From 7539a3b3d1f892dd97eaf094134d7de55c13befe Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sun, 13 Dec 2009 00:07:30 +0100 Subject: sched: Make wakeup side and atomic variants of completion API irq safe Alan Stern noticed that all the wakeup side (and atomic) variants of the completion APIs should be irq safe, but the newly introduced completion_done() and try_wait_for_completion() aren't. The use of the irq unsafe variants in IRQ contexts can cause crashes/hangs. Fix the problem by making them use spin_lock_irqsave() and spin_lock_irqrestore(). Reported-by: Alan Stern Signed-off-by: Rafael J. Wysocki Cc: Linus Torvalds Cc: Zhang Rui Cc: pm list Cc: Peter Zijlstra Cc: David Chinner Cc: Lachlan McIlroy LKML-Reference: <200912130007.30541.rjw@sisk.pl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index ff39cadf621e..8b3532f262d7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5908,14 +5908,15 @@ EXPORT_SYMBOL(wait_for_completion_killable); */ bool try_wait_for_completion(struct completion *x) { + unsigned long flags; int ret = 1; - spin_lock_irq(&x->wait.lock); + spin_lock_irqsave(&x->wait.lock, flags); if (!x->done) ret = 0; else x->done--; - spin_unlock_irq(&x->wait.lock); + spin_unlock_irqrestore(&x->wait.lock, flags); return ret; } EXPORT_SYMBOL(try_wait_for_completion); @@ -5930,12 +5931,13 @@ EXPORT_SYMBOL(try_wait_for_completion); */ bool completion_done(struct completion *x) { + unsigned long flags; int ret = 1; - spin_lock_irq(&x->wait.lock); + spin_lock_irqsave(&x->wait.lock, flags); if (!x->done) ret = 0; - spin_unlock_irq(&x->wait.lock); + spin_unlock_irqrestore(&x->wait.lock, flags); return ret; } EXPORT_SYMBOL(completion_done); -- cgit v1.2.2 From 663997d417330a59a566452f52cfa04c8ffd190b Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sat, 12 Dec 2009 13:57:27 -0800 Subject: sched: Use pr_fmt() and pr_() - Convert printk(KERN_ to pr_ (not KERN_DEBUG) - Add #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - Coalesce long format strings - Add missing \n to "ERROR: !SD_LOAD_BALANCE domain has parent" Signed-off-by: Joe Perches Cc: Peter Zijlstra LKML-Reference: <1260655047.2637.7.camel@Joe-Laptop.home> Signed-off-by: Ingo Molnar --- kernel/sched.c | 94 ++++++++++++++++++++++++++-------------------------------- 1 file changed, 42 insertions(+), 52 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 8b3532f262d7..258c73c6a2f3 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -26,6 +26,8 @@ * Thomas Gleixner, Mike Kravetz */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include #include #include @@ -5337,8 +5339,8 @@ static noinline void __schedule_bug(struct task_struct *prev) { struct pt_regs *regs = get_irq_regs(); - printk(KERN_ERR "BUG: scheduling while atomic: %s/%d/0x%08x\n", - prev->comm, prev->pid, preempt_count()); + pr_err("BUG: scheduling while atomic: %s/%d/0x%08x\n", + prev->comm, prev->pid, preempt_count()); debug_show_held_locks(prev); print_modules(); @@ -6906,23 +6908,23 @@ void sched_show_task(struct task_struct *p) unsigned state; state = p->state ? __ffs(p->state) + 1 : 0; - printk(KERN_INFO "%-13.13s %c", p->comm, + pr_info("%-13.13s %c", p->comm, state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'); #if BITS_PER_LONG == 32 if (state == TASK_RUNNING) - printk(KERN_CONT " running "); + pr_cont(" running "); else - printk(KERN_CONT " %08lx ", thread_saved_pc(p)); + pr_cont(" %08lx ", thread_saved_pc(p)); #else if (state == TASK_RUNNING) - printk(KERN_CONT " running task "); + pr_cont(" running task "); else - printk(KERN_CONT " %016lx ", thread_saved_pc(p)); + pr_cont(" %016lx ", thread_saved_pc(p)); #endif #ifdef CONFIG_DEBUG_STACK_USAGE free = stack_not_used(p); #endif - printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free, + pr_cont("%5lu %5d %6d 0x%08lx\n", free, task_pid_nr(p), task_pid_nr(p->real_parent), (unsigned long)task_thread_info(p)->flags); @@ -6934,11 +6936,9 @@ void show_state_filter(unsigned long state_filter) struct task_struct *g, *p; #if BITS_PER_LONG == 32 - printk(KERN_INFO - " task PC stack pid father\n"); + pr_info(" task PC stack pid father\n"); #else - printk(KERN_INFO - " task PC stack pid father\n"); + pr_info(" task PC stack pid father\n"); #endif read_lock(&tasklist_lock); do_each_thread(g, p) { @@ -7296,9 +7296,8 @@ again: * leave kernel. */ if (p->mm && printk_ratelimit()) { - printk(KERN_INFO "process %d (%s) no " - "longer affine to cpu%d\n", - task_pid_nr(p), p->comm, dead_cpu); + pr_info("process %d (%s) no longer affine to cpu%d\n", + task_pid_nr(p), p->comm, dead_cpu); } } @@ -7805,48 +7804,44 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, printk(KERN_DEBUG "%*s domain %d: ", level, "", level); if (!(sd->flags & SD_LOAD_BALANCE)) { - printk("does not load-balance\n"); + pr_cont("does not load-balance\n"); if (sd->parent) - printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain" - " has parent"); + pr_err("ERROR: !SD_LOAD_BALANCE domain has parent\n"); return -1; } - printk(KERN_CONT "span %s level %s\n", str, sd->name); + pr_cont("span %s level %s\n", str, sd->name); if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { - printk(KERN_ERR "ERROR: domain->span does not contain " - "CPU%d\n", cpu); + pr_err("ERROR: domain->span does not contain CPU%d\n", cpu); } if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) { - printk(KERN_ERR "ERROR: domain->groups does not contain" - " CPU%d\n", cpu); + pr_err("ERROR: domain->groups does not contain CPU%d\n", cpu); } printk(KERN_DEBUG "%*s groups:", level + 1, ""); do { if (!group) { - printk("\n"); - printk(KERN_ERR "ERROR: group is NULL\n"); + pr_cont("\n"); + pr_err("ERROR: group is NULL\n"); break; } if (!group->cpu_power) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: domain->cpu_power not " - "set\n"); + pr_cont("\n"); + pr_err("ERROR: domain->cpu_power not set\n"); break; } if (!cpumask_weight(sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: empty group\n"); + pr_cont("\n"); + pr_err("ERROR: empty group\n"); break; } if (cpumask_intersects(groupmask, sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: repeated CPUs\n"); + pr_cont("\n"); + pr_err("ERROR: repeated CPUs\n"); break; } @@ -7854,23 +7849,21 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); - printk(KERN_CONT " %s", str); + pr_cont(" %s", str); if (group->cpu_power != SCHED_LOAD_SCALE) { - printk(KERN_CONT " (cpu_power = %d)", - group->cpu_power); + pr_cont(" (cpu_power = %d)", group->cpu_power); } group = group->next; } while (group != sd->groups); - printk(KERN_CONT "\n"); + pr_cont("\n"); if (!cpumask_equal(sched_domain_span(sd), groupmask)) - printk(KERN_ERR "ERROR: groups don't span domain->span\n"); + pr_err("ERROR: groups don't span domain->span\n"); if (sd->parent && !cpumask_subset(groupmask, sched_domain_span(sd->parent))) - printk(KERN_ERR "ERROR: parent span is not a superset " - "of domain->span\n"); + pr_err("ERROR: parent span is not a superset of domain->span\n"); return 0; } @@ -8426,8 +8419,7 @@ static int build_numa_sched_groups(struct s_data *d, sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), GFP_KERNEL, num); if (!sg) { - printk(KERN_WARNING "Can not alloc domain group for node %d\n", - num); + pr_warning("Can not alloc domain group for node %d\n", num); return -ENOMEM; } d->sched_group_nodes[num] = sg; @@ -8456,8 +8448,8 @@ static int build_numa_sched_groups(struct s_data *d, sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), GFP_KERNEL, num); if (!sg) { - printk(KERN_WARNING - "Can not alloc domain group for node %d\n", j); + pr_warning("Can not alloc domain group for node %d\n", + j); return -ENOMEM; } sg->cpu_power = 0; @@ -8685,7 +8677,7 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, d->sched_group_nodes = kcalloc(nr_node_ids, sizeof(struct sched_group *), GFP_KERNEL); if (!d->sched_group_nodes) { - printk(KERN_WARNING "Can not alloc sched group node list\n"); + pr_warning("Can not alloc sched group node list\n"); return sa_notcovered; } sched_group_nodes_bycpu[cpumask_first(cpu_map)] = d->sched_group_nodes; @@ -8702,7 +8694,7 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, return sa_send_covered; d->rd = alloc_rootdomain(); if (!d->rd) { - printk(KERN_WARNING "Cannot alloc root domain\n"); + pr_warning("Cannot alloc root domain\n"); return sa_tmpmask; } return sa_rootdomain; @@ -9684,13 +9676,11 @@ void __might_sleep(char *file, int line, int preempt_offset) return; prev_jiffy = jiffies; - printk(KERN_ERR - "BUG: sleeping function called from invalid context at %s:%d\n", - file, line); - printk(KERN_ERR - "in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", - in_atomic(), irqs_disabled(), - current->pid, current->comm); + pr_err("BUG: sleeping function called from invalid context at %s:%d\n", + file, line); + pr_err("in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), + current->pid, current->comm); debug_show_held_locks(current); if (irqs_disabled()) -- cgit v1.2.2 From 5fe85be081edf0ac92d83f9c39e0ab5c1371eb82 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 Dec 2009 10:14:58 +0000 Subject: sched: Use rcu in sys_sched_getscheduler/sys_sched_getparam() read_lock(&tasklist_lock) does not protect sys_sched_getscheduler and sys_sched_getparam() against a concurrent update of the policy or scheduler parameters as do_sched_setscheduler() does not take the tasklist_lock. The accessed integers can be retrieved w/o locking and are snapshots anyway. Using rcu_read_lock() to protect find_task_by_vpid() and prevent the task struct from going away is not changing the above situation. