++++++++ kernel/sched/rt.c | 2 ++ kernel/sched/sched.h | 2 ++ kernel/time/posix-cpu-timers.c | 2 +- 7 files changed, 42 insertions(+), 48 deletions(-) diff --git a/include/linux/kernel_stat.h b/include/linux/kernel_stat.h index 8422b4ed6882..b9376cd5a187 100644 --- a/include/linux/kernel_stat.h +++ b/include/linux/kernel_stat.h @@ -77,11 +77,6 @@ static inline unsigned int kstat_cpu_irqs_sum(unsigned int cpu) return kstat_cpu(cpu).irqs_sum; } -/* - * Lock/unlock the current runqueue - to extract task statistics: - */ -extern unsigned long long task_delta_exec(struct task_struct *); - extern void account_user_time(struct task_struct *, cputime_t, cputime_t); extern void account_system_time(struct task_struct *, int, cputime_t, cputime_t); extern void account_steal_time(cputime_t); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 240157c13ddc..24beb9bb4c3e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2475,44 +2475,6 @@ EXPORT_PER_CPU_SYMBOL(kstat); EXPORT_PER_CPU_SYMBOL(kernel_cpustat); /* - * Return any ns on the sched_clock that have not yet been accounted in - * @p in case that task is currently running. - * - * Called with task_rq_lock() held on @rq. - */ -static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) -{ - u64 ns = 0; - - /* - * Must be ->curr _and_ ->on_rq. If dequeued, we would - * project cycles that may never be accounted to this - * thread, breaking clock_gettime(). - */ - if (task_current(rq, p) && task_on_rq_queued(p)) { - update_rq_clock(rq); - ns = rq_clock_task(rq) - p->se.exec_start; - if ((s64)ns < 0) - ns = 0; - } - - return ns; -} - -unsigned long long task_delta_exec(struct task_struct *p) -{ - unsigned long flags; - struct rq *rq; - u64 ns = 0; - - rq = task_rq_lock(p, &flags); - ns = do_task_delta_exec(p, rq); - task_rq_unlock(rq, p, &flags); - - return ns; -} - -/* * Return accounted runtime for the task. * In case the task is currently running, return the runtime plus current's * pending runtime that have not been accounted yet. @@ -2521,7 +2483,7 @@ unsigned long long task_sched_runtime(struct task_struct *p) { unsigned long flags; struct rq *rq; - u64 ns = 0; + u64 ns; #if defined(CONFIG_64BIT) && defined(CONFIG_SMP) /* @@ -2540,7 +2502,16 @@ unsigned long long task_sched_runtime(struct task_struct *p) #endif rq = task_rq_lock(p, &flags); - ns = p->se.sum_exec_runtime + do_task_delta_exec(p, rq); + /* + * Must be ->curr _and_ ->on_rq. If dequeued, we would + * project cycles that may never be accounted to this + * thread, breaking clock_gettime(). + */ + if (task_current(rq, p) && task_on_rq_queued(p)) { + update_rq_clock(rq); + p->sched_class->update_curr(rq); + } + ns = p->se.sum_exec_runtime; task_rq_unlock(rq, p, &flags); return ns; @@ -6368,6 +6339,10 @@ static void sched_init_numa(void) if (!sched_debug()) break; } + + if (!level) + return; + /* * 'level' contains the number of unique distances, excluding the * identity distance node_distance(i,i). @@ -7444,8 +7419,12 @@ void sched_move_task(struct task_struct *tsk) if (unlikely(running)) put_prev_task(rq, tsk); - tg = container_of(task_css_check(tsk, cpu_cgrp_id, - lockdep_is_held(&tsk->sighand->siglock)), + /* + * All callers are synchronized by task_rq_lock(); we do not use RCU + * which is pointless here. Thus, we pass "true" to task_css_check() + * to prevent lockdep warnings. + */ + tg = container_of(task_css_check(tsk, cpu_cgrp_id, true), struct task_group, css); tg = autogroup_task_group(tsk, tg); tsk->sched_task_group = tg; diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 5285332392d5..28fa9d9e9201 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -1701,4 +1701,6 @@ const struct sched_class dl_sched_class = { .prio_changed = prio_changed_dl, .switched_from = switched_from_dl, .switched_to = switched_to_dl, + + .update_curr = update_curr_dl, }; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 34baa60f8a7b..ef2b104b254c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -726,6 +726,11 @@ static void update_curr(struct cfs_rq *cfs_rq) account_cfs_rq_runtime(cfs_rq, delta_exec); } +static void update_curr_fair(struct rq *rq) +{ + update_curr(cfs_rq_of(&rq->curr->se)); +} + static inline void update_stats_wait_start(struct cfs_rq *cfs_rq, struct sched_entity *se) { @@ -1180,6 +1185,13 @@ static void task_numa_compare(struct task_numa_env *env, raw_spin_unlock_irq(&dst_rq->lock); /* + * Because we have preemption enabled we can get migrated around and + * end try selecting ourselves (current == env->p) as a swap candidate. + */ + if (cur == env->p) + goto unlock; + + /* * "imp" is the fault differential for the source task between the * source and destination node. Calculate the total differential for * the source task and potential destination task. The more negative @@ -7949,6 +7961,8 @@ const struct sched_class fair_sched_class = { .get_rr_interval = get_rr_interval_fair, + .update_curr = update_curr_fair, + #ifdef CONFIG_FAIR_GROUP_SCHED .task_move_group = task_move_group_fair, #endif diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index d024e6ce30ba..20bca398084a 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -2128,6 +2128,8 @@ const struct sched_class rt_sched_class = { .prio_changed = prio_changed_rt, .switched_to = switched_to_rt, + + .update_curr = update_curr_rt, }; #ifdef CONFIG_SCHED_DEBUG diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 24156c8434d1..2df8ef067cc5 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1135,6 +1135,8 @@ struct sched_class { unsigned int (*get_rr_interval) (struct rq *rq, struct task_struct *task); + void (*update_curr) (struct rq *rq); + #ifdef CONFIG_FAIR_GROUP_SCHED void (*task_move_group) (struct task_struct *p, int on_rq); #endif diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 492b986195d5..a16b67859e2a 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -553,7 +553,7 @@ static int cpu_timer_sample_group(const clockid_t which_clock, *sample = cputime_to_expires(cputime.utime); break; case CPUCLOCK_SCHED: - *sample = cputime.sum_exec_runtime + task_delta_exec(p); + *sample = cputime.sum_exec_runtime; break; } return 0;