diff options
Diffstat (limited to 'kernel')
50 files changed, 1877 insertions, 718 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index 0ae57a8..6c51128 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -94,6 +94,7 @@ obj-$(CONFIG_X86_DS) += trace/ obj-$(CONFIG_RING_BUFFER) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o obj-$(CONFIG_SLOW_WORK) += slow-work.o +obj-$(CONFIG_SLOW_WORK_DEBUG) += slow-work-debugfs.o obj-$(CONFIG_PERF_EVENTS) += perf_event.o obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 7ccba4b..0249f4b 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -703,7 +703,7 @@ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode); static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry); static int cgroup_populate_dir(struct cgroup *cgrp); static const struct inode_operations cgroup_dir_inode_operations; -static struct file_operations proc_cgroupstats_operations; +static const struct file_operations proc_cgroupstats_operations; static struct backing_dev_info cgroup_backing_dev_info = { .name = "cgroup", @@ -1710,14 +1710,13 @@ static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft, return -EFAULT; buffer[nbytes] = 0; /* nul-terminate */ - strstrip(buffer); if (cft->write_u64) { - u64 val = simple_strtoull(buffer, &end, 0); + u64 val = simple_strtoull(strstrip(buffer), &end, 0); if (*end) return -EINVAL; retval = cft->write_u64(cgrp, cft, val); } else { - s64 val = simple_strtoll(buffer, &end, 0); + s64 val = simple_strtoll(strstrip(buffer), &end, 0); if (*end) return -EINVAL; retval = cft->write_s64(cgrp, cft, val); @@ -1753,8 +1752,7 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, } buffer[nbytes] = 0; /* nul-terminate */ - strstrip(buffer); - retval = cft->write_string(cgrp, cft, buffer); + retval = cft->write_string(cgrp, cft, strstrip(buffer)); if (!retval) retval = nbytes; out: @@ -1863,7 +1861,7 @@ static int cgroup_seqfile_release(struct inode *inode, struct file *file) return single_release(inode, file); } -static struct file_operations cgroup_seqfile_operations = { +static const struct file_operations cgroup_seqfile_operations = { .read = seq_read, .write = cgroup_file_write, .llseek = seq_lseek, @@ -1922,7 +1920,7 @@ static int cgroup_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } -static struct file_operations cgroup_file_operations = { +static const struct file_operations cgroup_file_operations = { .read = cgroup_file_read, .write = cgroup_file_write, .llseek = generic_file_llseek, @@ -3369,7 +3367,7 @@ static int cgroup_open(struct inode *inode, struct file *file) return single_open(file, proc_cgroup_show, pid); } -struct file_operations proc_cgroup_operations = { +const struct file_operations proc_cgroup_operations = { .open = cgroup_open, .read = seq_read, .llseek = seq_lseek, @@ -3398,7 +3396,7 @@ static int cgroupstats_open(struct inode *inode, struct file *file) return single_open(file, proc_cgroupstats_show, NULL); } -static struct file_operations proc_cgroupstats_operations = { +static const struct file_operations proc_cgroupstats_operations = { .open = cgroupstats_open, .read = seq_read, .llseek = seq_lseek, @@ -3708,8 +3706,10 @@ static void check_for_release(struct cgroup *cgrp) void __css_put(struct cgroup_subsys_state *css) { struct cgroup *cgrp = css->cgroup; + int val; rcu_read_lock(); - if (atomic_dec_return(&css->refcnt) == 1) { + val = atomic_dec_return(&css->refcnt); + if (val == 1) { if (notify_on_release(cgrp)) { set_bit(CGRP_RELEASABLE, &cgrp->flags); check_for_release(cgrp); @@ -3717,6 +3717,7 @@ void __css_put(struct cgroup_subsys_state *css) cgroup_wakeup_rmdir_waiter(cgrp); } rcu_read_unlock(); + WARN_ON_ONCE(val < 1); } /* diff --git a/kernel/exit.c b/kernel/exit.c index 5859f59..f7864ac 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -359,10 +359,8 @@ void __set_special_pids(struct pid *pid) { struct task_struct *curr = current->group_leader; - if (task_session(curr) != pid) { + if (task_session(curr) != pid) change_pid(curr, PIDTYPE_SID, pid); - proc_sid_connector(curr); - } if (task_pgrp(curr) != pid) change_pid(curr, PIDTYPE_PGID, pid); @@ -991,8 +989,6 @@ NORET_TYPE void do_exit(long code) tsk->mempolicy = NULL; #endif #ifdef CONFIG_FUTEX - if (unlikely(!list_empty(&tsk->pi_state_list))) - exit_pi_state_list(tsk); if (unlikely(current->pi_state_cache)) kfree(current->pi_state_cache); #endif diff --git a/kernel/fork.c b/kernel/fork.c index 1b7512d..7d4a348e 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -92,7 +92,7 @@ int nr_processes(void) int cpu; int total = 0; - for_each_online_cpu(cpu) + for_each_possible_cpu(cpu) total += per_cpu(process_counts, cpu); return total; @@ -572,12 +572,18 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm) /* Get rid of any futexes when releasing the mm */ #ifdef CONFIG_FUTEX - if (unlikely(tsk->robust_list)) + if (unlikely(tsk->robust_list)) { exit_robust_list(tsk); + tsk->robust_list = NULL; + } #ifdef CONFIG_COMPAT - if (unlikely(tsk->compat_robust_list)) + if (unlikely(tsk->compat_robust_list)) { compat_exit_robust_list(tsk); + tsk->compat_robust_list = NULL; + } #endif + if (unlikely(!list_empty(&tsk->pi_state_list))) + exit_pi_state_list(tsk); #endif /* Get rid of any cached register state */ diff --git a/kernel/futex.c b/kernel/futex.c index b911adc..fb65e82 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -150,7 +150,8 @@ static struct futex_hash_bucket *hash_futex(union futex_key *key) */ static inline int match_futex(union futex_key *key1, union futex_key *key2) { - return (key1->both.word == key2->both.word + return (key1 && key2 + && key1->both.word == key2->both.word && key1->both.ptr == key2->both.ptr && key1->both.offset == key2->both.offset); } @@ -916,8 +917,8 @@ retry: hb1 = hash_futex(&key1); hb2 = hash_futex(&key2); - double_lock_hb(hb1, hb2); retry_private: + double_lock_hb(hb1, hb2); op_ret = futex_atomic_op_inuser(op, uaddr2); if (unlikely(op_ret < 0)) { @@ -1028,7 +1029,6 @@ static inline void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, struct futex_hash_bucket *hb) { - drop_futex_key_refs(&q->key); get_futex_key_refs(key); q->key = *key; @@ -1226,6 +1226,7 @@ retry_private: */ if (ret == 1) { WARN_ON(pi_state); + drop_count++; task_count++; ret = get_futex_value_locked(&curval2, uaddr2); if (!ret) @@ -1304,6 +1305,7 @@ retry_private: if (ret == 1) { /* We got the lock. */ requeue_pi_wake_futex(this, &key2, hb2); + drop_count++; continue; } else if (ret) { /* -EDEADLK */ @@ -1791,6 +1793,7 @@ static int futex_wait(u32 __user *uaddr, int fshared, current->timer_slack_ns); } +retry: /* Prepare to wait on uaddr. */ ret = futex_wait_setup(uaddr, val, fshared, &q, &hb); if (ret) @@ -1808,9 +1811,14 @@ static int futex_wait(u32 __user *uaddr, int fshared, goto out_put_key; /* - * We expect signal_pending(current), but another thread may - * have handled it for us already. + * We expect signal_pending(current), but we might be the + * victim of a spurious wakeup as well. */ + if (!signal_pending(current)) { + put_futex_key(fshared, &q.key); + goto retry; + } + ret = -ERESTARTSYS; if (!abs_time) goto out_put_key; @@ -2117,11 +2125,12 @@ int handle_early_requeue_pi_wakeup(struct futex_hash_bucket *hb, * Unqueue the futex_q and determine which it was. */ plist_del(&q->list, &q->list.plist); - drop_futex_key_refs(&q->key); + /* Handle spurious wakeups gracefully */ + ret = -EWOULDBLOCK; if (timeout && !timeout->task) ret = -ETIMEDOUT; - else + else if (signal_pending(current)) ret = -ERESTARTNOINTR; } return ret; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index 6d70204..3e1c36e 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -726,8 +726,6 @@ static int hrtimer_switch_to_hres(void) /* "Retrigger" the interrupt to get things going */ retrigger_next_event(NULL); local_irq_restore(flags); - printk(KERN_DEBUG "Switched to high resolution mode on CPU %d\n", - smp_processor_id()); return 1; } diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index a81cf80..17c71bb 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -11,6 +11,7 @@ */ #include <linux/irq.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/module.h> #include <linux/random.h> diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 114e704..bd7273e 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -121,7 +121,9 @@ static void poll_all_shared_irqs(void) if (!(status & IRQ_SPURIOUS_DISABLED)) continue; + local_irq_disable(); try_one_irq(i, desc); + local_irq_enable(); } } diff --git a/kernel/kprobes.c b/kernel/kprobes.c index cfadc12..5240d75 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -1333,7 +1333,7 @@ static int __kprobes kprobes_open(struct inode *inode, struct file *filp) return seq_open(filp, &kprobes_seq_ops); } -static struct file_operations debugfs_kprobes_operations = { +static const struct file_operations debugfs_kprobes_operations = { .open = kprobes_open, .read = seq_read, .llseek = seq_lseek, @@ -1515,7 +1515,7 @@ static ssize_t write_enabled_file_bool(struct file *file, return count; } -static struct file_operations fops_kp = { +static const struct file_operations fops_kp = { .read = read_enabled_file_bool, .write = write_enabled_file_bool, }; diff --git a/kernel/kthread.c b/kernel/kthread.c index 5fe7099..ab7ae57 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -150,29 +150,6 @@ struct task_struct *kthread_create(int (*threadfn)(void *data), EXPORT_SYMBOL(kthread_create); /** - * kthread_bind - bind a just-created kthread to a cpu. - * @k: thread created by kthread_create(). - * @cpu: cpu (might not be online, must be possible) for @k to run on. - * - * Description: This function is equivalent to set_cpus_allowed(), - * except that @cpu doesn't need to be online, and the thread must be - * stopped (i.e., just returned from kthread_create()). - */ -void kthread_bind(struct task_struct *k, unsigned int cpu) -{ - /* Must have done schedule() in kthread() before we set_task_cpu */ - if (!wait_task_inactive(k, TASK_UNINTERRUPTIBLE)) { - WARN_ON(1); - return; - } - set_task_cpu(k, cpu); - k->cpus_allowed = cpumask_of_cpu(cpu); - k->rt.nr_cpus_allowed = 1; - k->flags |= PF_THREAD_BOUND; -} -EXPORT_SYMBOL(kthread_bind); - -/** * kthread_stop - stop a thread created by kthread_create(). * @k: thread created by kthread_create(). * diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 3815ac1d..9af5672 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -142,6 +142,11 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock) #ifdef CONFIG_LOCK_STAT static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); +static inline u64 lockstat_clock(void) +{ + return cpu_clock(smp_processor_id()); +} + static int lock_point(unsigned long points[], unsigned long ip) { int i; @@ -158,7 +163,7 @@ static int lock_point(unsigned long points[], unsigned long ip) return i; } -static void lock_time_inc(struct lock_time *lt, s64 time) +static void lock_time_inc(struct lock_time *lt, u64 time) { if (time > lt->max) lt->max = time; @@ -234,12 +239,12 @@ static void put_lock_stats(struct lock_class_stats *stats) static void lock_release_holdtime(struct held_lock *hlock) { struct lock_class_stats *stats; - s64 holdtime; + u64 holdtime; if (!lock_stat) return; - holdtime = sched_clock() - hlock->holdtime_stamp; + holdtime = lockstat_clock() - hlock->holdtime_stamp; stats = get_lock_stats(hlock_class(hlock)); if (hlock->read) @@ -2792,7 +2797,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->references = references; #ifdef CONFIG_LOCK_STAT hlock->waittime_stamp = 0; - hlock->holdtime_stamp = sched_clock(); + hlock->holdtime_stamp = lockstat_clock(); #endif if (check == 2 && !mark_irqflags(curr, hlock)) @@ -3322,7 +3327,7 @@ found_it: if (hlock->instance != lock) return; - hlock->waittime_stamp = sched_clock(); + hlock->waittime_stamp = lockstat_clock(); contention_point = lock_point(hlock_class(hlock)->contention_point, ip); contending_point = lock_point(hlock_class(hlock)->contending_point, @@ -3345,8 +3350,7 @@ __lock_acquired(struct lockdep_map *lock, unsigned long ip) struct held_lock *hlock, *prev_hlock; struct lock_class_stats *stats; unsigned int depth; - u64 now; - s64 waittime = 0; + u64 now, waittime = 0; int i, cpu; depth = curr->lockdep_depth; @@ -3374,7 +3378,7 @@ found_it: cpu = smp_processor_id(); if (hlock->waittime_stamp) { - now = sched_clock(); + now = lockstat_clock(); waittime = now - hlock->waittime_stamp; hlock->holdtime_stamp = now; } diff --git a/kernel/module.c b/kernel/module.c index fe748a8..5842a71 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1187,7 +1187,8 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, /* Count loaded sections and allocate structures */ for (i = 0; i < nsect; i++) - if (sechdrs[i].sh_flags & SHF_ALLOC) + if (sechdrs[i].sh_flags & SHF_ALLOC + && sechdrs[i].sh_size) nloaded++; size[0] = ALIGN(sizeof(*sect_attrs) + nloaded * sizeof(sect_attrs->attrs[0]), @@ -1207,6 +1208,8 @@ static void add_sect_attrs(struct module *mod, unsigned int nsect, for (i = 0; i < nsect; i++) { if (! (sechdrs[i].sh_flags & SHF_ALLOC)) continue; + if (!sechdrs[i].sh_size) + continue; sattr->address = sechdrs[i].sh_addr; sattr->name = kstrdup(secstrings + sechdrs[i].sh_name, GFP_KERNEL); @@ -1992,12 +1995,14 @@ static inline unsigned long layout_symtab(struct module *mod, Elf_Shdr *sechdrs, unsigned int symindex, unsigned int strindex, - const Elf_Hdr *hdr, + const Elf_Ehdr *hdr, const char *secstrings, unsigned long *pstroffs, unsigned long *strmap) { + return 0; } + static inline void add_kallsyms(struct module *mod, Elf_Shdr *sechdrs, unsigned int shnum, @@ -2081,9 +2086,8 @@ static noinline struct module *load_module(void __user *umod, struct module *mod; long err = 0; void *percpu = NULL, *ptr = NULL; /* Stops spurious gcc warning */ -#ifdef CONFIG_KALLSYMS unsigned long symoffs, stroffs, *strmap; -#endif + mm_segment_t old_fs; DEBUGP("load_module: umod=%p, len=%lu, uargs=%p\n", diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 50d022e..ec815a9 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -16,6 +16,7 @@ #include <linux/delay.h> #include <linux/module.h> #include <linux/poison.h> +#include <linux/sched.h> #include <linux/spinlock.h> #include <linux/kallsyms.h> #include <linux/interrupt.h> diff --git a/kernel/panic.c b/kernel/panic.c index bcdef26..96b45d0 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -90,6 +90,8 @@ NORET_TYPE void panic(const char * fmt, ...) atomic_notifier_call_chain(&panic_notifier_list, 0, buf); + bust_spinlocks(0); + if (!panic_blink) panic_blink = no_blink; @@ -136,7 +138,6 @@ NORET_TYPE void panic(const char * fmt, ...) mdelay(1); i++; } - bust_spinlocks(0); } EXPORT_SYMBOL(panic); diff --git a/kernel/params.c b/kernel/params.c index 9da58ea..d656c27 100644 --- a/kernel/params.c +++ b/kernel/params.c @@ -218,15 +218,11 @@ int param_set_charp(const char *val, struct kernel_param *kp) return -ENOSPC; } - if (kp->flags & KPARAM_KMALLOCED) - kfree(*(char **)kp->arg); - /* This is a hack. We can't need to strdup in early boot, and we * don't need to; this mangled commandline is preserved. */ if (slab_is_available()) { - kp->flags |= KPARAM_KMALLOCED; *(char **)kp->arg = kstrdup(val, GFP_KERNEL); - if (!kp->arg) + if (!