diff options
Diffstat (limited to 'kernel')
| -rw-r--r-- | kernel/cgroup.c | 5 | ||||
| -rw-r--r-- | kernel/cpu.c | 24 | ||||
| -rw-r--r-- | kernel/cpuset.c | 351 | ||||
| -rw-r--r-- | kernel/dma-coherent.c | 2 | ||||
| -rw-r--r-- | kernel/exit.c | 96 | ||||
| -rw-r--r-- | kernel/hrtimer.c | 95 | ||||
| -rw-r--r-- | kernel/irq/manage.c | 9 | ||||
| -rw-r--r-- | kernel/kexec.c | 8 | ||||
| -rw-r--r-- | kernel/kgdb.c | 13 | ||||
| -rw-r--r-- | kernel/pid_namespace.c | 3 | ||||
| -rw-r--r-- | kernel/pm_qos_params.c | 25 | ||||
| -rw-r--r-- | kernel/posix-timers.c | 2 | ||||
| -rw-r--r-- | kernel/rcuclassic.c | 337 | ||||
| -rw-r--r-- | kernel/rcupreempt.c | 8 | ||||
| -rw-r--r-- | kernel/rcupreempt_trace.c | 7 | ||||
| -rw-r--r-- | kernel/resource.c | 68 | ||||
| -rw-r--r-- | kernel/sched.c | 467 | ||||
| -rw-r--r-- | kernel/sched_fair.c | 234 | ||||
| -rw-r--r-- | kernel/sched_features.h | 1 | ||||
| -rw-r--r-- | kernel/sched_idletask.c | 6 | ||||
| -rw-r--r-- | kernel/sched_rt.c | 58 | ||||
| -rw-r--r-- | kernel/sysctl.c | 9 | ||||
| -rw-r--r-- | kernel/time/clockevents.c | 15 | ||||
| -rw-r--r-- | kernel/time/ntp.c | 2 | ||||
| -rw-r--r-- | kernel/time/tick-broadcast.c | 99 | ||||
| -rw-r--r-- | kernel/time/tick-common.c | 15 | ||||
| -rw-r--r-- | kernel/time/tick-internal.h | 11 | ||||
| -rw-r--r-- | kernel/time/tick-oneshot.c | 44 | ||||
| -rw-r--r-- | kernel/time/tick-sched.c | 27 | ||||
| -rw-r--r-- | kernel/trace/trace_sysprof.c | 2 | ||||
| -rw-r--r-- | kernel/user.c | 4 |
31 files changed, 1322 insertions, 725 deletions
diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 13932abde159..a0123d75ec9a 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2738,14 +2738,15 @@ void cgroup_fork_callbacks(struct task_struct *child) */ void cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new) { - struct cgroup *oldcgrp, *newcgrp; + struct cgroup *oldcgrp, *newcgrp = NULL; if (need_mm_owner_callback) { int i; for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; oldcgrp = task_cgroup(old, ss->subsys_id); - newcgrp = task_cgroup(new, ss->subsys_id); + if (new) + newcgrp = task_cgroup(new, ss->subsys_id); if (oldcgrp == newcgrp) continue; if (ss->mm_owner_changed) diff --git a/kernel/cpu.c b/kernel/cpu.c index f17e9854c246..86d49045daed 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -199,13 +199,14 @@ static int __ref take_cpu_down(void *_param) struct take_cpu_down_param *param = _param; int err; - raw_notifier_call_chain(&cpu_chain, CPU_DYING | param->mod, - param->hcpu); /* Ensure this CPU doesn't handle any more interrupts. */ err = __cpu_disable(); if (err < 0) return err; + raw_notifier_call_chain(&cpu_chain, CPU_DYING | param->mod, + param->hcpu); + /* Force idle task to run as soon as we yield: it should immediately notice cpu is offline and die quickly. */ sched_idle_next(); @@ -453,6 +454,25 @@ out: } #endif /* CONFIG_PM_SLEEP_SMP */ +/** + * notify_cpu_starting(cpu) - call the CPU_STARTING notifiers + * @cpu: cpu that just started + * + * This function calls the cpu_chain notifiers with CPU_STARTING. + * It must be called by the arch code on the new cpu, before the new cpu + * enables interrupts and before the "boot" cpu returns from __cpu_up(). + */ +void notify_cpu_starting(unsigned int cpu) +{ + unsigned long val = CPU_STARTING; + +#ifdef CONFIG_PM_SLEEP_SMP + if (cpu_isset(cpu, frozen_cpus)) + val = CPU_STARTING_FROZEN; +#endif /* CONFIG_PM_SLEEP_SMP */ + raw_notifier_call_chain(&cpu_chain, val, (void *)(long)cpu); +} + #endif /* CONFIG_SMP */ /* diff --git a/kernel/cpuset.c b/kernel/cpuset.c index d5ab79cf516d..eab7bd6628e0 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -14,6 +14,8 @@ * 2003-10-22 Updates by Stephen Hemminger. * 2004 May-July Rework by Paul Jackson. * 2006 Rework by Paul Menage to use generic cgroups + * 2008 Rework of the scheduler domains and CPU hotplug handling + * by Max Krasnyansky * * This file is subject to the terms and conditions of the GNU General Public * License. See the file COPYING in the main directory of the Linux @@ -236,9 +238,11 @@ static struct cpuset top_cpuset = { static DEFINE_MUTEX(callback_mutex); -/* This is ugly, but preserves the userspace API for existing cpuset +/* + * This is ugly, but preserves the userspace API for existing cpuset * users. If someone tries to mount the "cpuset" filesystem, we - * silently switch it to mount "cgroup" instead */ + * silently switch it to mount "cgroup" instead + */ static int cpuset_get_sb(struct file_system_type *fs_type, int flags, const char *unused_dev_name, void *data, struct vfsmount *mnt) @@ -473,10 +477,9 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) } /* - * Helper routine for rebuild_sched_domains(). + * Helper routine for generate_sched_domains(). * Do cpusets a, b have overlapping cpus_allowed masks? */ - static int cpusets_overlap(struct cpuset *a, struct cpuset *b) { return cpus_intersects(a->cpus_allowed, b->cpus_allowed); @@ -518,26 +521,15 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) } /* - * rebuild_sched_domains() - * - * This routine will be called to rebuild the scheduler's dynamic - * sched domains: - * - if the flag 'sched_load_balance' of any cpuset with non-empty - * 'cpus' changes, - * - or if the 'cpus' allowed changes in any cpuset which has that - * flag enabled, - * - or if the 'sched_relax_domain_level' of any cpuset which has - * that flag enabled and with non-empty 'cpus' changes, - * - or if any cpuset with non-empty 'cpus' is removed, - * - or if a cpu gets offlined. - * - * This routine builds a partial partition of the systems CPUs - * (the set of non-overlappping cpumask_t's in the array 'part' - * below), and passes that partial partition to the kernel/sched.c - * partition_sched_domains() routine, which will rebuild the - * schedulers load balancing domains (sched domains) as specified - * by that partial partition. A 'partial partition' is a set of - * non-overlapping subsets whose union is a subset of that set. + * generate_sched_domains() + * + * This function builds a partial partition of the systems CPUs + * A 'partial partition' is a set of non-overlapping subsets whose + * union is a subset of that set. + * The output of this function needs to be passed to kernel/sched.c + * partition_sched_domains() routine, which will rebuild the scheduler's + * load balancing domains (sched domains) as specified by that partial + * partition. * * See "What is sched_load_balance" in Documentation/cpusets.txt * for a background explanation of this. @@ -547,13 +539,7 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) * domains when operating in the severe memory shortage situations * that could cause allocation failures below. * - * Call with cgroup_mutex held. May take callback_mutex during - * call due to the kfifo_alloc() and kmalloc() calls. May nest - * a call to the get_online_cpus()/put_online_cpus() pair. - * Must not be called holding callback_mutex, because we must not - * call get_online_cpus() while holding callback_mutex. Elsewhere - * the kernel nests callback_mutex inside get_online_cpus() calls. - * So the reverse nesting would risk an ABBA deadlock. + * Must be called with cgroup_lock held. * * The three key local variables below are: * q - a linked-list queue of cpuset pointers, used to implement a @@ -588,10 +574,10 @@ update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) * element of the partition (one sched domain) to be passed to * partition_sched_domains(). */ - -void rebuild_sched_domains(void) +static int generate_sched_domains(cpumask_t **domains, + struct sched_domain_attr **attributes) { - LIST_HEAD(q); /* queue of cpusets to be scanned*/ + LIST_HEAD(q); /* queue of cpusets to be scanned */ struct cpuset *cp; /* scans q */ struct cpuset **csa; /* array of all cpuset ptrs */ int csn; /* how many cpuset ptrs in csa so far */ @@ -601,23 +587,26 @@ void rebuild_sched_domains(void) int ndoms; /* number of sched domains in result */ int nslot; /* next empty doms[] cpumask_t slot */ - csa = NULL; + ndoms = 0; doms = NULL; dattr = NULL; + csa = NULL; /* Special case for the 99% of systems with one, full, sched domain */ if (is_sched_load_balance(&top_cpuset)) { - ndoms = 1; doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL); if (!doms) - goto rebuild; + goto done; + dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL); if (dattr) { *dattr = SD_ATTR_INIT; update_domain_attr_tree(dattr, &top_cpuset); } *doms = top_cpuset.cpus_allowed; - goto rebuild; + + ndoms = 1; + goto done; } csa = kmalloc(number_of_cpusets * sizeof(cp), GFP_KERNEL); @@ -680,61 +669,141 @@ restart: } } - /* Convert <csn, csa> to <ndoms, doms> */ + /* + * Now we know how many domains to create. + * Convert <csn, csa> to <ndoms, doms> and populate cpu masks. + */ doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL); - if (!doms) - goto rebuild; + if (!doms) { + ndoms = 0; + goto done; + } + + /* + * The rest of the code, including the scheduler, can deal with + * dattr==NULL case. No need to abort if alloc fails. + */ dattr = kmalloc(ndoms * sizeof(struct sched_domain_attr), GFP_KERNEL); for (nslot = 0, i = 0; i < csn; i++) { struct cpuset *a = csa[i]; + cpumask_t *dp; int apn = a->pn; - if (apn >= 0) { - cpumask_t *dp = doms + nslot; - - if (nslot == ndoms) { - static int warnings = 10; - if (warnings) { - printk(KERN_WARNING - "rebuild_sched_domains confused:" - " nslot %d, ndoms %d, csn %d, i %d," - " apn %d\n", - nslot, ndoms, csn, i, apn); - warnings--; - } - continue; + if (apn < 0) { + /* Skip completed partitions */ + continue; + } + + dp = doms + nslot; + + if (nslot == ndoms) { + static int warnings = 10; + if (warnings) { + printk(KERN_WARNING + "rebuild_sched_domains confused:" + " nslot %d, ndoms %d, csn %d, i %d," + " apn %d\n", + nslot, ndoms, csn, i, apn); + warnings--; } + continue; + } - cpus_clear(*dp); - if (dattr) - *(dattr + nslot) = SD_ATTR_INIT; - for (j = i; j < csn; j++) { - struct cpuset *b = csa[j]; - - if (apn == b->pn) { - cpus_or(*dp, *dp, b->cpus_allowed); - b->pn = -1; - if (dattr) - update_domain_attr_tree(dattr - + nslot, b); - } + cpus_clear(*dp); + if (dattr) + *(dattr + nslot) = SD_ATTR_INIT; + for (j = i; j < csn; j++) { + struct cpuset *b = csa[j]; + + if (apn == b->pn) { + cpus_or(*dp, *dp, b->cpus_allowed); + if (dattr) + update_domain_attr_tree(dattr + nslot, b); + + /* Done with this partition */ + b->pn = -1; } - nslot++; } + nslot++; } BUG_ON(nslot != ndoms); -rebuild: - /* Have scheduler rebuild sched domains */ +done: + kfree(csa); + + *domains = doms; + *attributes = dattr; + return ndoms; +} + +/* + * Rebuild scheduler domains. + * + * Call with neither cgroup_mutex held nor within get_online_cpus(). + * Takes both cgroup_mutex and get_online_cpus(). + * + * Cannot be directly called from cpuset code handling changes + * to the cpuset pseudo-filesystem, because it cannot be called + * from code that already holds cgroup_mutex. + */ +static void do_rebuild_sched_domains(struct work_struct *unused) +{ + struct sched_domain_attr *attr; + cpumask_t *doms; + int ndoms; + get_online_cpus(); - partition_sched_domains(ndoms, doms, dattr); + + /* Generate domain masks and attrs */ + cgroup_lock(); + ndoms = generate_sched_domains(&doms, &attr); + cgroup_unlock(); + + /* Have scheduler rebuild the domains */ + partition_sched_domains(ndoms, doms, attr); + put_online_cpus(); +} -done: - kfree(csa); - /* Don't kfree(doms) -- partition_sched_domains() does that. */ - /* Don't kfree(dattr) -- partition_sched_domains() does that. */ +static DECLARE_WORK(rebuild_sched_domains_work, do_rebuild_sched_domains); + +/* + * Rebuild scheduler domains, asynchronously via workqueue. + * + * If the flag 'sched_load_balance' of any cpuset with non-empty + * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset + * which has that flag enabled, or if any cpuset with a non-empty + * 'cpus' is removed, then call this routine to rebuild the + * scheduler's dynamic sched domains. + * + * The rebuild_sched_domains() and partition_sched_domains() + * routines must nest cgroup_lock() inside get_online_cpus(), + * but such cpuset changes as these must nest that locking the + * other way, holding cgroup_lock() for much of the code. + * + * So in order to avoid an ABBA deadlock, the cpuset code handling + * these user changes delegates the actual sched domain rebuilding + * to a separate workqueue thread, which ends up processing the + * above do_rebuild_sched_domains() function. + */ +static void async_rebuild_sched_domains(void) +{ + schedule_work(&rebuild_sched_domains_work); +} + +/* + * Accomplishes the same scheduler domain rebuild as the above + * async_rebuild_sched_domains(), however it directly calls the + * rebuild routine synchronously rather than calling it via an + * asynchronous work thread. + * + * This can only be called from code that is not holding + * cgroup_mutex (not nested in a cgroup_lock() call.) + */ +void rebuild_sched_domains(void) +{ + do_rebuild_sched_domains(NULL); } /** @@ -774,37 +843,25 @@ static void cpuset_change_cpumask(struct task_struct *tsk, /** * update_tasks_cpumask - Update the cpumasks of tasks in the cpuset. * @cs: the cpuset in which each task's cpus_allowed mask needs to be changed + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() * * Called with cgroup_mutex held * * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, * calling callback functions for each. * - * Return 0 if successful, -errno if not. + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. */ -static int update_tasks_cpumask(struct cpuset *cs) +static void update_tasks_cpumask(struct cpuset *cs, struct ptr_heap *heap) { struct cgroup_scanner scan; - struct ptr_heap heap; - int retval; - - /* - * cgroup_scan_tasks() will initialize heap->gt for us. - * heap_init() is still needed here for we should not change - * cs->cpus_allowed when heap_init() fails. - */ - retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); - if (retval) - return retval; scan.cg = cs->css.cgroup; scan.test_task = cpuset_test_cpumask; scan.process_task = cpuset_change_cpumask; - scan.heap = &heap; - retval = cgroup_scan_tasks(&scan); - - heap_free(&heap); - return retval; + scan.heap = heap; + cgroup_scan_tasks(&scan); } /** @@ -814,6 +871,7 @@ static int update_tasks_cpumask(struct cpuset *cs) */ static int