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra LKML-Reference: <20091209100706.753790977@linutronix.de> Signed-off-by: Ingo Molnar --- kernel/sched.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 258c73c6a2f3..1782beed2fa7 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6458,7 +6458,7 @@ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) return -EINVAL; retval = -ESRCH; - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_process_by_pid(pid); if (p) { retval = security_task_getscheduler(p); @@ -6466,7 +6466,7 @@ SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) retval = p->policy | (p->sched_reset_on_fork ? SCHED_RESET_ON_FORK : 0); } - read_unlock(&tasklist_lock); + rcu_read_unlock(); return retval; } @@ -6484,7 +6484,7 @@ SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) if (!param || pid < 0) return -EINVAL; - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_process_by_pid(pid); retval = -ESRCH; if (!p) @@ -6495,7 +6495,7 @@ SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) goto out_unlock; lp.sched_priority = p->rt_priority; - read_unlock(&tasklist_lock); + rcu_read_unlock(); /* * This one might sleep, we cannot do it with a spinlock held ... @@ -6505,7 +6505,7 @@ SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) return retval; out_unlock: - read_unlock(&tasklist_lock); + rcu_read_unlock(); return retval; } -- cgit v1.2.2 From 23f5d142519621b16cf2b378cf8adf4dcf01a616 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 Dec 2009 10:15:01 +0000 Subject: sched: Use rcu in sched_get/set_affinity() tasklist_lock is held read locked to protect the find_task_by_vpid() call and to prevent the task going away. sched_setaffinity acquires a task struct ref and drops tasklist lock right away. The access to the cpus_allowed mask is protected by rq->lock. rcu_read_lock() provides the same protection here. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra LKML-Reference: <20091209100706.789059966@linutronix.de> Signed-off-by: Ingo Molnar --- kernel/sched.c | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 1782beed2fa7..79893123325c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6516,22 +6516,18 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) int retval; get_online_cpus(); - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_process_by_pid(pid); if (!p) { - read_unlock(&tasklist_lock); + rcu_read_unlock(); put_online_cpus(); return -ESRCH; } - /* - * It is not safe to call set_cpus_allowed with the - * tasklist_lock held. We will bump the task_struct's - * usage count and then drop tasklist_lock. - */ + /* Prevent p going away */ get_task_struct(p); - read_unlock(&tasklist_lock); + rcu_read_unlock(); if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) { retval = -ENOMEM; @@ -6617,7 +6613,7 @@ long sched_getaffinity(pid_t pid, struct cpumask *mask) int retval; get_online_cpus(); - read_lock(&tasklist_lock); + rcu_read_lock(); retval = -ESRCH; p = find_process_by_pid(pid); @@ -6633,7 +6629,7 @@ long sched_getaffinity(pid_t pid, struct cpumask *mask) task_rq_unlock(rq, &flags); out_unlock: - read_unlock(&tasklist_lock); + rcu_read_unlock(); put_online_cpus(); return retval; -- cgit v1.2.2 From 1a551ae715825bb2a2107a2dd68de024a1fa4e32 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 9 Dec 2009 10:15:11 +0000 Subject: sched: Use rcu in sched_get_rr_param() read_lock(&tasklist_lock) does not protect sys_sched_get_rr_param() against a concurrent update of the policy or scheduler parameters as do_sched_scheduler() does not take the tasklist_lock. The access to task->sched_class->get_rr_interval is protected by task_rq_lock(task). Use rcu_read_lock() to protect find_task_by_vpid() and prevent the task struct from going away. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra LKML-Reference: <20091209100706.862897167@linutronix.de> Signed-off-by: Ingo Molnar --- kernel/sched.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 79893123325c..db5c26692dd5 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6873,7 +6873,7 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, return -EINVAL; retval = -ESRCH; - read_lock(&tasklist_lock); + rcu_read_lock(); p = find_process_by_pid(pid); if (!p) goto out_unlock; @@ -6886,13 +6886,13 @@ SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, time_slice = p->sched_class->get_rr_interval(rq, p); task_rq_unlock(rq, &flags); - read_unlock(&tasklist_lock); + rcu_read_unlock(); jiffies_to_timespec(time_slice, &t); retval = copy_to_user(interval, &t, sizeof(t)) ? -EFAULT : 0; return retval; out_unlock: - read_unlock(&tasklist_lock); + rcu_read_unlock(); return retval; } -- cgit v1.2.2 From c2f21ce2e31286a0a32f8da0a7856e9ca1122ef3 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 2 Dec 2009 20:02:59 +0100 Subject: locking: Implement new raw_spinlock Now that the raw_spin name space is freed up, we can implement raw_spinlock and the related functions which are used to annotate the locks which are not converted to sleeping spinlocks in preempt-rt. A side effect is that only such locks can be used with the low level lock fsunctions which circumvent lockdep. For !rt spin_* functions are mapped to the raw_spin* implementations. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index fd05861b2111..e6acf2d7b753 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -884,7 +884,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) { #ifdef CONFIG_DEBUG_SPINLOCK /* this is a valid case when another task releases the spinlock */ - rq->lock.owner = current; + rq->lock.rlock.owner = current; #endif /* * If we are tracking spinlock dependencies then we have to -- cgit v1.2.2 From 9828ea9d75c38fe3dce05d00566eed61c85732e6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 3 Dec 2009 20:55:53 +0100 Subject: locking: Further name space cleanups The name space hierarchy for the internal lock functions is now a bit backwards. raw_spin* functions map to _spin* which use __spin*, while we would like to have _raw_spin* and __raw_spin*. _raw_spin* is already used by lock debugging, so rename those funtions to do_raw_spin* to free up the _raw_spin* name space. No functional change. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index e6acf2d7b753..91c65dd91435 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6684,7 +6684,7 @@ SYSCALL_DEFINE0(sched_yield) */ __release(rq->lock); spin_release(&rq->lock.dep_map, 1, _THIS_IP_); - _raw_spin_unlock(&rq->lock); + do_raw_spin_unlock(&rq->lock); preempt_enable_no_resched(); schedule(); -- cgit v1.2.2 From 05fa785cf80c9b7c0254c3056037147aed3ea16b Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 14:28:38 +0100 Subject: sched: Convert rq->lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 183 +++++++++++++++++++++++++++++---------------------------- 1 file changed, 93 insertions(+), 90 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 91c65dd91435..3dbe6178ebfd 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -525,7 +525,7 @@ static struct root_domain def_root_domain; */ struct rq { /* runqueue lock: */ - spinlock_t lock; + raw_spinlock_t lock; /* * nr_running and cpu_load should be in the same cacheline because @@ -685,7 +685,7 @@ inline void update_rq_clock(struct rq *rq) */ int runqueue_is_locked(int cpu) { - return spin_is_locked(&cpu_rq(cpu)->lock); + return raw_spin_is_locked(&cpu_rq(cpu)->lock); } /* @@ -884,7 +884,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) { #ifdef CONFIG_DEBUG_SPINLOCK /* this is a valid case when another task releases the spinlock */ - rq->lock.rlock.owner = current; + rq->lock.owner = current; #endif /* * If we are tracking spinlock dependencies then we have to @@ -893,7 +893,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) */ spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); } #else /* __ARCH_WANT_UNLOCKED_CTXSW */ @@ -917,9 +917,9 @@ static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next) next->oncpu = 1; #endif #ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); #else - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); #endif } @@ -949,10 +949,10 @@ static inline struct rq *__task_rq_lock(struct task_struct *p) { for (;;) { struct rq *rq = task_rq(p); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); if (likely(rq == task_rq(p))) return rq; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } } @@ -969,10 +969,10 @@ static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) for (;;) { local_irq_save(*flags); rq = task_rq(p); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); if (likely(rq == task_rq(p))) return rq; - spin_unlock_irqrestore(&rq->lock, *flags); + raw_spin_unlock_irqrestore(&rq->lock, *flags); } } @@ -981,19 +981,19 @@ void task_rq_unlock_wait(struct task_struct *p) struct rq *rq = task_rq(p); smp_mb(); /* spin-unlock-wait is not a full memory barrier */ - spin_unlock_wait(&rq->lock); + raw_spin_unlock_wait(&rq->lock); } static void __task_rq_unlock(struct rq *rq) __releases(rq->lock) { - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } static inline void task_rq_unlock(struct rq *rq, unsigned long *flags) __releases(rq->lock) { - spin_unlock_irqrestore(&rq->lock, *flags); + raw_spin_unlock_irqrestore(&rq->lock, *flags); } /* @@ -1006,7 +1006,7 @@ static struct rq *this_rq_lock(void) local_irq_disable(); rq = this_rq(); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); return rq; } @@ -1053,10 +1053,10 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); update_rq_clock(rq); rq->curr->sched_class->task_tick(rq, rq->curr, 1); - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); return HRTIMER_NORESTART; } @@ -1069,10 +1069,10 @@ static void __hrtick_start(void *arg) { struct rq *rq = arg; - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); hrtimer_restart(&rq->hrtick_timer); rq->hrtick_csd_pending = 0; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } /* @@ -1179,7 +1179,7 @@ static void resched_task(struct task_struct *p) { int cpu; - assert_spin_locked(&task_rq(p)->lock); + assert_raw_spin_locked(&task_rq(p)->lock); if (test_tsk_need_resched(p)) return; @@ -1201,10 +1201,10 @@ static void resched_cpu(int cpu) struct rq *rq = cpu_rq(cpu); unsigned long flags; - if (!spin_trylock_irqsave(&rq->lock, flags)) + if (!raw_spin_trylock_irqsave(&rq->lock, flags)) return; resched_task(cpu_curr(cpu)); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } #ifdef CONFIG_NO_HZ @@ -1273,7 +1273,7 @@ static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) #else /* !CONFIG_SMP */ static void resched_task(struct task_struct *p) { - assert_spin_locked(&task_rq(p)->lock); + assert_raw_spin_locked(&task_rq(p)->lock); set_tsk_need_resched(p); } @@ -1600,11 +1600,11 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, struct rq *rq = cpu_rq(cpu); unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); tg->cfs_rq[cpu]->rq_weight = boost ? 0 : rq_weight; tg->cfs_rq[cpu]->shares = boost ? 0 : shares; __set_se_shares(tg->se[cpu], shares); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } } @@ -1706,9 +1706,9 @@ static void update_shares_locked(struct rq *rq, struct sched_domain *sd) if (root_task_group_empty()) return; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); update_shares(sd); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); } static void update_h_load(long cpu) @@ -1748,7 +1748,7 @@ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) __acquires(busiest->lock) __acquires(this_rq->lock) { - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); double_rq_lock(this_rq, busiest); return 1; @@ -1769,14 +1769,16 @@ static int _double_lock_balance(struct rq *this_rq, struct rq *busiest) { int ret = 0; - if (unlikely(!spin_trylock(&busiest->lock))) { + if (unlikely(!raw_spin_trylock(&busiest->lock))) { if (busiest < this_rq) { - spin_unlock(&this_rq->lock); - spin_lock(&busiest->lock); - spin_lock_nested(&this_rq->lock, SINGLE_DEPTH_NESTING); + raw_spin_unlock(&this_rq->lock); + raw_spin_lock(&busiest->lock); + raw_spin_lock_nested(&this_rq->lock, + SINGLE_DEPTH_NESTING); ret = 1; } else - spin_lock_nested(&busiest->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock_nested(&busiest->lock, + SINGLE_DEPTH_NESTING); } return ret; } @@ -1790,7 +1792,7 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) { if (unlikely(!irqs_disabled())) { /* printk() doesn't work good under rq->lock */ - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); BUG_ON(1); } @@ -1800,7 +1802,7 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest) __releases(busiest->lock) { - spin_unlock(&busiest->lock); + raw_spin_unlock(&busiest->lock); lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_); } #endif @@ -2023,13 +2025,13 @@ void kthread_bind(struct task_struct *p, unsigned int cpu) return; } - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); update_rq_clock(rq); set_task_cpu(p, cpu); p->cpus_allowed = cpumask_of_cpu(cpu); p->rt.nr_cpus_allowed = 1; p->flags |= PF_THREAD_BOUND; - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } EXPORT_SYMBOL(kthread_bind); @@ -2781,10 +2783,10 @@ static inline void post_schedule(struct rq *rq) if (rq->post_schedule) { unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->curr->sched_class->post_schedule) rq->curr->sched_class->post_schedule(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); rq->post_schedule = 0; } @@ -3066,15 +3068,15 @@ static void double_rq_lock(struct rq *rq1, struct rq *rq2) { BUG_ON(!irqs_disabled()); if (rq1 == rq2) { - spin_lock(&rq1->lock); + raw_spin_lock(&rq1->lock); __acquire(rq2->lock); /* Fake it out ;) */ } else { if (rq1 < rq2) { - spin_lock(&rq1->lock); - spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&rq1->lock); + raw_spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING); } else { - spin_lock(&rq2->lock); - spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); + raw_spin_lock(&rq2->lock); + raw_spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); } } update_rq_clock(rq1); @@ -3091,9 +3093,9 @@ static void double_rq_unlock(struct rq *rq1, struct rq *rq2) __releases(rq1->lock) __releases(rq2->lock) { - spin_unlock(&rq1->lock); + raw_spin_unlock(&rq1->lock); if (rq1 != rq2) - spin_unlock(&rq2->lock); + raw_spin_unlock(&rq2->lock); else __release(rq2->lock); } @@ -4186,14 +4188,15 @@ redo: if (unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2)) { - spin_lock_irqsave(&busiest->lock, flags); + raw_spin_lock_irqsave(&busiest->lock, flags); /* don't kick the migration_thread, if the curr * task on busiest cpu can't be moved to this_cpu */ if (!cpumask_test_cpu(this_cpu, &busiest->curr->cpus_allowed)) { - spin_unlock_irqrestore(&busiest->lock, flags); + raw_spin_unlock_irqrestore(&busiest->lock, + flags); all_pinned = 1; goto out_one_pinned; } @@ -4203,7 +4206,7 @@ redo: busiest->push_cpu = this_cpu; active_balance = 1; } - spin_unlock_irqrestore(&busiest->lock, flags); + raw_spin_unlock_irqrestore(&busiest->lock, flags); if (active_balance) wake_up_process(busiest->migration_thread); @@ -4385,10 +4388,10 @@ redo: /* * Should not call ttwu while holding a rq->lock */ - spin_unlock(&this_rq->lock); + raw_spin_unlock(&this_rq->lock); if (active_balance) wake_up_process(busiest->migration_thread); - spin_lock(&this_rq->lock); + raw_spin_lock(&this_rq->lock); } else sd->nr_balance_failed = 0; @@ -5257,11 +5260,11 @@ void scheduler_tick(void) sched_clock_tick(); - spin_lock(&rq->lock); + raw_spin_lock(&rq->lock); update_rq_clock(rq); update_cpu_load(rq); curr->sched_class->task_tick(rq, curr, 0); - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); perf_event_task_tick(curr, cpu); @@ -5455,7 +5458,7 @@ need_resched_nonpreemptible: if (sched_feat(HRTICK)) hrtick_clear(rq); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); update_rq_clock(rq); clear_tsk_need_resched(prev); @@ -5491,7 +5494,7 @@ need_resched_nonpreemptible: cpu = smp_processor_id(); rq = cpu_rq(cpu); } else - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); post_schedule(rq); @@ -6980,7 +6983,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) struct rq *rq = cpu_rq(cpu); unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __sched_fork(idle); idle->se.exec_start = sched_clock(); @@ -6992,7 +6995,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) #if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW) idle->oncpu = 1; #endif - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); /* Set the preempt count _outside_ the spinlocks! */ #if defined(CONFIG_PREEMPT) @@ -7209,10 +7212,10 @@ static int migration_thread(void *data) struct migration_req *req; struct list_head *head; - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); if (cpu_is_offline(cpu)) { - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); break; } @@ -7224,7 +7227,7 @@ static int migration_thread(void *data) head = &rq->migration_queue; if (list_empty(head)) { - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); schedule(); set_current_state(TASK_INTERRUPTIBLE); continue; @@ -7233,14 +7236,14 @@ static int migration_thread(void *data) list_del_init(head->next); if (req->task != NULL) { - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); __migrate_task(req->task, cpu, req->dest_cpu); } else if (likely(cpu == (badcpu = smp_processor_id()))) { req->dest_cpu = RCU_MIGRATION_GOT_QS; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); } else { req->dest_cpu = RCU_MIGRATION_MUST_SYNC; - spin_unlock(&rq->lock); + raw_spin_unlock(&rq->lock); WARN_ONCE(1, "migration_thread() on CPU %d, expected %d\n", badcpu, cpu); } local_irq_enable(); @@ -7363,14 +7366,14 @@ void sched_idle_next(void) * Strictly not necessary since rest of the CPUs are stopped by now * and interrupts disabled on the current cpu. */ - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __setscheduler(rq, p, SCHED_FIFO, MAX_RT_PRIO-1); update_rq_clock(rq); activate_task(rq, p, 0); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } /* @@ -7406,9 +7409,9 @@ static void migrate_dead(unsigned int dead_cpu, struct task_struct *p) * that's OK. No task can be added to this CPU, so iteration is * fine. */ - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); move_task_off_dead_cpu(dead_cpu, p); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); put_task_struct(p); } @@ -7674,13 +7677,13 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) /* Update our root-domain */ rq = cpu_rq(cpu); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_online(rq); } - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); break; #ifdef CONFIG_HOTPLUG_CPU @@ -7705,13 +7708,13 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) put_task_struct(rq->migration_thread); rq->migration_thread = NULL; /* Idle task back to normal (off runqueue, low prio) */ - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); update_rq_clock(rq); deactivate_task(rq, rq->idle, 0); __setscheduler(rq, rq->idle, SCHED_NORMAL, 0); rq->idle->sched_class = &idle_sched_class; migrate_dead_tasks(cpu); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); cpuset_unlock(); migrate_nr_uninterruptible(rq); BUG_ON(rq->nr_running != 0); @@ -7721,30 +7724,30 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) * they didn't take sched_hotcpu_mutex. Just wake up * the requestors. */ - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); while (!list_empty(&rq->migration_queue)) { struct migration_req *req; req = list_entry(rq->migration_queue.next, struct migration_req, list); list_del_init(&req->list); - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); complete(&req->done); - spin_lock_irq(&rq->lock); + raw_spin_lock_irq(&rq->lock); } - spin_unlock_irq(&rq->lock); + raw_spin_unlock_irq(&rq->lock); break; case CPU_DYING: case CPU_DYING_FROZEN: /* Update our root-domain */ rq = cpu_rq(cpu); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); break; #endif } @@ -7974,7 +7977,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) struct root_domain *old_rd = NULL; unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { old_rd = rq->rd; @@ -8000,7 +8003,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); if (old_rd) free_rootdomain(old_rd); @@ -9357,7 +9360,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) #ifdef CONFIG_SMP rt_rq->rt_nr_migratory = 0; rt_rq->overloaded = 0; - plist_head_init(&rt_rq->pushable_tasks, &rq->lock); + plist_head_init_raw(&rt_rq->pushable_tasks, &rq->lock); #endif rt_rq->rt_time = 0; @@ -9523,7 +9526,7 @@ void __init sched_init(void) struct rq *rq; rq = cpu_rq(i); - spin_lock_init(&rq->lock); + raw_spin_lock_init(&rq->lock); rq->nr_running = 0; rq->calc_load_active = 0; rq->calc_load_update = jiffies + LOAD_FREQ; @@ -10115,9 +10118,9 @@ static void set_se_shares(struct sched_entity *se, unsigned long shares) struct rq *rq = cfs_rq->rq; unsigned long flags; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); __set_se_shares(se, shares); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } static DEFINE_MUTEX(shares_mutex); @@ -10717,9 +10720,9 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu) /* * Take rq->lock to make 64-bit read safe on 32-bit platforms. */ - spin_lock_irq(&cpu_rq(cpu)->lock); + raw_spin_lock_irq(&cpu_rq(cpu)->lock); data = *cpuusage; - spin_unlock_irq(&cpu_rq(cpu)->lock); + raw_spin_unlock_irq(&cpu_rq(cpu)->lock); #else data = *cpuusage; #endif @@ -10735,9 +10738,9 @@ static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val) /* * Take rq->lock to make 64-bit write safe on 32-bit platforms. */ - spin_lock_irq(&cpu_rq(cpu)->lock); + raw_spin_lock_irq(&cpu_rq(cpu)->lock); *cpuusage = val; - spin_unlock_irq(&cpu_rq(cpu)->lock); + raw_spin_unlock_irq(&cpu_rq(cpu)->lock); #else *cpuusage = val; #endif @@ -10971,9 +10974,9 @@ void synchronize_sched_expedited(void) init_completion(&req->done); req->task = NULL; req->dest_cpu = RCU_MIGRATION_NEED_QS; - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); list_add(&req->list, &rq->migration_queue); - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); wake_up_process(rq->migration_thread); } for_each_online_cpu(cpu) { @@ -10981,11 +10984,11 @@ void synchronize_sched_expedited(void) req = &per_cpu(rcu_migration_req, cpu); rq = cpu_rq(cpu); wait_for_completion(&req->done); - spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&rq->lock, flags); if (unlikely(req->dest_cpu == RCU_MIGRATION_MUST_SYNC)) need_full_sync = 1; req->dest_cpu = RCU_MIGRATION_IDLE; - spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_unlock_irqrestore(&rq->lock, flags); } rcu_expedited_state = RCU_EXPEDITED_STATE_IDLE; synchronize_sched_expedited_count++; -- cgit v1.2.2 From 0986b11b12ad2baed338385f79fd0dec58a23fc6 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 15:32:06 +0100 Subject: sched: Convert rt_runtime_lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 3dbe6178ebfd..01c5016e57f1 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -141,7 +141,7 @@ struct rt_prio_array { struct rt_bandwidth { /* nests inside the rq lock: */ - spinlock_t rt_runtime_lock; + raw_spinlock_t rt_runtime_lock; ktime_t rt_period; u64 rt_runtime; struct hrtimer rt_period_timer; @@ -178,7 +178,7 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime) rt_b->rt_period = ns_to_ktime(period); rt_b->rt_runtime = runtime; - spin_lock_init(&rt_b->rt_runtime_lock); + raw_spin_lock_init(&rt_b->rt_runtime_lock); hrtimer_init(&rt_b->rt_period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); @@ -200,7 +200,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) if (hrtimer_active(&rt_b->rt_period_timer)) return; - spin_lock(&rt_b->rt_runtime_lock); + raw_spin_lock(&rt_b->rt_runtime_lock); for (;;) { unsigned long delta; ktime_t soft, hard; @@ -217,7 +217,7 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) __hrtimer_start_range_ns(&rt_b->rt_period_timer, soft, delta, HRTIMER_MODE_ABS_PINNED, 0); } - spin_unlock(&rt_b->rt_runtime_lock); + raw_spin_unlock(&rt_b->rt_runtime_lock); } #ifdef CONFIG_RT_GROUP_SCHED @@ -470,7 +470,7 @@ struct rt_rq { u64 rt_time; u64 rt_runtime; /* Nests inside the rq lock: */ - spinlock_t rt_runtime_lock; + raw_spinlock_t rt_runtime_lock; #ifdef CONFIG_RT_GROUP_SCHED unsigned long rt_nr_boosted; @@ -9366,7 +9366,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) rt_rq->rt_time = 0; rt_rq->rt_throttled = 0; rt_rq->rt_runtime = 0; - spin_lock_init(&rt_rq->rt_runtime_lock); + raw_spin_lock_init(&rt_rq->rt_runtime_lock); #ifdef CONFIG_RT_GROUP_SCHED rt_rq->rt_nr_boosted = 0; @@ -10305,18 +10305,18 @@ static int tg_set_bandwidth(struct task_group *tg, if (err) goto unlock; - spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); + raw_spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); tg->rt_bandwidth.rt_runtime = rt_runtime; for_each_possible_cpu(i) { struct rt_rq *rt_rq = tg->rt_rq[i]; - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = rt_runtime; - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } - spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); + raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); unlock: read_unlock(&tasklist_lock); mutex_unlock(&rt_constraints_mutex); @@ -10421,15 +10421,15 @@ static int sched_rt_global_constraints(void) if (sysctl_sched_rt_runtime == 0) return -EBUSY; - spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); + raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); for_each_possible_cpu(i) { struct rt_rq *rt_rq = &cpu_rq(i)->rt; - spin_lock(&rt_rq->rt_runtime_lock); + raw_spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = global_rt_runtime(); - spin_unlock(&rt_rq->rt_runtime_lock); + raw_spin_unlock(&rt_rq->rt_runtime_lock); } - spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); + raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); return 0; } -- cgit v1.2.2 From 1d615482547584b9a8bb6316a58fed6ce90dd9ff Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 17 Nov 2009 14:54:03 +0100 Subject: sched: Convert pi_lock to raw_spinlock Convert locks which cannot be sleeping locks in preempt-rt to raw_spinlocks. Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra Acked-by: Ingo Molnar --- kernel/sched.