*(char **)kp->arg) return -ENOMEM; } else *(const char **)kp->arg = val; @@ -304,6 +300,7 @@ static int param_array(const char *name, unsigned int min, unsigned int max, void *elem, int elemsize, int (*set)(const char *, struct kernel_param *kp), + u16 flags, unsigned int *num) { int ret; @@ -313,6 +310,7 @@ static int param_array(const char *name, /* Get the name right for errors. */ kp.name = name; kp.arg = elem; + kp.flags = flags; /* No equals sign? */ if (!val) { @@ -358,7 +356,8 @@ int param_array_set(const char *val, struct kernel_param *kp) unsigned int temp_num; return param_array(kp->name, val, 1, arr->max, arr->elem, - arr->elemsize, arr->set, arr->num ?: &temp_num); + arr->elemsize, arr->set, kp->flags, + arr->num ?: &temp_num); } int param_array_get(char *buffer, struct kernel_param *kp) @@ -605,11 +604,7 @@ void module_param_sysfs_remove(struct module *mod) void destroy_params(const struct kernel_param *params, unsigned num) { - unsigned int i; - - for (i = 0; i < num; i++) - if (params[i].flags & KPARAM_KMALLOCED) - kfree(*(char **)params[i].arg); + /* FIXME: This should free kmalloced charp parameters. It doesn't. */ } static void __init kernel_add_sysfs_param(const char *name, diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 0f86feb..7f29643 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -20,6 +20,7 @@ #include <linux/percpu.h> #include <linux/ptrace.h> #include <linux/vmstat.h> +#include <linux/vmalloc.h> #include <linux/hardirq.h> #include <linux/rculist.h> #include <linux/uaccess.h> @@ -1030,14 +1031,10 @@ void __perf_event_sched_out(struct perf_event_context *ctx, update_context_time(ctx); perf_disable(); - if (ctx->nr_active) { - list_for_each_entry(event, &ctx->group_list, group_entry) { - if (event != event->group_leader) - event_sched_out(event, cpuctx, ctx); - else - group_sched_out(event, cpuctx, ctx); - } - } + if (ctx->nr_active) + list_for_each_entry(event, &ctx->group_list, group_entry) + group_sched_out(event, cpuctx, ctx); + perf_enable(); out: spin_unlock(&ctx->lock); @@ -1258,12 +1255,8 @@ __perf_event_sched_in(struct perf_event_context *ctx, if (event->cpu != -1 && event->cpu != cpu) continue; - if (event != event->group_leader) - event_sched_in(event, cpuctx, ctx, cpu); - else { - if (group_can_go_on(event, cpuctx, 1)) - group_sched_in(event, cpuctx, ctx, cpu); - } + if (group_can_go_on(event, cpuctx, 1)) + group_sched_in(event, cpuctx, ctx, cpu); /* * If this pinned group hasn't been scheduled, @@ -1291,15 +1284,9 @@ __perf_event_sched_in(struct perf_event_context *ctx, if (event->cpu != -1 && event->cpu != cpu) continue; - if (event != event->group_leader) { - if (event_sched_in(event, cpuctx, ctx, cpu)) + if (group_can_go_on(event, cpuctx, can_add_hw)) + if (group_sched_in(event, cpuctx, ctx, cpu)) can_add_hw = 0; - } else { - if (group_can_go_on(event, cpuctx, can_add_hw)) { - if (group_sched_in(event, cpuctx, ctx, cpu)) - can_add_hw = 0; - } - } } perf_enable(); out: @@ -1368,7 +1355,7 @@ static void perf_ctx_adjust_freq(struct perf_event_context *ctx) u64 interrupts, freq; spin_lock(&ctx->lock); - list_for_each_entry(event, &ctx->group_list, group_entry) { + list_for_each_entry_rcu(event, &ctx->event_list, event_entry) { if (event->state != PERF_EVENT_STATE_ACTIVE) continue; @@ -2105,49 +2092,31 @@ unlock: rcu_read_unlock(); } -static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +static unsigned long perf_data_size(struct perf_mmap_data *data) { - struct perf_event *event = vma->vm_file->private_data; - struct perf_mmap_data *data; - int ret = VM_FAULT_SIGBUS; - - if (vmf->flags & FAULT_FLAG_MKWRITE) { - if (vmf->pgoff == 0) - ret = 0; - return ret; - } - - rcu_read_lock(); - data = rcu_dereference(event->data); - if (!data) - goto unlock; - - if (vmf->pgoff == 0) { - vmf->page = virt_to_page(data->user_page); - } else { - int nr = vmf->pgoff - 1; - - if ((unsigned)nr > data->nr_pages) - goto unlock; + return data->nr_pages << (PAGE_SHIFT + data->data_order); +} - if (vmf->flags & FAULT_FLAG_WRITE) - goto unlock; +#ifndef CONFIG_PERF_USE_VMALLOC - vmf->page = virt_to_page(data->data_pages[nr]); - } +/* + * Back perf_mmap() with regular GFP_KERNEL-0 pages. + */ - get_page(vmf->page); - vmf->page->mapping = vma->vm_file->f_mapping; - vmf->page->index = vmf->pgoff; +static struct page * +perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +{ + if (pgoff > data->nr_pages) + return NULL; - ret = 0; -unlock: - rcu_read_unlock(); + if (pgoff == 0) + return virt_to_page(data->user_page); - return ret; + return virt_to_page(data->data_pages[pgoff - 1]); } -static int perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +static struct perf_mmap_data * +perf_mmap_data_alloc(struct perf_event *event, int nr_pages) { struct perf_mmap_data *data; unsigned long size; @@ -2172,19 +2141,10 @@ static int perf_mmap_data_alloc(struct perf_event *event, int nr_pages) goto fail_data_pages; } + data->data_order = 0; data->nr_pages = nr_pages; - atomic_set(&data->lock, -1); - - if (event->attr.watermark) { - data->watermark = min_t(long, PAGE_SIZE * nr_pages, - event->attr.wakeup_watermark); - } - if (!data->watermark) - data->watermark = max(PAGE_SIZE, PAGE_SIZE * nr_pages / 4); - - rcu_assign_pointer(event->data, data); - return 0; + return data; fail_data_pages: for (i--; i >= 0; i--) @@ -2196,7 +2156,7 @@ fail_user_page: kfree(data); fail: - return -ENOMEM; + return NULL; } static void perf_mmap_free_page(unsigned long addr) @@ -2207,28 +2167,169 @@ static void perf_mmap_free_page(unsigned long addr) __free_page(page); } -static void __perf_mmap_data_free(struct rcu_head *rcu_head) +static void perf_mmap_data_free(struct perf_mmap_data *data) { - struct perf_mmap_data *data; int i; - data = container_of(rcu_head, struct perf_mmap_data, rcu_head); - perf_mmap_free_page((unsigned long)data->user_page); for (i = 0; i < data->nr_pages; i++) perf_mmap_free_page((unsigned long)data->data_pages[i]); +} + +#else + +/* + * Back perf_mmap() with vmalloc memory. + * + * Required for architectures that have d-cache aliasing issues. + */ + +static struct page * +perf_mmap_to_page(struct perf_mmap_data *data, unsigned long pgoff) +{ + if (pgoff > (1UL << data->data_order)) + return NULL; + + return vmalloc_to_page((void *)data->user_page + pgoff * PAGE_SIZE); +} + +static void perf_mmap_unmark_page(void *addr) +{ + struct page *page = vmalloc_to_page(addr); + + page->mapping = NULL; +} + +static void perf_mmap_data_free_work(struct work_struct *work) +{ + struct perf_mmap_data *data; + void *base; + int i, nr; + + data = container_of(work, struct perf_mmap_data, work); + nr = 1 << data->data_order; + base = data->user_page; + for (i = 0; i < nr + 1; i++) + perf_mmap_unmark_page(base + (i * PAGE_SIZE)); + + vfree(base); +} + +static void perf_mmap_data_free(struct perf_mmap_data *data) +{ + schedule_work(&data->work); +} + +static struct perf_mmap_data * +perf_mmap_data_alloc(struct perf_event *event, int nr_pages) +{ + struct perf_mmap_data *data; + unsigned long size; + void *all_buf; + + WARN_ON(atomic_read(&event->mmap_count)); + + size = sizeof(struct perf_mmap_data); + size += sizeof(void *); + + data = kzalloc(size, GFP_KERNEL); + if (!data) + goto fail; + + INIT_WORK(&data->work, perf_mmap_data_free_work); + + all_buf = vmalloc_user((nr_pages + 1) * PAGE_SIZE); + if (!all_buf) + goto fail_all_buf; + + data->user_page = all_buf; + data->data_pages[0] = all_buf + PAGE_SIZE; + data->data_order = ilog2(nr_pages); + data->nr_pages = 1; + + return data; + +fail_all_buf: kfree(data); + +fail: + return NULL; } -static void perf_mmap_data_free(struct perf_event *event) +#endif + +static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct perf_event *event = vma->vm_file->private_data; + struct perf_mmap_data *data; + int ret = VM_FAULT_SIGBUS; + + if (vmf->flags & FAULT_FLAG_MKWRITE) { + if (vmf->pgoff == 0) + ret = 0; + return ret; + } + + rcu_read_lock(); + data = rcu_dereference(event->data); + if (!data) + goto unlock; + + if (vmf->pgoff && (vmf->flags & FAULT_FLAG_WRITE)) + goto unlock; + + vmf->page = perf_mmap_to_page(data, vmf->pgoff); + if (!vmf->page) + goto unlock; + + get_page(vmf->page); + vmf->page->mapping = vma->vm_file->f_mapping; + vmf->page->index = vmf->pgoff; + + ret = 0; +unlock: + rcu_read_unlock(); + + return ret; +} + +static void +perf_mmap_data_init(struct perf_event *event, struct perf_mmap_data *data) +{ + long max_size = perf_data_size(data); + + atomic_set(&data->lock, -1); + + if (event->attr.watermark) { + data->watermark = min_t(long, max_size, + event->attr.wakeup_watermark); + } + + if (!data->watermark) + data->watermark = max_t(long, PAGE_SIZE, max_size / 2); + + + rcu_assign_pointer(event->data, data); +} + +static void perf_mmap_data_free_rcu(struct rcu_head *rcu_head) +{ + struct perf_mmap_data *data; + + data = container_of(rcu_head, struct perf_mmap_data, rcu_head); + perf_mmap_data_free(data); + kfree(data); +} + +static void perf_mmap_data_release(struct perf_event *event) { struct perf_mmap_data *data = event->data; WARN_ON(atomic_read(&event->mmap_count)); rcu_assign_pointer(event->data, NULL); - call_rcu(&data->rcu_head, __perf_mmap_data_free); + call_rcu(&data->rcu_head, perf_mmap_data_free_rcu); } static void perf_mmap_open(struct vm_area_struct *vma) @@ -2244,11 +2345,12 @@ static void perf_mmap_close(struct vm_area_struct *vma) WARN_ON_ONCE(event->ctx->parent_ctx); if (atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex)) { + unsigned long size = perf_data_size(event->data); struct user_struct *user = current_user(); - atomic_long_sub(event->data->nr_pages + 1, &user->locked_vm); + atomic_long_sub((size >> PAGE_SHIFT) + 1, &user->locked_vm); vma->vm_mm->locked_vm -= event->data->nr_locked; - perf_mmap_data_free(event); + perf_mmap_data_release(event); mutex_unlock(&event->mmap_mutex); } } @@ -2266,6 +2368,7 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) unsigned long user_locked, user_lock_limit; struct user_struct *user = current_user(); unsigned long locked, lock_limit; + struct perf_mmap_data *data; unsigned long vma_size; unsigned long nr_pages; long user_extra, extra; @@ -2328,10 +2431,15 @@ static int perf_mmap(struct file *file, struct vm_area_struct *vma) } WARN_ON(event->data); - ret = perf_mmap_data_alloc(event, nr_pages); - if (ret) + + data = perf_mmap_data_alloc(event, nr_pages); + ret = -ENOMEM; + if (!data) goto unlock; + ret = 0; + perf_mmap_data_init(event, data); + atomic_set(&event->mmap_count, 1); atomic_long_add(user_extra, &user->locked_vm); vma->vm_mm->locked_vm += extra; @@ -2519,7 +2627,7 @@ static bool perf_output_space(struct perf_mmap_data *data, unsigned long tail, if (!data->writable) return true; - mask = (data->nr_pages << PAGE_SHIFT) - 1; + mask = perf_data_size(data) - 1; offset = (offset - tail) & mask; head = (head - tail) & mask; @@ -2624,7 +2732,7 @@ void perf_output_copy(struct perf_output_handle *handle, const void *buf, unsigned int len) { unsigned int pages_mask; - unsigned int offset; + unsigned long offset; unsigned int size; void **pages; @@ -2633,12 +2741,14 @@ void perf_output_copy(struct perf_output_handle *handle, pages = handle->data->data_pages; do { - unsigned int page_offset; + unsigned long page_offset; + unsigned long page_size; int nr; nr = (offset >> PAGE_SHIFT) & pages_mask; - page_offset = offset & (PAGE_SIZE - 1); - size = min_t(unsigned int, PAGE_SIZE - page_offset, len); + page_size = 1UL << (handle->data->data_order + PAGE_SHIFT); + page_offset = offset & (page_size - 1); + size = min_t(unsigned int, page_size - page_offset, len); memcpy(pages[nr] + page_offset, buf, size); @@ -3849,8 +3959,9 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) regs = task_pt_regs(current); if (regs) { - if (perf_event_overflow(event, 0, &data, regs)) - ret = HRTIMER_NORESTART; + if (!(event->attr.exclude_idle && current->pid == 0)) + if (perf_event_overflow(event, 0, &data, regs)) + ret = HRTIMER_NORESTART; } period = max_t(u64, 10000, event->hw.sample_period); @@ -3859,6 +3970,42 @@ static enum hrtimer_restart perf_swevent_hrtimer(struct hrtimer *hrtimer) return ret; } +static void perf_swevent_start_hrtimer(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hwc->hrtimer.function = perf_swevent_hrtimer; + if (hwc->sample_period) { + u64 period; + + if (hwc->remaining) { + if (hwc->remaining < 0) + period = 10000; + else + period = hwc->remaining; + hwc->remaining = 0; + } else { + period = max_t(u64, 10000, hwc->sample_period); + } + __hrtimer_start_range_ns(&hwc->hrtimer, + ns_to_ktime(period), 0, + HRTIMER_MODE_REL, 0); + } +} + +static void perf_swevent_cancel_hrtimer(struct perf_event *event) +{ + struct hw_perf_event *hwc = &event->hw; + + if (hwc->sample_period) { + ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer); + hwc->remaining = ktime_to_ns(remaining); + + hrtimer_cancel(&hwc->hrtimer); + } +} + /* * Software event: cpu wall time clock */ @@ -3881,22 +4028,14 @@ static int cpu_clock_perf_event_enable(struct perf_event *event) int cpu = raw_smp_processor_id(); atomic64_set(&hwc->prev_count, cpu_clock(cpu)); - hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - hwc->hrtimer.function = perf_swevent_hrtimer; - if (hwc->sample_period) { - u64 period = max_t(u64, 10000, hwc->sample_period); - __hrtimer_start_range_ns(&hwc->hrtimer, - ns_to_ktime(period), 0, - HRTIMER_MODE_REL, 0); - } + perf_swevent_start_hrtimer(event); return 0; } static void cpu_clock_perf_event_disable(struct perf_event *event) { - if (event->hw.sample_period) - hrtimer_cancel(&event->hw.hrtimer); + perf_swevent_cancel_hrtimer(event); cpu_clock_perf_event_update(event); } @@ -3933,22 +4072,15 @@ static int task_clock_perf_event_enable(struct perf_event *event) now = event->ctx->time; atomic64_set(&hwc->prev_count, now); - hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - hwc->hrtimer.function = perf_swevent_hrtimer; - if (hwc->sample_period) { - u64 period = max_t(u64, 10000, hwc->sample_period); - __hrtimer_start_range_ns(&hwc->hrtimer, - ns_to_ktime(period), 0, - HRTIMER_MODE_REL, 0); - } + + perf_swevent_start_hrtimer(event); return 0; } static void task_clock_perf_event_disable(struct perf_event *event) { - if (event->hw.sample_period) - hrtimer_cancel(&event->hw.hrtimer); + perf_swevent_cancel_hrtimer(event); task_clock_perf_event_update(event, event->ctx->time); } @@ -4781,9 +4913,7 @@ int perf_event_init_task(struct task_struct *child) * We dont have to disable NMIs - we are only looking at * the list, not manipulating it: */ - list_for_each_entry_rcu(event, &parent_ctx->event_list, event_entry) { - if (event != event->group_leader) - continue; + list_for_each_entry(event, &parent_ctx->group_list, group_entry) { if (!event->attr.inherit) { inherited_all = 0; diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 04b3a83..04a9e90 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -693,21 +693,22 @@ static int software_resume(void) /* The snapshot device should not be opened while we're running */ if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { error = -EBUSY; + swsusp_close(FMODE_READ); goto Unlock; } pm_prepare_console(); error = pm_notifier_call_chain(PM_RESTORE_PREPARE); if (error) - goto Finish; + goto close_finish; error = usermodehelper_disable(); if (error) - goto Finish; + goto close_finish; error = create_basic_memory_bitmaps(); if (error) - goto Finish; + goto close_finish; pr_debug("PM: Preparing processes for restore.\n"); error = prepare_processes(); @@ -719,6 +720,7 @@ static int software_resume(void) pr_debug("PM: Reading hibernation image.\n"); error = swsusp_read(&flags); + swsusp_close(FMODE_READ); if (!error) hibernation_restore(flags & SF_PLATFORM_MODE); @@ -737,6 +739,9 @@ static int software_resume(void) mutex_unlock(&pm_mutex); pr_debug("PM: Resume from disk failed.\n"); return error; +close_finish: + swsusp_close(FMODE_READ); + goto Finish; } late_initcall(software_resume); diff --git a/kernel/power/suspend_test.c b/kernel/power/suspend_test.c index 17d8bb1a..25596e4 100644 --- a/kernel/power/suspend_test.c +++ b/kernel/power/suspend_test.c @@ -19,7 +19,7 @@ * The time it takes is system-specific though, so when we test this * during system bootup we allow a LOT of time. */ -#define TEST_SUSPEND_SECONDS 5 +#define TEST_SUSPEND_SECONDS 10 static unsigned long suspend_test_start_time; @@ -49,7 +49,8 @@ void suspend_test_finish(const char *label) * has some performance issues. The stack dump of a WARN_ON * is more likely to get the right attention than a printk... */ - WARN(msec > (TEST_SUSPEND_SECONDS * 1000), "Component: %s\n", label); + WARN(msec > (TEST_SUSPEND_SECONDS * 1000), + "Component: %s, time: %u\n", label, msec); } /* diff --git a/kernel/power/swap.c b/kernel/power/swap.c index b101cdc..890f6b1 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -314,7 +314,6 @@ static int save_image(struct swap_map_handle *handle, { unsigned int m; int ret; - int error = 0; int nr_pages; int err2; struct bio *bio; @@ -329,26 +328,27 @@ static int save_image(struct swap_map_handle *handle, nr_pages = 0; bio = NULL; do_gettimeofday(&start); - do { + while (1) { ret = snapshot_read_next(snapshot, PAGE_SIZE); - if (ret > 0) { - error = swap_write_page(handle, data_of(*snapshot), - &bio); - if (error) - break; - if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); - nr_pages++; - } - } while (ret > 0); + if (ret <= 0) + break; + ret = swap_write_page(handle, data_of(*snapshot), &bio); + if (ret) + break; + if (!