update_cpumask(struct cpuset *cs, const char *buf) { + struct ptr_heap heap; struct cpuset trialcs; int retval; int is_load_balanced; @@ -848,6 +906,10 @@ static int update_cpumask(struct cpuset *cs, const char *buf) if (cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed)) return 0; + retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); + if (retval) + return retval; + is_load_balanced = is_sched_load_balance(&trialcs); mutex_lock(&callback_mutex); @@ -858,12 +920,12 @@ static int update_cpumask(struct cpuset *cs, const char *buf) * Scan tasks in the cpuset, and update the cpumasks of any * that need an update. */ - retval = update_tasks_cpumask(cs); - if (retval < 0) - return retval; + update_tasks_cpumask(cs, &heap); + + heap_free(&heap); if (is_load_balanced) - rebuild_sched_domains(); + async_rebuild_sched_domains(); return 0; } @@ -1090,7 +1152,7 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val) if (val != cs->relax_domain_level) { cs->relax_domain_level = val; if (!cpus_empty(cs->cpus_allowed) && is_sched_load_balance(cs)) - rebuild_sched_domains(); + async_rebuild_sched_domains(); } return 0; @@ -1131,7 +1193,7 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, mutex_unlock(&callback_mutex); if (cpus_nonempty && balance_flag_changed) - rebuild_sched_domains(); + async_rebuild_sched_domains(); return 0; } @@ -1492,6 +1554,9 @@ static u64 cpuset_read_u64(struct cgroup *cont, struct cftype *cft) default: BUG(); } + + /* Unreachable but makes gcc happy */ + return 0; } static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft) @@ -1504,6 +1569,9 @@ static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft) default: BUG(); } + + /* Unrechable but makes gcc happy */ + return 0; } @@ -1692,15 +1760,9 @@ static struct cgroup_subsys_state *cpuset_create( } /* - * Locking note on the strange update_flag() call below: - * * If the cpuset being removed has its flag 'sched_load_balance' * enabled, then simulate turning sched_load_balance off, which - * will call rebuild_sched_domains(). The get_online_cpus() - * call in rebuild_sched_domains() must not be made while holding - * callback_mutex. Elsewhere the kernel nests callback_mutex inside - * get_online_cpus() calls. So the reverse nesting would risk an - * ABBA deadlock. + * will call async_rebuild_sched_domains(). */ static void cpuset_destroy(struct cgroup_subsys *ss, struct cgroup *cont) @@ -1719,7 +1781,7 @@ static void cpuset_destroy(struct cgroup_subsys *ss, struct cgroup *cont) struct cgroup_subsys cpuset_subsys = { .name = "cpuset", .create = cpuset_create, - .destroy = cpuset_destroy, + .destroy = cpuset_destroy, .can_attach = cpuset_can_attach, .attach = cpuset_attach, .populate = cpuset_populate, @@ -1811,7 +1873,7 @@ static void move_member_tasks_to_cpuset(struct cpuset *from, struct cpuset *to) } /* - * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs + * If CPU and/or memory hotplug handlers, below, unplug any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, * removing that CPU or node from all cpusets. If this removes the * last CPU or node from a cpuset, then move the tasks in the empty @@ -1859,7 +1921,7 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs) * that has tasks along with an empty 'mems'. But if we did see such * a cpuset, we'd handle it just like we do if its 'cpus' was empty. */ -static void scan_for_empty_cpusets(const struct cpuset *root) +static void scan_for_empty_cpusets(struct cpuset *root) { LIST_HEAD(queue); struct cpuset *cp; /* scans cpusets being updated */ @@ -1896,42 +1958,13 @@ static void scan_for_empty_cpusets(const struct cpuset *root) nodes_empty(cp->mems_allowed)) remove_tasks_in_empty_cpuset(cp); else { - update_tasks_cpumask(cp); + update_tasks_cpumask(cp, NULL); update_tasks_nodemask(cp, &oldmems); } } } /* - * The cpus_allowed and mems_allowed nodemasks in the top_cpuset track - * cpu_online_map and node_states[N_HIGH_MEMORY]. Force the top cpuset to - * track what's online after any CPU or memory node hotplug or unplug event. - * - * Since there are two callers of this routine, one for CPU hotplug - * events and one for memory node hotplug events, we could have coded - * two separate routines here. We code it as a single common routine - * in order to minimize text size. - */ - -static void common_cpu_mem_hotplug_unplug(int rebuild_sd) -{ - cgroup_lock(); - - top_cpuset.cpus_allowed = cpu_online_map; - top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - scan_for_empty_cpusets(&top_cpuset); - - /* - * Scheduler destroys domains on hotplug events. - * Rebuild them based on the current settings. - */ - if (rebuild_sd) - rebuild_sched_domains(); - - cgroup_unlock(); -} - -/* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent * (of no affect) on systems that are actively using CPU hotplug @@ -1939,40 +1972,52 @@ static void common_cpu_mem_hotplug_unplug(int rebuild_sd) * * This routine ensures that top_cpuset.cpus_allowed tracks * cpu_online_map on each CPU hotplug (cpuhp) event. + * + * Called within get_online_cpus(). Needs to call cgroup_lock() + * before calling generate_sched_domains(). */ - -static int cpuset_handle_cpuhp(struct notifier_block *unused_nb, +static int cpuset_track_online_cpus(struct notifier_block *unused_nb, unsigned long phase, void *unused_cpu) { + struct sched_domain_attr *attr; + cpumask_t *doms; + int ndoms; + switch (phase) { - case CPU_UP_CANCELED: - case CPU_UP_CANCELED_FROZEN: - case CPU_DOWN_FAILED: - case CPU_DOWN_FAILED_FROZEN: case CPU_ONLINE: case CPU_ONLINE_FROZEN: case CPU_DEAD: case CPU_DEAD_FROZEN: - common_cpu_mem_hotplug_unplug(1); break; + default: return NOTIFY_DONE; } + cgroup_lock(); + top_cpuset.cpus_allowed = cpu_online_map; + scan_for_empty_cpusets(&top_cpuset); + ndoms = generate_sched_domains(&doms, &attr); + cgroup_unlock(); + + /* Have scheduler rebuild the domains */ + partition_sched_domains(ndoms, doms, attr); + return NOTIFY_OK; } #ifdef CONFIG_MEMORY_HOTPLUG /* * Keep top_cpuset.mems_allowed tracking node_states[N_HIGH_MEMORY]. - * Call this routine anytime after you change - * node_states[N_HIGH_MEMORY]. - * See also the previous routine cpuset_handle_cpuhp(). + * Call this routine anytime after node_states[N_HIGH_MEMORY] changes. + * See also the previous routine cpuset_track_online_cpus(). */ - void cpuset_track_online_nodes(void) { - common_cpu_mem_hotplug_unplug(0); + cgroup_lock(); + top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; + scan_for_empty_cpusets(&top_cpuset); + cgroup_unlock(); } #endif @@ -1987,7 +2032,7 @@ void __init cpuset_init_smp(void) top_cpuset.cpus_allowed = cpu_online_map; top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - hotcpu_notifier(cpuset_handle_cpuhp, 0); + hotcpu_notifier(cpuset_track_online_cpus, 0); } /** diff --git a/kernel/dma-coherent.c b/kernel/dma-coherent.c index c1d4d5b4c61c..f013a0c2e111 100644 --- a/kernel/dma-coherent.c +++ b/kernel/dma-coherent.c @@ -124,6 +124,7 @@ int dma_alloc_from_coherent(struct device *dev, ssize_t size, } return (mem != NULL); } +EXPORT_SYMBOL(dma_alloc_from_coherent); /** * dma_release_from_coherent() - try to free the memory allocated from per-device coherent memory pool @@ -151,3 +152,4 @@ int dma_release_from_coherent(struct device *dev, int order, void *vaddr) } return 0; } +EXPORT_SYMBOL(dma_release_from_coherent); diff --git a/kernel/exit.c b/kernel/exit.c index 75c647387639..85a83c831856 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -112,9 +112,9 @@ static void __exit_signal(struct task_struct *tsk) * We won't ever get here for the group leader, since it * will have been the last reference on the signal_struct. */ - sig->utime = cputime_add(sig->utime, tsk->utime); - sig->stime = cputime_add(sig->stime, tsk->stime); - sig->gtime = cputime_add(sig->gtime, tsk->gtime); + sig->utime = cputime_add(sig->utime, task_utime(tsk)); + sig->stime = cputime_add(sig->stime, task_stime(tsk)); + sig->gtime = cputime_add(sig->gtime, task_gtime(tsk)); sig->min_flt += tsk->min_flt; sig->maj_flt += tsk->maj_flt; sig->nvcsw += tsk->nvcsw; @@ -583,8 +583,6 @@ mm_need_new_owner(struct mm_struct *mm, struct task_struct *p) * If there are other users of the mm and the owner (us) is exiting * we need to find a new owner to take on the responsibility. */ - if (!mm) - return 0; if (atomic_read(&mm->mm_users) <= 1) return 0; if (mm->owner != p) @@ -627,6 +625,16 @@ retry: } while_each_thread(g, c); read_unlock(&tasklist_lock); + /* + * We found no owner yet mm_users > 1: this implies that we are + * most likely racing with swapoff (try_to_unuse()) or /proc or + * ptrace or page migration (get_task_mm()). Mark owner as NULL, + * so that subsystems can understand the callback and take action. + */ + down_write(&mm->mmap_sem); + cgroup_mm_owner_callbacks(mm->owner, NULL); + mm->owner = NULL; + up_write(&mm->mmap_sem); return; assign_new_owner: @@ -831,26 +839,50 @@ static void reparent_thread(struct task_struct *p, struct task_struct *father) * the child reaper process (ie "init") in our pid * space. */ +static struct task_struct *find_new_reaper(struct task_struct *father) +{ + struct pid_namespace *pid_ns = task_active_pid_ns(father); + struct task_struct *thread; + + thread = father; + while_each_thread(father, thread) { + if (thread->flags & PF_EXITING) + continue; + if (unlikely(pid_ns->child_reaper == father)) + pid_ns->child_reaper = thread; + return thread; + } + + if (unlikely(pid_ns->child_reaper == father)) { + write_unlock_irq(&tasklist_lock); + if (unlikely(pid_ns == &init_pid_ns)) + panic("Attempted to kill init!"); + + zap_pid_ns_processes(pid_ns); + write_lock_irq(&tasklist_lock); + /* + * We can not clear ->child_reaper or leave it alone. + * There may by stealth EXIT_DEAD tasks on ->children, + * forget_original_parent() must move them somewhere. + */ + pid_ns->child_reaper = init_pid_ns.child_reaper; + } + + return pid_ns->child_reaper; +} + static void forget_original_parent(struct task_struct *father) { - struct task_struct *p, *n, *reaper = father; + struct task_struct *p, *n, *reaper; LIST_HEAD(ptrace_dead); write_lock_irq(&tasklist_lock); - + reaper = find_new_reaper(father); /* * First clean up ptrace if we were using it. */ ptrace_exit(father, &ptrace_dead); - do { - reaper = next_thread(reaper); - if (reaper == father) { - reaper = task_child_reaper(father); - break; - } - } while (reaper->flags & PF_EXITING); - list_for_each_entry_safe(p, n, &father->children, sibling) { p->real_parent = reaper; if (p->parent == father) { @@ -959,39 +991,6 @@ static void check_stack_usage(void) static inline void check_stack_usage(void) {} #endif -static inline void exit_child_reaper(struct task_struct *tsk) -{ - if (likely(tsk->group_leader != task_child_reaper(tsk))) - return; - - if (tsk->nsproxy->pid_ns == &init_pid_ns) - panic("Attempted to kill init!"); - - /* - * @tsk is the last thread in the 'cgroup-init' and is exiting. - * Terminate all remaining processes in the namespace and reap them - * before exiting @tsk. - * - * Note that @tsk (last thread of cgroup-init) may not necessarily - * be the child-reaper (i.e main thread of cgroup-init) of the - * namespace i.e the child_reaper may have already exited. - * - * Even after a child_reaper exits, we let it inherit orphaned children, - * because, pid_ns->child_reaper remains valid as long as there is - * at least one living sub-thread in the cgroup init. - - * This living sub-thread of the cgroup-init will be notified when - * a child inherited by the 'child-reaper' exits (do_notify_parent() - * uses __group_send_sig_info()). Further, when reaping child processes, - * do_wait() iterates over children of all living sub threads. - - * i.e even though 'child_reaper' thread is listed as the parent of the - * orphaned children, any living sub-thread in the cgroup-init can - * perform the role of the child_reaper. - */ - zap_pid_ns_processes(tsk->nsproxy->pid_ns); -} - NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; @@ -1051,7 +1050,6 @@ NORET_TYPE void do_exit(long code) } group_dead = atomic_dec_and_test(&tsk->signal->live); if (group_dead) { - exit_child_reaper(tsk); hrtimer_cancel(&tsk->signal->real_timer); exit_itimers(tsk->signal); } diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index b8e4dce80a74..cdec83e722fa 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -672,13 +672,14 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer, */ BUG_ON(timer->function(timer) != HRTIMER_NORESTART); return 1; - case HRTIMER_CB_IRQSAFE_NO_SOFTIRQ: + case HRTIMER_CB_IRQSAFE_PERCPU: + case HRTIMER_CB_IRQSAFE_UNLOCKED: /* * This is solely for the sched tick emulation with * dynamic tick support to ensure that we do not * restart the tick right on the edge and end up with * the tick timer in the softirq ! The calling site - * takes care of this. + * takes care of this. Also used for hrtimer sleeper ! */ debug_hrtimer_deactivate(timer); return 1; @@ -1245,7 +1246,8 @@ static void __run_hrtimer(struct hrtimer *timer) timer_stats_account_hrtimer(timer); fn = timer->function; - if (timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ) { + if (timer->cb_mode == HRTIMER_CB_IRQSAFE_PERCPU || + timer->cb_mode == HRTIMER_CB_IRQSAFE_UNLOCKED) { /* * Used for scheduler timers, avoid lock inversion with * rq->lock and tasklist_lock. @@ -1452,7 +1454,7 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task) sl->timer.function = hrtimer_wakeup; sl->task = task; #ifdef CONFIG_HIGH_RES_TIMERS - sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ; + sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_UNLOCKED; #endif } @@ -1591,29 +1593,95 @@ static void __cpuinit init_hrtimers_cpu(int cpu) #ifdef CONFIG_HOTPLUG_CPU -static void migrate_hrtimer_list(struct hrtimer_clock_base *old_base, - struct hrtimer_clock_base *new_base) +static int migrate_hrtimer_list(struct hrtimer_clock_base *old_base, + struct hrtimer_clock_base *new_base, int dcpu) { struct hrtimer *timer; struct rb_node *node; + int raise = 0; while ((node = rb_first(&old_base->active))) { timer = rb_entry(node, struct hrtimer, node); BUG_ON(hrtimer_callback_running(timer)); debug_hrtimer_deactivate(timer); - __remove_hrtimer(timer, old_base, HRTIMER_STATE_INACTIVE, 0); + + /* + * Should not happen. Per CPU timers should be + * canceled _before_ the migration code is called + */ + if (timer->cb_mode == HRTIMER_CB_IRQSAFE_PERCPU) { + __remove_hrtimer(timer, old_base, + HRTIMER_STATE_INACTIVE, 0); + WARN(1, "hrtimer (%p %p)active but cpu %d dead\n", + timer, timer->function, dcpu); + continue; + } + + /* + * Mark it as STATE_MIGRATE not INACTIVE otherwise the + * timer could be seen as !active and just vanish away + * under us on another CPU + */ + __remove_hrtimer(timer, old_base, HRTIMER_STATE_MIGRATE, 0); timer->base = new_base; /* * Enqueue the timer. Allow reprogramming of the event device */ enqueue_hrtimer(timer, new_base, 1); + +#ifdef CONFIG_HIGH_RES_TIMERS + /* + * Happens with high res enabled when the timer was + * already expired and the callback mode is + * HRTIMER_CB_IRQSAFE_UNLOCKED (hrtimer_sleeper). The + * enqueue code does not move them to the soft irq + * pending list for performance/latency reasons, but + * in the migration state, we need to do that + * otherwise we end up with a stale timer. + */ + if (timer->state == HRTIMER_STATE_MIGRATE) { + timer->state = HRTIMER_STATE_PENDING; + list_add_tail(&timer->cb_entry, + &new_base->cpu_base->cb_pending); + raise = 1; + } +#endif + /* Clear the migration state bit */ + timer->state &= ~HRTIMER_STATE_MIGRATE; + } + return raise; +} + +#ifdef CONFIG_HIGH_RES_TIMERS +static int migrate_hrtimer_pending(struct hrtimer_cpu_base *old_base, + struct hrtimer_cpu_base *new_base) +{ + struct hrtimer *timer; + int raise = 0; + + while (!list_empty(&old_base->cb_pending)) { + timer = list_entry(old_base->cb_pending.next, + struct hrtimer, cb_entry); + + __remove_hrtimer(timer, timer->base, HRTIMER_STATE_PENDING, 0); + timer->base = &new_base->clock_base[timer->base->index]; + list_add_tail(&timer->cb_entry, &new_base->cb_pending); + raise = 1; } + return raise; +} +#else +static int migrate_hrtimer_pending(struct hrtimer_cpu_base *old_base, + struct hrtimer_cpu_base *new_base) +{ + return 0; } +#endif static void migrate_hrtimers(int cpu) { struct hrtimer_cpu_base *old_base, *new_base; - int i; + int i, raise = 0; BUG_ON(cpu_online(cpu)); old_base = &per_cpu(hrtimer_bases, cpu); @@ -1626,14 +1694,21 @@ static void migrate_hrtimers(int cpu) spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING); for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) { - migrate_hrtimer_list(&old_base->clock_base[i], - &new_base->clock_base[i]); + if (migrate_hrtimer_list(&old_base->clock_base[i], + &new_base->clock_base[i], cpu)) + raise = 1; } + if (migrate_hrtimer_pending(old_base, new_base)) + raise = 1; + spin_unlock(&old_base->lock); spin_unlock(&new_base->lock); local_irq_enable(); put_cpu_var(hrtimer_bases); + + if (raise) + hrtimer_raise_softirq(); } #endif /* CONFIG_HOTPLUG_CPU */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 0314074fa232..60c49e324390 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -89,7 +89,14 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask) set_balance_irq_affinity(irq, cpumask); #ifdef CONFIG_GENERIC_PENDING_IRQ - set_pending_irq(irq, cpumask); + if (desc->status & IRQ_MOVE_PCNTXT) { + unsigned long flags; + + spin_lock_irqsave(&desc->lock, flags); + desc->chip->set_affinity(irq, cpumask); + spin_unlock_irqrestore(&desc->lock, flags); + } else + set_pending_irq(irq, cpumask); #else desc->affinity = cpumask; desc->chip->set_affinity(irq, cpumask); diff --git a/kernel/kexec.c b/kernel/kexec.c index 59f3f0df35d4..aef265325cd3 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -753,8 +753,14 @@ static struct page *kimage_alloc_page(struct kimage *image, *old = addr | (*old & ~PAGE_MASK); /* The old page I have found cannot be a - * destination page, so return it. + * destination page, so return it if it's + * gfp_flags honor the ones passed in. */ + if (!(gfp_mask & __GFP_HIGHMEM) && + PageHighMem(old_page)) { + kimage_free_pages(old_page); + continue; + } addr = old_addr; page = old_page; break; diff --git a/kernel/kgdb.c b/kernel/kgdb.c index eaa21fc9ad1d..e4dcfb2272a4 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -488,7 +488,7 @@ static int write_mem_msg(int binary) if (err) return err; if (CACHE_FLUSH_IS_SAFE) - flush_icache_range(addr, addr + length + 1); + flush_icache_range(addr, addr + length); return 0; } @@ -590,6 +590,7 @@ static void kgdb_wait(struct pt_regs *regs) /* Signal the primary CPU that we are done: */ atomic_set(&cpu_in_kgdb[cpu], 0); + touch_softlockup_watchdog(); clocksource_touch_watchdog(); local_irq_restore(flags); } @@ -1432,6 +1433,7 @@ acquirelock: atomic_read(&kgdb_cpu_doing_single_step) != cpu) { atomic_set(&kgdb_active, -1); + touch_softlockup_watchdog(); clocksource_touch_watchdog(); local_irq_restore(flags); @@ -1462,7 +1464,7 @@ acquirelock: * Get the passive CPU lock which will hold all the non-primary * CPU in a spin state while the debugger is active */ - if (!kgdb_single_step || !kgdb_contthread) { + if (!kgdb_single_step) { for (i = 0; i < NR_CPUS; i++) atomic_set(&passive_cpu_wait[i], 1); } @@ -1475,7 +1477,7 @@ acquirelock: #ifdef CONFIG_SMP /* Signal the other CPUs to enter kgdb_wait() */ - if ((!kgdb_single_step || !kgdb_contthread) && kgdb_do_roundup) + if ((!kgdb_single_step) && kgdb_do_roundup) kgdb_roundup_cpus(flags); #endif @@ -1494,7 +1496,7 @@ acquirelock: kgdb_post_primary_code(ks->linux_regs, ks->ex_vector, ks->err_code); kgdb_deactivate_sw_breakpoints(); kgdb_single_step = 0; - kgdb_contthread = NULL; + kgdb_contthread = current; exception_level = 0; /* Talk to debugger with gdbserial protocol */ @@ -1508,7 +1510,7 @@ acquirelock: kgdb_info[ks->cpu].task = NULL; atomic_set(&cpu_in_kgdb[ks->cpu], 0); - if (!kgdb_single_step || !kgdb_contthread) { + if (!kgdb_single_step) { for (i = NR_CPUS-1; i >= 0; i--) atomic_set(&passive_cpu_wait[i], 0); /* @@ -1524,6 +1526,7 @@ acquirelock: kgdb_restore: /* Free kgdb_active */ atomic_set(&kgdb_active, -1); + touch_softlockup_watchdog(); clocksource_touch_watchdog(); local_irq_restore(flags); diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index ea567b78d1aa..fab8ea86fac3 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -179,9 +179,6 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) rc = sys_wait4(-1, NULL, __WALL, NULL); } while (rc != -ECHILD); - - /* Child reaper for the pid namespace is going away */ - pid_ns->child_reaper = NULL; acct_exit_ns(pid_ns); return; } diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c index da9c2dda6a4e..dfdec524d1b7 100644 --- a/kernel/pm_qos_params.c +++ b/kernel/pm_qos_params.c @@ -43,7 +43,7 @@ #include <linux/uaccess.h> /* - * locking rule: all changes to target_value or requirements or notifiers lists + * locking rule: all changes to requirements or notifiers lists * or pm_qos_object list and pm_qos_objects need to happen with pm_qos_lock * held, taken with _irqsave. One lock to rule them all */ @@ -66,7 +66,7 @@ struct pm_qos_object { struct miscdevice pm_qos_power_miscdev; char *name; s32 default_value; - s32 target_value; + atomic_t target_value; s32 (*comparitor)(s32, s32); }; @@ -77,7 +77,7 @@ static struct pm_qos_object cpu_dma_pm_qos = { .notifiers = &cpu_dma_lat_notifier, .name = "cpu_dma_latency", .default_value = 2000 * USEC_PER_SEC, - .target_value = 2000 * USEC_PER_SEC, + .target_value = ATOMIC_INIT(2000 * USEC_PER_SEC), .comparitor = min_compare }; @@ -87,7 +87,7 @@ static struct pm_qos_object network_lat_pm_qos = { .notifiers = &network_lat_notifier, .name = "network_latency", .default_value = 2000 * USEC_PER_SEC, - .target_value = 2000 * USEC_PER_SEC, + .target_value = ATOMIC_INIT(2000 * USEC_PER_SEC), .comparitor = min_compare }; @@ -99,7 +99,7 @@ static struct pm_qos_object network_throughput_pm_qos = { .notifiers = &network_throughput_notifier, .name = "network_throughput", .default_value = 0, - .target_value = 0, + .target_value = ATOMIC_INIT(0), .comparitor = max_compare }; @@ -150,11 +150,11 @@ static void update_target(int target) extreme_value = pm_qos_array[target]->comparitor( extreme_value, node->value); } - if (pm_qos_array[target]->target_value != extreme_value) { + if (atomic_read(&pm_qos_array[target]->target_value) != extreme_value) { call_notifier = 1; - pm_qos_array[target]->target_value = extreme_value; + atomic_set(&pm_qos_array[target]->target_value, extreme_value); pr_debug(KERN_ERR "new target for qos %d is %d\n", target, - pm_qos_array[target]->target_value); + atomic_read(&pm_qos_array[target]->target_value)); } spin_unlock_irqrestore(&pm_qos_lock, flags); @@ -193,14 +193,7 @@ static int find_pm_qos_object_by_minor(int minor) */ int pm_qos_requirement(int pm_qos_class) { - int ret_val; - unsigned long flags; - - spin_lock_irqsave(&pm_qos_lock, flags); - ret_val = pm_qos_array[pm_qos_class]->target_value; - spin_unlock_irqrestore(&pm_qos_lock, flags); - - return ret_val; + return atomic_read(&pm_qos_array[pm_qos_class]->target_value); } EXPORT_SYMBOL_GPL(pm_qos_requirement); diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index e36d5798cbff..5131e5471169 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -441,7 +441,7 @@ static struct k_itimer * alloc_posix_timer(void) return tmr; if (unlikely(!(tmr->sigq = sigqueue_alloc()))) { kmem_cache_free(posix_timers_cache, tmr); - tmr = NULL; + return NULL; } memset(&tmr->sigq->info, 0, sizeof(siginfo_t)); return tmr; diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index aad93cdc9f68..37f72e551542 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -47,6 +47,7 @@ #include <linux/notifier.h> #include <linux/cpu.h> #include <linux/mutex.h> +#include <linux/time.h> #ifdef CONFIG_DEBUG_LOCK_ALLOC static struct lock_class_key rcu_lock_key; @@ -60,12 +61,14 @@ EXPORT_SYMBOL_GPL(rcu_lock_map); static struct rcu_ctrlblk rcu_ctrlblk = { .cur = -300, .completed = -300, + .pending = -300, .lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock), .cpumask = CPU_MASK_NONE, }; static struct rcu_ctrlblk rcu_bh_ctrlblk = { .cur = -300, .completed = -300, + .pending = -300, .lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock), .cpumask = CPU_MASK_NONE, }; @@ -83,7 +86,10 @@ static void force_quiescent_state(struct rcu_data *rdp, { int cpu; cpumask_t cpumask; + unsigned long flags; + set_need_resched(); + spin_lock_irqsave(&rcp->lock, flags); if (unlikely(!rcp->signaled)) { rcp->signaled = 1; /* @@ -109,6 +115,7 @@ static void force_quiescent_state(struct rcu_data *rdp, for_each_cpu_mask_nr(cpu, cpumask) smp_send_reschedule(cpu); } + spin_unlock_irqrestore(&rcp->lock, flags); } #else static inline void force_quiescent_state(struct rcu_data *rdp, @@ -118,6 +125,126 @@ static inline void force_quiescent_state(struct rcu_data *rdp, } #endif +static void __call_rcu(struct rcu_head *head, struct rcu_ctrlblk *rcp, + struct rcu_data *rdp) +{ + long batch; + + head->next = NULL; + smp_mb(); /* Read of rcu->cur must happen after any change by caller. */ + + /* + * Determine the batch number of this callback. + * + * Using ACCESS_ONCE to avoid the following error when gcc eliminates + * local variable "batch" and emits codes like this: + * 1) rdp->batch = rcp->cur + 1 # gets old value + * ...... + * 2)rcu_batch_after(rcp->cur + 1, rdp->batch) # gets new value + * then [*nxttail[0], *nxttail[1]) may contain callbacks + * that batch# = rdp->batch, see the comment of struct rcu_data. + */ + batch = ACCESS_ONCE(rcp->cur) + 1; + + if (rdp->nxtlist && rcu_batch_after(batch, rdp->batch)) { + /* process callbacks */ + rdp->nxttail[0] = rdp->nxttail[1]; + rdp->nxttail[1] = rdp->nxttail[2]; + if (rcu_batch_after(batch - 1, rdp->batch)) + rdp->nxttail[0] = rdp->nxttail[2]; + } + + rdp->batch = batch; + *rdp->nxttail[2] = head; + rdp->nxttail[2] = &head->next; + + if (unlikely(++rdp->qlen > qhimark)) { + rdp->blimit = INT_MAX; + force_quiescent_state(rdp, &rcu_ctrlblk); + } +} + +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR + +static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) +{ + rcp->gp_start = jiffies; + rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_CHECK; +} + +static void print_other_cpu_stall(struct rcu_ctrlblk *rcp) +{ + int cpu; + long delta; + unsigned long flags; + + /* Only let one CPU complain about others per time interval. */ + + spin_lock_irqsave(&rcp->lock, flags); + delta = jiffies - rcp->jiffies_stall; + if (delta < 2 || rcp->cur != rcp->completed) { + spin_unlock_irqrestore(&rcp->lock, flags); + return; + } + rcp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; + spin_unlock_irqrestore(&rcp->lock, flags); + + /* OK, time to rat on our buddy... */ + + printk(KERN_ERR "RCU detected CPU stalls:"); + for_each_possible_cpu(cpu) { + if (cpu_isset(cpu, rcp->cpumask)) + printk(" %d", cpu); + } + printk(" (detected by %d, t=%ld jiffies)\n", + smp_processor_id(), (long)(jiffies - rcp->gp_start)); +} + +static void print_cpu_stall(struct rcu_ctrlblk *rcp) +{ + unsigned long flags; + + printk(KERN_ERR "RCU detected CPU %d stall (t=%lu/%lu jiffies)\n", + smp_processor_id(), jiffies, + jiffies - rcp->gp_start); + dump_stack(); + spin_lock_irqsave(&rcp->lock, flags); + if ((long)(jiffies - rcp->jiffies_stall) >= 0) + rcp->jiffies_stall = + jiffies + RCU_SECONDS_TILL_STALL_RECHECK; + spin_unlock_irqrestore(&rcp->lock, flags); + set_need_resched(); /* kick ourselves to get things going. */ +} + +static void check_cpu_stall(struct rcu_ctrlblk *rcp) +{ + long delta; + + delta = jiffies - rcp->jiffies_stall; + if (cpu_isset(smp_processor_id(), rcp->cpumask) && delta >= 0) { + + /* We haven't checked in, so go dump stack. */ + print_cpu_stall(rcp); + + } else if (rcp->cur != rcp->completed && delta >= 2) { + + /* They had two seconds to dump stack, so complain. */ + print_other_cpu_stall(rcp); + } +} + +#else /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + +static void record_gp_stall_check_time(struct rcu_ctrlblk *rcp) +{ +} + +static inline void check_cpu_stall(struct rcu_ctrlblk *rcp) +{ +} + +#endif /* #else #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ + /** * call_rcu - Queue an RCU callback for invocation after a grace period. * @head: structure to be used for queueing the RCU updates. @@ -133,18 +260,10 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { unsigned long flags; - struct rcu_data *rdp; head->func = func; - head->next = NULL; local_irq_save(flags); - rdp = &__get_cpu_var(rcu_data); - *rdp->nxttail = head; - rdp->nxttail = &head->next; - if (unlikely(++rdp->qlen > qhimark)) { - rdp->blimit = INT_MAX; - force_quiescent_state(rdp, &rcu_ctrlblk); - } + __call_rcu(head, &rcu_ctrlblk, &__get_cpu_var(rcu_data)); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(call_rcu); @@ -169,20 +288,10 @@ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { unsigned long flags; - struct rcu_data *rdp; head->func = func; - head->next = NULL; local_irq_save(flags); - rdp = &__get_cpu_var(rcu_bh_data); - *rdp->nxttail = head; - rdp->nxttail = &head->next; - - if (unlikely(++rdp->qlen > qhimark)) { - rdp->blimit = INT_MAX; - force_quiescent_state(rdp, &rcu_bh_ctrlblk); - } - + __call_rcu(head, &rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); local_irq_restore(flags); } EXPORT_SYMBOL_GPL(call_rcu_bh); @@ -211,12 +320,6 @@ EXPORT_SYMBOL_GPL(rcu_batches_completed_bh); static inline void raise_rcu_softirq(void) { raise_softirq(RCU_SOFTIRQ); - /* - * The smp_mb() here is required to ensure that this cpu's - * __rcu_process_callbacks() reads the most recently updated - * value of rcu->cur. - */ - smp_mb(); } /* @@ -225,6 +328,7 @@ static inline void raise_rcu_softirq(void) */ static void rcu_do_batch(struct rcu_data *rdp) { + unsigned long flags; struct rcu_head *next, *list; int count = 0; @@ -239,9 +343,9 @@ static void rcu_do_batch(struct rcu_data *rdp) } rdp->donelist = list; - local_irq_disable(); + local_irq_save(flags); rdp->qlen -= count; - local_irq_enable(); + local_irq_restore(flags); if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark) rdp->blimit = blimit; @@ -269,6 +373,7 @@ static void rcu_do_batch(struct rcu_data *rdp) * rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace * period (if necessary). */ + /* * Register a new batch of callbacks, and start it up if there is currently no * active batch and the batch to be registered has not already occurred. @@ -276,15 +381,10 @@ static void rcu_do_batch(struct rcu_data *rdp) */ static void rcu_start_batch(struct rcu_ctrlblk *rcp) { - if (rcp->next_pending && + if (rcp->cur != rcp->pending && rcp->completed == rcp->cur) { - rcp->next_pending = 0; - /* - * next_pending == 0 must be visible in - * __rcu_process_callbacks() before it can see new value of cur. - */ - smp_wmb(); rcp->cur++; + record_gp_stall_check_time(rcp); /* * Accessing nohz_cpu_mask before incrementing rcp->cur needs a @@ -322,6 +422,8 @@ static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp) static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, struct rcu_data *rdp) { + unsigned long flags; + if (rdp->quiescbatch != rcp->cur) { /* start new grace period: */ rdp->qs_pending = 1; @@ -345,7 +447,7 @@ static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, return; rdp->qs_pending = 0; - spin_lock(&rcp->lock); + spin_lock_irqsave(&rcp->lock, flags); /* * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync * during cpu startup. Ignore the quiescent state. @@ -353,7 +455,7 @@ static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, if (likely(rdp->quiescbatch == rcp->cur)) cpu_quiet(rdp->cpu, rcp); - spin_unlock(&rcp->lock); + spin_unlock_irqrestore(&rcp->lock, flags); } @@ -364,33 +466,38 @@ static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp, * which is dead and hence not processing interrupts. */ static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list, - struct rcu_head **tail) + struct rcu_head **tail, long batch) { - local_irq_disable(); - *this_rdp->nxttail = list; - if (list) - this_rdp->nxttail = tail; - local_irq_enable(); + unsigned long flags; + + if (list) { + local_irq_save(flags); + this_rdp->batch = batch; + *this_rdp->nxttail[2] = list; + this_rdp->nxttail[2] = tail; + local_irq_restore(flags); + } } static void __rcu_offline_cpu(struct rcu_data *this_rdp, struct rcu_ctrlblk *rcp, struct rcu_data *rdp) { - /* if the cpu going offline owns the grace period + unsigned long flags; + + /* + * if the cpu going offline owns the grace period * we can block indefinitely waiting for it, so flush * it here */ - spin_lock_bh(&rcp->lock); + spin_lock_irqsave(&rcp->lock, flags); if (rcp->cur != rcp->completed) cpu_quiet(rdp->cpu, rcp); - spin_unlock_bh(&rcp->lock); - rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail); - rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail); - rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail); + rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail, rcp->cur + 1); + rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail[2], rcp->cur + 1); + spin_unlock(&rcp->lock); - local_irq_disable(); this_rdp->qlen += rdp->qlen; - local_irq_enable(); + local_irq_restore(flags); } static void rcu_offline_cpu(int cpu) @@ -420,38 +527,52 @@ static void rcu_offline_cpu(int cpu) static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp, struct rcu_data *rdp) { - if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) { - *rdp->donetail = rdp->curlist; - rdp->donetail = rdp->curtail; - rdp->curlist = NULL; - rdp->curtail = &rdp->curlist; - } + unsigned long flags; + long completed_snap; - if (rdp->nxtlist && !rdp->curlist) { - local_irq_disable(); - rdp->curlist = rdp->nxtlist; - rdp->curtail = rdp->nxttail; - rdp->nxtlist = NULL; - rdp->nxttail = &rdp->nxtlist; - local_irq_enable(); + if (rdp->nxtlist) { + local_irq_save(flags); + completed_snap = ACCESS_ONCE(rcp->completed); /* - * start the next batch of callbacks + * move the other grace-period-completed entries to + * [rdp->nxtlist, *rdp->nxttail[0]) temporarily */ + if (!rcu_batch_before(completed_snap, rdp->batch)) + rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2]; + else if (!rcu_batch_before(completed_snap, rdp->batch - 1)) + rdp->nxttail[0] = rdp->nxttail[1]; - /* determine batch number */ - rdp->batch = rcp->cur + 1; - /* see the comment and corresponding wmb() in - * the rcu_start_batch() + /* + * the grace period for entries in + * [rdp->nxtlist, *rdp->nxttail[0]) has completed and + * move these entries to donelist */ - smp_rmb(); + if (rdp->nxttail[0] != &rdp->nxtlist) { + *rdp->donetail = rdp->nxtlist; + rdp->donetail = rdp->nxttail[0]; + rdp->nxtlist = *rdp->nxttail[0]; + *rdp->donetail = NULL; + + if (rdp->nxttail[1] == rdp->nxttail[0]) + rdp->nxttail[1] = &rdp->nxtlist; + if (rdp->nxttail[2] == rdp->nxttail[0]) + rdp->nxttail[2] = &rdp->nxtlist; + rdp->nxttail[0] = &rdp->nxtlist; + } + + local_irq_restore(flags); + + if (rcu_batch_after(rdp->batch, rcp->pending)) { + unsigned long flags2; - if (!rcp->next_pending) { /* and start it/schedule start if it's a new batch */ - spin_lock(&rcp->lock); - rcp->next_pending = 1; - rcu_start_batch(rcp); - spin_unlock(&rcp->lock); + spin_lock_irqsave(&rcp->lock, flags2); + if (rcu_batch_after(rdp->batch, rcp->pending)) { + rcp->pending = rdp->batch; + rcu_start_batch(rcp); + } + spin_unlock_irqrestore(&rcp->lock, flags2); } } @@ -462,21 +583,53 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp, static void rcu_process_callbacks(struct softirq_action *unused) { + /* + * Memory references from any prior RCU read-side critical sections + * executed by the interrupted code must be see before any RCU + * grace-period manupulations below. + */ + + smp_mb(); /* See above block comment. */ + __rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data)); __rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data)); + + /* + * Memory references from any later RCU read-side critical sections + * executed by the interrupted code must be see after any RCU + * grace-period manupulations above. + */ + + smp_mb(); /* See above block comment. */ } static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp) { - /* This cpu has pending rcu entries and the grace period - * for them has completed. - */ - if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) - return 1; + /* Check for CPU stalls, if enabled. */ + check_cpu_stall(rcp); - /* This cpu has no pending entries, but there are new entries */ - if (!rdp->curlist && rdp->nxtlist) - return 1; + if (rdp->nxtlist) { + long completed_snap = ACCESS_ONCE(rcp->completed); + + /* + * This cpu has pending rcu entries and the grace period + * for them has completed. + */ + if (!rcu_batch_before(completed_snap, rdp->batch)) + return 1; + if (!rcu_batch_before(completed_snap, rdp->batch - 1) && + rdp->nxttail[0] != rdp->nxttail[1]) + return 1; + if (rdp->nxttail[0] != &rdp->nxtlist) + return 1; + + /* + * This cpu has pending rcu entries and the new batch + * for then hasn't been started nor scheduled start + */ + if (rcu_batch_after(rdp->batch, rcp->pending)) + return 1; + } /* This cpu has finished callbacks to invoke */ if (rdp->donelist) @@ -512,9 +665,15 @@ int rcu_needs_cpu(int cpu) struct rcu_data *rdp = &per_cpu(rcu_data, cpu); struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu); - return (!!rdp->curlist || !!rdp_bh->curlist || rcu_pending(cpu)); + return !!rdp->nxtlist || !!rdp_bh->nxtlist || rcu_pending(cpu); } +/* + * Top-level function driving RCU grace-period detection, normally + * invoked from the scheduler-clock interrupt. This function simply + * increments counters that are read only from softirq by this same + * CPU, so there are no memory barriers required. + */ void rcu_check_callbacks(int cpu, int user) { if (user || @@ -558,14 +717,17 @@ void rcu_check_callbacks(int cpu, int user) static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp, struct rcu_data *rdp) { + unsigned long flags; + + spin_lock_irqsave(&rcp->lock, flags); memset(rdp, 0, sizeof(*rdp)); - rdp->curtail = &rdp->curlist; - rdp->nxttail = &rdp->nxtlist; + rdp->nxttail[0] = rdp->nxttail[1] = rdp->nxttail[2] = &rdp->nxtlist; rdp->donetail = &rdp->donelist; rdp->quiescbatch = rcp->completed; rdp->qs_pending = 0; rdp->cpu = cpu; rdp->blimit = blimit; + spin_unlock_irqrestore(&rcp->lock, flags); } static void __cpuinit rcu_online_cpu(int cpu) @@ -610,6 +772,9 @@ static struct notifier_block __cpuinitdata rcu_nb = { */ void __init __rcu_init(void) { +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR + printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long)smp_processor_id()); /* Register notifier for non-boot CPUs */ diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c index 27827931ca0d..ca4bbbe04aa4 100644 --- a/kernel/rcupreempt.c +++ b/kernel/rcupreempt.c @@ -59,14 +59,6 @@ #include <linux/rcupreempt_trace.h> /* - * Macro that prevents the compiler from reordering accesses, but does - * absolutely -nothing- to prevent CPUs from reordering. This is used - * only to mediate communication between mainline code and hardware - * interrupt and NMI handlers. - */ -#define ACCESS_ONCE(x) (*(volatile typeof(x) *)&(x)) - -/* * PREEMPT_RCU data structures. */ diff --git a/kernel/rcupreempt_trace.c b/kernel/rcupreempt_trace.c index 5edf82c34bbc..35c2d3360ecf 100644 --- a/kernel/rcupreempt_trace.c +++ b/kernel/rcupreempt_trace.c @@ -308,11 +308,16 @@ out: static int __init rcupreempt_trace_init(void) { + int ret; + mutex_init(&rcupreempt_trace_mutex); rcupreempt_trace_buf = kmalloc(RCUPREEMPT_TRACE_BUF_SIZE, GFP_KERNEL); if (!rcupreempt_trace_buf) return 1; - return rcupreempt_debugfs_init(); + ret = rcupreempt_debugfs_init(); + if (ret) + kfree(rcupreempt_trace_buf); + return ret; } static void __exit rcupreempt_trace_cleanup(void) diff --git a/kernel/resource.c b/kernel/resource.c index 03d796c1b2e9..414d6fc9131e 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -516,6 +516,74 @@ int adjust_resource(struct resource *res, resource_size_t start, resource_size_t return result; } +static void __init __reserve_region_with_split(struct resource *root, + resource_size_t start, resource_size_t end, + const char *name) +{ + struct resource *parent = root; + struct resource *conflict; + struct resource *res = kzalloc(sizeof(*res), GFP_KERNEL); + + if (!res) + return; + + res->name = name; + res->start = start; + res->end = end; + res->flags = IORESOURCE_BUSY; + + for (;;) { + conflict = __request_resource(parent, res); + if (!conflict) + break; + if (conflict != parent) { + parent = conflict; + if (!(conflict->flags & IORESOURCE_BUSY)) + continue; + } + + /* Uhhuh, that didn't work out.. */ + kfree(res); + res = NULL; + break; + } + + if (!res) { + printk(KERN_DEBUG " __reserve_region_with_split: (%s) [%llx, %llx], res: (%s) [%llx, %llx]\n", + conflict->name, conflict->start, conflict->end, + name, start, end); + + /* failed, split and try again */ + + /* conflict coverred whole area */ + if (conflict->start <= start && conflict->end >= end) + return; + + if (conflict->start > start) + __reserve_region_with_split(root, start, conflict->start-1, name); + if (!