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 01c5016e57f1..18cceeecce35 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6323,7 +6323,7 @@ recheck: * make sure no PI-waiters arrive (or leave) while we are * changing the priority of the task: */ - spin_lock_irqsave(&p->pi_lock, flags); + raw_spin_lock_irqsave(&p->pi_lock, flags); /* * To be able to change p->policy safely, the apropriate * runqueue lock must be held. @@ -6333,7 +6333,7 @@ recheck: if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; __task_rq_unlock(rq); - spin_unlock_irqrestore(&p->pi_lock, flags); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); goto recheck; } update_rq_clock(rq); @@ -6357,7 +6357,7 @@ recheck: check_class_changed(rq, p, prev_class, oldprio, running); } __task_rq_unlock(rq); - spin_unlock_irqrestore(&p->pi_lock, flags); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); rt_mutex_adjust_pi(p); @@ -9624,7 +9624,7 @@ void __init sched_init(void) #endif #ifdef CONFIG_RT_MUTEXES - plist_head_init(&init_task.pi_waiters, &init_task.pi_lock); + plist_head_init_raw(&init_task.pi_waiters, &init_task.pi_lock); #endif /* @@ -9749,13 +9749,13 @@ void normalize_rt_tasks(void) continue; } - spin_lock(&p->pi_lock); + raw_spin_lock(&p->pi_lock); rq = __task_rq_lock(p); normalize_task(rq, p); __task_rq_unlock(rq); - spin_unlock(&p->pi_lock); + raw_spin_unlock(&p->pi_lock); } while_each_thread(g, p); read_unlock_irqrestore(&tasklist_lock, flags); -- cgit v1.2.2 From e6c8fba7771563b2f3dfb96a78f36ec17e15bdf0 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:33 +0100 Subject: sched: Fix task_hot() test order Make sure not to access sched_fair fields before verifying it is indeed a sched_fair task. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith CC: stable@kernel.org LKML-Reference: <20091216170517.577998058@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 9c30858b6463..1d8ca25dd6fb 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2046,6 +2046,9 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) { s64 delta; + if (p->sched_class != &fair_sched_class) + return 0; + /* * Buddy candidates are cache hot: */ @@ -2054,9 +2057,6 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) &p->se == cfs_rq_of(&p->se)->last)) return 1; - if (p->sched_class != &fair_sched_class) - return 0; - if (sysctl_sched_migration_cost == -1) return 1; if (sysctl_sched_migration_cost == 0) -- cgit v1.2.2 From 06b83b5fbea273672822b6ee93e16781046553ec Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:35 +0100 Subject: sched: Use TASK_WAKING for fork wakups For later convenience use TASK_WAKING for fresh tasks. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170517.732561278@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 1d8ca25dd6fb..1672823aabfe 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2540,14 +2540,6 @@ static void __sched_fork(struct task_struct *p) #ifdef CONFIG_PREEMPT_NOTIFIERS INIT_HLIST_HEAD(&p->preempt_notifiers); #endif - - /* - * We mark the process as running here, but have not actually - * inserted it onto the runqueue yet. This guarantees that - * nobody will actually run it, and a signal or other external - * event cannot wake it up and insert it on the runqueue either. - */ - p->state = TASK_RUNNING; } /* @@ -2558,6 +2550,12 @@ void sched_fork(struct task_struct *p, int clone_flags) int cpu = get_cpu(); __sched_fork(p); + /* + * We mark the process as waking here. This guarantees that + * nobody will actually run it, and a signal or other external + * event cannot wake it up and insert it on the runqueue either. + */ + p->state = TASK_WAKING; /* * Revert to default priority/policy on fork if requested. @@ -2626,7 +2624,8 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) struct rq *rq; rq = task_rq_lock(p, &flags); - BUG_ON(p->state != TASK_RUNNING); + BUG_ON(p->state != TASK_WAKING); + p->state = TASK_RUNNING; update_rq_clock(rq); activate_task(rq, p, 0); trace_sched_wakeup_new(rq, p, 1); @@ -6984,6 +6983,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) raw_spin_lock_irqsave(&rq->lock, flags); __sched_fork(idle); + idle->state = TASK_RUNNING; idle->se.exec_start = sched_clock(); cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu)); -- cgit v1.2.2 From e2912009fb7b715728311b0d8fe327a1432b3f79 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:36 +0100 Subject: sched: Ensure set_task_cpu() is never called on blocked tasks In order to clean up the set_task_cpu() rq dependencies we need to ensure it is never called on blocked tasks because such usage does not pair with consistent rq->lock usage. This puts the migration burden on ttwu(). Furthermore we need to close a race against changing ->cpus_allowed, since select_task_rq() runs with only preemption disabled. For sched_fork() this is safe because the child isn't in the tasklist yet, for wakeup we fix this by synchronizing set_cpus_allowed_ptr() against TASK_WAKING, which leaves sched_exec to be a problem This also closes a hole in (6ad4c1888 sched: Fix balance vs hotplug race) where ->select_task_rq() doesn't validate the result against the sched_domain/root_domain. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170517.807938893@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 85 +++++++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 66 insertions(+), 19 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 1672823aabfe..33d7965f63f0 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2018,22 +2018,15 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, */ void kthread_bind(struct task_struct *p, unsigned int cpu) { - struct rq *rq = cpu_rq(cpu); - unsigned long flags; - /* Must have done schedule() in kthread() before we set_task_cpu */ if (!wait_task_inactive(p, TASK_UNINTERRUPTIBLE)) { WARN_ON(1); return; } - raw_spin_lock_irqsave(&rq->lock, flags); - update_rq_clock(rq); - set_task_cpu(p, cpu); p->cpus_allowed = cpumask_of_cpu(cpu); p->rt.nr_cpus_allowed = 1; p->flags |= PF_THREAD_BOUND; - raw_spin_unlock_irqrestore(&rq->lock, flags); } EXPORT_SYMBOL(kthread_bind); @@ -2074,6 +2067,14 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) struct cfs_rq *old_cfsrq = task_cfs_rq(p), *new_cfsrq = cpu_cfs_rq(old_cfsrq, new_cpu); +#ifdef CONFIG_SCHED_DEBUG + /* + * We should never call set_task_cpu() on a blocked task, + * ttwu() will sort out the placement. + */ + WARN_ON(p->state != TASK_RUNNING && p->state != TASK_WAKING); +#endif + trace_sched_migrate_task(p, new_cpu); if (old_cpu != new_cpu) { @@ -2107,13 +2108,10 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req) /* * If the task is not on a runqueue (and not running), then - * it is sufficient to simply update the task's cpu field. + * the next wake-up will properly place the task. */ - if (!p->se.on_rq && !task_running(rq, p)) { - update_rq_clock(rq); - set_task_cpu(p, dest_cpu); + if (!p->se.on_rq && !task_running(rq, p)) return 0; - } init_completion(&req->done); req->task = p; @@ -2319,10 +2317,42 @@ void task_oncpu_function_call(struct task_struct *p, } #ifdef CONFIG_SMP +/* + * Called from: + * + * - fork, @p is stable because it isn't on the tasklist yet + * + * - exec, @p is unstable XXX + * + * - wake-up, we serialize ->cpus_allowed against TASK_WAKING so + * we should be good. + */ static inline int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) { - return p->sched_class->select_task_rq(p, sd_flags, wake_flags); + int cpu = p->sched_class->select_task_rq(p, sd_flags, wake_flags); + + /* + * In order not to call set_task_cpu() on a blocking task we need + * to rely on ttwu() to place the task on a valid ->cpus_allowed + * cpu. + * + * Since this is common to all placement strategies, this lives here. + * + * [ this allows ->select_task() to simply return task_cpu(p) and + * not worry about this generic constraint ] + */ + if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) || + !cpu_active(cpu))) { + + cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); + /* + * XXX: race against hot-plug modifying cpu_active_mask + */ + BUG_ON(cpu >= nr_cpu_ids); + } + + return cpu; } #endif @@ -7098,7 +7128,23 @@ int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) struct rq *rq; int ret = 0; + /* + * Since we rely on wake-ups to migrate sleeping tasks, don't change + * the ->cpus_allowed mask from under waking tasks, which would be + * possible when we change rq->lock in ttwu(), so synchronize against + * TASK_WAKING to avoid that. + */ +again: + while (p->state == TASK_WAKING) + cpu_relax(); + rq = task_rq_lock(p, &flags); + + if (p->state == TASK_WAKING) { + task_rq_unlock(rq, &flags); + goto again; + } + if (!cpumask_intersects(new_mask, cpu_active_mask)) { ret = -EINVAL; goto out; @@ -7154,7 +7200,7 @@ EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr); static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) { struct rq *rq_dest, *rq_src; - int ret = 0, on_rq; + int ret = 0; if (unlikely(!cpu_active(dest_cpu))) return ret; @@ -7170,12 +7216,13 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) goto fail; - on_rq = p->se.on_rq; - if (on_rq) + /* + * If we're not on a rq, the next wake-up will ensure we're + * placed properly. + */ + if (p->se.on_rq) { deactivate_task(rq_src, p, 0); - - set_task_cpu(p, dest_cpu); - if (on_rq) { + set_task_cpu(p, dest_cpu); activate_task(rq_dest, p, 0); check_preempt_curr(rq_dest, p, 0); } -- cgit v1.2.2 From 3802290628348674985d14914f9bfee7b9084548 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:37 +0100 Subject: sched: Fix sched_exec() balancing Since we access ->cpus_allowed without holding rq->lock we need a retry loop to validate the result, this comes for near free when we merge sched_migrate_task() into sched_exec() since that already does the needed check. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170517.884743662@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 45 +++++++++++++++++++++++---------------------- 1 file changed, 23 insertions(+), 22 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 33d7965f63f0..63e55ac242d1 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2322,7 +2322,7 @@ void task_oncpu_function_call(struct task_struct *p, * * - fork, @p is stable because it isn't on the tasklist yet * - * - exec, @p is unstable XXX + * - exec, @p is unstable, retry loop * * - wake-up, we serialize ->cpus_allowed against TASK_WAKING so * we should be good. @@ -3132,21 +3132,36 @@ static void double_rq_unlock(struct rq *rq1, struct rq *rq2) } /* - * If dest_cpu is allowed for this process, migrate the task to it. - * This is accomplished by forcing the cpu_allowed mask to only - * allow dest_cpu, which will force the cpu onto dest_cpu. Then - * the cpu_allowed mask is restored. + * sched_exec - execve() is a valuable balancing opportunity, because at + * this point the task has the smallest effective memory and cache footprint. */ -static void sched_migrate_task(struct task_struct *p, int dest_cpu) +void sched_exec(void) { + struct task_struct *p = current; struct migration_req req; + int dest_cpu, this_cpu; unsigned long flags; struct rq *rq; +again: + this_cpu = get_cpu(); + dest_cpu = select_task_rq(p, SD_BALANCE_EXEC, 0); + if (dest_cpu == this_cpu) { + put_cpu(); + return; + } + rq = task_rq_lock(p, &flags); + put_cpu(); + + /* + * select_task_rq() can race against ->cpus_allowed + */ if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed) - || unlikely(!cpu_active(dest_cpu))) - goto out; + || unlikely(!cpu_active(dest_cpu))) { + task_rq_unlock(rq, &flags); + goto again; + } /* force the process onto the specified CPU */ if (migrate_task(p, dest_cpu, &req)) { @@ -3161,23 +3176,9 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu) return; } -out: task_rq_unlock(rq, &flags); } -/* - * sched_exec - execve() is a valuable balancing opportunity, because at - * this point the task has the smallest effective memory and cache footprint. - */ -void sched_exec(void) -{ - int new_cpu, this_cpu = get_cpu(); - new_cpu = select_task_rq(current, SD_BALANCE_EXEC, 0); - put_cpu(); - if (new_cpu != this_cpu) - sched_migrate_task(current, new_cpu); -} - /* * pull_task - move a task from a remote runqueue to the local runqueue. * Both runqueues must be locked. -- cgit v1.2.2 From 5da9a0fb673a0ea0a093862f95f6b89b3390c31e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:38 +0100 Subject: sched: Fix select_task_rq() vs hotplug issues Since select_task_rq() is now responsible for guaranteeing ->cpus_allowed and cpu_active_mask, we need to verify this. select_task_rq_rt() can blindly return smp_processor_id()/task_cpu() without checking the valid masks, select_task_rq_fair() can do the same in the rare case that all SD_flags are disabled. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170517.961475466@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 75 +++++++++++++++++++++++++++++++--------------------------- 1 file changed, 40 insertions(+), 35 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 63e55ac242d1..cc40bdadee7a 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2317,6 +2317,43 @@ void task_oncpu_function_call(struct task_struct *p, } #ifdef CONFIG_SMP +static int select_fallback_rq(int cpu, struct task_struct *p) +{ + int dest_cpu; + const struct cpumask *nodemask = cpumask_of_node(cpu_to_node(cpu)); + + /* Look for allowed, online CPU in same node. */ + for_each_cpu_and(dest_cpu, nodemask, cpu_active_mask) + if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) + return dest_cpu; + + /* Any allowed, online CPU? */ + dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); + if (dest_cpu < nr_cpu_ids) + return dest_cpu; + + /* No more Mr. Nice Guy. */ + if (dest_cpu >= nr_cpu_ids) { + rcu_read_lock(); + cpuset_cpus_allowed_locked(p, &p->cpus_allowed); + rcu_read_unlock(); + dest_cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed); + + /* + * Don't tell them about moving exiting tasks or + * kernel threads (both mm NULL), since they never + * leave kernel. + */ + if (p->mm && printk_ratelimit()) { + printk(KERN_INFO "process %d (%s) no " + "longer affine to cpu%d\n", + task_pid_nr(p), p->comm, cpu); + } + } + + return dest_cpu; +} + /* * Called from: * @@ -2343,14 +2380,8 @@ int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) * not worry about this generic constraint ] */ if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) || - !cpu_active(cpu))) { - - cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); - /* - * XXX: race against hot-plug modifying cpu_active_mask - */ - BUG_ON(cpu >= nr_cpu_ids); - } + !cpu_active(cpu))) + cpu = select_fallback_rq(task_cpu(p), p); return cpu; } @@ -7319,36 +7350,10 @@ static int __migrate_task_irq(struct task_struct *p, int src_cpu, int dest_cpu) static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) { int dest_cpu; - const struct cpumask *nodemask = cpumask_of_node(cpu_to_node(dead_cpu)); again: - /* Look for allowed, online CPU in same node. */ - for_each_cpu_and(dest_cpu, nodemask, cpu_active_mask) - if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) - goto move; - - /* Any allowed, online CPU? */ - dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_active_mask); - if (dest_cpu < nr_cpu_ids) - goto move; - - /* No more Mr. Nice Guy. */ - if (dest_cpu >= nr_cpu_ids) { - cpuset_cpus_allowed_locked(p, &p->cpus_allowed); - dest_cpu = cpumask_any_and(cpu_active_mask, &p->cpus_allowed); - - /* - * Don't tell them about moving exiting tasks or - * kernel threads (both mm NULL), since they never - * leave kernel. - */ - if (p->mm && printk_ratelimit()) { - pr_info("process %d (%s) no longer affine to cpu%d\n", - task_pid_nr(p), p->comm, dead_cpu); - } - } + dest_cpu = select_fallback_rq(dead_cpu, p); -move: /* It can have affinity changed while we were choosing. */ if (unlikely(!__migrate_task_irq(p, dead_cpu, dest_cpu))) goto again; -- cgit v1.2.2 From 881232b70b195768a71cd74ff4b4e8ab9502997b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:39 +0100 Subject: sched: Move kthread_bind() back to kthread.c Since kthread_bind() lost its dependencies on sched.c, move it back where it came from. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170518.039524041@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 26 -------------------------- 1 file changed, 26 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index cc40bdadee7a..297dc441ff96 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2004,32 +2004,6 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, p->sched_class->prio_changed(rq, p, oldprio, running); } -/** - * kthread_bind - bind a just-created kthread to a cpu. - * @p: thread created by kthread_create(). - * @cpu: cpu (might not be online, must be possible) for @k to run on. - * - * Description: This function is equivalent to set_cpus_allowed(), - * except that @cpu doesn't need to be online, and the thread must be - * stopped (i.e., just returned from kthread_create()). - * - * Function lives here instead of kthread.c because it messes with - * scheduler internals which require locking. - */ -void kthread_bind(struct task_struct *p, unsigned int cpu) -{ - /* Must have done schedule() in kthread() before we set_task_cpu */ - if (!wait_task_inactive(p, TASK_UNINTERRUPTIBLE)) { - WARN_ON(1); - return; - } - - p->cpus_allowed = cpumask_of_cpu(cpu); - p->rt.nr_cpus_allowed = 1; - p->flags |= PF_THREAD_BOUND; -} -EXPORT_SYMBOL(kthread_bind); - #ifdef CONFIG_SMP /* * Is this task likely cache-hot: -- cgit v1.2.2 From efbbd05a595343a413964ad85a2ad359b7b7efbd Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:40 +0100 Subject: sched: Add pre and post wakeup hooks As will be apparent in the next patch, we need a pre wakeup hook for sched_fair task migration, hence rename the post wakeup hook and one pre wakeup. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170518.114746117@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 297dc441ff96..6c571bdd5658 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2412,6 +2412,10 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, if (task_contributes_to_load(p)) rq->nr_uninterruptible--; p->state = TASK_WAKING; + + if (p->sched_class->task_waking) + p->sched_class->task_waking(rq, p); + __task_rq_unlock(rq); cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags); @@ -2475,8 +2479,8 @@ out_running: p->state = TASK_RUNNING; #ifdef CONFIG_SMP - if (p->sched_class->task_wake_up) - p->sched_class->task_wake_up(rq, p); + if (p->sched_class->task_woken) + p->sched_class->task_woken(rq, p); if (unlikely(rq->idle_stamp)) { u64 delta = rq->clock - rq->idle_stamp; @@ -2666,8 +2670,8 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) trace_sched_wakeup_new(rq, p, 1); check_preempt_curr(rq, p, WF_FORK); #ifdef CONFIG_SMP - if (p->sched_class->task_wake_up) - p->sched_class->task_wake_up(rq, p); + if (p->sched_class->task_woken) + p->sched_class->task_woken(rq, p); #endif task_rq_unlock(rq, &flags); } -- cgit v1.2.2 From 88ec22d3edb72b261f8628226cd543589a6d5e1b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:41 +0100 Subject: sched: Remove the cfs_rq dependency from set_task_cpu() In order to remove the cfs_rq dependency from set_task_cpu() we need to ensure the task is cfs_rq invariant for all callsites. The simple approach is to substract cfs_rq->min_vruntime from se->vruntime on dequeue, and add cfs_rq->min_vruntime on enqueue. However, this has the downside of breaking FAIR_SLEEPERS since we loose the old vruntime as we only maintain the relative position. To solve this, we observe that we only migrate runnable tasks, we do this using deactivate_task(.sleep=0) and activate_task(.wakeup=0), therefore we can restrain the min_vruntime invariance to that state. The only other case is wakeup balancing, since we want to maintain the old vruntime we cannot make it relative on dequeue, but since we don't migrate inactive tasks, we can do so right before we activate it again. This is where we need the new pre-wakeup hook, we need to call this while still holding the old rq->lock. We could fold it into ->select_task_rq(), but since that has multiple callsites and would obfuscate the locking requirements, that seems like a fudge. This leaves the fork() case, simply make sure that ->task_fork() leaves the ->vruntime in a relative state. This covers all cases where set_task_cpu() gets called, and ensures it sees a relative vruntime. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170518.191697025@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 6c571bdd5658..f92ce63edfff 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2038,8 +2038,6 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { int old_cpu = task_cpu(p); - struct cfs_rq *old_cfsrq = task_cfs_rq(p), - *new_cfsrq = cpu_cfs_rq(old_cfsrq, new_cpu); #ifdef CONFIG_SCHED_DEBUG /* @@ -2056,8 +2054,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0); } - p->se.vruntime -= old_cfsrq->min_vruntime - - new_cfsrq->min_vruntime; __set_task_cpu(p, new_cpu); } @@ -10102,7 +10098,7 @@ void sched_move_task(struct task_struct *tsk) #ifdef CONFIG_FAIR_GROUP_SCHED if (tsk->sched_class->moved_group) - tsk->sched_class->moved_group(tsk); + tsk->sched_class->moved_group(tsk, on_rq); #endif if (unlikely(running)) -- cgit v1.2.2 From 738d2be4301007f054541c5c4bf7fb6a361c9b3a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 16 Dec 2009 18:04:42 +0100 Subject: sched: Simplify set_task_cpu() Rearrange code a bit now that its a simpler function. Signed-off-by: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170518.269101883@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index f92ce63edfff..8a2bfd37ab4f 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2034,11 +2034,8 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) return delta < (s64)sysctl_sched_migration_cost; } - void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { - int old_cpu = task_cpu(p); - #ifdef CONFIG_SCHED_DEBUG /* * We should never call set_task_cpu() on a blocked task, @@ -2049,11 +2046,11 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) trace_sched_migrate_task(p, new_cpu); - if (old_cpu != new_cpu) { - p->se.nr_migrations++; - perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, - 1, 1, NULL, 0); - } + if (task_cpu(p) == new_cpu) + return; + + p->se.nr_migrations++; + perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0); __set_task_cpu(p, new_cpu); } -- cgit v1.2.2 From 416eb39556a03d1c7e52b0791e9052ccd71db241 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 17 Dec 2009 06:05:49 +0100 Subject: sched: Make warning less noisy Cc: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20091216170517.807938893@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 8a2bfd37ab4f..af7dfa74e6bb 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2041,7 +2041,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) * We should never call set_task_cpu() on a blocked task, * ttwu() will sort out the placement. */ - WARN_ON(p->state != TASK_RUNNING && p->state != TASK_WAKING); + WARN_ON_ONCE(p->state != TASK_RUNNING && p->state != TASK_WAKING); #endif trace_sched_migrate_task(p, new_cpu); -- cgit v1.2.2 From 234da7bcdc7aaa935846534c3b726dbc79a9cdd5 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 16 Dec 2009 20:21:05 +0100 Subject: sched: Teach might_sleep() about preemptible RCU In practice, it is harmless to voluntarily sleep in a rcu_read_lock() section if we are running under preempt rcu, but it is illegal if we build a kernel running non-preemptable rcu. Currently, might_sleep() doesn't notice sleepable operations under rcu_read_lock() sections if we are running under preemptable rcu because preempt_count() is left untouched after rcu_read_lock() in this case. But we want developers who test their changes under such config to notice the "sleeping while atomic" issues. So we add rcu_read_lock_nesting to prempt_count() in might_sleep() checks. [ v2: Handle rcu-tiny ] Signed-off-by: Frederic Weisbecker Reviewed-by: Paul E. McKenney Cc: Peter Zijlstra LKML-Reference: <1260991265-8451-1-git-send-regression-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index af7dfa74e6bb..7be88a7be047 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -9682,7 +9682,7 @@ void __init sched_init(void) #ifdef CONFIG_DEBUG_SPINLOCK_SLEEP static inline int preempt_count_equals(int preempt_offset) { - int nested = preempt_count() & ~PREEMPT_ACTIVE; + int nested = (preempt_count() & ~PREEMPT_ACTIVE) + rcu_preempt_depth(); return (nested == PREEMPT_INATOMIC_BASE + preempt_offset); } -- cgit v1.2.2 From 077614ee1e93245a3b9a4e1213659405dbeb0ba6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 17 Dec 2009 13:16:31 +0100 Subject: sched: Fix broken assertion There's a preemption race in the set_task_cpu() debug check in that when we get preempted after setting task->state we'd still be on the rq proper, but fail the test. Check for preempted tasks, since those are always on the RQ. Signed-off-by: Peter Zijlstra LKML-Reference: <20091217121830.137155561@chello.nl> Signed-off-by: Ingo Molnar --- kernel/sched.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 7be88a7be047..720df108a2d6 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2041,7 +2041,8 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) * We should never call set_task_cpu() on a blocked task, * ttwu() will sort out the placement. */ - WARN_ON_ONCE(p->state != TASK_RUNNING && p->state != TASK_WAKING); + WARN_ON_ONCE(p->state != TASK_RUNNING && p->state != TASK_WAKING && + !(task_thread_info(p)->preempt_count & PREEMPT_ACTIVE)); #endif trace_sched_migrate_task(p, new_cpu); -- cgit v1.2.2 From 3df0fc5b2e9d8092dcaeb5ae0b6753d85c851d66 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sun, 20 Dec 2009 14:23:57 +0100 Subject: sched: Restore printk sanity Revert the braindead pr_* crap. (Commit 663997d "sched: Use pr_fmt() and pr_()") It's dumb and causes stupid "sched: " strings all over the place. Signed-off-by: Peter Zijlstra Acked-by: Mike Galbraith Cc: Joe Perches Cc: Linus Torvalds Cc: Andrew Morton LKML-Reference: <1261315437.4314.6.camel@laptop> [ i dont mind the pr_*() patterns that much - but Peter dislikes them with a vengence. ] [ - v2: remove spurious diffstat from changelog :-/ ] Signed-off-by: Ingo Molnar --- kernel/sched.c | 89 ++++++++++++++++++++++++++++++++-------------------------- 1 file changed, 49 insertions(+), 40 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 720df108a2d6..7ffde2ae7868 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -26,8 +26,6 @@ * Thomas Gleixner, Mike Kravetz */ -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - #include #include #include @@ -5375,8 +5373,8 @@ static noinline void __schedule_bug(struct task_struct *prev) { struct pt_regs *regs = get_irq_regs(); - pr_err("BUG: scheduling while atomic: %s/%d/0x%08x\n", - prev->comm, prev->pid, preempt_count()); + printk(KERN_ERR "BUG: scheduling while atomic: %s/%d/0x%08x\n", + prev->comm, prev->pid, preempt_count()); debug_show_held_locks(prev); print_modules(); @@ -6940,23 +6938,23 @@ void sched_show_task(struct task_struct *p) unsigned state; state = p->state ? __ffs(p->state) + 1 : 0; - pr_info("%-13.13s %c", p->comm, + printk(KERN_INFO "%-13.13s %c", p->comm, state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'); #if BITS_PER_LONG == 32 if (state == TASK_RUNNING) - pr_cont(" running "); + printk(KERN_CONT " running "); else - pr_cont(" %08lx ", thread_saved_pc(p)); + printk(KERN_CONT " %08lx ", thread_saved_pc(p)); #else if (state == TASK_RUNNING) - pr_cont(" running task "); + printk(KERN_CONT " running task "); else - pr_cont(" %016lx ", thread_saved_pc(p)); + printk(KERN_CONT " %016lx ", thread_saved_pc(p)); #endif #ifdef CONFIG_DEBUG_STACK_USAGE free = stack_not_used(p); #endif - pr_cont("%5lu %5d %6d 0x%08lx\n", free, + printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free, task_pid_nr(p), task_pid_nr(p->real_parent), (unsigned long)task_thread_info(p)->flags); @@ -6968,9 +6966,11 @@ void show_state_filter(unsigned long state_filter) struct task_struct *g, *p; #if BITS_PER_LONG == 32 - pr_info(" task PC stack pid father\n"); + printk(KERN_INFO + " task PC stack pid father\n"); #else - pr_info(" task PC stack pid father\n"); + printk(KERN_INFO + " task PC stack pid father\n"); #endif read_lock(&tasklist_lock); do_each_thread(g, p) { @@ -7828,44 +7828,48 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, printk(KERN_DEBUG "%*s domain %d: ", level, "", level); if (!