(nr_pages % m)) + printk("\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + } err2 = wait_on_bio_chain(&bio); do_gettimeofday(&stop); - if (!error) - error = err2; - if (!error) + if (!ret) + ret = err2; + if (!ret) printk("\b\b\b\bdone\n"); + else + printk("\n"); swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); - return error; + return ret; } /** @@ -536,7 +536,8 @@ static int load_image(struct swap_map_handle *handle, snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; - } + } else + printk("\n"); swsusp_show_speed(&start, &stop, nr_to_read, "Read"); return error; } @@ -572,8 +573,6 @@ int swsusp_read(unsigned int *flags_p) error = load_image(&handle, &snapshot, header->pages - 1); release_swap_reader(&handle); - blkdev_put(resume_bdev, FMODE_READ); - if (!error) pr_debug("PM: Image successfully loaded\n"); else @@ -596,7 +595,7 @@ int swsusp_check(void) error = bio_read_page(swsusp_resume_block, swsusp_header, NULL); if (error) - return error; + goto put; if (!memcmp(SWSUSP_SIG, swsusp_header->sig, 10)) { memcpy(swsusp_header->sig, swsusp_header->orig_sig, 10); @@ -604,8 +603,10 @@ int swsusp_check(void) error = bio_write_page(swsusp_resume_block, swsusp_header, NULL); } else { - return -EINVAL; + error = -EINVAL; } + +put: if (error) blkdev_put(resume_bdev, FMODE_READ); else diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 37ac454..4001833 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -46,22 +46,15 @@ #include <linux/module.h> #include <linux/kernel_stat.h> -enum rcu_barrier { - RCU_BARRIER_STD, - RCU_BARRIER_BH, - RCU_BARRIER_SCHED, -}; +#ifdef CONFIG_DEBUG_LOCK_ALLOC +static struct lock_class_key rcu_lock_key; +struct lockdep_map rcu_lock_map = + STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); +EXPORT_SYMBOL_GPL(rcu_lock_map); +#endif -static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; -static atomic_t rcu_barrier_cpu_count; -static DEFINE_MUTEX(rcu_barrier_mutex); -static struct completion rcu_barrier_completion; int rcu_scheduler_active __read_mostly; -static atomic_t rcu_migrate_type_count = ATOMIC_INIT(0); -static struct rcu_head rcu_migrate_head[3]; -static DECLARE_WAIT_QUEUE_HEAD(rcu_migrate_wq); - /* * Awaken the corresponding synchronize_rcu() instance now that a * grace period has elapsed. @@ -164,129 +157,10 @@ void synchronize_rcu_bh(void) } EXPORT_SYMBOL_GPL(synchronize_rcu_bh); -static void rcu_barrier_callback(struct rcu_head *notused) -{ - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) - complete(&rcu_barrier_completion); -} - -/* - * Called with preemption disabled, and from cross-cpu IRQ context. - */ -static void rcu_barrier_func(void *type) -{ - int cpu = smp_processor_id(); - struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); - - atomic_inc(&rcu_barrier_cpu_count); - switch ((enum rcu_barrier)type) { - case RCU_BARRIER_STD: - call_rcu(head, rcu_barrier_callback); - break; - case RCU_BARRIER_BH: - call_rcu_bh(head, rcu_barrier_callback); - break; - case RCU_BARRIER_SCHED: - call_rcu_sched(head, rcu_barrier_callback); - break; - } -} - -static inline void wait_migrated_callbacks(void) -{ - wait_event(rcu_migrate_wq, !atomic_read(&rcu_migrate_type_count)); - smp_mb(); /* In case we didn't sleep. */ -} - -/* - * Orchestrate the specified type of RCU barrier, waiting for all - * RCU callbacks of the specified type to complete. - */ -static void _rcu_barrier(enum rcu_barrier type) -{ - BUG_ON(in_interrupt()); - /* Take cpucontrol mutex to protect against CPU hotplug */ - mutex_lock(&rcu_barrier_mutex); - init_completion(&rcu_barrier_completion); - /* - * Initialize rcu_barrier_cpu_count to 1, then invoke - * rcu_barrier_func() on each CPU, so that each CPU also has - * incremented rcu_barrier_cpu_count. Only then is it safe to - * decrement rcu_barrier_cpu_count -- otherwise the first CPU - * might complete its grace period before all of the other CPUs - * did their increment, causing this function to return too - * early. - */ - atomic_set(&rcu_barrier_cpu_count, 1); - on_each_cpu(rcu_barrier_func, (void *)type, 1); - if (atomic_dec_and_test(&rcu_barrier_cpu_count)) - complete(&rcu_barrier_completion); - wait_for_completion(&rcu_barrier_completion); - mutex_unlock(&rcu_barrier_mutex); - wait_migrated_callbacks(); -} - -/** - * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. - */ -void rcu_barrier(void) -{ - _rcu_barrier(RCU_BARRIER_STD); -} -EXPORT_SYMBOL_GPL(rcu_barrier); - -/** - * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete. - */ -void rcu_barrier_bh(void) -{ - _rcu_barrier(RCU_BARRIER_BH); -} -EXPORT_SYMBOL_GPL(rcu_barrier_bh); - -/** - * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks. - */ -void rcu_barrier_sched(void) -{ - _rcu_barrier(RCU_BARRIER_SCHED); -} -EXPORT_SYMBOL_GPL(rcu_barrier_sched); - -static void rcu_migrate_callback(struct rcu_head *notused) -{ - if (atomic_dec_and_test(&rcu_migrate_type_count)) - wake_up(&rcu_migrate_wq); -} - -extern int rcu_cpu_notify(struct notifier_block *self, - unsigned long action, void *hcpu); - static int __cpuinit rcu_barrier_cpu_hotplug(struct notifier_block *self, unsigned long action, void *hcpu) { - rcu_cpu_notify(self, action, hcpu); - if (action == CPU_DYING) { - /* - * preempt_disable() in on_each_cpu() prevents stop_machine(), - * so when "on_each_cpu(rcu_barrier_func, (void *)type, 1);" - * returns, all online cpus have queued rcu_barrier_func(), - * and the dead cpu(if it exist) queues rcu_migrate_callback()s. - * - * These callbacks ensure _rcu_barrier() waits for all - * RCU callbacks of the specified type to complete. - */ - atomic_set(&rcu_migrate_type_count, 3); - call_rcu_bh(rcu_migrate_head, rcu_migrate_callback); - call_rcu_sched(rcu_migrate_head + 1, rcu_migrate_callback); - call_rcu(rcu_migrate_head + 2, rcu_migrate_callback); - } else if (action == CPU_DOWN_PREPARE) { - /* Don't need to wait until next removal operation. */ - /* rcu_migrate_head is protected by cpu_add_remove_lock */ - wait_migrated_callbacks(); - } - - return NOTIFY_OK; + return rcu_cpu_notify(self, action, hcpu); } void __init rcu_init(void) diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index 233768f..697c0a0 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -606,8 +606,6 @@ static struct rcu_torture_ops sched_ops_sync = { .name = "sched_sync" }; -extern int rcu_expedited_torture_stats(char *page); - static struct rcu_torture_ops sched_expedited_ops = { .init = rcu_sync_torture_init, .cleanup = NULL, @@ -650,7 +648,7 @@ rcu_torture_writer(void *arg) old_rp = rcu_torture_current; rp->rtort_mbtest = 1; rcu_assign_pointer(rcu_torture_current, rp); - smp_wmb(); + smp_wmb(); /* Mods to old_rp must follow rcu_assign_pointer() */ if (old_rp) { i = old_rp->rtort_pipe_count; if (i > RCU_TORTURE_PIPE_LEN) diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 52b06f6..f3077c0 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -49,13 +49,6 @@ #include "rcutree.h" -#ifdef CONFIG_DEBUG_LOCK_ALLOC -static struct lock_class_key rcu_lock_key; -struct lockdep_map rcu_lock_map = - STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key); -EXPORT_SYMBOL_GPL(rcu_lock_map); -#endif - /* Data structures. */ #define RCU_STATE_INITIALIZER(name) { \ @@ -66,10 +59,13 @@ EXPORT_SYMBOL_GPL(rcu_lock_map); NUM_RCU_LVL_2, \ NUM_RCU_LVL_3, /* == MAX_RCU_LVLS */ \ }, \ - .signaled = RCU_SIGNAL_INIT, \ + .signaled = RCU_GP_IDLE, \ .gpnum = -300, \ .completed = -300, \ .onofflock = __SPIN_LOCK_UNLOCKED(&name.onofflock), \ + .orphan_cbs_list = NULL, \ + .orphan_cbs_tail = &name.orphan_cbs_list, \ + .orphan_qlen = 0, \ .fqslock = __SPIN_LOCK_UNLOCKED(&name.fqslock), \ .n_force_qs = 0, \ .n_force_qs_ngp = 0, \ @@ -81,24 +77,16 @@ DEFINE_PER_CPU(struct rcu_data, rcu_sched_data); struct rcu_state rcu_bh_state = RCU_STATE_INITIALIZER(rcu_bh_state); DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); -extern long rcu_batches_completed_sched(void); -static struct rcu_node *rcu_get_root(struct rcu_state *rsp); -static void cpu_quiet_msk(unsigned long mask, struct rcu_state *rsp, - struct rcu_node *rnp, unsigned long flags); -static void cpu_quiet_msk_finish(struct rcu_state *rsp, unsigned long flags); -#ifdef CONFIG_HOTPLUG_CPU -static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp); -#endif /* #ifdef CONFIG_HOTPLUG_CPU */ -static void __rcu_process_callbacks(struct rcu_state *rsp, - struct rcu_data *rdp); -static void __call_rcu(struct rcu_head *head, - void (*func)(struct rcu_head *rcu), - struct rcu_state *rsp); -static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp); -static void __cpuinit rcu_init_percpu_data(int cpu, struct rcu_state *rsp, - int preemptable); -#include "rcutree_plugin.h" +/* + * Return true if an RCU grace period is in progress. The ACCESS_ONCE()s + * permit this function to be invoked without holding the root rcu_node + * structure's ->lock, but of course results can be subject to change. + */ +static int rcu_gp_in_progress(struct rcu_state *rsp) +{ + return ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum); +} /* * Note a quiescent state. Because we do not need to know @@ -137,6 +125,10 @@ static int blimit = 10; /* Maximum callbacks per softirq. */ static int qhimark = 10000; /* If this many pending, ignore blimit. */ static int qlowmark = 100; /* Once only this many pending, use blimit. */ +module_param(blimit, int, 0); +module_param(qhimark, int, 0); +module_param(qlowmark, int, 0); + static void force_quiescent_state(struct rcu_state *rsp, int relaxed); static int rcu_pending(int cpu); @@ -173,9 +165,7 @@ cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp) static int cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp) { - /* ACCESS_ONCE() because we are accessing outside of lock. */ - return *rdp->nxttail[RCU_DONE_TAIL] && - ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum); + return *rdp->nxttail[RCU_DONE_TAIL] && !rcu_gp_in_progress(rsp); } /* @@ -369,7 +359,7 @@ static long dyntick_recall_completed(struct rcu_state *rsp) /* * Snapshot the specified CPU's dynticks counter so that we can later * credit them with an implicit quiescent state. Return 1 if this CPU - * is already in a quiescent state courtesy of dynticks idle mode. + * is in dynticks idle mode, which is an extended quiescent state. */ static int dyntick_save_progress_counter(struct rcu_data *rdp) { @@ -475,30 +465,34 @@ static void print_other_cpu_stall(struct rcu_state *rsp) long delta; unsigned long flags; struct rcu_node *rnp = rcu_get_root(rsp); - struct rcu_node *rnp_cur = rsp->level[NUM_RCU_LVLS - 1]; - struct rcu_node *rnp_end = &rsp->node[NUM_RCU_NODES]; /* Only let one CPU complain about others per time interval. */ spin_lock_irqsave(&rnp->lock, flags); delta = jiffies - rsp->jiffies_stall; - if (delta < RCU_STALL_RAT_DELAY || rsp->gpnum == rsp->completed) { + if (delta < RCU_STALL_RAT_DELAY || !rcu_gp_in_progress(rsp)) { spin_unlock_irqrestore(&rnp->lock, flags); return; } rsp->jiffies_stall = jiffies + RCU_SECONDS_TILL_STALL_RECHECK; + + /* + * Now rat on any tasks that got kicked up to the root rcu_node + * due to CPU offlining. + */ + rcu_print_task_stall(rnp); spin_unlock_irqrestore(&rnp->lock, flags); /* OK, time to rat on our buddy... */ printk(KERN_ERR "INFO: RCU detected CPU stalls:"); - for (; rnp_cur < rnp_end; rnp_cur++) { + rcu_for_each_leaf_node(rsp, rnp) { rcu_print_task_stall(rnp); - if (rnp_cur->qsmask == 0) + if (rnp->qsmask == 0) continue; - for (cpu = 0; cpu <= rnp_cur->grphi - rnp_cur->grplo; cpu++) - if (rnp_cur->qsmask & (1UL << cpu)) - printk(" %d", rnp_cur->grplo + cpu); + for (cpu = 0; cpu <= rnp->grphi - rnp->grplo; cpu++) + if (rnp->qsmask & (1UL << cpu)) + printk(" %d", rnp->grplo + cpu); } printk(" (detected by %d, t=%ld jiffies)\n", smp_processor_id(), (long)(jiffies - rsp->gp_start)); @@ -537,8 +531,7 @@ static void check_cpu_stall(struct rcu_state *rsp, struct rcu_data *rdp) /* We haven't checked in, so go dump stack. */ print_cpu_stall(rsp); - } else if (rsp->gpnum != rsp->completed && - delta >= RCU_STALL_RAT_DELAY) { + } else if (rcu_gp_in_progress(rsp) && delta >= RCU_STALL_RAT_DELAY) { /* They had two time units to dump stack, so complain. */ print_other_cpu_stall(rsp); @@ -617,9 +610,15 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) note_new_gpnum(rsp, rdp); /* - * Because we are first, we know that all our callbacks will - * be covered by this upcoming grace period, even the ones - * that were registered arbitrarily recently. + * Because this CPU just now started the new grace period, we know + * that all of its callbacks will be covered by this upcoming grace + * period, even the ones that were registered arbitrarily recently. + * Therefore, advance all outstanding callbacks to RCU_WAIT_TAIL. + * + * Other CPUs cannot be sure exactly when the grace period started. + * Therefore, their recently registered callbacks must pass through + * an additional RCU_NEXT_READY stage, so that they will be handled + * by the next RCU grace period. */ rdp->nxttail[RCU_NEXT_READY_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; rdp->nxttail[RCU_WAIT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; @@ -657,15 +656,18 @@ rcu_start_gp(struct rcu_state *rsp, unsigned long flags) * one corresponding to this CPU, due to the fact that we have * irqs disabled. */ - for (rnp = &rsp->node[0]; rnp < &rsp->node[NUM_RCU_NODES]; rnp++) { - spin_lock(&rnp->lock); /* irqs already disabled. */ + rcu_for_each_node_breadth_first(rsp, rnp) { + spin_lock(&rnp->lock); /* irqs already disabled. */ rcu_preempt_check_blocked_tasks(rnp); rnp->qsmask = rnp->qsmaskinit; rnp->gpnum = rsp->gpnum; - spin_unlock(&rnp->lock); /* irqs already disabled. */ + spin_unlock(&rnp->lock); /* irqs remain disabled. */ } + rnp = rcu_get_root(rsp); + spin_lock(&rnp->lock); /* irqs already disabled. */ rsp->signaled = RCU_SIGNAL_INIT; /* force_quiescent_state now OK. */ + spin_unlock(&rnp->lock); /* irqs remain disabled. */ spin_unlock_irqrestore(&rsp->onofflock, flags); } @@ -703,10 +705,11 @@ rcu_process_gp_end(struct rcu_state *rsp, struct rcu_data *rdp) * hold rnp->lock, as required by rcu_start_gp(), which will release it. */ static void cpu_quiet_msk_finish(struct rcu_state *rsp, unsigned long flags) - __releases(rnp->lock) + __releases(rcu_get_root(rsp)->lock) { - WARN_ON_ONCE(rsp->completed == rsp->gpnum); + WARN_ON_ONCE(!rcu_gp_in_progress(rsp)); rsp->completed = rsp->gpnum; + rsp->signaled = RCU_GP_IDLE; rcu_process_gp_end(rsp, rsp->rda[smp_processor_id()]); rcu_start_gp(rsp, flags); /* releases root node's rnp->lock. */ } @@ -842,17 +845,63 @@ rcu_check_quiescent_state(struct rcu_state *rsp, struct rcu_data *rdp) #ifdef CONFIG_HOTPLUG_CPU /* + * Move a dying CPU's RCU callbacks to the ->orphan_cbs_list for the + * specified flavor of RCU. The callbacks will be adopted by the next + * _rcu_barrier() invocation or by the CPU_DEAD notifier, whichever + * comes first. Because this is invoked from the CPU_DYING notifier, + * irqs are already disabled. + */ +static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +{ + int i; + struct rcu_data *rdp = rsp->rda[smp_processor_id()]; + + if (rdp->nxtlist == NULL) + return; /* irqs disabled, so comparison is stable. */ + spin_lock(&rsp->onofflock); /* irqs already disabled. */ + *rsp->orphan_cbs_tail = rdp->nxtlist; + rsp->orphan_cbs_tail = rdp->nxttail[RCU_NEXT_TAIL]; + rdp->nxtlist = NULL; + for (i = 0; i < RCU_NEXT_SIZE; i++) + rdp->nxttail[i] = &rdp->nxtlist; + rsp->orphan_qlen += rdp->qlen; + rdp->qlen = 0; + spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ +} + +/* + * Adopt previously orphaned RCU callbacks. + */ +static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) +{ + unsigned long flags; + struct rcu_data *rdp; + + spin_lock_irqsave(&rsp->onofflock, flags); + rdp = rsp->rda[smp_processor_id()]; + if (rsp->orphan_cbs_list == NULL) { + spin_unlock_irqrestore(&rsp->onofflock, flags); + return; + } + *rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_list; + rdp->nxttail[RCU_NEXT_TAIL] = rsp->orphan_cbs_tail; + rdp->qlen += rsp->orphan_qlen; + rsp->orphan_cbs_list = NULL; + rsp->orphan_cbs_tail = &rsp->orphan_cbs_list; + rsp->orphan_qlen = 0; + spin_unlock_irqrestore(&rsp->onofflock, flags); +} + +/* * Remove the outgoing CPU from the bitmasks in the rcu_node hierarchy * and move all callbacks from the outgoing CPU to the current one. */ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) { - int i; unsigned long flags; long lastcomp; unsigned long mask; struct rcu_data *rdp = rsp->rda[cpu]; - struct rcu_data *rdp_me; struct rcu_node *rnp; /* Exclude any attempts to start a new grace period. */ @@ -868,39 +917,29 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp) spin_unlock(&rnp->lock); /* irqs remain disabled. */ break; } - rcu_preempt_offline_tasks(rsp, rnp, rdp); + + /* + * If there was a task blocking the current grace period, + * and if all CPUs have checked in, we need to propagate + * the quiescent state up the rcu_node hierarchy. But that + * is inconvenient at the moment due to deadlock issues if + * this should end the current grace period. So set the + * offlined CPU's bit in ->qsmask in order to force the + * next force_quiescent_state() invocation to clean up this + * mess in a deadlock-free manner. + */ + if (rcu_preempt_offline_tasks(rsp, rnp, rdp) && !rnp->qsmask) + rnp->qsmask |= mask; + mask = rnp->grpmask; spin_unlock(&rnp->lock); /* irqs remain disabled. */ rnp = rnp->parent; } while (rnp != NULL); lastcomp = rsp->completed; - spin_unlock(&rsp->onofflock); /* irqs remain disabled. */ + spin_unlock_irqrestore(&rsp->onofflock, flags); - /* - * Move callbacks from the outgoing CPU to the running CPU. - * Note that the outgoing CPU is now quiscent, so it is now - * (uncharacteristically) safe to access its rcu_data structure. - * Note also that we must carefully retain the order of the - * outgoing CPU's callbacks in order for rcu_barrier() to work - * correctly. Finally, note that we start all the callbacks - * afresh, even those that have passed through a grace period - * and are therefore ready to invoke. The theory is that hotplug - * events are rare, and that if they are frequent enough to - * indefinitely delay callbacks, you have far worse things to - * be worrying about. - */ - rdp_me = rsp->rda[smp_processor_id()]; - if (rdp->nxtlist != NULL) { - *rdp_me->nxttail[RCU_NEXT_TAIL] = rdp->nxtlist; - rdp_me->nxttail[RCU_NEXT_TAIL] = rdp->nxttail[RCU_NEXT_TAIL]; - rdp->nxtlist = NULL; - for (i = 0; i < RCU_NEXT_SIZE; i++) - rdp->nxttail[i] = &rdp->nxtlist; - rdp_me->qlen += rdp->qlen; - rdp->qlen = 0; - } - local_irq_restore(flags); + rcu_adopt_orphan_cbs(rsp); } /* @@ -918,6 +957,14 @@ static void rcu_offline_cpu(int cpu) #else /* #ifdef CONFIG_HOTPLUG_CPU */ +static void rcu_send_cbs_to_orphanage(struct rcu_state *rsp) +{ +} + +static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) +{ +} + static void rcu_offline_cpu(int cpu) { } @@ -928,7 +975,7 @@ static void rcu_offline_cpu(int cpu) * Invoke any RCU callbacks that have made it to the end of their grace * period. Thottle as specified by rdp->blimit. */ -static void rcu_do_batch(struct rcu_data *rdp) +static void rcu_do_batch(struct rcu_state *rsp, struct rcu_data *rdp) { unsigned long flags; struct rcu_head *next, *list, **tail; @@ -981,6 +1028,13 @@ static void rcu_do_batch(struct rcu_data *rdp) if (rdp->blimit == LONG_MAX && rdp->qlen <= qlowmark) rdp->blimit = blimit; + /* Reset ->qlen_last_fqs_check trigger if enough CBs have drained. */ + if (rdp->qlen == 0 && rdp->qlen_last_fqs_check != 0) { + rdp->qlen_last_fqs_check = 0; + rdp->n_force_qs_snap = rsp->n_force_qs; + } else if (rdp->qlen < rdp->qlen_last_fqs_check - qhimark) + rdp->qlen_last_fqs_check = rdp->qlen; + local_irq_restore(flags); /* Re-raise the RCU softirq if there are callbacks remaining. */ @@ -1050,33 +1104,32 @@ static int rcu_process_dyntick(struct rcu_state *rsp, long lastcomp, int cpu; unsigned long flags; unsigned long mask; - struct rcu_node *rnp_cur = rsp->level[NUM_RCU_LVLS - 1]; - struct rcu_node *rnp_end = &rsp->node[NUM_RCU_NODES]; + struct rcu_node *rnp; - for (; rnp_cur < rnp_end; rnp_cur++) { + rcu_for_each_leaf_node(rsp, rnp) { mask = 