(conflict->flags & IORESOURCE_BUSY)) { + resource_size_t common_start, common_end; + + common_start = max(conflict->start, start); + common_end = min(conflict->end, end); + if (common_start < common_end) + __reserve_region_with_split(root, common_start, common_end, name); + } + if (conflict->end < end) + __reserve_region_with_split(root, conflict->end+1, end, name); + } + +} + +void reserve_region_with_split(struct resource *root, + resource_size_t start, resource_size_t end, + const char *name) +{ + write_lock(&resource_lock); + __reserve_region_with_split(root, start, end, name); + write_unlock(&resource_lock); +} + EXPORT_SYMBOL(adjust_resource); /** diff --git a/kernel/sched.c b/kernel/sched.c index 9a1ddb84e26d..6f230596bd0c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -201,14 +201,19 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime) hrtimer_init(&rt_b->rt_period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); rt_b->rt_period_timer.function = sched_rt_period_timer; - rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ; + rt_b->rt_period_timer.cb_mode = HRTIMER_CB_IRQSAFE_UNLOCKED; +} + +static inline int rt_bandwidth_enabled(void) +{ + return sysctl_sched_rt_runtime >= 0; } static void start_rt_bandwidth(struct rt_bandwidth *rt_b) { ktime_t now; - if (rt_b->rt_runtime == RUNTIME_INF) + if (rt_bandwidth_enabled() && rt_b->rt_runtime == RUNTIME_INF) return; if (hrtimer_active(&rt_b->rt_period_timer)) @@ -298,9 +303,9 @@ static DEFINE_PER_CPU(struct cfs_rq, init_cfs_rq) ____cacheline_aligned_in_smp; static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity); static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp; #endif /* CONFIG_RT_GROUP_SCHED */ -#else /* !CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_USER_SCHED */ #define root_task_group init_task_group -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* CONFIG_USER_SCHED */ /* task_group_lock serializes add/remove of task groups and also changes to * a task group's cpu shares. @@ -604,9 +609,9 @@ struct rq { static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); -static inline void check_preempt_curr(struct rq *rq, struct task_struct *p) +static inline void check_preempt_curr(struct rq *rq, struct task_struct *p, int sync) { - rq->curr->sched_class->check_preempt_curr(rq, p); + rq->curr->sched_class->check_preempt_curr(rq, p, sync); } static inline int cpu_of(struct rq *rq) @@ -1087,7 +1092,7 @@ hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu) return NOTIFY_DONE; } -static void init_hrtick(void) +static __init void init_hrtick(void) { hotcpu_notifier(hotplug_hrtick, 0); } @@ -1102,7 +1107,7 @@ static void hrtick_start(struct rq *rq, u64 delay) hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), HRTIMER_MODE_REL); } -static void init_hrtick(void) +static inline void init_hrtick(void) { } #endif /* CONFIG_SMP */ @@ -1119,9 +1124,9 @@ static void init_rq_hrtick(struct rq *rq) hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); rq->hrtick_timer.function = hrtick; - rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ; + rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU; } -#else +#else /* CONFIG_SCHED_HRTICK */ static inline void hrtick_clear(struct rq *rq) { } @@ -1133,7 +1138,7 @@ static inline void init_rq_hrtick(struct rq *rq) static inline void init_hrtick(void) { } -#endif +#endif /* CONFIG_SCHED_HRTICK */ /* * resched_task - mark a task 'to be rescheduled now'. @@ -1380,38 +1385,24 @@ static inline void dec_cpu_load(struct rq *rq, unsigned long load) update_load_sub(&rq->load, load); } -#ifdef CONFIG_SMP -static unsigned long source_load(int cpu, int type); -static unsigned long target_load(int cpu, int type); -static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); - -static unsigned long cpu_avg_load_per_task(int cpu) -{ - struct rq *rq = cpu_rq(cpu); - - if (rq->nr_running) - rq->avg_load_per_task = rq->load.weight / rq->nr_running; - - return rq->avg_load_per_task; -} - -#ifdef CONFIG_FAIR_GROUP_SCHED - -typedef void (*tg_visitor)(struct task_group *, int, struct sched_domain *); +#if (defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED)) || defined(CONFIG_RT_GROUP_SCHED) +typedef int (*tg_visitor)(struct task_group *, void *); /* * Iterate the full tree, calling @down when first entering a node and @up when * leaving it for the final time. */ -static void -walk_tg_tree(tg_visitor down, tg_visitor up, int cpu, struct sched_domain *sd) +static int walk_tg_tree(tg_visitor down, tg_visitor up, void *data) { struct task_group *parent, *child; + int ret; rcu_read_lock(); parent = &root_task_group; down: - (*down)(parent, cpu, sd); + ret = (*down)(parent, data); + if (ret) + goto out_unlock; list_for_each_entry_rcu(child, &parent->children, siblings) { parent = child; goto down; @@ -1419,15 +1410,43 @@ down: up: continue; } - (*up)(parent, cpu, sd); + ret = (*up)(parent, data); + if (ret) + goto out_unlock; child = parent; parent = parent->parent; if (parent) goto up; +out_unlock: rcu_read_unlock(); + + return ret; +} + +static int tg_nop(struct task_group *tg, void *data) +{ + return 0; +} +#endif + +#ifdef CONFIG_SMP +static unsigned long source_load(int cpu, int type); +static unsigned long target_load(int cpu, int type); +static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); + +static unsigned long cpu_avg_load_per_task(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + + if (rq->nr_running) + rq->avg_load_per_task = rq->load.weight / rq->nr_running; + + return rq->avg_load_per_task; } +#ifdef CONFIG_FAIR_GROUP_SCHED + static void __set_se_shares(struct sched_entity *se, unsigned long shares); /* @@ -1486,11 +1505,11 @@ __update_group_shares_cpu(struct task_group *tg, int cpu, * This needs to be done in a bottom-up fashion because the rq weight of a * parent group depends on the shares of its child groups. */ -static void -tg_shares_up(struct task_group *tg, int cpu, struct sched_domain *sd) +static int tg_shares_up(struct task_group *tg, void *data) { unsigned long rq_weight = 0; unsigned long shares = 0; + struct sched_domain *sd = data; int i; for_each_cpu_mask(i, sd->span) { @@ -1515,6 +1534,8 @@ tg_shares_up(struct task_group *tg, int cpu, struct sched_domain *sd) __update_group_shares_cpu(tg, i, shares, rq_weight); spin_unlock_irqrestore(&rq->lock, flags); } + + return 0; } /* @@ -1522,10 +1543,10 @@ tg_shares_up(struct task_group *tg, int cpu, struct sched_domain *sd) * This needs to be done in a top-down fashion because the load of a child * group is a fraction of its parents load. */ -static void -tg_load_down(struct task_group *tg, int cpu, struct sched_domain *sd) +static int tg_load_down(struct task_group *tg, void *data) { unsigned long load; + long cpu = (long)data; if (!tg->parent) { load = cpu_rq(cpu)->load.weight; @@ -1536,11 +1557,8 @@ tg_load_down(struct task_group *tg, int cpu, struct sched_domain *sd) } tg->cfs_rq[cpu]->h_load = load; -} -static void -tg_nop(struct task_group *tg, int cpu, struct sched_domain *sd) -{ + return 0; } static void update_shares(struct sched_domain *sd) @@ -1550,7 +1568,7 @@ static void update_shares(struct sched_domain *sd) if (elapsed >= (s64)(u64)sysctl_sched_shares_ratelimit) { sd->last_update = now; - walk_tg_tree(tg_nop, tg_shares_up, 0, sd); + walk_tg_tree(tg_nop, tg_shares_up, sd); } } @@ -1561,9 +1579,9 @@ static void update_shares_locked(struct rq *rq, struct sched_domain *sd) spin_lock(&rq->lock); } -static void update_h_load(int cpu) +static void update_h_load(long cpu) { - walk_tg_tree(tg_load_down, tg_nop, cpu, NULL); + walk_tg_tree(tg_load_down, tg_nop, (void *)cpu); } #else @@ -1921,11 +1939,8 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) running = task_running(rq, p); on_rq = p->se.on_rq; ncsw = 0; - if (!match_state || p->state == match_state) { - ncsw = p->nivcsw + p->nvcsw; - if (unlikely(!ncsw)) - ncsw = 1; - } + if (!match_state || p->state == match_state) + ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ task_rq_unlock(rq, &flags); /* @@ -2285,7 +2300,7 @@ out_running: trace_mark(kernel_sched_wakeup, "pid %d state %ld ## rq %p task %p rq->curr %p", p->pid, p->state, rq, p, rq->curr); - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, sync); p->state = TASK_RUNNING; #ifdef CONFIG_SMP @@ -2420,7 +2435,7 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) trace_mark(kernel_sched_wakeup_new, "pid %d state %ld ## rq %p task %p rq->curr %p", p->pid, p->state, rq, p, rq->curr); - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, 0); #ifdef CONFIG_SMP if (p->sched_class->task_wake_up) p->sched_class->task_wake_up(rq, p); @@ -2880,7 +2895,7 @@ static void pull_task(struct rq *src_rq, struct task_struct *p, * Note that idle threads have a prio of MAX_PRIO, for this test * to be always true for them. */ - check_preempt_curr(this_rq, p); + check_preempt_curr(this_rq, p, 0); } /* @@ -4179,6 +4194,65 @@ void account_steal_time(struct task_struct *p, cputime_t steal) } /* + * Use precise platform statistics if available: + */ +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +cputime_t task_utime(struct task_struct *p) +{ + return p->utime; +} + +cputime_t task_stime(struct task_struct *p) +{ + return p->stime; +} +#else +cputime_t task_utime(struct task_struct *p) +{ + clock_t utime = cputime_to_clock_t(p->utime), + total = utime + cputime_to_clock_t(p->stime); + u64 temp; + + /* + * Use CFS's precise accounting: + */ + temp = (u64)nsec_to_clock_t(p->se.sum_exec_runtime); + + if (total) { + temp *= utime; + do_div(temp, total); + } + utime = (clock_t)temp; + + p->prev_utime = max(p->prev_utime, clock_t_to_cputime(utime)); + return p->prev_utime; +} + +cputime_t task_stime(struct task_struct *p) +{ + clock_t stime; + + /* + * Use CFS's precise accounting. (we subtract utime from + * the total, to make sure the total observed by userspace + * grows monotonically - apps rely on that): + */ + stime = nsec_to_clock_t(p->se.sum_exec_runtime) - + cputime_to_clock_t(task_utime(p)); + + if (stime >= 0) + p->prev_stime = max(p->prev_stime, clock_t_to_cputime(stime)); + + return p->prev_stime; +} +#endif + +inline cputime_t task_gtime(struct task_struct *p) +{ + return p->gtime; +} + +/* * This function gets called by the timer code, with HZ frequency. * We call it with interrupts disabled. * @@ -4568,6 +4642,15 @@ __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) } EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ +/** + * complete: - signals a single thread waiting on this completion + * @x: holds the state of this particular completion + * + * This will wake up a single thread waiting on this completion. Threads will be + * awakened in the same order in which they were queued. + * + * See also complete_all(), wait_for_completion() and related routines. + */ void complete(struct completion *x) { unsigned long flags; @@ -4579,6 +4662,12 @@ void complete(struct completion *x) } EXPORT_SYMBOL(complete); +/** + * complete_all: - signals all threads waiting on this completion + * @x: holds the state of this particular completion + * + * This will wake up all threads waiting on this particular completion event. + */ void complete_all(struct completion *x) { unsigned long flags; @@ -4599,10 +4688,7 @@ do_wait_for_common(struct completion *x, long timeout, int state) wait.flags |= WQ_FLAG_EXCLUSIVE; __add_wait_queue_tail(&x->wait, &wait); do { - if ((state == TASK_INTERRUPTIBLE && - signal_pending(current)) || - (state == TASK_KILLABLE && - fatal_signal_pending(current))) { + if (signal_pending_state(state, current)) { timeout = -ERESTARTSYS; break; } @@ -4630,12 +4716,31 @@ wait_for_common(struct completion *x, long timeout, int state) return timeout; } +/** + * wait_for_completion: - waits for completion of a task + * @x: holds the state of this particular completion + * + * This waits to be signaled for completion of a specific task. It is NOT + * interruptible and there is no timeout. + * + * See also similar routines (i.e. wait_for_completion_timeout()) with timeout + * and interrupt capability. Also see complete(). + */ void __sched wait_for_completion(struct completion *x) { wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE); } EXPORT_SYMBOL(wait_for_completion); +/** + * wait_for_completion_timeout: - waits for completion of a task (w/timeout) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be signaled or for a + * specified timeout to expire. The timeout is in jiffies. It is not + * interruptible. + */ unsigned long __sched wait_for_completion_timeout(struct completion *x, unsigned long timeout) { @@ -4643,6 +4748,13 @@ wait_for_completion_timeout(struct completion *x, unsigned long timeout) } EXPORT_SYMBOL(wait_for_completion_timeout); +/** + * wait_for_completion_interruptible: - waits for completion of a task (w/intr) + * @x: holds the state of this particular completion + * + * This waits for completion of a specific task to be signaled. It is + * interruptible. + */ int __sched wait_for_completion_interruptible(struct completion *x) { long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_INTERRUPTIBLE); @@ -4652,6 +4764,14 @@ int __sched wait_for_completion_interruptible(struct completion *x) } EXPORT_SYMBOL(wait_for_completion_interruptible); +/** + * wait_for_completion_interruptible_timeout: - waits for completion (w/(to,intr)) + * @x: holds the state of this particular completion + * @timeout: timeout value in jiffies + * + * This waits for either a completion of a specific task to be signaled or for a + * specified timeout to expire. It is interruptible. The timeout is in jiffies. + */ unsigned long __sched wait_for_completion_interruptible_timeout(struct completion *x, unsigned long timeout) @@ -4660,6 +4780,13 @@ wait_for_completion_interruptible_timeout(struct completion *x, } EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); +/** + * wait_for_completion_killable: - waits for completion of a task (killable) + * @x: holds the state of this particular completion + * + * This waits to be signaled for completion of a specific task. It can be + * interrupted by a kill signal. + */ int __sched wait_for_completion_killable(struct completion *x) { long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_KILLABLE); @@ -5062,7 +5189,8 @@ recheck: * Do not allow realtime tasks into groups that have no runtime * assigned. */ - if (rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0) + if (rt_bandwidth_enabled() && rt_policy(policy) && + task_group(p)->rt_bandwidth.rt_runtime == 0) return -EPERM; #endif @@ -5898,7 +6026,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) set_task_cpu(p, dest_cpu); if (on_rq) { activate_task(rq_dest, p, 0); - check_preempt_curr(rq_dest, p); + check_preempt_curr(rq_dest, p, 0); } done: ret = 1; @@ -6223,7 +6351,7 @@ set_table_entry(struct ctl_table *entry, static struct ctl_table * sd_alloc_ctl_domain_table(struct sched_domain *sd) { - struct ctl_table *table = sd_alloc_ctl_entry(12); + struct ctl_table *table = sd_alloc_ctl_entry(13); if (table == NULL) return NULL; @@ -6251,7 +6379,9 @@ sd_alloc_ctl_domain_table(struct sched_domain *sd) sizeof(int), 0644, proc_dointvec_minmax); set_table_entry(&table[10], "flags", &sd->flags, sizeof(int), 0644, proc_dointvec_minmax); - /* &table[11] is terminator */ + set_table_entry(&table[11], "name", sd->name, + CORENAME_MAX_SIZE, 0444, proc_dostring); + /* &table[12] is terminator */ return table; } @@ -7135,13 +7265,21 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd) * Non-inlined to reduce accumulated stack pressure in build_sched_domains() */ +#ifdef CONFIG_SCHED_DEBUG +# define SD_INIT_NAME(sd, type) sd->name = #type +#else +# define SD_INIT_NAME(sd, type) do { } while (0) +#endif + #define SD_INIT(sd, type) sd_init_##type(sd) + #define SD_INIT_FUNC(type) \ static noinline void sd_init_##type(struct sched_domain *sd) \ { \ memset(sd, 0, sizeof(*sd)); \ *sd = SD_##type##_INIT; \ sd->level = SD_LV_##type; \ + SD_INIT_NAME(sd, type); \ } SD_INIT_FUNC(CPU) @@ -7637,24 +7775,27 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * and partition_sched_domains() will fallback to the single partition * 'fallback_doms', it also forces the domains to be rebuilt. * + * If doms_new==NULL it will be replaced with cpu_online_map. + * ndoms_new==0 is a special case for destroying existing domains. + * It will not create the default domain. + * * Call with hotplug lock held */ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, struct sched_domain_attr *dattr_new) { - int i, j; + int i, j, n; mutex_lock(&sched_domains_mutex); /* always unregister in case we don't destroy any domains */ unregister_sched_domain_sysctl(); - if (doms_new == NULL) - ndoms_new = 0; + n = doms_new ? ndoms_new : 0; /* Destroy deleted domains */ for (i = 0; i < ndoms_cur; i++) { - for (j = 0; j < ndoms_new; j++) { + for (j = 0; j < n; j++) { if (cpus_equal(doms_cur[i], doms_new[j]) && dattrs_equal(dattr_cur, i, dattr_new, j)) goto match1; @@ -7667,7 +7808,6 @@ match1: if (doms_new == NULL) { ndoms_cur = 0; - ndoms_new = 1; doms_new = &fallback_doms; cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map); dattr_new = NULL; @@ -7704,8 +7844,13 @@ match2: int arch_reinit_sched_domains(void) { get_online_cpus(); + + /* Destroy domains first to force the rebuild */ + partition_sched_domains(0, NULL, NULL); + rebuild_sched_domains(); put_online_cpus(); + return 0; } @@ -7789,7 +7934,7 @@ static int update_sched_domains(struct notifier_block *nfb, case CPU_ONLINE_FROZEN: case CPU_DEAD: case CPU_DEAD_FROZEN: - partition_sched_domains(0, NULL, NULL); + partition_sched_domains(1, NULL, NULL); return NOTIFY_OK; default: @@ -8176,20 +8321,25 @@ void __might_sleep(char *file, int line) #ifdef in_atomic static unsigned long prev_jiffy; /* ratelimiting */ - if ((in_atomic() || irqs_disabled()) && - system_state == SYSTEM_RUNNING && !oops_in_progress) { - if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) - return; - prev_jiffy = jiffies; - printk(KERN_ERR "BUG: sleeping function called from invalid" - " context at %s:%d\n", file, line); - printk("in_atomic():%d, irqs_disabled():%d\n", - in_atomic(), irqs_disabled()); - debug_show_held_locks(current); - if (irqs_disabled()) - print_irqtrace_events(current); - dump_stack(); - } + if ((!in_atomic() && !irqs_disabled()) || + system_state != SYSTEM_RUNNING || oops_in_progress) + return; + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) + 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); + + debug_show_held_locks(current); + if (irqs_disabled()) + print_irqtrace_events(current); + dump_stack(); #endif } EXPORT_SYMBOL(__might_sleep); @@ -8687,73 +8837,95 @@ static DEFINE_MUTEX(rt_constraints_mutex); static unsigned long to_ratio(u64 period, u64 runtime) { if (runtime == RUNTIME_INF) - return 1ULL << 16; + return 1ULL << 20; - return div64_u64(runtime << 16, period); + return div64_u64(runtime << 20, period); } -#ifdef CONFIG_CGROUP_SCHED -static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) +/* Must be called with tasklist_lock held */ +static inline int tg_has_rt_tasks(struct task_group *tg) { - struct task_group *tgi, *parent = tg->parent; - unsigned long total = 0; + struct task_struct *g, *p; - if (!parent) { - if (global_rt_period() < period) - return 0; + do_each_thread(g, p) { + if (rt_task(p) && rt_rq_of_se(&p->rt)->tg == tg) + return 1; + } while_each_thread(g, p); - return to_ratio(period, runtime) < - to_ratio(global_rt_period(), global_rt_runtime()); - } + return 0; +} - if (ktime_to_ns(parent->rt_bandwidth.rt_period) < period) - return 0; +struct rt_schedulable_data { + struct task_group *tg; + u64 rt_period; + u64 rt_runtime; +}; - rcu_read_lock(); - list_for_each_entry_rcu(tgi, &parent->children, siblings) { - if (tgi == tg) - continue; +static int tg_schedulable(struct task_group *tg, void *data) +{ + struct rt_schedulable_data *d = data; + struct task_group *child; + unsigned long total, sum = 0; + u64 period, runtime; + + period = ktime_to_ns(tg->rt_bandwidth.rt_period); + runtime = tg->rt_bandwidth.rt_runtime; - total += to_ratio(ktime_to_ns(tgi->rt_bandwidth.rt_period), - tgi->rt_bandwidth.rt_runtime); + if (tg == d->tg) { + period = d->rt_period; + runtime = d->rt_runtime; } - rcu_read_unlock(); - return total + to_ratio(period, runtime) <= - to_ratio(ktime_to_ns(parent->rt_bandwidth.rt_period), - parent->rt_bandwidth.rt_runtime); -} -#elif defined CONFIG_USER_SCHED -static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) -{ - struct task_group *tgi; - unsigned long total = 0; - unsigned long global_ratio = - to_ratio(global_rt_period(), global_rt_runtime()); + /* + * Cannot have more runtime than the period. + */ + if (runtime > period && runtime != RUNTIME_INF) + return -EINVAL; - rcu_read_lock(); - list_for_each_entry_rcu(tgi, &task_groups, list) { - if (tgi == tg) - continue; + /* + * Ensure we don't starve existing RT tasks. + */ + if (rt_bandwidth_enabled() && !runtime && tg_has_rt_tasks(tg)) + return -EBUSY; + + total = to_ratio(period, runtime); - total += to_ratio(ktime_to_ns(tgi->rt_bandwidth.rt_period), - tgi->rt_bandwidth.rt_runtime); + /* + * Nobody can have more than the global setting allows. + */ + if (total > to_ratio(global_rt_period(), global_rt_runtime())) + return -EINVAL; + + /* + * The sum of our children's runtime should not exceed our own. + */ + list_for_each_entry_rcu(child, &tg->children, siblings) { + period = ktime_to_ns(child->rt_bandwidth.rt_period); + runtime = child->rt_bandwidth.rt_runtime; + + if (child == d->tg) { + period = d->rt_period; + runtime = d->rt_runtime; + } + + sum += to_ratio(period, runtime); } - rcu_read_unlock(); - return total + to_ratio(period, runtime) < global_ratio; + if (sum > total) + return -EINVAL; + + return 0; } -#endif -/* Must be called with tasklist_lock held */ -static inline int tg_has_rt_tasks(struct task_group *tg) +static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) { - struct task_struct *g, *p; - do_each_thread(g, p) { - if (rt_task(p) && rt_rq_of_se(&p->rt)->tg == tg) - return 1; - } while_each_thread(g, p); - return 0; + struct rt_schedulable_data data = { + .tg = tg, + .rt_period = period, + .rt_runtime = runtime, + }; + + return walk_tg_tree(tg_schedulable, tg_nop, &data); } static int tg_set_bandwidth(struct task_group *tg, @@ -8763,14 +8935,9 @@ static int tg_set_bandwidth(struct task_group *tg, mutex_lock(&rt_constraints_mutex); read_lock(&tasklist_lock); - if (rt_runtime == 0 && tg_has_rt_tasks(tg)) { - err = -EBUSY; + err = __rt_schedulable(tg, rt_period, rt_runtime); + if (err) goto unlock; - } - if (!__rt_schedulable(tg, rt_period, rt_runtime)) { - err = -EINVAL; - goto unlock; - } spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); @@ -8839,16 +9006,25 @@ long sched_group_rt_period(struct task_group *tg) static int sched_rt_global_constraints(void) { - struct task_group *tg = &root_task_group; - u64 rt_runtime, rt_period; + u64 runtime, period; int ret = 0; - rt_period = ktime_to_ns(tg->rt_bandwidth.rt_period); - rt_runtime = tg->rt_bandwidth.rt_runtime; + if (sysctl_sched_rt_period <= 0) + return -EINVAL; + + runtime = global_rt_runtime(); + period = global_rt_period(); + + /* + * Sanity check on the sysctl variables. + */ + if (runtime > period && runtime != RUNTIME_INF) + return -EINVAL; mutex_lock(&rt_constraints_mutex); - if (!__rt_schedulable(tg, rt_period, rt_runtime)) - ret = -EINVAL; + read_lock(&tasklist_lock); + ret = __rt_schedulable(NULL, 0, 0); + read_unlock(&tasklist_lock); mutex_unlock(&rt_constraints_mutex); return ret; @@ -8859,6 +9035,9 @@ static int sched_rt_global_constraints(void) unsigned long flags; int i; + if (sysctl_sched_rt_period <= 0) + return -EINVAL; + spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); for_each_possible_cpu(i) { struct rt_rq *rt_rq = &cpu_rq(i)->rt; @@ -8919,7 +9098,6 @@ cpu_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cgrp) if (!cgrp->parent) { /* This is early initialization for the top cgroup */ - init_task_group.css.cgroup = cgrp; return &init_task_group.css; } @@ -8928,9 +9106,6 @@ cpu_cgroup_create(struct cgroup_subsys *ss, struct cgroup *cgrp) if (IS_ERR(tg)) return ERR_PTR(-ENOMEM); - /* Bind the cgroup to task_group object we just created */ - tg->css.cgroup = cgrp; - return &tg->css; } diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index fb8994c6d4bb..18fd17172eb6 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -409,64 +409,6 @@ static u64 sched_vslice_add(struct cfs_rq *cfs_rq, struct sched_entity *se) } /* - * The goal of calc_delta_asym() is to be asymmetrically around NICE_0_LOAD, in - * that it favours >=0 over <0. - * - * -20 | - * | - * 0 --------+------- - * .' - * 19 .' - * - */ -static unsigned long -calc_delta_asym(unsigned long delta, struct sched_entity *se) -{ - struct load_weight lw = { - .weight = NICE_0_LOAD, - .inv_weight = 1UL << (WMULT_SHIFT-NICE_0_SHIFT) - }; - - for_each_sched_entity(se) { - struct load_weight *se_lw = &se->load; - unsigned long rw = cfs_rq_of(se)->load.weight; - -#ifdef CONFIG_FAIR_SCHED_GROUP - struct cfs_rq *cfs_rq = se->my_q; - struct task_group *tg = NULL - - if (cfs_rq) - tg = cfs_rq->tg; - - if (tg && tg->shares < NICE_0_LOAD) { - /* - * scale shares to what it would have been had - * tg->weight been NICE_0_LOAD: - * - * weight = 1024 * shares / tg->weight - */ - lw.weight *= se->load.weight; - lw.weight /= tg->shares; - - lw.inv_weight = 0; - - se_lw = &lw; - rw += lw.weight - se->load.weight; - } else -#endif - - if (se->load.weight < NICE_0_LOAD) { - se_lw = &lw; - rw += NICE_0_LOAD - se->load.weight; - } - - delta = calc_delta_mine(delta, rw, se_lw); - } - - return delta; -} - -/* * Update the current task's runtime statistics. Skip current tasks that * are not in our scheduling class. */ @@ -586,11 +528,12 @@ account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) update_load_add(&cfs_rq->load, se->load.weight); if (!parent_entity(se)) inc_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) + if (entity_is_task(se)) { add_cfs_task_weight(cfs_rq, se->load.weight); + list_add(&se->group_node, &cfs_rq->tasks); + } cfs_rq->nr_running++; se->on_rq = 1; - list_add(&se->group_node, &cfs_rq->tasks); } static void @@ -599,11 +542,12 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) update_load_sub(&cfs_rq->load, se->load.weight); if (!parent_entity(se)) dec_cpu_load(rq_of(cfs_rq), se->load.weight); - if (entity_is_task(se)) + if (entity_is_task(se)) { add_cfs_task_weight(cfs_rq, -se->load.weight); + list_del_init(&se->group_node); + } cfs_rq->nr_running--; se->on_rq = 0; - list_del_init(&se->group_node); } static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) @@ -1085,7 +1029,6 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg) { struct sched_entity *se = tg->se[cpu]; - long more_w; if (!tg->parent) return wl; @@ -1097,18 +1040,17 @@ static long effective_load(struct task_group *tg, int cpu, if (!wl && sched_feat(ASYM_EFF_LOAD)) return wl; - /* - * Instead of using this increment, also add the difference - * between when the shares were last updated and now. - */ - more_w = se->my_q->load.weight - se->my_q->rq_weight; - wl += more_w; - wg += more_w; - for_each_sched_entity(se) { -#define D(n) (likely(n) ? (n) : 1) - long S, rw, s, a, b; + long more_w; + + /* + * Instead of using this increment, also add the difference + * between when the shares were last updated and now. + */ + more_w = se->my_q->load.weight - se->my_q->rq_weight; + wl += more_w; + wg += more_w; S = se->my_q->tg->shares; s = se->my_q->shares; @@ -1117,7 +1059,11 @@ static long effective_load(struct task_group *tg, int cpu, a = S*(rw + wl); b = S*rw + s*wg; - wl = s*(a-b)/D(b); + wl = s*(a-b); + + if (likely(b)) + wl /= b; + /* * Assume the group is already running and will * thus already be accounted for in the weight. @@ -1126,7 +1072,6 @@ static long effective_load(struct task_group *tg, int cpu, * alter the group weight. */ wg = 0; -#undef D } return wl; @@ -1143,7 +1088,7 @@ static inline unsigned long effective_load(struct task_group *tg, int cpu, #endif static int -wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, +wake_affine(struct sched_domain *this_sd, struct rq *this_rq, struct task_struct *p, int prev_cpu, int this_cpu, int sync, int idx, unsigned long load, unsigned long this_load, unsigned int imbalance) @@ -1158,6 +1103,11 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, if (!