(sd->flags & SD_LOAD_BALANCE)) { - pr_cont("does not load-balance\n"); + printk("does not load-balance\n"); if (sd->parent) - pr_err("ERROR: !SD_LOAD_BALANCE domain has parent\n"); + printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain" + " has parent"); return -1; } - pr_cont("span %s level %s\n", str, sd->name); + printk(KERN_CONT "span %s level %s\n", str, sd->name); if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { - pr_err("ERROR: domain->span does not contain CPU%d\n", cpu); + printk(KERN_ERR "ERROR: domain->span does not contain " + "CPU%d\n", cpu); } if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) { - pr_err("ERROR: domain->groups does not contain CPU%d\n", cpu); + printk(KERN_ERR "ERROR: domain->groups does not contain" + " CPU%d\n", cpu); } printk(KERN_DEBUG "%*s groups:", level + 1, ""); do { if (!group) { - pr_cont("\n"); - pr_err("ERROR: group is NULL\n"); + printk("\n"); + printk(KERN_ERR "ERROR: group is NULL\n"); break; } if (!group->cpu_power) { - pr_cont("\n"); - pr_err("ERROR: domain->cpu_power not set\n"); + printk(KERN_CONT "\n"); + printk(KERN_ERR "ERROR: domain->cpu_power not " + "set\n"); break; } if (!cpumask_weight(sched_group_cpus(group))) { - pr_cont("\n"); - pr_err("ERROR: empty group\n"); + printk(KERN_CONT "\n"); + printk(KERN_ERR "ERROR: empty group\n"); break; } if (cpumask_intersects(groupmask, sched_group_cpus(group))) { - pr_cont("\n"); - pr_err("ERROR: repeated CPUs\n"); + printk(KERN_CONT "\n"); + printk(KERN_ERR "ERROR: repeated CPUs\n"); break; } @@ -7873,21 +7877,23 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); - pr_cont(" %s", str); + printk(KERN_CONT " %s", str); if (group->cpu_power != SCHED_LOAD_SCALE) { - pr_cont(" (cpu_power = %d)", group->cpu_power); + printk(KERN_CONT " (cpu_power = %d)", + group->cpu_power); } group = group->next; } while (group != sd->groups); - pr_cont("\n"); + printk(KERN_CONT "\n"); if (!cpumask_equal(sched_domain_span(sd), groupmask)) - pr_err("ERROR: groups don't span domain->span\n"); + printk(KERN_ERR "ERROR: groups don't span domain->span\n"); if (sd->parent && !cpumask_subset(groupmask, sched_domain_span(sd->parent))) - pr_err("ERROR: parent span is not a superset of domain->span\n"); + printk(KERN_ERR "ERROR: parent span is not a superset " + "of domain->span\n"); return 0; } @@ -8443,7 +8449,8 @@ static int build_numa_sched_groups(struct s_data *d, sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), GFP_KERNEL, num); if (!sg) { - pr_warning("Can not alloc domain group for node %d\n", num); + printk(KERN_WARNING "Can not alloc domain group for node %d\n", + num); return -ENOMEM; } d->sched_group_nodes[num] = sg; @@ -8472,8 +8479,8 @@ static int build_numa_sched_groups(struct s_data *d, sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), GFP_KERNEL, num); if (!sg) { - pr_warning("Can not alloc domain group for node %d\n", - j); + printk(KERN_WARNING + "Can not alloc domain group for node %d\n", j); return -ENOMEM; } sg->cpu_power = 0; @@ -8701,7 +8708,7 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, d->sched_group_nodes = kcalloc(nr_node_ids, sizeof(struct sched_group *), GFP_KERNEL); if (!d->sched_group_nodes) { - pr_warning("Can not alloc sched group node list\n"); + printk(KERN_WARNING "Can not alloc sched group node list\n"); return sa_notcovered; } sched_group_nodes_bycpu[cpumask_first(cpu_map)] = d->sched_group_nodes; @@ -8718,7 +8725,7 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, return sa_send_covered; d->rd = alloc_rootdomain(); if (!d->rd) { - pr_warning("Cannot alloc root domain\n"); + printk(KERN_WARNING "Cannot alloc root domain\n"); return sa_tmpmask; } return sa_rootdomain; @@ -9700,11 +9707,13 @@ void __might_sleep(char *file, int line, int preempt_offset) return; prev_jiffy = jiffies; - pr_err("BUG: sleeping function called from invalid context at %s:%d\n", - file, line); - pr_err("in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", - in_atomic(), irqs_disabled(), - current->pid, current->comm); + printk(KERN_ERR + "BUG: sleeping function called from invalid context at %s:%d\n", + file, line); + printk(KERN_ERR + "in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), + current->pid, current->comm); debug_show_held_locks(current); if (irqs_disabled()) -- cgit v1.2.2 From 70f1120527797adb31c68bdc6f1b45e182c342c7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Sun, 20 Dec 2009 17:36:27 +0100 Subject: sched: Fix hotplug hang The hot-unplug kstopmachine usage does a wakeup after deactivating the cpu, hence we cannot use cpu_active() here but must rely on the good olde online. Reported-by: Sachin Sant Reported-by: Jens Axboe Signed-off-by: Peter Zijlstra Tested-by: Jens Axboe Cc: Heiko Carstens Cc: Benjamin Herrenschmidt LKML-Reference: <1261326987.4314.24.camel@laptop> Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 7ffde2ae7868..87f1f47beffe 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2346,7 +2346,7 @@ int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) * not worry about this generic constraint ] */ if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) || - !cpu_active(cpu))) + !cpu_online(cpu))) cpu = select_fallback_rq(task_cpu(p), p); return cpu; -- cgit v1.2.2 From 0c69774e6ce94364cfaa8bdeb18061edc414bc5a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 22 Dec 2009 15:43:19 +0100 Subject: sched: Revert 738d2be, simplify set_task_cpu() Effectively reverts 738d2be4301007f054541c5c4bf7fb6a361c9b3a. As demonstrated by Eric, we really need to call __set_task_cpu() early in the fork() path to properly initialize the various task state -- specifically the cgroup state through set_task_rq(). [ we could probably fix this by explicitly calling __set_task_cpu() from sched_fork(), but lets try that for the next cycle and simply revert to the old behaviour for now. ] Reported-by: Eric Paris Tested-by: Eric Paris , Signed-off-by: Peter Zijlstra Cc: efault@gmx.de LKML-Reference: <1261492999.4937.36.camel@laptop> Signed-off-by: Ingo Molnar --- kernel/sched.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index 87f1f47beffe..c535cc4f6428 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2045,11 +2045,10 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) trace_sched_migrate_task(p, new_cpu); - if (task_cpu(p) == new_cpu) - return; - - p->se.nr_migrations++; - perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0); + if (task_cpu(p) != new_cpu) { + p->se.nr_migrations++; + perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, 1, NULL, 0); + } __set_task_cpu(p, new_cpu); } -- cgit v1.2.2 From 6d558c3ac9b6508d26fd5cadccce51fc9d726b1c Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Mon, 11 Jan 2010 14:21:25 +0800 Subject: sched: Reassign prev and switch_count when reacquire_kernel_lock() fail Assume A->B schedule is processing, if B have acquired BKL before and it need reschedule this time. Then on B's context, it will go to need_resched_nonpreemptible for reschedule. But at this time, prev and switch_count are related to A. It's wrong and will lead to incorrect scheduler statistics. Signed-off-by: Yong Zhang Signed-off-by: Peter Zijlstra LKML-Reference: <2674af741001102238w7b0ddcadref00d345e2181d11@mail.gmail.com> Signed-off-by: Ingo Molnar --- kernel/sched.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'kernel/sched.c') diff --git a/kernel/sched.c b/kernel/sched.c index c535cc4f6428..4508fe7048be 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -5530,8 +5530,11 @@ need_resched_nonpreemptible: post_schedule(rq); - if (unlikely(reacquire_kernel_lock(current) < 0)) + if (unlikely(reacquire_kernel_lock(current) < 0)) { + prev = rq->curr; + switch_count = &prev->nivcsw; goto need_resched_nonpreemptible; + } preempt_enable_no_resched(); if (need_resched()) -- cgit v1.2.2