0; - spin_lock_irqsave(&rnp_cur->lock, flags); + spin_lock_irqsave(&rnp->lock, flags); if (rsp->completed != lastcomp) { - spin_unlock_irqrestore(&rnp_cur->lock, flags); + spin_unlock_irqrestore(&rnp->lock, flags); return 1; } - if (rnp_cur->qsmask == 0) { - spin_unlock_irqrestore(&rnp_cur->lock, flags); + if (rnp->qsmask == 0) { + spin_unlock_irqrestore(&rnp->lock, flags); continue; } - cpu = rnp_cur->grplo; + cpu = rnp->grplo; bit = 1; - for (; cpu <= rnp_cur->grphi; cpu++, bit <<= 1) { - if ((rnp_cur->qsmask & bit) != 0 && f(rsp->rda[cpu])) + for (; cpu <= rnp->grphi; cpu++, bit <<= 1) { + if ((rnp->qsmask & bit) != 0 && f(rsp->rda[cpu])) mask |= bit; } if (mask != 0 && rsp->completed == lastcomp) { - /* cpu_quiet_msk() releases rnp_cur->lock. */ - cpu_quiet_msk(mask, rsp, rnp_cur, flags); + /* cpu_quiet_msk() releases rnp->lock. */ + cpu_quiet_msk(mask, rsp, rnp, flags); continue; } - spin_unlock_irqrestore(&rnp_cur->lock, flags); + spin_unlock_irqrestore(&rnp->lock, flags); } return 0; } @@ -1092,7 +1145,7 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) struct rcu_node *rnp = rcu_get_root(rsp); u8 signaled; - if (ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum)) + if (!rcu_gp_in_progress(rsp)) return; /* No grace period in progress, nothing to force. */ if (!spin_trylock_irqsave(&rsp->fqslock, flags)) { rsp->n_force_qs_lh++; /* Inexact, can lose counts. Tough! */ @@ -1113,9 +1166,10 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) } spin_unlock(&rnp->lock); switch (signaled) { + case RCU_GP_IDLE: case RCU_GP_INIT: - break; /* grace period still initializing, ignore. */ + break; /* grace period idle or initializing, ignore. */ case RCU_SAVE_DYNTICK: @@ -1129,7 +1183,8 @@ static void force_quiescent_state(struct rcu_state *rsp, int relaxed) /* Update state, record completion counter. */ spin_lock(&rnp->lock); - if (lastcomp == rsp->completed) { + if (lastcomp == rsp->completed && + rsp->signaled == RCU_SAVE_DYNTICK) { rsp->signaled = RCU_FORCE_QS; dyntick_record_completed(rsp, lastcomp); } @@ -1195,7 +1250,7 @@ __rcu_process_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) } /* If there are callbacks ready, invoke them. */ - rcu_do_batch(rdp); + rcu_do_batch(rsp, rdp); } /* @@ -1251,7 +1306,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rdp->nxttail[RCU_NEXT_TAIL] = &head->next; /* Start a new grace period if one not already started. */ - if (ACCESS_ONCE(rsp->completed) == ACCESS_ONCE(rsp->gpnum)) { + if (!rcu_gp_in_progress(rsp)) { unsigned long nestflag; struct rcu_node *rnp_root = rcu_get_root(rsp); @@ -1259,10 +1314,20 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rcu_start_gp(rsp, nestflag); /* releases rnp_root->lock. */ } - /* Force the grace period if too many callbacks or too long waiting. */ - if (unlikely(++rdp->qlen > qhimark)) { + /* + * Force the grace period if too many callbacks or too long waiting. + * Enforce hysteresis, and don't invoke force_quiescent_state() + * if some other CPU has recently done so. Also, don't bother + * invoking force_quiescent_state() if the newly enqueued callback + * is the only one waiting for a grace period to complete. + */ + if (unlikely(++rdp->qlen > rdp->qlen_last_fqs_check + qhimark)) { rdp->blimit = LONG_MAX; - force_quiescent_state(rsp, 0); + if (rsp->n_force_qs == rdp->n_force_qs_snap && + *rdp->nxttail[RCU_DONE_TAIL] != head) + force_quiescent_state(rsp, 0); + rdp->n_force_qs_snap = rsp->n_force_qs; + rdp->qlen_last_fqs_check = rdp->qlen; } else if ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0) force_quiescent_state(rsp, 1); local_irq_restore(flags); @@ -1331,7 +1396,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp) } /* Has an RCU GP gone long enough to send resched IPIs &c? */ - if (ACCESS_ONCE(rsp->completed) != ACCESS_ONCE(rsp->gpnum) && + if (rcu_gp_in_progress(rsp) && ((long)(ACCESS_ONCE(rsp->jiffies_force_qs) - jiffies) < 0)) { rdp->n_rp_need_fqs++; return 1; @@ -1368,6 +1433,82 @@ int rcu_needs_cpu(int cpu) rcu_preempt_needs_cpu(cpu); } +static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL}; +static atomic_t rcu_barrier_cpu_count; +static DEFINE_MUTEX(rcu_barrier_mutex); +static struct completion rcu_barrier_completion; + +static void rcu_barrier_callback(struct rcu_head *notused) +{ + if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + complete(&rcu_barrier_completion); +} + +/* + * Called with preemption disabled, and from cross-cpu IRQ context. + */ +static void rcu_barrier_func(void *type) +{ + int cpu = smp_processor_id(); + struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu); + void (*call_rcu_func)(struct rcu_head *head, + void (*func)(struct rcu_head *head)); + + atomic_inc(&rcu_barrier_cpu_count); + call_rcu_func = type; + call_rcu_func(head, rcu_barrier_callback); +} + +/* + * Orchestrate the specified type of RCU barrier, waiting for all + * RCU callbacks of the specified type to complete. + */ +static void _rcu_barrier(struct rcu_state *rsp, + void (*call_rcu_func)(struct rcu_head *head, + void (*func)(struct rcu_head *head))) +{ + BUG_ON(in_interrupt()); + /* Take mutex to serialize concurrent rcu_barrier() requests. */ + mutex_lock(&rcu_barrier_mutex); + init_completion(&rcu_barrier_completion); + /* + * Initialize rcu_barrier_cpu_count to 1, then invoke + * rcu_barrier_func() on each CPU, so that each CPU also has + * incremented rcu_barrier_cpu_count. Only then is it safe to + * decrement rcu_barrier_cpu_count -- otherwise the first CPU + * might complete its grace period before all of the other CPUs + * did their increment, causing this function to return too + * early. + */ + atomic_set(&rcu_barrier_cpu_count, 1); + preempt_disable(); /* stop CPU_DYING from filling orphan_cbs_list */ + rcu_adopt_orphan_cbs(rsp); + on_each_cpu(rcu_barrier_func, (void *)call_rcu_func, 1); + preempt_enable(); /* CPU_DYING can again fill orphan_cbs_list */ + if (atomic_dec_and_test(&rcu_barrier_cpu_count)) + complete(&rcu_barrier_completion); + wait_for_completion(&rcu_barrier_completion); + mutex_unlock(&rcu_barrier_mutex); +} + +/** + * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete. + */ +void rcu_barrier_bh(void) +{ + _rcu_barrier(&rcu_bh_state, call_rcu_bh); +} +EXPORT_SYMBOL_GPL(rcu_barrier_bh); + +/** + * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks. + */ +void rcu_barrier_sched(void) +{ + _rcu_barrier(&rcu_sched_state, call_rcu_sched); +} +EXPORT_SYMBOL_GPL(rcu_barrier_sched); + /* * Do boot-time initialization of a CPU's per-CPU RCU data. */ @@ -1418,6 +1559,8 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp, int preemptable) rdp->beenonline = 1; /* We have now been online. */ rdp->preemptable = preemptable; rdp->passed_quiesc_completed = lastcomp - 1; + rdp->qlen_last_fqs_check = 0; + rdp->n_force_qs_snap = rsp->n_force_qs; rdp->blimit = blimit; spin_unlock(&rnp->lock); /* irqs remain disabled. */ @@ -1464,6 +1607,22 @@ int __cpuinit rcu_cpu_notify(struct notifier_block *self, case CPU_UP_PREPARE_FROZEN: rcu_online_cpu(cpu); break; + case CPU_DYING: + case CPU_DYING_FROZEN: + /* + * preempt_disable() in _rcu_barrier() prevents stop_machine(), + * so when "on_each_cpu(rcu_barrier_func, (void *)type, 1);" + * returns, all online cpus have queued rcu_barrier_func(). + * The dying CPU clears its cpu_online_mask bit and + * moves all of its RCU callbacks to ->orphan_cbs_list + * in the context of stop_machine(), so subsequent calls + * to _rcu_barrier() will adopt these callbacks and only + * then queue rcu_barrier_func() on all remaining CPUs. + */ + rcu_send_cbs_to_orphanage(&rcu_bh_state); + rcu_send_cbs_to_orphanage(&rcu_sched_state); + rcu_preempt_send_cbs_to_orphanage(); + break; case CPU_DEAD: case CPU_DEAD_FROZEN: case CPU_UP_CANCELED: @@ -1526,7 +1685,8 @@ static void __init rcu_init_one(struct rcu_state *rsp) cpustride *= rsp->levelspread[i]; rnp = rsp->level[i]; for (j = 0; j < rsp->levelcnt[i]; j++, rnp++) { - spin_lock_init(&rnp->lock); + if (rnp != rcu_get_root(rsp)) + spin_lock_init(&rnp->lock); rnp->gpnum = 0; rnp->qsmask = 0; rnp->qsmaskinit = 0; @@ -1549,6 +1709,7 @@ static void __init rcu_init_one(struct rcu_state *rsp) INIT_LIST_HEAD(&rnp->blocked_tasks[1]); } } + spin_lock_init(&rcu_get_root(rsp)->lock); } /* @@ -1558,6 +1719,10 @@ static void __init rcu_init_one(struct rcu_state *rsp) */ #define RCU_INIT_FLAVOR(rsp, rcu_data) \ do { \ + int i; \ + int j; \ + struct rcu_node *rnp; \ + \ rcu_init_one(rsp); \ rnp = (rsp)->level[NUM_RCU_LVLS - 1]; \ j = 0; \ @@ -1570,31 +1735,8 @@ do { \ } \ } while (0) -#ifdef CONFIG_TREE_PREEMPT_RCU - -void __init __rcu_init_preempt(void) -{ - int i; /* All used by RCU_INIT_FLAVOR(). */ - int j; - struct rcu_node *rnp; - - RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_data); -} - -#else /* #ifdef CONFIG_TREE_PREEMPT_RCU */ - -void __init __rcu_init_preempt(void) -{ -} - -#endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ - void __init __rcu_init(void) { - int i; /* All used by RCU_INIT_FLAVOR(). */ - int j; - struct rcu_node *rnp; - rcu_bootup_announce(); #ifdef CONFIG_RCU_CPU_STALL_DETECTOR printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); @@ -1605,6 +1747,4 @@ void __init __rcu_init(void) open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); } -module_param(blimit, int, 0); -module_param(qhimark, int, 0); -module_param(qlowmark, int, 0); +#include "rcutree_plugin.h" diff --git a/kernel/rcutree.h b/kernel/rcutree.h index 8e8287a..1899023 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -48,14 +48,14 @@ #elif NR_CPUS <= RCU_FANOUT_SQ # define NUM_RCU_LVLS 2 # define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 (((NR_CPUS) + RCU_FANOUT - 1) / RCU_FANOUT) +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) # define NUM_RCU_LVL_2 (NR_CPUS) # define NUM_RCU_LVL_3 0 #elif NR_CPUS <= RCU_FANOUT_CUBE # define NUM_RCU_LVLS 3 # define NUM_RCU_LVL_0 1 -# define NUM_RCU_LVL_1 (((NR_CPUS) + RCU_FANOUT_SQ - 1) / RCU_FANOUT_SQ) -# define NUM_RCU_LVL_2 (((NR_CPUS) + (RCU_FANOUT) - 1) / (RCU_FANOUT)) +# define NUM_RCU_LVL_1 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT_SQ) +# define NUM_RCU_LVL_2 DIV_ROUND_UP(NR_CPUS, RCU_FANOUT) # define NUM_RCU_LVL_3 NR_CPUS #else # error "CONFIG_RCU_FANOUT insufficient for NR_CPUS" @@ -79,15 +79,21 @@ struct rcu_dynticks { * Definition for node within the RCU grace-period-detection hierarchy. */ struct rcu_node { - spinlock_t lock; + spinlock_t lock; /* Root rcu_node's lock protects some */ + /* rcu_state fields as well as following. */ long gpnum; /* Current grace period for this node. */ /* This will either be equal to or one */ /* behind the root rcu_node's gpnum. */ unsigned long qsmask; /* CPUs or groups that need to switch in */ /* order for current grace period to proceed.*/ + /* In leaf rcu_node, each bit corresponds to */ + /* an rcu_data structure, otherwise, each */ + /* bit corresponds to a child rcu_node */ + /* structure. */ unsigned long qsmaskinit; /* Per-GP initialization for qsmask. */ unsigned long grpmask; /* Mask to apply to parent qsmask. */ + /* Only one bit will be set in this mask. */ int grplo; /* lowest-numbered CPU or group here. */ int grphi; /* highest-numbered CPU or group here. */ u8 grpnum; /* CPU/group number for next level up. */ @@ -95,8 +101,23 @@ struct rcu_node { struct rcu_node *parent; struct list_head blocked_tasks[2]; /* Tasks blocked in RCU read-side critsect. */ + /* Grace period number (->gpnum) x blocked */ + /* by tasks on the (x & 0x1) element of the */ + /* blocked_tasks[] array. */ } ____cacheline_internodealigned_in_smp; +/* + * Do a full breadth-first scan of the rcu_node structures for the + * specified rcu_state structure. + */ +#define rcu_for_each_node_breadth_first(rsp, rnp) \ + for ((rnp) = &(rsp)->node[0]; \ + (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + +#define rcu_for_each_leaf_node(rsp, rnp) \ + for ((rnp) = (rsp)->level[NUM_RCU_LVLS - 1]; \ + (rnp) < &(rsp)->node[NUM_RCU_NODES]; (rnp)++) + /* Index values for nxttail array in struct rcu_data. */ #define RCU_DONE_TAIL 0 /* Also RCU_WAIT head. */ #define RCU_WAIT_TAIL 1 /* Also RCU_NEXT_READY head. */ @@ -126,23 +147,30 @@ struct rcu_data { * Any of the partitions might be empty, in which case the * pointer to that partition will be equal to the pointer for * the following partition. When the list is empty, all of - * the nxttail elements point to nxtlist, which is NULL. + * the nxttail elements point to the ->nxtlist pointer itself, + * which in that case is NULL. * - * [*nxttail[RCU_NEXT_READY_TAIL], NULL = *nxttail[RCU_NEXT_TAIL]): - * Entries that might have arrived after current GP ended - * [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]): - * Entries known to have arrived before current GP ended - * [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]): - * Entries that batch # <= ->completed - 1: waiting for current GP * [nxtlist, *nxttail[RCU_DONE_TAIL]): * Entries that batch # <= ->completed * The grace period for these entries has completed, and * the other grace-period-completed entries may be moved * here temporarily in rcu_process_callbacks(). + * [*nxttail[RCU_DONE_TAIL], *nxttail[RCU_WAIT_TAIL]): + * Entries that batch # <= ->completed - 1: waiting for current GP + * [*nxttail[RCU_WAIT_TAIL], *nxttail[RCU_NEXT_READY_TAIL]): + * Entries known to have arrived before current GP ended + * [*nxttail[RCU_NEXT_READY_TAIL], *nxttail[RCU_NEXT_TAIL]): + * Entries that might have arrived after current GP ended + * Note that the value of *nxttail[RCU_NEXT_TAIL] will + * always be NULL, as this is the end of the list. */ struct rcu_head *nxtlist; struct rcu_head **nxttail[RCU_NEXT_SIZE]; long qlen; /* # of queued callbacks */ + long qlen_last_fqs_check; + /* qlen at last check for QS forcing */ + unsigned long n_force_qs_snap; + /* did other CPU force QS recently? */ long blimit; /* Upper limit on a processed batch */ #ifdef CONFIG_NO_HZ @@ -173,9 +201,10 @@ struct rcu_data { }; /* Values for signaled field in struct rcu_state. */ -#define RCU_GP_INIT 0 /* Grace period being initialized. */ -#define RCU_SAVE_DYNTICK 1 /* Need to scan dyntick state. */ -#define RCU_FORCE_QS 2 /* Need to force quiescent state. */ +#define RCU_GP_IDLE 0 /* No grace period in progress. */ +#define RCU_GP_INIT 1 /* Grace period being initialized. */ +#define RCU_SAVE_DYNTICK 2 /* Need to scan dyntick state. */ +#define RCU_FORCE_QS 3 /* Need to force quiescent state. */ #ifdef CONFIG_NO_HZ #define RCU_SIGNAL_INIT RCU_SAVE_DYNTICK #else /* #ifdef CONFIG_NO_HZ */ @@ -216,8 +245,19 @@ struct rcu_state { /* Force QS state. */ long gpnum; /* Current gp number. */ long completed; /* # of last completed gp. */ + + /* End of fields guarded by root rcu_node's lock. */ + spinlock_t onofflock; /* exclude on/offline and */ - /* starting new GP. */ + /* starting new GP. Also */ + /* protects the following */ + /* orphan_cbs fields. */ + struct rcu_head *orphan_cbs_list; /* list of rcu_head structs */ + /* orphaned by all CPUs in */ + /* a given leaf rcu_node */ + /* going offline. */ + struct rcu_head **orphan_cbs_tail; /* And tail pointer. */ + long orphan_qlen; /* Number of orphaned cbs. */ spinlock_t fqslock; /* Only one task forcing */ /* quiescent states. */ unsigned long jiffies_force_qs; /* Time at which to invoke */ @@ -255,5 +295,30 @@ extern struct rcu_state rcu_preempt_state; DECLARE_PER_CPU(struct rcu_data, rcu_preempt_data); #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ -#endif /* #ifdef RCU_TREE_NONCORE */ +#else /* #ifdef RCU_TREE_NONCORE */ + +/* Forward declarations for rcutree_plugin.h */ +static inline void rcu_bootup_announce(void); +long rcu_batches_completed(void); +static void rcu_preempt_note_context_switch(int cpu); +static int rcu_preempted_readers(struct rcu_node *rnp); +#ifdef CONFIG_RCU_CPU_STALL_DETECTOR +static void rcu_print_task_stall(struct rcu_node *rnp); +#endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ +static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); +#ifdef CONFIG_HOTPLUG_CPU +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp); +static void rcu_preempt_offline_cpu(int cpu); +#endif /* #ifdef CONFIG_HOTPLUG_CPU */ +static void rcu_preempt_check_callbacks(int cpu); +static void rcu_preempt_process_callbacks(void); +void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); +static int rcu_preempt_pending(int cpu); +static int rcu_preempt_needs_cpu(int cpu); +static void __cpuinit rcu_preempt_init_percpu_data(int cpu); +static void rcu_preempt_send_cbs_to_orphanage(void); +static void __init __rcu_init_preempt(void); +#endif /* #else #ifdef RCU_TREE_NONCORE */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 1cee04f..ef2a58c 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -150,6 +150,16 @@ void __rcu_read_lock(void) } EXPORT_SYMBOL_GPL(__rcu_read_lock); +/* + * Check for preempted RCU readers blocking the current grace period + * for the specified rcu_node structure. If the caller needs a reliable + * answer, it must hold the rcu_node's ->lock. + */ +static int rcu_preempted_readers(struct rcu_node *rnp) +{ + return !list_empty(&rnp->blocked_tasks[rnp->gpnum & 0x1]); +} + static void rcu_read_unlock_special(struct task_struct *t) { int empty; @@ -196,7 +206,7 @@ static void rcu_read_unlock_special(struct task_struct *t) break; spin_unlock(&rnp->lock); /* irqs remain disabled. */ } - empty = list_empty(&rnp->blocked_tasks[rnp->gpnum & 0x1]); + empty = !rcu_preempted_readers(rnp); list_del_init(&t->rcu_node_entry); t->rcu_blocked_node = NULL; @@ -207,7 +217,7 @@ static void rcu_read_unlock_special(struct task_struct *t) * drop rnp->lock and restore irq. */ if (!empty && rnp->qsmask == 0 && - list_empty(&rnp->blocked_tasks[rnp->gpnum & 0x1])) { + !rcu_preempted_readers(rnp)) { struct rcu_node *rnp_p; if (rnp->parent == NULL) { @@ -257,12 +267,12 @@ static void rcu_print_task_stall(struct rcu_node *rnp) { unsigned long flags; struct list_head *lp; - int phase = rnp->gpnum & 0x1; + int phase; struct task_struct *t; - if (!list_empty(&rnp->blocked_tasks[phase])) { + if (rcu_preempted_readers(rnp)) { spin_lock_irqsave(&rnp->lock, flags); - phase = rnp->gpnum & 0x1; /* re-read under lock. */ + phase = rnp->gpnum & 0x1; lp = &rnp->blocked_tasks[phase]; list_for_each_entry(t, lp, rcu_node_entry) printk(" P%d", t->pid); @@ -281,20 +291,10 @@ static void rcu_print_task_stall(struct rcu_node *rnp) */ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) { - WARN_ON_ONCE(!list_empty(&rnp->blocked_tasks[rnp->gpnum & 0x1])); + WARN_ON_ONCE(rcu_preempted_readers(rnp)); WARN_ON_ONCE(rnp->qsmask); } -/* - * Check for preempted RCU readers for the specified rcu_node structure. - * If the caller needs a reliable answer, it must hold the rcu_node's - * >lock. - */ -static int rcu_preempted_readers(struct rcu_node *rnp) -{ - return !list_empty(&rnp->blocked_tasks[rnp->gpnum & 0x1]); -} - #ifdef CONFIG_HOTPLUG_CPU /* @@ -304,21 +304,25 @@ static int rcu_preempted_readers(struct rcu_node *rnp) * parent is to remove the need for rcu_read_unlock_special() to * make more than two attempts to acquire the target rcu_node's lock. * + * Returns 1 if there was previously a task blocking the current grace + * period on the specified rcu_node structure. + * * The caller must hold rnp->lock with irqs disabled. */ -static void rcu_preempt_offline_tasks(struct rcu_state *rsp, - struct rcu_node *rnp, - struct rcu_data *rdp) +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp) { int i; struct list_head *lp; struct list_head *lp_root; + int retval = rcu_preempted_readers(rnp); struct rcu_node *rnp_root = rcu_get_root(rsp); struct task_struct *tp; if (rnp == rnp_root) { WARN_ONCE(1, "Last CPU thought to be offlined?"); - return; /* Shouldn't happen: at least one CPU online. */ + return 0; /* Shouldn't happen: at least one CPU online. */ } WARN_ON_ONCE(rnp != rdp->mynode && (!list_empty(&rnp->blocked_tasks[0]) || @@ -342,6 +346,8 @@ static void rcu_preempt_offline_tasks(struct rcu_state *rsp, spin_unlock(&rnp_root->lock); /* irqs remain disabled */ } } + + return retval; } /* @@ -393,6 +399,17 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) EXPORT_SYMBOL_GPL(call_rcu); /* + * Wait for an rcu-preempt grace period. We are supposed to expedite the + * grace period, but this is the crude slow compatability hack, so just + * invoke synchronize_rcu(). + */ +void synchronize_rcu_expedited(void) +{ + synchronize_rcu(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); + +/* * Check to see if there is any immediate preemptable-RCU-related work * to be done. */ @@ -410,6 +427,15 @@ static int rcu_preempt_needs_cpu(int cpu) return !!per_cpu(rcu_preempt_data, cpu).nxtlist; } +/** + * rcu_barrier - Wait until all in-flight call_rcu() callbacks complete. + */ +void rcu_barrier(void) +{ + _rcu_barrier(&rcu_preempt_state, call_rcu); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + /* * Initialize preemptable RCU's per-CPU data. */ @@ -419,6 +445,22 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) } /* + * Move preemptable RCU's callbacks to ->orphan_cbs_list. + */ +static void rcu_preempt_send_cbs_to_orphanage(void) +{ + rcu_send_cbs_to_orphanage(&rcu_preempt_state); +} + +/* + * Initialize preemptable RCU's state structures. + */ +static void __init __rcu_init_preempt(void) +{ + RCU_INIT_FLAVOR(&rcu_preempt_state, rcu_preempt_data); +} + +/* * Check for a task exiting while in a preemptable-RCU read-side * critical section, clean up if so. No need to issue warnings, * as debug_check_no_locks_held() already does this if lockdep @@ -461,6 +503,15 @@ static void rcu_preempt_note_context_switch(int cpu) { } +/* + * Because preemptable RCU does not exist, there are never any preempted + * RCU readers. + */ +static int rcu_preempted_readers(struct rcu_node *rnp) +{ + return 0; +} + #ifdef CONFIG_RCU_CPU_STALL_DETECTOR /* @@ -483,25 +534,19 @@ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) WARN_ON_ONCE(rnp->qsmask); } -/* - * Because preemptable RCU does not exist, there are never any preempted - * RCU readers. - */ -static int rcu_preempted_readers(struct rcu_node *rnp) -{ - return 0; -} - #ifdef CONFIG_HOTPLUG_CPU /* * Because preemptable RCU does not exist, it never needs to migrate - * tasks that were blocked within RCU read-side critical sections. + * tasks that were blocked within RCU read-side critical sections, and + * such non-existent tasks cannot possibly have been blocking the current + * grace period. */ -static void rcu_preempt_offline_tasks(struct rcu_state *rsp, - struct rcu_node *rnp, - struct rcu_data *rdp) +static int rcu_preempt_offline_tasks(struct rcu_state *rsp, + struct rcu_node *rnp, + struct rcu_data *rdp) { + return 0; } /* @@ -518,7 +563,7 @@ static void rcu_preempt_offline_cpu(int cpu) * Because preemptable RCU does not exist, it never has any callbacks * to check. */ -void rcu_preempt_check_callbacks(int cpu) +static void rcu_preempt_check_callbacks(int cpu) { } @@ -526,7 +571,7 @@ void rcu_preempt_check_callbacks(int cpu) * Because preemptable RCU does not exist, it never has any callbacks * to process. */ -void rcu_preempt_process_callbacks(void) +static void rcu_preempt_process_callbacks(void) { } @@ -540,6 +585,16 @@ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) EXPORT_SYMBOL_GPL(call_rcu); /* + * Wait for an rcu-preempt grace period, but make it happen quickly. + * But because preemptable RCU does not exist, map to rcu-sched. + */ +void synchronize_rcu_expedited(void) +{ + synchronize_sched_expedited(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_expedited); + +/* * Because preemptable RCU does not exist, it never has any work to do. */ static int rcu_preempt_pending(int cpu) @@ -556,6 +611,16 @@ static int rcu_preempt_needs_cpu(int cpu) } /* + * Because preemptable RCU does not exist, rcu_barrier() is just + * another name for rcu_barrier_sched(). + */ +void rcu_barrier(void) +{ + rcu_barrier_sched(); +} +EXPORT_SYMBOL_GPL(rcu_barrier); + +/* * Because preemptable RCU does not exist, there is no per-CPU * data to initialize. */ @@ -563,4 +628,18 @@ static void __cpuinit rcu_preempt_init_percpu_data(int cpu) { } +/* + * Because there is no preemptable RCU, there are no callbacks to move. + */ +static void rcu_preempt_send_cbs_to_orphanage(void) +{ +} + +/* + * Because preemptable RCU does not exist, it need not be initialized. + */ +static void __init __rcu_init_preempt(void) +{ +} + #endif /* #else #ifdef CONFIG_TREE_PREEMPT_RCU */ diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index c89f5e9..4b31c77 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -93,7 +93,7 @@ static int rcudata_open(struct inode *inode, struct file *file) return single_open(file, show_rcudata, NULL); } -static struct file_operations rcudata_fops = { +static const struct file_operations rcudata_fops = { .owner = THIS_MODULE, .open = rcudata_open, .read = seq_read, @@ -145,7 +145,7 @@ static int rcudata_csv_open(struct inode *inode, struct file *file) return single_open(file, show_rcudata_csv, NULL); } -static struct file_operations rcudata_csv_fops = { +static const struct file_operations rcudata_csv_fops = { .owner = THIS_MODULE, .open = rcudata_csv_open, .read = seq_read, @@ -159,13 +159,13 @@ static void print_one_rcu_state(struct seq_file *m, struct rcu_state *rsp) struct rcu_node *rnp; seq_printf(m, "c=%ld g=%ld s=%d jfq=%ld j=%x " - "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu\n", + "nfqs=%lu/nfqsng=%lu(%lu) fqlh=%lu oqlen=%ld\n", rsp->completed, rsp->gpnum, rsp->signaled, (long)(rsp->jiffies_force_qs - jiffies), (int)(jiffies & 0xffff), rsp->n_force_qs, rsp->n_force_qs_ngp, rsp->n_force_qs - rsp->n_force_qs_ngp, - rsp->n_force_qs_lh); + rsp->n_force_qs_lh, rsp->orphan_qlen); for (rnp = &rsp->node[0]; rnp - &rsp->node[0] < NUM_RCU_NODES; rnp++) { if (rnp->level != level) { seq_puts(m, "\n"); @@ -196,7 +196,7 @@ static int rcuhier_open(struct inode *inode, struct file *file) return single_open(file, show_rcuhier, NULL); } -static struct file_operations rcuhier_fops = { +static const struct file_operations rcuhier_fops = { .owner = THIS_MODULE, .open = rcuhier_open, .read = seq_read, @@ -222,7 +222,7 @@ static int rcugp_open(struct inode *inode, struct file *file) return single_open(file, show_rcugp, NULL); } -static struct file_operations rcugp_fops = { +static const struct file_operations rcugp_fops = { .owner = THIS_MODULE, .open = rcugp_open, .read = seq_read, @@ -276,7 +276,7 @@ static int rcu_pending_open(struct inode *inode, struct file *file) return single_open(file, show_rcu_pending, NULL); } -static struct file_operations rcu_pending_fops = { +static const struct file_operations rcu_pending_fops = { .owner = THIS_MODULE, .open = rcu_pending_open, .read = seq_read, diff --git a/kernel/res_counter.c b/kernel/res_counter.c index 88faec2..bcdabf3 100644 --- a/kernel/res_counter.c +++ b/kernel/res_counter.c @@ -37,27 +37,17 @@ int res_counter_charge_locked(struct res_counter *counter, unsigned long val) } int res_counter_charge(struct res_counter *counter, unsigned long val, - struct res_counter **limit_fail_at, - struct res_counter **soft_limit_fail_at) + struct res_counter **limit_fail_at) { int ret; unsigned long flags; struct res_counter *c, *u; *limit_fail_at = NULL; - if (soft_limit_fail_at) - *soft_limit_fail_at = NULL; local_irq_save(flags); for (c = counter; c != NULL; c = c->parent) { spin_lock(&c->lock); ret = res_counter_charge_locked(c, val); - /* - * With soft limits, we return the highest ancestor - * that exceeds its soft limit - */ - if (soft_limit_fail_at && - !res_counter_soft_limit_check_locked(c)) - *soft_limit_fail_at = c; spin_unlock(&c->lock); if (ret < 0) { *limit_fail_at = c; @@ -85,8 +75,7 @@ void res_counter_uncharge_locked(struct res_counter *counter, unsigned long val) counter->usage -= val; } -void res_counter_uncharge(struct res_counter *counter, unsigned long val, - bool *was_soft_limit_excess) +void res_counter_uncharge(struct res_counter *counter, unsigned long val) { unsigned long flags; struct res_counter *c; @@ -94,9 +83,6 @@ void res_counter_uncharge(struct res_counter *counter, unsigned long val, local_irq_save(flags); for (c = counter; c != NULL; c = c->parent) { spin_lock(&c->lock); - if (was_soft_limit_excess) - *was_soft_limit_excess = - !res_counter_soft_limit_check_locked(c); res_counter_uncharge_locked(c, val); spin_unlock(&c->lock); } diff --git a/kernel/sched.c b/kernel/sched.c index ee61f45..3c11ae0 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -309,6 +309,8 @@ static DEFINE_PER_CPU_SHARED_ALIGNED(struct rt_rq, init_rt_rq); */ static DEFINE_SPINLOCK(task_group_lock); +#ifdef CONFIG_FAIR_GROUP_SCHED + #ifdef CONFIG_SMP static int root_task_group_empty(void) { @@ -316,7 +318,6 @@ static int root_task_group_empty(void) } #endif -#ifdef CONFIG_FAIR_GROUP_SCHED #ifdef CONFIG_USER_SCHED # define INIT_TASK_GROUP_LOAD (2*NICE_0_LOAD) #else /* !CONFIG_USER_SCHED */ @@ -676,6 +677,7 @@ inline void update_rq_clock(struct rq *rq) /** * runqueue_is_locked + * @cpu: the processor in question. * * Returns true if the current cpu runqueue is locked. * This interface allows printk to be called with the runqueue lock @@ -780,7 +782,7 @@ static int sched_feat_open(struct inode *inode, struct file *filp) return single_open(filp, sched_feat_show, NULL); } -static struct file_operations sched_feat_fops = { +static const struct file_operations sched_feat_fops = { .open = sched_feat_open, .write = sched_feat_write, .read = seq_read, @@ -1563,11 +1565,7 @@ static unsigned long cpu_avg_load_per_task(int cpu) #ifdef CONFIG_FAIR_GROUP_SCHED -struct update_shares_data { - unsigned long rq_weight[NR_CPUS]; -}; - -static DEFINE_PER_CPU(struct update_shares_data, update_shares_data); +static __read_mostly unsigned long *update_shares_data; static void __set_se_shares(struct sched_entity *se, unsigned long shares); @@ -1577,12 +1575,12 @@ static void __set_se_shares(struct sched_entity *se, unsigned long shares); static void update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long sd_shares, unsigned long sd_rq_weight, - struct update_shares_data *usd) + unsigned long *usd_rq_weight) { unsigned long shares, rq_weight; int boost = 0; - rq_weight = usd->rq_weight[cpu]; + rq_weight = usd_rq_weight[cpu]; if (!rq_weight) { boost = 1; rq_weight = NICE_0_LOAD; @@ -1617,7 +1615,7 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, static int tg_shares_up(struct task_group *tg, void *data) { unsigned long weight, rq_weight = 0, shares = 0; - struct update_shares_data *usd; + unsigned long *usd_rq_weight; struct sched_domain *sd = data; unsigned long flags; int i; @@ -1626,11 +1624,11 @@ static int tg_shares_up(struct task_group *tg, void *data) return 0; local_irq_save(flags); - usd = &__get_cpu_var(update_shares_data); + usd_rq_weight = per_cpu_ptr(update_shares_data, smp_processor_id()); for_each_cpu(i, sched_domain_span(sd)) { weight = tg->cfs_rq[i]->load.weight; - usd->rq_weight[i] = weight; + usd_rq_weight[i] = weight; /* * If there are currently no tasks on the cpu pretend there @@ -1651,7 +1649,7 @@ static int tg_shares_up(struct task_group *tg, void *data) shares = tg->shares; for_each_cpu(i, sched_domain_span(sd)) - update_group_shares_cpu(tg, i, shares, rq_weight, usd); + update_group_shares_cpu(tg, i, shares, rq_weight, usd_rq_weight); local_irq_restore(flags); @@ -1995,6 +1993,38 @@ static inline void check_class_changed(struct rq *rq, struct task_struct *p, p->sched_class->prio_changed(rq, p, oldprio, running); } +/** + * kthread_bind - bind a just-created kthread to a cpu. + * @p: thread created by kthread_create(). + * @cpu: cpu (might not be online, must be possible) for @k to run on. + * + * Description: This function is equivalent to set_cpus_allowed(), + * except that @cpu doesn't need to be online, and the thread must be + * stopped (i.e., just returned from kthread_create()). + * + * Function lives here instead of kthread.c because it messes with + * scheduler internals which require locking. + */ +void kthread_bind(struct task_struct *p, unsigned int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + /* Must have done schedule() in kthread() before we set_task_cpu */ + if (!wait_task_inactive(p, TASK_UNINTERRUPTIBLE)) { + WARN_ON(1); + return; + } + + spin_lock_irqsave(&rq->lock, flags); + set_task_cpu(p, cpu); + p->cpus_allowed = cpumask_of_cpu(cpu); + p->rt.nr_cpus_allowed = 1; + p->flags |= PF_THREAD_BOUND; + spin_unlock_irqrestore(&rq->lock, flags); +} +EXPORT_SYMBOL(kthread_bind); + #ifdef CONFIG_SMP /* * Is this task likely cache-hot: @@ -2007,7 +2037,7 @@ task_hot(struct task_struct *p, u64 now, struct sched_domain *sd) /* * Buddy candidates are cache hot: */ - if (sched_feat(CACHE_HOT_BUDDY) && + if (sched_feat(CACHE_HOT_BUDDY) && this_rq()->nr_running && (&p->se == cfs_rq_of(&p->se)->next || &p->se == cfs_rq_of(&p->se)->last)) return 1; @@ -2311,7 +2341,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, { int cpu, orig_cpu, this_cpu, success = 0; unsigned long flags; - struct rq *rq; + struct rq *rq, *orig_rq; if (!sched_feat(SYNC_WAKEUPS)) wake_flags &= ~WF_SYNC; @@ -2319,7 +2349,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, this_cpu = get_cpu(); smp_wmb(); - rq = task_rq_lock(p, &flags); + rq = orig_rq = task_rq_lock(p, &flags); update_rq_clock(rq); if (!