(this_sd->flags & SD_WAKE_AFFINE) || !sched_feat(AFFINE_WAKEUPS)) return 0; + if (!sync && sched_feat(SYNC_WAKEUPS) && + curr->se.avg_overlap < sysctl_sched_migration_cost && + p->se.avg_overlap < sysctl_sched_migration_cost) + sync = 1; + /* * If sync wakeup then subtract the (maximum possible) * effect of the currently running task from the load @@ -1182,17 +1132,14 @@ wake_affine(struct rq *rq, struct sched_domain *this_sd, struct rq *this_rq, * a reasonable amount of time then attract this newly * woken task: */ - if (sync && balanced) { - if (curr->se.avg_overlap < sysctl_sched_migration_cost && - p->se.avg_overlap < sysctl_sched_migration_cost) - return 1; - } + if (sync && balanced) + return 1; schedstat_inc(p, se.nr_wakeups_affine_attempts); tl_per_task = cpu_avg_load_per_task(this_cpu); - if ((tl <= load && tl + target_load(prev_cpu, idx) <= tl_per_task) || - balanced) { + if (balanced || (tl <= load && tl + target_load(prev_cpu, idx) <= + tl_per_task)) { /* * This domain has SD_WAKE_AFFINE and * p is cache cold in this domain, and @@ -1211,16 +1158,17 @@ static int select_task_rq_fair(struct task_struct *p, int sync) struct sched_domain *sd, *this_sd = NULL; int prev_cpu, this_cpu, new_cpu; unsigned long load, this_load; - struct rq *rq, *this_rq; + struct rq *this_rq; unsigned int imbalance; int idx; prev_cpu = task_cpu(p); - rq = task_rq(p); this_cpu = smp_processor_id(); this_rq = cpu_rq(this_cpu); new_cpu = prev_cpu; + if (prev_cpu == this_cpu) + goto out; /* * 'this_sd' is the first domain that both * this_cpu and prev_cpu are present in: @@ -1248,13 +1196,10 @@ static int select_task_rq_fair(struct task_struct *p, int sync) load = source_load(prev_cpu, idx); this_load = target_load(this_cpu, idx); - if (wake_affine(rq, this_sd, this_rq, p, prev_cpu, this_cpu, sync, idx, + if (wake_affine(this_sd, this_rq, p, prev_cpu, this_cpu, sync, idx, load, this_load, imbalance)) return this_cpu; - if (prev_cpu == this_cpu) - goto out; - /* * Start passive balancing when half the imbalance_pct * limit is reached. @@ -1281,62 +1226,20 @@ static unsigned long wakeup_gran(struct sched_entity *se) * + nice tasks. */ if (sched_feat(ASYM_GRAN)) - gran = calc_delta_asym(sysctl_sched_wakeup_granularity, se); - else - gran = calc_delta_fair(sysctl_sched_wakeup_granularity, se); + gran = calc_delta_mine(gran, NICE_0_LOAD, &se->load); return gran; } /* - * Should 'se' preempt 'curr'. - * - * |s1 - * |s2 - * |s3 - * g - * |<--->|c - * - * w(c, s1) = -1 - * w(c, s2) = 0 - * w(c, s3) = 1 - * - */ -static int -wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se) -{ - s64 gran, vdiff = curr->vruntime - se->vruntime; - - if (vdiff < 0) - return -1; - - gran = wakeup_gran(curr); - if (vdiff > gran) - return 1; - - return 0; -} - -/* return depth at which a sched entity is present in the hierarchy */ -static inline int depth_se(struct sched_entity *se) -{ - int depth = 0; - - for_each_sched_entity(se) - depth++; - - return depth; -} - -/* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_wakeup(struct rq *rq, struct task_struct *p) +static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int sync) { struct task_struct *curr = rq->curr; struct cfs_rq *cfs_rq = task_cfs_rq(curr); struct sched_entity *se = &curr->se, *pse = &p->se; - int se_depth, pse_depth; + s64 delta_exec; if (unlikely(rt_prio(p->prio))) { update_rq_clock(rq); @@ -1351,6 +1254,13 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p) cfs_rq_of(pse)->next = pse; /* + * We can come here with TIF_NEED_RESCHED already set from new task + * wake up path. + */ + if (test_tsk_need_resched(curr)) + return; + + /* * Batch tasks do not preempt (their preemption is driven by * the tick): */ @@ -1360,33 +1270,15 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p) if (!sched_feat(WAKEUP_PREEMPT)) return; - /* - * preemption test can be made between sibling entities who are in the - * same cfs_rq i.e who have a common parent. Walk up the hierarchy of - * both tasks until we find their ancestors who are siblings of common - * parent. - */ - - /* First walk up until both entities are at same depth */ - se_depth = depth_se(se); - pse_depth = depth_se(pse); - - while (se_depth > pse_depth) { - se_depth--; - se = parent_entity(se); - } - - while (pse_depth > se_depth) { - pse_depth--; - pse = parent_entity(pse); - } - - while (!is_same_group(se, pse)) { - se = parent_entity(se); - pse = parent_entity(pse); + if (sched_feat(WAKEUP_OVERLAP) && (sync || + (se->avg_overlap < sysctl_sched_migration_cost && + pse->avg_overlap < sysctl_sched_migration_cost))) { + resched_task(curr); + return; } - if (wakeup_preempt_entity(se, pse) == 1) + delta_exec = se->sum_exec_runtime - se->prev_sum_exec_runtime; + if (delta_exec > wakeup_gran(pse)) resched_task(curr); } @@ -1445,19 +1337,9 @@ __load_balance_iterator(struct cfs_rq *cfs_rq, struct list_head *next) if (next == &cfs_rq->tasks) return NULL; - /* Skip over entities that are not tasks */ - do { - se = list_entry(next, struct sched_entity, group_node); - next = next->next; - } while (next != &cfs_rq->tasks && !entity_is_task(se)); - - if (next == &cfs_rq->tasks) - return NULL; - - cfs_rq->balance_iterator = next; - - if (entity_is_task(se)) - p = task_of(se); + se = list_entry(next, struct sched_entity, group_node); + p = task_of(se); + cfs_rq->balance_iterator = next->next; return p; } @@ -1507,7 +1389,7 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest, rcu_read_lock(); update_h_load(busiest_cpu); - list_for_each_entry(tg, &task_groups, list) { + list_for_each_entry_rcu(tg, &task_groups, list) { struct cfs_rq *busiest_cfs_rq = tg->cfs_rq[busiest_cpu]; unsigned long busiest_h_load = busiest_cfs_rq->h_load; unsigned long busiest_weight = busiest_cfs_rq->load.weight; @@ -1620,10 +1502,10 @@ static void task_new_fair(struct rq *rq, struct task_struct *p) * 'current' within the tree based on its new key value. */ swap(curr->vruntime, se->vruntime); + resched_task(rq->curr); } enqueue_task_fair(rq, p, 0); - resched_task(rq->curr); } /* @@ -1642,7 +1524,7 @@ static void prio_changed_fair(struct rq *rq, struct task_struct *p, if (p->prio > oldprio) resched_task(rq->curr); } else - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, 0); } /* @@ -1659,7 +1541,7 @@ static void switched_to_fair(struct rq *rq, struct task_struct *p, if (running) resched_task(rq->curr); else - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, 0); } /* Account for a task changing its policy or group. diff --git a/kernel/sched_features.h b/kernel/sched_features.h index 9353ca78154e..7c9e8f4a049f 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -11,3 +11,4 @@ SCHED_FEAT(ASYM_GRAN, 1) SCHED_FEAT(LB_BIAS, 1) SCHED_FEAT(LB_WAKEUP_UPDATE, 1) SCHED_FEAT(ASYM_EFF_LOAD, 1) +SCHED_FEAT(WAKEUP_OVERLAP, 0) diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c index 3a4f92dbbe66..dec4ccabe2f5 100644 --- a/kernel/sched_idletask.c +++ b/kernel/sched_idletask.c @@ -14,7 +14,7 @@ static int select_task_rq_idle(struct task_struct *p, int sync) /* * Idle tasks are unconditionally rescheduled: */ -static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p) +static void check_preempt_curr_idle(struct rq *rq, struct task_struct *p, int sync) { resched_task(rq->idle); } @@ -76,7 +76,7 @@ static void switched_to_idle(struct rq *rq, struct task_struct *p, if (running) resched_task(rq->curr); else - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, 0); } static void prio_changed_idle(struct rq *rq, struct task_struct *p, @@ -93,7 +93,7 @@ static void prio_changed_idle(struct rq *rq, struct task_struct *p, if (p->prio > oldprio) resched_task(rq->curr); } else - check_preempt_curr(rq, p); + check_preempt_curr(rq, p, 0); } /* diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 552310798dad..cdf5740ab03e 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -102,12 +102,12 @@ static void dequeue_rt_entity(struct sched_rt_entity *rt_se); static void sched_rt_rq_enqueue(struct rt_rq *rt_rq) { + struct task_struct *curr = rq_of_rt_rq(rt_rq)->curr; struct sched_rt_entity *rt_se = rt_rq->rt_se; - if (rt_se && !on_rt_rq(rt_se) && rt_rq->rt_nr_running) { - struct task_struct *curr = rq_of_rt_rq(rt_rq)->curr; - - enqueue_rt_entity(rt_se); + if (rt_rq->rt_nr_running) { + if (rt_se && !on_rt_rq(rt_se)) + enqueue_rt_entity(rt_se); if (rt_rq->highest_prio < curr->prio) resched_task(curr); } @@ -231,6 +231,9 @@ static inline struct rt_bandwidth *sched_rt_bandwidth(struct rt_rq *rt_rq) #endif /* CONFIG_RT_GROUP_SCHED */ #ifdef CONFIG_SMP +/* + * We ran out of runtime, see if we can borrow some from our neighbours. + */ static int do_balance_runtime(struct rt_rq *rt_rq) { struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq); @@ -250,9 +253,18 @@ static int do_balance_runtime(struct rt_rq *rt_rq) continue; spin_lock(&iter->rt_runtime_lock); + /* + * Either all rqs have inf runtime and there's nothing to steal + * or __disable_runtime() below sets a specific rq to inf to + * indicate its been disabled and disalow stealing. + */ if (iter->rt_runtime == RUNTIME_INF) goto next; + /* + * From runqueues with spare time, take 1/n part of their + * spare time, but no more than our period. + */ diff = iter->rt_runtime - iter->rt_time; if (diff > 0) { diff = div_u64((u64)diff, weight); @@ -274,6 +286,9 @@ next: return more; } +/* + * Ensure this RQ takes back all the runtime it lend to its neighbours. + */ static void __disable_runtime(struct rq *rq) { struct root_domain *rd = rq->rd; @@ -289,17 +304,33 @@ static void __disable_runtime(struct rq *rq) spin_lock(&rt_b->rt_runtime_lock); spin_lock(&rt_rq->rt_runtime_lock); + /* + * Either we're all inf and nobody needs to borrow, or we're + * already disabled and thus have nothing to do, or we have + * exactly the right amount of runtime to take out. + */ if (rt_rq->rt_runtime == RUNTIME_INF || rt_rq->rt_runtime == rt_b->rt_runtime) goto balanced; spin_unlock(&rt_rq->rt_runtime_lock); + /* + * Calculate the difference between what we started out with + * and what we current have, that's the amount of runtime + * we lend and now have to reclaim. + */ want = rt_b->rt_runtime - rt_rq->rt_runtime; + /* + * Greedy reclaim, take back as much as we can. + */ for_each_cpu_mask(i, rd->span) { struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i); s64 diff; + /* + * Can't reclaim from ourselves or disabled runqueues. + */ if (iter == rt_rq || iter->rt_runtime == RUNTIME_INF) continue; @@ -319,8 +350,16 @@ static void __disable_runtime(struct rq *rq) } spin_lock(&rt_rq->rt_runtime_lock); + /* + * We cannot be left wanting - that would mean some runtime + * leaked out of the system. + */ BUG_ON(want); balanced: + /* + * Disable all the borrow logic by pretending we have inf + * runtime - in which case borrowing doesn't make sense. + */ rt_rq->rt_runtime = RUNTIME_INF; spin_unlock(&rt_rq->rt_runtime_lock); spin_unlock(&rt_b->rt_runtime_lock); @@ -343,6 +382,9 @@ static void __enable_runtime(struct rq *rq) if (unlikely(!scheduler_running)) return; + /* + * Reset each runqueue's bandwidth settings + */ for_each_leaf_rt_rq(rt_rq, rq) { struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq); @@ -350,6 +392,7 @@ static void __enable_runtime(struct rq *rq) spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_runtime = rt_b->rt_runtime; rt_rq->rt_time = 0; + rt_rq->rt_throttled = 0; spin_unlock(&rt_rq->rt_runtime_lock); spin_unlock(&rt_b->rt_runtime_lock); } @@ -388,7 +431,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) int i, idle = 1; cpumask_t span; - if (rt_b->rt_runtime == RUNTIME_INF) + if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) return 1; span = sched_rt_period_mask(); @@ -486,6 +529,9 @@ static void update_curr_rt(struct rq *rq) curr->se.exec_start = rq->clock; cpuacct_charge(curr, delta_exec); + if (!rt_bandwidth_enabled()) + return; + for_each_sched_rt_entity(rt_se) { rt_rq = rt_rq_of_se(rt_se); @@ -783,7 +829,7 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p) +static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p, int sync) { if (p->prio < rq->curr->prio) { resched_task(rq->curr); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index fe4713347275..1bf369bd4423 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -118,10 +118,8 @@ extern char modprobe_path[]; extern int sg_big_buff; #endif -#ifdef __sparc__ -extern char reboot_command []; -extern int stop_a_enabled; -extern int scons_pwroff; +#ifdef CONFIG_SPARC +#include <asm/system.h> #endif #ifdef __hppa__ @@ -159,6 +157,7 @@ static int proc_dointvec_taint(struct ctl_table *table, int write, struct file * static struct ctl_table root_table[]; static struct ctl_table_root sysctl_table_root; static struct ctl_table_header root_table_header = { + .count = 1, .ctl_table = root_table, .ctl_entry = LIST_HEAD_INIT(sysctl_table_root.default_set.list), .root = &sysctl_table_root, @@ -414,7 +413,7 @@ static struct ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, -#ifdef __sparc__ +#ifdef CONFIG_SPARC { .ctl_name = KERN_SPARC_REBOOT, .procname = "reboot-cmd", diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 3d1e3e1a1971..f8d968063cea 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -72,6 +72,16 @@ void clockevents_set_mode(struct clock_event_device *dev, } /** + * clockevents_shutdown - shutdown the device and clear next_event + * @dev: device to shutdown + */ +void clockevents_shutdown(struct clock_event_device *dev) +{ + clockevents_set_mode(dev, CLOCK_EVT_MODE_SHUTDOWN); + dev->next_event.tv64 = KTIME_MAX; +} + +/** * clockevents_program_event - Reprogram the clock event device. * @expires: absolute expiry time (monotonic clock) * @@ -177,7 +187,7 @@ void clockevents_register_device(struct clock_event_device *dev) /* * Noop handler when we shut down an event device */ -static void clockevents_handle_noop(struct clock_event_device *dev) +void clockevents_handle_noop(struct clock_event_device *dev) { } @@ -199,7 +209,6 @@ void clockevents_exchange_device(struct clock_event_device *old, * released list and do a notify add later. */ if (old) { - old->event_handler = clockevents_handle_noop; clockevents_set_mode(old, CLOCK_EVT_MODE_UNUSED); list_del(&old->list); list_add(&old->list, &clockevents_released); @@ -207,7 +216,7 @@ void clockevents_exchange_device(struct clock_event_device *old, if (new) { BUG_ON(new->mode != CLOCK_EVT_MODE_UNUSED); - clockevents_set_mode(new, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_shutdown(new); } local_irq_restore(flags); } diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 5125ddd8196b..1ad46f3df6e7 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -245,7 +245,7 @@ static void sync_cmos_clock(unsigned long dummy) if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec / 2) fail = update_persistent_clock(now); - next.tv_nsec = (NSEC_PER_SEC / 2) - now.tv_nsec; + next.tv_nsec = (NSEC_PER_SEC / 2) - now.tv_nsec - (TICK_NSEC / 2); if (next.tv_nsec <= 0) next.tv_nsec += NSEC_PER_SEC; diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 31463d370b94..cb01cd8f919b 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -175,6 +175,8 @@ static void tick_do_periodic_broadcast(void) */ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) { + ktime_t next; + tick_do_periodic_broadcast(); /* @@ -185,10 +187,13 @@ static void tick_handle_periodic_broadcast(struct clock_event_device *dev) /* * Setup the next period for devices, which do not have - * periodic mode: + * periodic mode. We read dev->next_event first and add to it + * when the event alrady expired. clockevents_program_event() + * sets dev->next_event only when the event is really + * programmed to the device. */ - for (;;) { - ktime_t next = ktime_add(dev->next_event, tick_period); + for (next = dev->next_event; ;) { + next = ktime_add(next, tick_period); if (!clockevents_program_event(dev, next, ktime_get())) return; @@ -205,7 +210,7 @@ static void tick_do_broadcast_on_off(void *why) struct clock_event_device *bc, *dev; struct tick_device *td; unsigned long flags, *reason = why; - int cpu; + int cpu, bc_stopped; spin_lock_irqsave(&tick_broadcast_lock, flags); @@ -223,14 +228,16 @@ static void tick_do_broadcast_on_off(void *why) if (!tick_device_is_functional(dev)) goto out; + bc_stopped = cpus_empty(tick_broadcast_mask); + switch (*reason) { case CLOCK_EVT_NOTIFY_BROADCAST_ON: case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: if (!cpu_isset(cpu, tick_broadcast_mask)) { cpu_set(cpu, tick_broadcast_mask); - if (td->mode == TICKDEV_MODE_PERIODIC) - clockevents_set_mode(dev, - CLOCK_EVT_MODE_SHUTDOWN); + if (tick_broadcast_device.mode == + TICKDEV_MODE_PERIODIC) + clockevents_shutdown(dev); } if (*reason == CLOCK_EVT_NOTIFY_BROADCAST_FORCE) tick_broadcast_force = 1; @@ -239,15 +246,17 @@ static void tick_do_broadcast_on_off(void *why) if (!tick_broadcast_force && cpu_isset(cpu, tick_broadcast_mask)) { cpu_clear(cpu, tick_broadcast_mask); - if (td->mode == TICKDEV_MODE_PERIODIC) + if (tick_broadcast_device.mode == + TICKDEV_MODE_PERIODIC) tick_setup_periodic(dev, 0); } break; } - if (cpus_empty(tick_broadcast_mask)) - clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN); - else { + if (cpus_empty(tick_broadcast_mask)) { + if (!bc_stopped) + clockevents_shutdown(bc); + } else if (bc_stopped) { if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) tick_broadcast_start_periodic(bc); else @@ -298,7 +307,7 @@ void tick_shutdown_broadcast(unsigned int *cpup) if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) { if (bc && cpus_empty(tick_broadcast_mask)) - clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_shutdown(bc); } spin_unlock_irqrestore(&tick_broadcast_lock, flags); @@ -313,7 +322,7 @@ void tick_suspend_broadcast(void) bc = tick_broadcast_device.evtdev; if (bc) - clockevents_set_mode(bc, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_shutdown(bc); spin_unlock_irqrestore(&tick_broadcast_lock, flags); } @@ -364,16 +373,8 @@ cpumask_t *tick_get_broadcast_oneshot_mask(void) static int tick_broadcast_set_event(ktime_t expires, int force) { struct clock_event_device *bc = tick_broadcast_device.evtdev; - ktime_t now = ktime_get(); - int res; - - for(;;) { - res = clockevents_program_event(bc, expires, now); - if (!res || !force) - return res; - now = ktime_get(); - expires = ktime_add(now, ktime_set(0, bc->min_delta_ns)); - } + + return tick_dev_program_event(bc, expires, force); } int tick_resume_broadcast_oneshot(struct clock_event_device *bc) @@ -491,14 +492,52 @@ static void tick_broadcast_clear_oneshot(int cpu) cpu_clear(cpu, tick_broadcast_oneshot_mask); } +static void tick_broadcast_init_next_event(cpumask_t *mask, ktime_t expires) +{ + struct tick_device *td; + int cpu; + + for_each_cpu_mask_nr(cpu, *mask) { + td = &per_cpu(tick_cpu_device, cpu); + if (td->evtdev) + td->evtdev->next_event = expires; + } +} + /** * tick_broadcast_setup_oneshot - setup the broadcast device */ void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { - bc->event_handler = tick_handle_oneshot_broadcast; - clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT); - bc->next_event.tv64 = KTIME_MAX; + /* Set it up only once ! */ + if (bc->event_handler != tick_handle_oneshot_broadcast) { + int was_periodic = bc->mode == CLOCK_EVT_MODE_PERIODIC; + int cpu = smp_processor_id(); + cpumask_t mask; + + bc->event_handler = tick_handle_oneshot_broadcast; + clockevents_set_mode(bc, CLOCK_EVT_MODE_ONESHOT); + + /* Take the do_timer update */ + tick_do_timer_cpu = cpu; + + /* + * We must be careful here. There might be other CPUs + * waiting for periodic broadcast. We need to set the + * oneshot_mask bits for those and program the + * broadcast device to fire. + */ + mask = tick_broadcast_mask; + cpu_clear(cpu, mask); + cpus_or(tick_broadcast_oneshot_mask, + tick_broadcast_oneshot_mask, mask); + + if (was_periodic && !cpus_empty(mask)) { + tick_broadcast_init_next_event(&mask, tick_next_period); + tick_broadcast_set_event(tick_next_period, 1); + } else + bc->next_event.tv64 = KTIME_MAX; + } } /* @@ -538,4 +577,12 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup) spin_unlock_irqrestore(&tick_broadcast_lock, flags); } +/* + * Check, whether the broadcast device is in one shot mode + */ +int tick_broadcast_oneshot_active(void) +{ + return tick_broadcast_device.mode == TICKDEV_MODE_ONESHOT; +} + #endif diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 80c4336f4188..df12434b43ca 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -33,7 +33,7 @@ DEFINE_PER_CPU(struct tick_device, tick_cpu_device); */ ktime_t tick_next_period; ktime_t tick_period; -int tick_do_timer_cpu __read_mostly = -1; +int tick_do_timer_cpu __read_mostly = TICK_DO_TIMER_BOOT; DEFINE_SPINLOCK(tick_device_lock); /* @@ -109,7 +109,8 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast) if (!tick_device_is_functional(dev)) return; - if (dev->features & CLOCK_EVT_FEAT_PERIODIC) { + if ((dev->features & CLOCK_EVT_FEAT_PERIODIC) && + !tick_broadcast_oneshot_active()) { clockevents_set_mode(dev, CLOCK_EVT_MODE_PERIODIC); } else { unsigned long seq; @@ -148,7 +149,7 @@ static void tick_setup_device(struct tick_device *td, * If no cpu took the do_timer update, assign it to * this cpu: */ - if (tick_do_timer_cpu == -1) { + if (tick_do_timer_cpu == TICK_DO_TIMER_BOOT) { tick_do_timer_cpu = cpu; tick_next_period = ktime_get(); tick_period = ktime_set(0, NSEC_PER_SEC / HZ); @@ -161,6 +162,7 @@ static void tick_setup_device(struct tick_device *td, } else { handler = td->evtdev->event_handler; next_event = td->evtdev->next_event; + td->evtdev->event_handler = clockevents_handle_noop; } td->evtdev = newdev; @@ -248,7 +250,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) * not give it back to the clockevents layer ! */ if (tick_is_broadcast_device(curdev)) { - clockevents_set_mode(curdev, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_shutdown(curdev); curdev = NULL; } clockevents_exchange_device(curdev, newdev); @@ -299,7 +301,8 @@ static void tick_shutdown(unsigned int *cpup) if (*cpup == tick_do_timer_cpu) { int cpu = first_cpu(cpu_online_map); - tick_do_timer_cpu = (cpu != NR_CPUS) ? cpu : -1; + tick_do_timer_cpu = (cpu != NR_CPUS) ? cpu : + TICK_DO_TIMER_NONE; } spin_unlock_irqrestore(&tick_device_lock, flags); } @@ -310,7 +313,7 @@ static void tick_suspend(void) unsigned long flags; spin_lock_irqsave(&tick_device_lock, flags); - clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_SHUTDOWN); + clockevents_shutdown(td->evtdev); spin_unlock_irqrestore(&tick_device_lock, flags); } diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index f13f2b7f4fd4..469248782c23 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -1,6 +1,10 @@ /* * tick internal variable and functions used by low/high res code */ + +#define TICK_DO_TIMER_NONE -1 +#define TICK_DO_TIMER_BOOT -2 + DECLARE_PER_CPU(struct tick_device, tick_cpu_device); extern spinlock_t tick_device_lock; extern ktime_t tick_next_period; @@ -10,6 +14,8 @@ extern int tick_do_timer_cpu __read_mostly; extern void tick_setup_periodic(struct clock_event_device *dev, int broadcast); extern void tick_handle_periodic(struct clock_event_device *dev); +extern void clockevents_shutdown(struct clock_event_device *dev); + /* * NO_HZ / high resolution timer shared code */ @@ -17,6 +23,8 @@ extern void tick_handle_periodic(struct clock_event_device *dev); extern void tick_setup_oneshot(struct clock_event_device *newdev, void (*handler)(struct clock_event_device *), ktime_t nextevt); +extern int tick_dev_program_event(struct clock_event_device *dev, + ktime_t expires, int force); extern int tick_program_event(ktime_t expires, int force); extern void tick_oneshot_notify(void); extern int tick_switch_to_oneshot(void (*handler)(struct clock_event_device *)); @@ -27,6 +35,7 @@ extern void tick_broadcast_oneshot_control(unsigned long reason); extern void tick_broadcast_switch_to_oneshot(void); extern void tick_shutdown_broadcast_oneshot(unsigned int *cpup); extern int tick_resume_broadcast_oneshot(struct clock_event_device *bc); +extern int tick_broadcast_oneshot_active(void); # else /* BROADCAST */ static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) { @@ -35,6 +44,7 @@ static inline void tick_broadcast_setup_oneshot(struct clock_event_device *bc) static inline void tick_broadcast_oneshot_control(unsigned long reason) { } static inline void tick_broadcast_switch_to_oneshot(void) { } static inline void tick_shutdown_broadcast_oneshot(unsigned int *cpup) { } +static inline int tick_broadcast_oneshot_active(void) { return 0; } # endif /* !BROADCAST */ #else /* !ONESHOT */ @@ -64,6 +74,7 @@ static inline int tick_resume_broadcast_oneshot(struct clock_event_device *bc) { return 0; } +static inline int tick_broadcast_oneshot_active(void) { return 0; } #endif /* !TICK_ONESHOT */ /* diff --git a/kernel/time/tick-oneshot.c b/kernel/time/tick-oneshot.c index 450c04935b66..2e8de678e767 100644 --- a/kernel/time/tick-oneshot.c +++ b/kernel/time/tick-oneshot.c @@ -23,24 +23,56 @@ #include "tick-internal.h" /** - * tick_program_event + * tick_program_event internal worker function */ -int tick_program_event(ktime_t expires, int force) +int tick_dev_program_event(struct clock_event_device *dev, ktime_t expires, + int force) { - struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev; ktime_t now = ktime_get(); + int i; - while (1) { + for (i = 0;;) { int ret = clockevents_program_event(dev, expires, now); if (!ret || !force) return ret; + + /* + * We tried 2 times to program the device with the given + * min_delta_ns. If that's not working then we double it + * and emit a warning. + */ + if (++i > 2) { + /* Increase the min. delta and try again */ + if (!dev->min_delta_ns) + dev->min_delta_ns = 5000; + else + dev->min_delta_ns += dev->min_delta_ns >> 1; + + printk(KERN_WARNING + "CE: %s increasing min_delta_ns to %lu nsec\n", + dev->name ? dev->name : "?", + dev->min_delta_ns << 1); + + i = 0; + } + now = ktime_get(); - expires = ktime_add(now, ktime_set(0, dev->min_delta_ns)); + expires = ktime_add_ns(now, dev->min_delta_ns); } } /** + * tick_program_event + */ +int tick_program_event(ktime_t expires, int force) +{ + struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev; + + return tick_dev_program_event(dev, expires, force); +} + +/** * tick_resume_onshot - resume oneshot mode */ void tick_resume_oneshot(void) @@ -61,7 +93,7 @@ void tick_setup_oneshot(struct clock_event_device *newdev, { newdev->event_handler = handler; clockevents_set_mode(newdev, CLOCK_EVT_MODE_ONESHOT); - clockevents_program_event(newdev, next_event, ktime_get()); + tick_dev_program_event(newdev, next_event, 1); } /** diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 7a46bde78c66..a4d219398167 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -20,6 +20,7 @@ #include <linux/profile.h> #include <linux/sched.h> #include <linux/tick.h> +#include <linux/module.h> #include <asm/irq_regs.h> @@ -75,6 +76,9 @@ static void tick_do_update_jiffies64(ktime_t now) incr * ticks); } do_timer(++ticks); + + /* Keep the tick_next_period variable up to date */ + tick_next_period = ktime_add(last_jiffies_update, tick_period); } write_sequnlock(&xtime_lock); } @@ -162,6 +166,8 @@ void tick_nohz_stop_idle(int cpu) ts->idle_lastupdate = now; ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta); ts->idle_active = 0; + + sched_clock_idle_wakeup_event(0); } } @@ -177,6 +183,7 @@ static ktime_t tick_nohz_start_idle(struct tick_sched *ts) } ts->idle_entrytime = now; ts->idle_active = 1; + sched_clock_idle_sleep_event(); return now; } @@ -184,9 +191,17 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time) { struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu); - *last_update_time = ktime_to_us(ts->idle_lastupdate); + if (!tick_nohz_enabled) + return -1; + + if (ts->idle_active) + *last_update_time = ktime_to_us(ts->idle_lastupdate); + else + *last_update_time = ktime_to_us(ktime_get()); + return ktime_to_us(ts->idle_sleeptime); } +EXPORT_SYMBOL_GPL(get_cpu_idle_time_us); /** * tick_nohz_stop_sched_tick - stop the idle tick from the idle task @@ -218,7 +233,7 @@ void tick_nohz_stop_sched_tick(int inidle) */ if (unlikely(!cpu_online(cpu))) { if (cpu == tick_do_timer_cpu) - tick_do_timer_cpu = -1; + tick_do_timer_cpu = TICK_DO_TIMER_NONE; } if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE)) @@ -300,7 +315,7 @@ void tick_nohz_stop_sched_tick(int inidle) * invoked. */ if (cpu == tick_do_timer_cpu) - tick_do_timer_cpu = -1; + tick_do_timer_cpu = TICK_DO_TIMER_NONE; ts->idle_sleeps++; @@ -465,7 +480,7 @@ static void tick_nohz_handler(struct clock_event_device *dev) * this duty, then the jiffies update is still serialized by * xtime_lock. */ - if (unlikely(tick_do_timer_cpu == -1)) + if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_NONE)) tick_do_timer_cpu = cpu; /* Check, if the jiffies need an update */ @@ -567,7 +582,7 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer) * this duty, then the jiffies update is still serialized by * xtime_lock. */ - if (unlikely(tick_do_timer_cpu == -1)) + if (unlikely(tick_do_timer_cpu == TICK_DO_TIMER_NONE)) tick_do_timer_cpu = cpu; #endif @@ -619,7 +634,7 @@ void tick_setup_sched_timer(void) */ hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); ts->sched_timer.function = tick_sched_timer; - ts->sched_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ; + ts->sched_timer.cb_mode = HRTIMER_CB_IRQSAFE_PERCPU; /* Get the next period (per cpu) */ ts->sched_timer.expires = tick_init_jiffy_update(); diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c index bb948e52ce20..db58fb66a135 100644 --- a/kernel/trace/trace_sysprof.c +++ b/kernel/trace/trace_sysprof.c @@ -202,7 +202,7 @@ static void start_stack_timer(int cpu) hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hrtimer->function = stack_trace_timer_fn; - hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ; + hrtimer->cb_mode = HRTIMER_CB_IRQSAFE_PERCPU; hrtimer_start(hrtimer, ns_to_ktime(sample_period), HRTIMER_MODE_REL); } diff --git a/kernel/user.c b/kernel/user.c index 865ecf57a096..39d6159fae43 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -169,7 +169,7 @@ static ssize_t cpu_rt_runtime_show(struct kobject *kobj, { struct user_struct *up = container_of(kobj, struct user_struct, kobj); - return sprintf(buf, "%lu\n", sched_group_rt_runtime(up->tg)); + return sprintf(buf, "%ld\n", sched_group_rt_runtime(up->tg)); } static ssize_t cpu_rt_runtime_store(struct kobject *kobj, @@ -180,7 +180,7 @@ static ssize_t cpu_rt_runtime_store(struct kobject *kobj, unsigned long rt_runtime; int rc; - sscanf(buf, "%lu", &rt_runtime); + sscanf(buf, "%ld", &rt_runtime); rc = sched_group_set_rt_runtime(up->tg, rt_runtime); |