(p->state & state)) goto out; @@ -2350,6 +2380,10 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, set_task_cpu(p, cpu); rq = task_rq_lock(p, &flags); + + if (rq != orig_rq) + update_rq_clock(rq); + WARN_ON(p->state != TASK_WAKING); cpu = task_cpu(p); @@ -2515,22 +2549,17 @@ void sched_fork(struct task_struct *p, int clone_flags) __sched_fork(p); /* - * Make sure we do not leak PI boosting priority to the child. - */ - p->prio = current->normal_prio; - - /* * Revert to default priority/policy on fork if requested. */ if (unlikely(p->sched_reset_on_fork)) { - if (p->policy == SCHED_FIFO || p->policy == SCHED_RR) + if (p->policy == SCHED_FIFO || p->policy == SCHED_RR) { p->policy = SCHED_NORMAL; - - if (p->normal_prio < DEFAULT_PRIO) - p->prio = DEFAULT_PRIO; + p->normal_prio = p->static_prio; + } if (PRIO_TO_NICE(p->static_prio) < 0) { p->static_prio = NICE_TO_PRIO(0); + p->normal_prio = p->static_prio; set_load_weight(p); } @@ -2541,6 +2570,11 @@ void sched_fork(struct task_struct *p, int clone_flags) p->sched_reset_on_fork = 0; } + /* + * Make sure we do not leak PI boosting priority to the child. + */ + p->prio = current->normal_prio; + if (!rt_prio(p->prio)) p->sched_class = &fair_sched_class; @@ -2581,8 +2615,6 @@ void wake_up_new_task(struct task_struct *p, unsigned long clone_flags) BUG_ON(p->state != TASK_RUNNING); update_rq_clock(rq); - p->prio = effective_prio(p); - if (!p->sched_class->task_new || !current->se.on_rq) { activate_task(rq, p, 0); } else { @@ -3658,6 +3690,7 @@ static void update_group_power(struct sched_domain *sd, int cpu) /** * update_sg_lb_stats - Update sched_group's statistics for load balancing. + * @sd: The sched_domain whose statistics are to be updated. * @group: sched_group whose statistics are to be updated. * @this_cpu: Cpu for which load balance is currently performed. * @idle: Idle status of this_cpu @@ -6720,9 +6753,6 @@ EXPORT_SYMBOL(yield); /* * This task is about to go to sleep on IO. Increment rq->nr_iowait so * that process accounting knows that this is a task in IO wait state. - * - * But don't do that if it is a deliberate, throttling IO wait (this task - * has set its backing_dev_info: the queue against which it should throttle) */ void __sched io_schedule(void) { @@ -9406,6 +9436,10 @@ void __init sched_init(void) #endif /* CONFIG_USER_SCHED */ #endif /* CONFIG_GROUP_SCHED */ +#if defined CONFIG_FAIR_GROUP_SCHED && defined CONFIG_SMP + update_shares_data = __alloc_percpu(nr_cpu_ids * sizeof(unsigned long), + __alignof__(unsigned long)); +#endif for_each_possible_cpu(i) { struct rq *rq; @@ -9531,13 +9565,13 @@ void __init sched_init(void) current->sched_class = &fair_sched_class; /* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */ - alloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT); + zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT); #ifdef CONFIG_SMP #ifdef CONFIG_NO_HZ - alloc_cpumask_var(&nohz.cpu_mask, GFP_NOWAIT); + zalloc_cpumask_var(&nohz.cpu_mask, GFP_NOWAIT); alloc_cpumask_var(&nohz.ilb_grp_nohz_mask, GFP_NOWAIT); #endif - alloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); + zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); #endif /* SMP */ perf_event_init(); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 4e777b4..37087a7 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -822,6 +822,26 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) * re-elected due to buddy favours. */ clear_buddies(cfs_rq, curr); + return; + } + + /* + * Ensure that a task that missed wakeup preemption by a + * narrow margin doesn't have to wait for a full slice. + * This also mitigates buddy induced latencies under load. + */ + if (!sched_feat(WAKEUP_PREEMPT)) + return; + + if (delta_exec < sysctl_sched_min_granularity) + return; + + if (cfs_rq->nr_running > 1) { + struct sched_entity *se = __pick_next_entity(cfs_rq); + s64 delta = curr->vruntime - se->vruntime; + + if (delta > ideal_runtime) + resched_task(rq_of(cfs_rq)->curr); } } @@ -861,12 +881,18 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); static struct sched_entity *pick_next_entity(struct cfs_rq *cfs_rq) { struct sched_entity *se = __pick_next_entity(cfs_rq); + struct sched_entity *left = se; - if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, se) < 1) - return cfs_rq->next; + if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) + se = cfs_rq->next; - if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, se) < 1) - return cfs_rq->last; + /* + * Prefer last buddy, try to return the CPU to a preempted task. + */ + if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1) + se = cfs_rq->last; + + clear_buddies(cfs_rq, se); return se; } @@ -1568,6 +1594,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ struct sched_entity *se = &curr->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(curr); int sync = wake_flags & WF_SYNC; + int scale = cfs_rq->nr_running >= sched_nr_latency; update_curr(cfs_rq); @@ -1582,18 +1609,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ if (unlikely(se == pse)) return; - /* - * Only set the backward buddy when the current task is still on the - * rq. This can happen when a wakeup gets interleaved with schedule on - * the ->pre_schedule() or idle_balance() point, either of which can - * drop the rq lock. - * - * Also, during early boot the idle thread is in the fair class, for - * obvious reasons its a bad idea to schedule back to the idle thread. - */ - if (sched_feat(LAST_BUDDY) && likely(se->on_rq && curr != rq->idle)) - set_last_buddy(se); - if (sched_feat(NEXT_BUDDY) && !(wake_flags & WF_FORK)) + if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK)) set_next_buddy(pse); /* @@ -1639,8 +1655,22 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ BUG_ON(!pse); - if (wakeup_preempt_entity(se, pse) == 1) + if (wakeup_preempt_entity(se, pse) == 1) { resched_task(curr); + /* + * Only set the backward buddy when the current task is still + * on the rq. This can happen when a wakeup gets interleaved + * with schedule on the ->pre_schedule() or idle_balance() + * point, either of which can * drop the rq lock. + * + * Also, during early boot the idle thread is in the fair class, + * for obvious reasons its a bad idea to schedule back to it. + */ + if (unlikely(!se->on_rq || curr == rq->idle)) + return; + if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) + set_last_buddy(se); + } } static struct task_struct *pick_next_task_fair(struct rq *rq) @@ -1654,16 +1684,6 @@ static struct task_struct *pick_next_task_fair(struct rq *rq) do { se = pick_next_entity(cfs_rq); - /* - * If se was a buddy, clear it so that it will have to earn - * the favour again. - * - * If se was not a buddy, clear the buddies because neither - * was elegible to run, let them earn it again. - * - * IOW. unconditionally clear buddies. - */ - __clear_buddies(cfs_rq, NULL); set_next_entity(cfs_rq, se); cfs_rq = group_cfs_rq(se); } while (cfs_rq); diff --git a/kernel/slow-work-debugfs.c b/kernel/slow-work-debugfs.c new file mode 100644 index 0000000..e45c436 --- /dev/null +++ b/kernel/slow-work-debugfs.c @@ -0,0 +1,227 @@ +/* Slow work debugging + * + * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#include <linux/module.h> +#include <linux/slow-work.h> +#include <linux/fs.h> +#include <linux/time.h> +#include <linux/seq_file.h> +#include "slow-work.h" + +#define ITERATOR_SHIFT (BITS_PER_LONG - 4) +#define ITERATOR_SELECTOR (0xfUL << ITERATOR_SHIFT) +#define ITERATOR_COUNTER (~ITERATOR_SELECTOR) + +void slow_work_new_thread_desc(struct slow_work *work, struct seq_file *m) +{ + seq_puts(m, "Slow-work: New thread"); +} + +/* + * Render the time mark field on a work item into a 5-char time with units plus + * a space + */ +static void slow_work_print_mark(struct seq_file *m, struct slow_work *work) +{ + struct timespec now, diff; + + now = CURRENT_TIME; + diff = timespec_sub(now, work->mark); + + if (diff.tv_sec < 0) + seq_puts(m, " -ve "); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000) + seq_printf(m, "%3luns ", diff.tv_nsec); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000000) + seq_printf(m, "%3luus ", diff.tv_nsec / 1000); + else if (diff.tv_sec == 0 && diff.tv_nsec < 1000000000) + seq_printf(m, "%3lums ", diff.tv_nsec / 1000000); + else if (diff.tv_sec <= 1) + seq_puts(m, " 1s "); + else if (diff.tv_sec < 60) + seq_printf(m, "%4lus ", diff.tv_sec); + else if (diff.tv_sec < 60 * 60) + seq_printf(m, "%4lum ", diff.tv_sec / 60); + else if (diff.tv_sec < 60 * 60 * 24) + seq_printf(m, "%4luh ", diff.tv_sec / 3600); + else + seq_puts(m, "exces "); +} + +/* + * Describe a slow work item for debugfs + */ +static int slow_work_runqueue_show(struct seq_file *m, void *v) +{ + struct slow_work *work; + struct list_head *p = v; + unsigned long id; + + switch ((unsigned long) v) { + case 1: + seq_puts(m, "THR PID ITEM ADDR FL MARK DESC\n"); + return 0; + case 2: + seq_puts(m, "=== ===== ================ == ===== ==========\n"); + return 0; + + case 3 ... 3 + SLOW_WORK_THREAD_LIMIT - 1: + id = (unsigned long) v - 3; + + read_lock(&slow_work_execs_lock); + work = slow_work_execs[id]; + if (work) { + smp_read_barrier_depends(); + + seq_printf(m, "%3lu %5d %16p %2lx ", + id, slow_work_pids[id], work, work->flags); + slow_work_print_mark(m, work); + + if (work->ops->desc) + work->ops->desc(work, m); + seq_putc(m, '\n'); + } + read_unlock(&slow_work_execs_lock); + return 0; + + default: + work = list_entry(p, struct slow_work, link); + seq_printf(m, "%3s - %16p %2lx ", + work->flags & SLOW_WORK_VERY_SLOW ? "vsq" : "sq", + work, work->flags); + slow_work_print_mark(m, work); + + if (work->ops->desc) + work->ops->desc(work, m); + seq_putc(m, '\n'); + return 0; + } +} + +/* + * map the iterator to a work item + */ +static void *slow_work_runqueue_index(struct seq_file *m, loff_t *_pos) +{ + struct list_head *p; + unsigned long count, id; + + switch (*_pos >> ITERATOR_SHIFT) { + case 0x0: + if (*_pos == 0) + *_pos = 1; + if (*_pos < 3) + return (void *)(unsigned long) *_pos; + if (*_pos < 3 + SLOW_WORK_THREAD_LIMIT) + for (id = *_pos - 3; + id < SLOW_WORK_THREAD_LIMIT; + id++, (*_pos)++) + if (slow_work_execs[id]) + return (void *)(unsigned long) *_pos; + *_pos = 0x1UL << ITERATOR_SHIFT; + + case 0x1: + count = *_pos & ITERATOR_COUNTER; + list_for_each(p, &slow_work_queue) { + if (count == 0) + return p; + count--; + } + *_pos = 0x2UL << ITERATOR_SHIFT; + + case 0x2: + count = *_pos & ITERATOR_COUNTER; + list_for_each(p, &vslow_work_queue) { + if (count == 0) + return p; + count--; + } + *_pos = 0x3UL << ITERATOR_SHIFT; + + default: + return NULL; + } +} + +/* + * set up the iterator to start reading from the first line + */ +static void *slow_work_runqueue_start(struct seq_file *m, loff_t *_pos) +{ + spin_lock_irq(&slow_work_queue_lock); + return slow_work_runqueue_index(m, _pos); +} + +/* + * move to the next line + */ +static void *slow_work_runqueue_next(struct seq_file *m, void *v, loff_t *_pos) +{ + struct list_head *p = v; + unsigned long selector = *_pos >> ITERATOR_SHIFT; + + (*_pos)++; + switch (selector) { + case 0x0: + return slow_work_runqueue_index(m, _pos); + + case 0x1: + if (*_pos >> ITERATOR_SHIFT == 0x1) { + p = p->next; + if (p != &slow_work_queue) + return p; + } + *_pos = 0x2UL << ITERATOR_SHIFT; + p = &vslow_work_queue; + + case 0x2: + if (*_pos >> ITERATOR_SHIFT == 0x2) { + p = p->next; + if (p != &vslow_work_queue) + return p; + } + *_pos = 0x3UL << ITERATOR_SHIFT; + + default: + return NULL; + } +} + +/* + * clean up after reading + */ +static void slow_work_runqueue_stop(struct seq_file *m, void *v) +{ + spin_unlock_irq(&slow_work_queue_lock); +} + +static const struct seq_operations slow_work_runqueue_ops = { + .start = slow_work_runqueue_start, + .stop = slow_work_runqueue_stop, + .next = slow_work_runqueue_next, + .show = slow_work_runqueue_show, +}; + +/* + * open "/sys/kernel/debug/slow_work/runqueue" to list queue contents + */ +static int slow_work_runqueue_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &slow_work_runqueue_ops); +} + +const struct file_operations slow_work_runqueue_fops = { + .owner = THIS_MODULE, + .open = slow_work_runqueue_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; diff --git a/kernel/slow-work.c b/kernel/slow-work.c index 0d31135..00889bd 100644 --- a/kernel/slow-work.c +++ b/kernel/slow-work.c @@ -16,11 +16,8 @@ #include <linux/kthread.h> #include <linux/freezer.h> #include <linux/wait.h> - -#define SLOW_WORK_CULL_TIMEOUT (5 * HZ) /* cull threads 5s after running out of - * things to do */ -#define SLOW_WORK_OOM_TIMEOUT (5 * HZ) /* can't start new threads for 5s after - * OOM */ +#include <linux/debugfs.h> +#include "slow-work.h" static void slow_work_cull_timeout(unsigned long); static void slow_work_oom_timeout(unsigned long); @@ -46,7 +43,7 @@ static unsigned vslow_work_proportion = 50; /* % of threads that may process #ifdef CONFIG_SYSCTL static const int slow_work_min_min_threads = 2; -static int slow_work_max_max_threads = 255; +static int slow_work_max_max_threads = SLOW_WORK_THREAD_LIMIT; static const int slow_work_min_vslow = 1; static const int slow_work_max_vslow = 99; @@ -98,6 +95,56 @@ static DEFINE_TIMER(slow_work_oom_timer, slow_work_oom_timeout, 0, 0); static struct slow_work slow_work_new_thread; /* new thread starter */ /* + * slow work ID allocation (use slow_work_queue_lock) + */ +static DECLARE_BITMAP(slow_work_ids, SLOW_WORK_THREAD_LIMIT); + +/* + * Unregistration tracking to prevent put_ref() from disappearing during module + * unload + */ +#ifdef CONFIG_MODULES +static struct module *slow_work_thread_processing[SLOW_WORK_THREAD_LIMIT]; +static struct module *slow_work_unreg_module; +static struct slow_work *slow_work_unreg_work_item; +static DECLARE_WAIT_QUEUE_HEAD(slow_work_unreg_wq); +static DEFINE_MUTEX(slow_work_unreg_sync_lock); + +static void slow_work_set_thread_processing(int id, struct slow_work *work) +{ + if (work) + slow_work_thread_processing[id] = work->owner; +} +static void slow_work_done_thread_processing(int id, struct slow_work *work) +{ + struct module *module = slow_work_thread_processing[id]; + + slow_work_thread_processing[id] = NULL; + smp_mb(); + if (slow_work_unreg_work_item == work || + slow_work_unreg_module == module) + wake_up_all(&slow_work_unreg_wq); +} +static void slow_work_clear_thread_processing(int id) +{ + slow_work_thread_processing[id] = NULL; +} +#else +static void slow_work_set_thread_processing(int id, struct slow_work *work) {} +static void slow_work_done_thread_processing(int id, struct slow_work *work) {} +static void slow_work_clear_thread_processing(int id) {} +#endif + +/* + * Data for tracking currently executing items for indication through /proc + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +struct slow_work *slow_work_execs[SLOW_WORK_THREAD_LIMIT]; +pid_t slow_work_pids[SLOW_WORK_THREAD_LIMIT]; +DEFINE_RWLOCK(slow_work_execs_lock); +#endif + +/* * The queues of work items and the lock governing access to them. These are * shared between all the CPUs. It doesn't make sense to have per-CPU queues * as the number of threads bears no relation to the number of CPUs. @@ -105,9 +152,18 @@ static struct slow_work slow_work_new_thread; /* new thread starter */ * There are two queues of work items: one for slow work items, and one for * very slow work items. */ -static LIST_HEAD(slow_work_queue); -static LIST_HEAD(vslow_work_queue); -static DEFINE_SPINLOCK(slow_work_queue_lock); +LIST_HEAD(slow_work_queue); +LIST_HEAD(vslow_work_queue); +DEFINE_SPINLOCK(slow_work_queue_lock); + +/* + * The following are two wait queues that get pinged when a work item is placed + * on an empty queue. These allow work items that are hogging a thread by + * sleeping in a way that could be deferred to yield their thread and enqueue + * themselves. + */ +static DECLARE_WAIT_QUEUE_HEAD(slow_work_queue_waits_for_occupation); +static DECLARE_WAIT_QUEUE_HEAD(vslow_work_queue_waits_for_occupation); /* * The thread controls. A variable used to signal to the threads that they @@ -126,6 +182,20 @@ static DECLARE_COMPLETION(slow_work_last_thread_exited); static int slow_work_user_count; static DEFINE_MUTEX(slow_work_user_lock); +static inline int slow_work_get_ref(struct slow_work *work) +{ + if (work->ops->get_ref) + return work->ops->get_ref(work); + + return 0; +} + +static inline void slow_work_put_ref(struct slow_work *work) +{ + if (work->ops->put_ref) + work->ops->put_ref(work); +} + /* * Calculate the maximum number of active threads in the pool that are * permitted to process very slow work items. @@ -149,7 +219,7 @@ static unsigned slow_work_calc_vsmax(void) * Attempt to execute stuff queued on a slow thread. Return true if we managed * it, false if there was nothing to do. */ -static bool slow_work_execute(void) +static noinline bool slow_work_execute(int id) { struct slow_work *work = NULL; unsigned vsmax; @@ -186,6 +256,13 @@ static bool slow_work_execute(void) } else { very_slow = false; /* avoid the compiler warning */ } + + slow_work_set_thread_processing(id, work); + if (work) { + slow_work_mark_time(work); + slow_work_begin_exec(id, work); + } + spin_unlock_irq(&slow_work_queue_lock); if (!work) @@ -194,12 +271,19 @@ static bool slow_work_execute(void) if (!test_and_clear_bit(SLOW_WORK_PENDING, &work->flags)) BUG(); - work->ops->execute(work); + /* don't execute if the work is in the process of being cancelled */ + if (!test_bit(SLOW_WORK_CANCELLING, &work->flags)) + work->ops->execute(work); if (very_slow) atomic_dec(&vslow_work_executing_count); clear_bit_unlock(SLOW_WORK_EXECUTING, &work->flags); + /* wake up anyone waiting for this work to be complete */ + wake_up_bit(&work->flags, SLOW_WORK_EXECUTING); + + slow_work_end_exec(id, work); + /* if someone tried to enqueue the item whilst we were executing it, * then it'll be left unenqueued to avoid multiple threads trying to * execute it simultaneously @@ -219,7 +303,10 @@ static bool slow_work_execute(void) spin_unlock_irq(&slow_work_queue_lock); } - work->ops->put_ref(work); + /* sort out the race between module unloading and put_ref() */ + slow_work_put_ref(work); + slow_work_done_thread_processing(id, work); + return true; auto_requeue: @@ -227,15 +314,61 @@ auto_requeue: * - we transfer our ref on the item back to the appropriate queue * - don't wake another thread up as we're awake already */ + slow_work_mark_time(work); if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) list_add_tail(&work->link, &vslow_work_queue); else list_add_tail(&work->link, &slow_work_queue); spin_unlock_irq(&slow_work_queue_lock); + slow_work_clear_thread_processing(id); return true; } /** + * slow_work_sleep_till_thread_needed - Sleep till thread needed by other work + * work: The work item under execution that wants to sleep + * _timeout: Scheduler sleep timeout + * + * Allow a requeueable work item to sleep on a slow-work processor thread until + * that thread is needed to do some other work or the sleep is interrupted by + * some other event. + * + * The caller must set up a wake up event before calling this and must have set + * the appropriate sleep mode (such as TASK_UNINTERRUPTIBLE) and tested its own + * condition before calling this function as no test is made here. + * + * False is returned if there is nothing on the queue; true is returned if the + * work item should be requeued + */ +bool slow_work_sleep_till_thread_needed(struct slow_work *work, + signed long *_timeout) +{ + wait_queue_head_t *wfo_wq; + struct list_head *queue; + + DEFINE_WAIT(wait); + + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + + if (!list_empty(queue)) + return true; + + add_wait_queue_exclusive(wfo_wq, &wait); + if (list_empty(queue)) + *_timeout = schedule_timeout(*_timeout); + finish_wait(wfo_wq, &wait); + + return !list_empty(queue); +} +EXPORT_SYMBOL(slow_work_sleep_till_thread_needed); + +/** * slow_work_enqueue - Schedule a slow work item for processing * @work: The work item to queue * @@ -260,16 +393,22 @@ auto_requeue: * allowed to pick items to execute. This ensures that very slow items won't * overly block ones that are just ordinarily slow. * - * Returns 0 if successful, -EAGAIN if not. + * Returns 0 if successful, -EAGAIN if not (or -ECANCELED if cancelled work is + * attempted queued) */ int slow_work_enqueue(struct slow_work *work) { + wait_queue_head_t *wfo_wq; + struct list_head *queue; unsigned long flags; + int ret; + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + return -ECANCELED; BUG_ON(slow_work_user_count <= 0); BUG_ON(!work); BUG_ON(!work->ops); - BUG_ON(!work->ops->get_ref); /* when honouring an enqueue request, we only promise that we will run * the work function in the future; we do not promise to run it once @@ -280,8 +419,19 @@ int slow_work_enqueue(struct slow_work *work) * maintaining our promise */ if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + spin_lock_irqsave(&slow_work_queue_lock, flags); + if (unlikely(test_bit(SLOW_WORK_CANCELLING, &work->flags))) + goto cancelled; + /* we promise that we will not attempt to execute the work * function in more than one thread simultaneously * @@ -299,25 +449,221 @@ int slow_work_enqueue(struct slow_work *work) if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); } else { - if (work->ops->get_ref(work) < 0) - goto cant_get_ref; - if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) - list_add_tail(&work->link, &vslow_work_queue); - else - list_add_tail(&work->link, &slow_work_queue); + ret = slow_work_get_ref(work); + if (ret < 0) + goto failed; + slow_work_mark_time(work); + list_add_tail(&work->link, queue); wake_up(&slow_work_thread_wq); + + /* if someone who could be requeued is sleeping on a + * thread, then ask them to yield their thread */ + if (work->link.prev == queue) + wake_up(wfo_wq); } spin_unlock_irqrestore(&slow_work_queue_lock, flags); } return 0; -cant_get_ref: +cancelled: + ret = -ECANCELED; +failed: spin_unlock_irqrestore(&slow_work_queue_lock, flags); - return -EAGAIN; + return ret; } EXPORT_SYMBOL(slow_work_enqueue); +static int slow_work_wait(void *word) +{ + schedule(); + return 0; +} + +/** + * slow_work_cancel - Cancel a slow work item + * @work: The work item to cancel + * + * This function will cancel a previously enqueued work item. If we cannot + * cancel the work item, it is guarenteed to have run when this function + * returns. + */ +void slow_work_cancel(struct slow_work *work) +{ + bool wait = true, put = false; + + set_bit(SLOW_WORK_CANCELLING, &work->flags); + smp_mb(); + + /* if the work item is a delayed work item with an active timer, we + * need to wait for the timer to finish _before_ getting the spinlock, + * lest we deadlock against the timer routine + * + * the timer routine will leave DELAYED set if it notices the + * CANCELLING flag in time + */ + if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { + struct delayed_slow_work *dwork = + container_of(work, struct delayed_slow_work, work); + del_timer_sync(&dwork->timer); + } + + spin_lock_irq(&slow_work_queue_lock); + + if (test_bit(SLOW_WORK_DELAYED, &work->flags)) { + /* the timer routine aborted or never happened, so we are left + * holding the timer's reference on the item and should just + * drop the pending flag and wait for any ongoing execution to + * finish */ + struct delayed_slow_work *dwork = + container_of(work, struct delayed_slow_work, work); + + BUG_ON(timer_pending(&dwork->timer)); + BUG_ON(!list_empty(&work->link)); + + clear_bit(SLOW_WORK_DELAYED, &work->flags); + put = true; + clear_bit(SLOW_WORK_PENDING, &work->flags); + + } else if (test_bit(SLOW_WORK_PENDING, &work->flags) && + !list_empty(&work->link)) { + /* the link in the pending queue holds a reference on the item + * that we will need to release */ + list_del_init(&work->link); + wait = false; + put = true; + clear_bit(SLOW_WORK_PENDING, &work->flags); + + } else if (test_and_clear_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags)) { + /* the executor is holding our only reference on the item, so + * we merely need to wait for it to finish executing */ + clear_bit(SLOW_WORK_PENDING, &work->flags); + } + + spin_unlock_irq(&slow_work_queue_lock); + + /* the EXECUTING flag is set by the executor whilst the spinlock is set + * and before the item is dequeued - so assuming the above doesn't + * actually dequeue it, simply waiting for the EXECUTING flag to be + * released here should be sufficient */ + if (wait) + wait_on_bit(&work->flags, SLOW_WORK_EXECUTING, slow_work_wait, + TASK_UNINTERRUPTIBLE); + + clear_bit(SLOW_WORK_CANCELLING, &work->flags); + if (put) + slow_work_put_ref(work); +} +EXPORT_SYMBOL(slow_work_cancel); + +/* + * Handle expiry of the delay timer, indicating that a delayed slow work item + * should now be queued if not cancelled + */ +static void delayed_slow_work_timer(unsigned long data) +{ + wait_queue_head_t *wfo_wq; + struct list_head *queue; + struct slow_work *work = (struct slow_work *) data; + unsigned long flags; + bool queued = false, put = false, first = false; + + if (test_bit(SLOW_WORK_VERY_SLOW, &work->flags)) { + wfo_wq = &vslow_work_queue_waits_for_occupation; + queue = &vslow_work_queue; + } else { + wfo_wq = &slow_work_queue_waits_for_occupation; + queue = &slow_work_queue; + } + + spin_lock_irqsave(&slow_work_queue_lock, flags); + if (likely(!test_bit(SLOW_WORK_CANCELLING, &work->flags))) { + clear_bit(SLOW_WORK_DELAYED, &work->flags); + + if (test_bit(SLOW_WORK_EXECUTING, &work->flags)) { + /* we discard the reference the timer was holding in + * favour of the one the executor holds */ + set_bit(SLOW_WORK_ENQ_DEFERRED, &work->flags); + put = true; + } else { + slow_work_mark_time(work); + list_add_tail(&work->link, queue); + queued = true; + if (work->link.prev == queue) + first = true; + } + } + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + if (put) + slow_work_put_ref(work); + if (first) + wake_up(wfo_wq); + if (queued) + wake_up(&slow_work_thread_wq); +} + +/** + * delayed_slow_work_enqueue - Schedule a delayed slow work item for processing + * @dwork: The delayed work item to queue + * @delay: When to start executing the work, in jiffies from now + * + * This is similar to slow_work_enqueue(), but it adds a delay before the work + * is actually queued for processing. + * + * The item can have delayed processing requested on it whilst it is being + * executed. The delay will begin immediately, and if it expires before the + * item finishes executing, the item will be placed back on the queue when it + * has done executing. + */ +int delayed_slow_work_enqueue(struct delayed_slow_work *dwork, + unsigned long delay) +{ + struct slow_work *work = &dwork->work; + unsigned long flags; + int ret; + + if (delay == 0) + return slow_work_enqueue(&dwork->work); + + BUG_ON(slow_work_user_count <= 0); + BUG_ON(!work); + BUG_ON(!work->ops); + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + return -ECANCELED; + + if (!test_and_set_bit_lock(SLOW_WORK_PENDING, &work->flags)) { + spin_lock_irqsave(&slow_work_queue_lock, flags); + + if (test_bit(SLOW_WORK_CANCELLING, &work->flags)) + goto cancelled; + + /* the timer holds a reference whilst it is pending */ + ret = work->ops->get_ref(work); + if (ret < 0) + goto cant_get_ref; + + if (test_and_set_bit(SLOW_WORK_DELAYED, &work->flags)) + BUG(); + dwork->timer.expires = jiffies + delay; + dwork->timer.data = (unsigned long) work; + dwork->timer.function = delayed_slow_work_timer; + add_timer(&dwork->timer); + + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + } + + return 0; + +cancelled: + ret = -ECANCELED; +cant_get_ref: + spin_unlock_irqrestore(&slow_work_queue_lock, flags); + return ret; +} +EXPORT_SYMBOL(delayed_slow_work_enqueue); + /* * Schedule a cull of the thread pool at some time in the near future */ @@ -368,13 +714,23 @@ static inline bool slow_work_available(int vsmax) */ static int slow_work_thread(void *_data) { - int vsmax; + int vsmax, id; DEFINE_WAIT(wait); set_freezable(); set_user_nice(current, -5); + /* allocate ourselves an ID */ + spin_lock_irq(&slow_work_queue_lock); + id = find_first_zero_bit(slow_work_ids, SLOW_WORK_THREAD_LIMIT); + BUG_ON(id < 0 || id >= SLOW_WORK_THREAD_LIMIT); + __set_bit(id, slow_work_ids); + slow_work_set_thread_pid(id, current->pid); + spin_unlock_irq(&slow_work_queue_lock); + + sprintf(current->comm, "kslowd%03u", id); + for (;;) { vsmax = vslow_work_proportion; vsmax *= atomic_read(&slow_work_thread_count); @@ -395,7 +751,7 @@ static int slow_work_thread(void *_data) vsmax *= atomic_read(&slow_work_thread_count); vsmax /= 100; - if (slow_work_available(vsmax) && slow_work_execute()) { + if (slow_work_available(vsmax) && slow_work_execute(id)) { cond_resched(); if (list_empty(&slow_work_queue) && list_empty(&vslow_work_queue) && @@ -412,6 +768,11 @@ static int slow_work_thread(void *_data) break; } + spin_lock_irq(&slow_work_queue_lock); + slow_work_set_thread_pid(id, 0); + __clear_bit(id, slow_work_ids); + spin_unlock_irq(&slow_work_queue_lock); + if (atomic_dec_and_test(&slow_work_thread_count)) complete_and_exit(&slow_work_last_thread_exited, 0); return 0; @@ -427,21 +788,6 @@ static void slow_work_cull_timeout(unsigned long data) } /* - * Get a reference on slow work thread starter - */ -static int slow_work_new_thread_get_ref(struct slow_work *work) -{ - return 0; -} - -/* - * Drop a reference on slow work thread starter - */ -static void slow_work_new_thread_put_ref(struct slow_work *work) -{ -} - -/* * Start a new slow work thread */ static void slow_work_new_thread_execute(struct slow_work *work) @@ -475,9 +821,11 @@ static void slow_work_new_thread_execute(struct slow_work *work) } static const struct slow_work_ops slow_work_new_thread_ops = { - .get_ref = slow_work_new_thread_get_ref, - .put_ref = slow_work_new_thread_put_ref, + .owner = THIS_MODULE, .execute = slow_work_new_thread_execute, +#ifdef CONFIG_SLOW_WORK_DEBUG + .desc = slow_work_new_thread_desc, +#endif }; /* @@ -546,12 +894,13 @@ static int slow_work_max_threads_sysctl(struct ctl_table *table, int write, /** * slow_work_register_user - Register a user of the facility + * @module: The module about to make use of the facility * * Register a user of the facility, starting up the initial threads if there * aren't any other users at this point. This will return 0 if successful, or * an error if not. */ -int slow_work_register_user(void) +int slow_work_register_user(struct module *module) { struct task_struct *p; int loop; @@ -598,14 +947,81 @@ error: } EXPORT_SYMBOL(slow_work_register_user); +/* + * wait for all outstanding items from the calling module to complete + * - note that more items may be queued whilst we're waiting + */ +static void slow_work_wait_for_items(struct module *module) +{ +#ifdef CONFIG_MODULES + DECLARE_WAITQUEUE(myself, current); + struct slow_work *work; + int loop; + + mutex_lock(&slow_work_unreg_sync_lock); + add_wait_queue(&slow_work_unreg_wq, &myself); + + for (;;) { + spin_lock_irq(&slow_work_queue_lock); + + /* first of all, we wait for the last queued item in each list + * to be processed */ + list_for_each_entry_reverse(work, &vslow_work_queue, link) { + if (work->owner == module) { + set_current_state(TASK_UNINTERRUPTIBLE); + slow_work_unreg_work_item = work; + goto do_wait; + } + } + list_for_each_entry_reverse(work, &slow_work_queue, link) { + if (work->owner == module) { + set_current_state(TASK_UNINTERRUPTIBLE); + slow_work_unreg_work_item = work; + goto do_wait; + } + } + + /* then we wait for the items being processed to finish */ + slow_work_unreg_module = module; + smp_mb(); + for (loop = 0; loop < SLOW_WORK_THREAD_LIMIT; loop++) { + if (slow_work_thread_processing[loop] == module) + goto do_wait; + } + spin_unlock_irq(&slow_work_queue_lock); + break; /* okay, we're done */ + + do_wait: + spin_unlock_irq(&slow_work_queue_lock); + schedule(); + slow_work_unreg_work_item = NULL; + slow_work_unreg_module = NULL; + } + + remove_wait_queue(&slow_work_unreg_wq, &myself); + mutex_unlock(&slow_work_unreg_sync_lock); +#endif /* CONFIG_MODULES */ +} + /** * slow_work_unregister_user - Unregister a user of the facility + * @module: The module whose items should be cleared * * Unregister a user of the facility, killing all the threads if this was the * last one. + * + * This waits for all the work items belonging to the nominated module to go + * away before proceeding. */ -void slow_work_unregister_user(void) +void slow_work_unregister_user(struct module *module) { + /* first of all, wait for all outstanding items from the calling module + * to complete */ + if (module) + slow_work_wait_for_items(module); + + /* then we can actually go about shutting down the facility if need + * be */ mutex_lock(&slow_work_user_lock); BUG_ON(slow_work_user_count <= 0); @@ -639,6 +1055,16 @@ static int __init init_slow_work(void) if (slow_work_max_max_threads < nr_cpus * 2) slow_work_max_max_threads = nr_cpus * 2; #endif +#ifdef CONFIG_SLOW_WORK_DEBUG + { + struct dentry *dbdir; + + dbdir = debugfs_create_dir("slow_work", NULL); + if (dbdir && !IS_ERR(dbdir)) + debugfs_create_file("runqueue", S_IFREG | 0400, dbdir, + NULL, &slow_work_runqueue_fops); + } +#endif return 0; } diff --git a/kernel/slow-work.h b/kernel/slow-work.h new file mode 100644 index 0000000..321f3c5 --- /dev/null +++ b/kernel/slow-work.h @@ -0,0 +1,72 @@ +/* Slow work private definitions + * + * Copyright (C) 2009 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public Licence + * as published by the Free Software Foundation; either version + * 2 of the Licence, or (at your option) any later version. + */ + +#define SLOW_WORK_CULL_TIMEOUT (5 * HZ) /* cull threads 5s after running out of + * things to do */ +#define SLOW_WORK_OOM_TIMEOUT (5 * HZ) /* can't start new threads for 5s after + * OOM */ + +#define SLOW_WORK_THREAD_LIMIT 255 /* abs maximum number of slow-work threads */ + +/* + * slow-work.c + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +extern struct slow_work *slow_work_execs[]; +extern pid_t slow_work_pids[]; +extern rwlock_t slow_work_execs_lock; +#endif + +extern struct list_head slow_work_queue; +extern struct list_head vslow_work_queue; +extern spinlock_t slow_work_queue_lock; + +/* + * slow-work-debugfs.c + */ +#ifdef CONFIG_SLOW_WORK_DEBUG +extern const struct file_operations slow_work_runqueue_fops; + +extern void slow_work_new_thread_desc(struct slow_work *, struct seq_file *); +#endif + +/* + * Helper functions + */ +static inline void slow_work_set_thread_pid(int id, pid_t pid) +{ +#ifdef CONFIG_SLOW_WORK_PROC + slow_work_pids[id] = pid; +#endif +} + +static inline void slow_work_mark_time(struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + work->mark = CURRENT_TIME; +#endif +} + +static inline void slow_work_begin_exec(int id, struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + slow_work_execs[id] = work; +#endif +} + +static inline void slow_work_end_exec(int id, struct slow_work *work) +{ +#ifdef CONFIG_SLOW_WORK_PROC + write_lock(&slow_work_execs_lock); + slow_work_execs[id] = NULL; + write_unlock(&slow_work_execs_lock); +#endif +} diff --git a/kernel/sys.c b/kernel/sys.c index 255475d..ce17760 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1110,6 +1110,8 @@ SYSCALL_DEFINE0(setsid) err = session; out: write_unlock_irq(&tasklist_lock); + if (err > 0) + proc_sid_connector(group_leader); return err; } @@ -1546,24 +1548,37 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, if (arg4 | arg5) return -EINVAL; switch (arg2) { - case 0: + case PR_MCE_KILL_CLEAR: if (arg3 != 0) return -EINVAL; current->flags &= ~PF_MCE_PROCESS; break; - case 1: + case PR_MCE_KILL_SET: current->flags |= PF_MCE_PROCESS; - if (arg3 != 0) + if (arg3 == PR_MCE_KILL_EARLY) current->flags |= PF_MCE_EARLY; - else + else if (arg3 == PR_MCE_KILL_LATE) current->flags &= ~PF_MCE_EARLY; + else if (arg3 == PR_MCE_KILL_DEFAULT) + current->flags &= + ~(PF_MCE_EARLY|PF_MCE_PROCESS); + else + return -EINVAL; break; default: return -EINVAL; } error = 0; break; - + case PR_MCE_KILL_GET: + if (arg2 | arg3 | arg4 | arg5) + return -EINVAL; + if (current->flags & PF_MCE_PROCESS) + error = (current->flags & PF_MCE_EARLY) ? + PR_MCE_KILL_EARLY : PR_MCE_KILL_LATE; + else + error = PR_MCE_KILL_DEFAULT; + break; default: error = -EINVAL; break; diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c index b38423c..b6e7aaea 100644 --- a/kernel/sysctl_check.c +++ b/kernel/sysctl_check.c @@ -1521,7 +1521,7 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table) if (!table->ctl_name && table->strategy) set_fail(&fail, table, "Strategy without ctl_name"); #endif -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL if (table->procname && !table->proc_handler) set_fail(&fail, table, "No proc_handler"); #endif diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index e0f59a2..89aed59 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -231,6 +231,13 @@ void tick_nohz_stop_sched_tick(int inidle) if (!inidle && !ts->inidle) goto end; + /* + * Set ts->inidle unconditionally. Even if the system did not + * switch to NOHZ mode the cpu frequency governers rely on the + * update of the idle time accounting in tick_nohz_start_idle(). + */ + ts->inidle = 1; + now = tick_nohz_start_idle(ts); /* @@ -248,8 +255,6 @@ void tick_nohz_stop_sched_tick(int inidle) if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE)) goto end; - ts->inidle = 1; - if (need_resched()) goto end; diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index fb0f46f..c3a4e29 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -13,6 +13,7 @@ #include <linux/percpu.h> #include <linux/init.h> #include <linux/mm.h> +#include <linux/sched.h> #include <linux/sysdev.h> #include <linux/clocksource.h> #include <linux/jiffies.h> diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index fddd69d..1b5b7aa 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -275,7 +275,7 @@ static int timer_list_open(struct inode *inode, struct file *filp) return single_open(filp, timer_list_show, NULL); } -static struct file_operations timer_list_fops = { +static const struct file_operations timer_list_fops = { .open = timer_list_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c index 4cde8b9..ee5681f 100644 --- a/kernel/time/timer_stats.c +++ b/kernel/time/timer_stats.c @@ -395,7 +395,7 @@ static int tstats_open(struct inode *inode, struct file *filp) return single_open(filp, tstats_show, NULL); } -static struct file_operations tstats_fops = { +static const struct file_operations tstats_fops = { .open = tstats_open, .read = seq_read, .write = tstats_write, diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 3eb159c..d9d6206 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -856,6 +856,37 @@ static void blk_add_trace_remap(struct request_queue *q, struct bio *bio, } /** + * blk_add_trace_rq_remap - Add a trace for a request-remap operation + * @q: queue the io is for + * @rq: the source request + * @dev: target device + * @from: source sector + * + * Description: + * Device mapper remaps request to other devices. + * Add a trace for that action. + * + **/ +static void blk_add_trace_rq_remap(struct request_queue *q, + struct request *rq, dev_t dev, + sector_t from) +{ + struct blk_trace *bt = q->blk_trace; + struct blk_io_trace_remap r; + + if (likely(!bt)) + return; + + r.device_from = cpu_to_be32(dev); + r.device_to = cpu_to_be32(disk_devt(rq->rq_disk)); + r.sector_from = cpu_to_be64(from); + + __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), + rq_data_dir(rq), BLK_TA_REMAP, !!rq->errors, + sizeof(r), &r); +} + +/** * blk_add_driver_data - Add binary message with driver-specific data * @q: queue the io is for * @rq: io request @@ -922,10 +953,13 @@ static void blk_register_tracepoints(void) WARN_ON(ret); ret = register_trace_block_remap(blk_add_trace_remap); WARN_ON(ret); + ret = register_trace_block_rq_remap(blk_add_trace_rq_remap); + WARN_ON(ret); } static void blk_unregister_tracepoints(void) { + unregister_trace_block_rq_remap(blk_add_trace_rq_remap); unregister_trace_block_remap(blk_add_trace_remap); unregister_trace_block_split(blk_add_trace_split); unregister_trace_block_unplug_io(blk_add_trace_unplug_io); @@ -1657,6 +1691,11 @@ int blk_trace_init_sysfs(struct device *dev) return sysfs_create_group(&dev->kobj, &blk_trace_attr_group); } +void blk_trace_remove_sysfs(struct device *dev) +{ + sysfs_remove_group(&dev->kobj, &blk_trace_attr_group); +} + #endif /* CONFIG_BLK_DEV_IO_TRACE */ #ifdef CONFIG_EVENT_TRACING diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 46592fe..6dc4e5e 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -225,7 +225,11 @@ static void ftrace_update_pid_func(void) if (ftrace_trace_function == ftrace_stub) return; +#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST func = ftrace_trace_function; +#else + func = __ftrace_trace_function; +#endif if (ftrace_pid_trace) { set_ftrace_pid_function(func); @@ -736,7 +740,7 @@ ftrace_profile_write(struct file *filp, const char __user *ubuf, out: mutex_unlock(&ftrace_profile_lock); - filp->f_pos += cnt; + *ppos += cnt; return cnt; } @@ -1074,14 +1078,9 @@ static void ftrace_replace_code(int enable) failed = __ftrace_replace_code(rec, enable); if (failed) { rec->flags |= FTRACE_FL_FAILED; - if ((system_state == SYSTEM_BOOTING) || - !core_kernel_text(rec->ip)) { - ftrace_free_rec(rec); - } else { - ftrace_bug(failed, rec->ip); - /* Stop processing */ - return; - } + ftrace_bug(failed, rec->ip); + /* Stop processing */ + return; } } while_for_each_ftrace_rec(); } @@ -2223,15 +2222,15 @@ ftrace_regex_write(struct file *file, const char __user *ubuf, ret = ftrace_process_regex(parser->buffer, parser->idx, enable); if (ret) - goto out; + goto out_unlock; trace_parser_clear(parser); } ret = read; - +out_unlock: mutex_unlock(&ftrace_regex_lock); -out: + return ret; } @@ -2658,19 +2657,17 @@ static int ftrace_convert_nops(struct module *mod, } #ifdef CONFIG_MODULES -void ftrace_release(void *start, void *end) +void ftrace_release_mod(struct module *mod) { struct dyn_ftrace *rec; struct ftrace_page *pg; - unsigned long s = (unsigned long)start; - unsigned long e = (unsigned long)end; - if (ftrace_disabled || !start || start == end) + if (ftrace_disabled) return; mutex_lock(&ftrace_lock); do_for_each_ftrace_rec(pg, rec) { - if ((rec->ip >= s) && (rec->ip < e)) { + if (within_module_core(rec->ip, mod)) { /* * rec->ip is changed in ftrace_free_rec() * It should not between s and e if record was freed. @@ -2702,9 +2699,7 @@ static int ftrace_module_notify(struct notifier_block *self, mod->num_ftrace_callsites); break; case MODULE_STATE_GOING: - ftrace_release(mod->ftrace_callsites, - mod->ftrace_callsites + - mod->num_ftrace_callsites); + ftrace_release_mod(mod); break; } diff --git a/kernel/trace/kmemtrace.c b/kernel/trace/kmemtrace.c index 81b1645..a91da69 100644 --- a/kernel/trace/kmemtrace.c +++ b/kernel/trace/kmemtrace.c @@ -501,7 +501,7 @@ static int __init init_kmem_tracer(void) return 1; } - if (!register_tracer(&kmem_tracer)) { + if (register_tracer(&kmem_tracer) != 0) { pr_warning("Warning: could not register the kmem tracer\n"); return 1; } diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index d4ff019..5dd017f 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -483,7 +483,7 @@ struct ring_buffer_iter { /* Up this if you want to test the TIME_EXTENTS and normalization */ #define DEBUG_SHIFT 0 -static inline u64 rb_time_stamp(struct ring_buffer *buffer, int cpu) +static inline u64 rb_time_stamp(struct ring_buffer *buffer) { /* shift to debug/test normalization and TIME_EXTENTS */ return buffer->clock() << DEBUG_SHIFT; @@ -494,7 +494,7 @@ u64 ring_buffer_time_stamp(struct ring_buffer *buffer, int cpu) u64 time; preempt_disable_notrace(); - time = rb_time_stamp(buffer, cpu); + time = rb_time_stamp(buffer); preempt_enable_no_resched_notrace(); return time; @@ -599,7 +599,7 @@ static struct list_head *rb_list_head(struct list_head *list) } /* - * rb_is_head_page - test if the give page is the head page + * rb_is_head_page - test if the given page is the head page * * Because the reader may move the head_page pointer, we can * not trust what the head page is (it may be pointing to @@ -1193,6 +1193,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); + spin_lock_irq(&cpu_buffer->reader_lock); rb_head_page_deactivate(cpu_buffer); for (i = 0; i < nr_pages; i++) { @@ -1207,6 +1208,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) return; rb_reset_cpu(cpu_buffer); + spin_unlock_irq(&cpu_buffer->reader_lock); rb_check_pages(cpu_buffer); @@ -1868,7 +1870,7 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, * Nested commits always have zero deltas, so * just reread the time stamp */ - *ts = rb_time_stamp(buffer, cpu_buffer->cpu); + *ts = rb_time_stamp(buffer); next_page->page->time_stamp = *ts; } @@ -2111,7 +2113,7 @@ rb_reserve_next_event(struct ring_buffer *buffer, if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000)) goto out_fail; - ts = rb_time_stamp(cpu_buffer->buffer, cpu_buffer->cpu); + ts = rb_time_stamp(cpu_buffer->buffer); /* * Only the first commit can update the timestamp. @@ -2681,7 +2683,7 @@ unsigned long ring_buffer_entries(struct ring_buffer *buffer) EXPORT_SYMBOL_GPL(ring_buffer_entries); /** - * ring_buffer_overrun_cpu - get the number of overruns in buffer + * ring_buffer_overruns - get the number of overruns in buffer * @buffer: The ring buffer * * Returns the total number of overruns in the ring buffer diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 4506826..b20d3ec 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1393,7 +1393,7 @@ int trace_array_vprintk(struct trace_array *tr, int trace_vprintk(unsigned long ip, const char *fmt, va_list args) { - return trace_array_printk(&global_trace, ip, fmt, args); + return trace_array_vprintk(&global_trace, ip, fmt, args); } EXPORT_SYMBOL_GPL(trace_vprintk); @@ -2440,7 +2440,7 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, return ret; } - filp->f_pos += cnt; + *ppos += cnt; return cnt; } @@ -2582,7 +2582,7 @@ tracing_ctrl_write(struct file *filp, const char __user *ubuf, } mutex_unlock(&trace_types_lock); - filp->f_pos += cnt; + *ppos += cnt; return cnt; } @@ -2764,7 +2764,7 @@ tracing_set_trace_write(struct file *filp, const char __user *ubuf, if (err) return err; - filp->f_pos += ret; + *ppos += ret; return ret; } @@ -3299,7 +3299,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, } } - filp->f_pos += cnt; + *ppos += cnt; /* If check pages failed, return ENOMEM */ if (tracing_disabled) diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index 7a7a9fd..4a194f0 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -34,6 +34,7 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) struct trace_array *tr = branch_tracer; struct ring_buffer_event *event; struct trace_branch *entry; + struct ring_buffer *buffer; unsigned long flags; int cpu, pc; const char *p; @@ -54,7 +55,8 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) goto out; pc = preempt_count(); - event = trace_buffer_lock_reserve(tr, TRACE_BRANCH, + buffer = tr->buffer; + event = trace_buffer_lock_reserve(buffer, TRACE_BRANCH, sizeof(*entry), flags, pc); if (!event) goto out; @@ -74,8 +76,8 @@ probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) entry->line = f->line; entry->correct = val == expect; - if (!filter_check_discard(call, entry, tr->buffer, event)) - ring_buffer_unlock_commit(tr->buffer, event); + if (!filter_check_discard(call, entry, buffer, event)) + ring_buffer_unlock_commit(buffer, event); out: atomic_dec(&tr->data[cpu]->disabled); diff --git a/kernel/trace/trace_event_profile.c b/kernel/trace/trace_event_profile.c index dd44b87..8d5c171 100644 --- a/kernel/trace/trace_event_profile.c +++ b/kernel/trace/trace_event_profile.c @@ -31,7 +31,7 @@ static int ftrace_profile_enable_event(struct ftrace_event_call *event) if (atomic_inc_return(&event->profile_count)) return 0; - if (!total_profile_count++) { + if (!total_profile_count) { buf = (char *)alloc_percpu(profile_buf_t); if (!buf) goto fail_buf; @@ -46,14 +46,19 @@ static int ftrace_profile_enable_event(struct ftrace_event_call *event) } ret = event->profile_enable(); - if (!ret) + if (!ret) { + total_profile_count++; return 0; + } - kfree(trace_profile_buf_nmi); fail_buf_nmi: - kfree(trace_profile_buf); + if (!total_profile_count) { + free_percpu(trace_profile_buf_nmi); + free_percpu(trace_profile_buf); + trace_profile_buf_nmi = NULL; + trace_profile_buf = NULL; + } fail_buf: - total_profile_count--; atomic_dec(&event->profile_count); return ret; diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 2324578..98a6cc5 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -933,8 +933,9 @@ static void postfix_clear(struct filter_parse_state *ps) while (!list_empty(&ps->postfix)) { elt = list_first_entry(&ps->postfix, struct postfix_elt, list); - kfree(elt->operand); list_del(&elt->list); + kfree(elt->operand); + kfree(elt); } } diff --git a/kernel/trace/trace_hw_branches.c b/kernel/trace/trace_hw_branches.c index 23b6385..69543a9 100644 --- a/kernel/trace/trace_hw_branches.c +++ b/kernel/trace/trace_hw_branches.c @@ -165,6 +165,7 @@ void trace_hw_branch(u64 from, u64 to) struct ftrace_event_call *call = &event_hw_branch; struct trace_array *tr = hw_branch_trace; struct ring_buffer_event *event; + struct ring_buffer *buf; struct hw_branch_entry *entry; unsigned long irq1; int cpu; @@ -180,7 +181,8 @@ void trace_hw_branch(u64 from, u64 to) if (atomic_inc_return(&tr->data[cpu]->disabled) != 1) goto out; - event = trace_buffer_lock_reserve(tr, TRACE_HW_BRANCHES, + buf = tr->buffer; + event = trace_buffer_lock_reserve(buf, TRACE_HW_BRANCHES, sizeof(*entry), 0, 0); if (!event) goto out; @@ -189,8 +191,8 @@ void trace_hw_branch(u64 from, u64 to) entry->ent.type = TRACE_HW_BRANCHES; entry->from = from; entry->to = to; - if (!filter_check_discard(call, entry, tr->buffer, event)) - trace_buffer_unlock_commit(tr, event, 0, 0); + if (!filter_check_discard(call, entry, buf, event)) + trace_buffer_unlock_commit(buf, event, 0, 0); out: atomic_dec(&tr->data[cpu]->disabled); diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index f572f44..b6c12c6 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -69,6 +69,9 @@ enum print_line_t trace_print_printk_msg_only(struct trace_iterator *iter) * @s: trace sequence descriptor * @fmt: printf format string * + * It returns 0 if the trace oversizes the buffer's free + * space, 1 otherwise. + * * The tracer may use either sequence operations or its own * copy to user routines. To simplify formating of a trace * trace_seq_printf is used to store strings into a special @@ -95,7 +98,7 @@ trace_seq_printf(struct trace_seq *s, const char *fmt, ...) s->len += ret; - return len; + return 1; } EXPORT_SYMBOL_GPL(trace_seq_printf); @@ -486,16 +489,18 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry) hardirq ? 'h' : softirq ? 's' : '.')) return 0; - if (entry->lock_depth < 0) - ret = trace_seq_putc(s, '.'); + if (entry->preempt_count) + ret = trace_seq_printf(s, "%x", entry->preempt_count); else - ret = trace_seq_printf(s, "%d", entry->lock_depth); + ret = trace_seq_putc(s, '.'); + if (!ret) return 0; - if (entry->preempt_count) - return trace_seq_printf(s, "%x", entry->preempt_count); - return trace_seq_putc(s, '.'); + if (entry->lock_depth < 0) + return trace_seq_putc(s, '.'); + + return trace_seq_printf(s, "%d", entry->lock_depth); } static int @@ -883,7 +888,7 @@ static int trace_ctxwake_raw(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - task_state_char(field->prev_state); + S = task_state_char(field->prev_state); T = task_state_char(field->next_state); if (!trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n", field->prev_pid, @@ -918,7 +923,7 @@ static int trace_ctxwake_hex(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - task_state_char(field->prev_state); + S = task_state_char(field->prev_state); T = task_state_char(field->next_state); SEQ_PUT_HEX_FIELD_RET(s, field->prev_pid); diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 9fbce6c..527e17e 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -166,7 +166,7 @@ int syscall_exit_format(struct ftrace_event_call *call, struct trace_seq *s) "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n" "\tfield:%s %s;\toffset:%zu;\tsize:%zu;\n", SYSCALL_FIELD(int, nr), - SYSCALL_FIELD(unsigned long, ret)); + SYSCALL_FIELD(long, ret)); if (!ret) return 0; @@ -212,7 +212,7 @@ int syscall_exit_define_fields(struct ftrace_event_call *call) if (ret) return ret; - ret = trace_define_field(call, SYSCALL_FIELD(unsigned long, ret), 0, + ret = trace_define_field(call, SYSCALL_FIELD(long, ret), 0, FILTER_OTHER); return ret; diff --git a/kernel/user.c b/kernel/user.c index 2c000e7..46d0165 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -330,9 +330,9 @@ done: */ static void free_user(struct user_struct *up, unsigned long flags) { - spin_unlock_irqrestore(&uidhash_lock, flags); INIT_DELAYED_WORK(&up->work, cleanup_user_struct); schedule_delayed_work(&up->work, msecs_to_jiffies(1000)); + spin_unlock_irqrestore(&uidhash_lock, flags); } #else /* CONFIG_USER_SCHED && CONFIG_SYSFS */ diff --git a/kernel/workqueue.c b/kernel/workqueue.c index addfe2d..67e526b 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -640,6 +640,24 @@ int schedule_delayed_work(struct delayed_work *dwork, EXPORT_SYMBOL(schedule_delayed_work); /** + * flush_delayed_work - block until a dwork_struct's callback has terminated + * @dwork: the delayed work which is to be flushed + * + * Any timeout is cancelled, and any pending work is run immediately. + */ +void flush_delayed_work(struct delayed_work *dwork) +{ + if (del_timer_sync(&dwork->timer)) { + struct cpu_workqueue_struct *cwq; + cwq = wq_per_cpu(keventd_wq, get_cpu()); + __queue_work(cwq, &dwork->work); + put_cpu(); + } + flush_work(&dwork->work); +} +EXPORT_SYMBOL(flush_delayed_work); + +/** * schedule_delayed_work_on - queue work in global workqueue on CPU after delay * @cpu: cpu to use * @dwork: job to be done @@ -667,6 +685,7 @@ EXPORT_SYMBOL(schedule_delayed_work_on); int schedule_on_each_cpu(work_func_t func) { int cpu; + int orig = -1; struct work_struct *works; works = alloc_percpu(struct work_struct); @@ -674,14 +693,28 @@ int schedule_on_each_cpu(work_func_t func) return -ENOMEM; get_online_cpus(); + + /* + * When running in keventd don't schedule a work item on + * itself. Can just call directly because the work queue is + * already bound. This also is faster. + */ + if (current_is_keventd()) + orig = raw_smp_processor_id(); + for_each_online_cpu(cpu) { struct work_struct *work = per_cpu_ptr(works, cpu); INIT_WORK(work, func); - schedule_work_on(cpu, work); + if (cpu != orig) + schedule_work_on(cpu, work); } + if (orig >= 0) + func(per_cpu_ptr(works, orig)); + for_each_online_cpu(cpu) flush_work(per_cpu_ptr(works, cpu)); + put_online_cpus(); free_percpu(works); return 0; |