diff options
Diffstat (limited to 'kernel')
68 files changed, 4116 insertions, 872 deletions
diff --git a/kernel/audit.c b/kernel/audit.c index 91e53d04b6a9..6def25f1b351 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -103,7 +103,8 @@ static int audit_rate_limit; /* Number of outstanding audit_buffers allowed. */ static int audit_backlog_limit = 64; -static int audit_backlog_wait_time = 60 * HZ; +#define AUDIT_BACKLOG_WAIT_TIME (60 * HZ) +static int audit_backlog_wait_time = AUDIT_BACKLOG_WAIT_TIME; static int audit_backlog_wait_overflow = 0; /* The identity of the user shutting down the audit system. */ @@ -613,7 +614,7 @@ static int audit_log_common_recv_msg(struct audit_buffer **ab, u16 msg_type) int rc = 0; uid_t uid = from_kuid(&init_user_ns, current_uid()); - if (!audit_enabled) { + if (!audit_enabled && msg_type != AUDIT_USER_AVC) { *ab = NULL; return rc; } @@ -659,6 +660,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) switch (msg_type) { case AUDIT_GET: + status_set.mask = 0; status_set.enabled = audit_enabled; status_set.failure = audit_failure; status_set.pid = audit_pid; @@ -670,7 +672,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) &status_set, sizeof(status_set)); break; case AUDIT_SET: - if (nlh->nlmsg_len < sizeof(struct audit_status)) + if (nlmsg_len(nlh) < sizeof(struct audit_status)) return -EINVAL; status_get = (struct audit_status *)data; if (status_get->mask & AUDIT_STATUS_ENABLED) { @@ -832,7 +834,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) memset(&s, 0, sizeof(s)); /* guard against past and future API changes */ - memcpy(&s, data, min(sizeof(s), (size_t)nlh->nlmsg_len)); + memcpy(&s, data, min_t(size_t, sizeof(s), nlmsg_len(nlh))); if ((s.enabled != 0 && s.enabled != 1) || (s.log_passwd != 0 && s.log_passwd != 1)) return -EINVAL; @@ -1117,9 +1119,10 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, sleep_time = timeout_start + audit_backlog_wait_time - jiffies; - if ((long)sleep_time > 0) + if ((long)sleep_time > 0) { wait_for_auditd(sleep_time); - continue; + continue; + } } if (audit_rate_check() && printk_ratelimit()) printk(KERN_WARNING @@ -1133,6 +1136,8 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, return NULL; } + audit_backlog_wait_time = AUDIT_BACKLOG_WAIT_TIME; + ab = audit_buffer_alloc(ctx, gfp_mask, type); if (!ab) { audit_log_lost("out of memory in audit_log_start"); @@ -1535,6 +1540,26 @@ void audit_log_name(struct audit_context *context, struct audit_names *n, } } + /* log the audit_names record type */ + audit_log_format(ab, " nametype="); + switch(n->type) { + case AUDIT_TYPE_NORMAL: + audit_log_format(ab, "NORMAL"); + break; + case AUDIT_TYPE_PARENT: + audit_log_format(ab, "PARENT"); + break; + case AUDIT_TYPE_CHILD_DELETE: + audit_log_format(ab, "DELETE"); + break; + case AUDIT_TYPE_CHILD_CREATE: + audit_log_format(ab, "CREATE"); + break; + default: + audit_log_format(ab, "UNKNOWN"); + break; + } + audit_log_fcaps(ab, n); audit_log_end(ab); } diff --git a/kernel/audit.h b/kernel/audit.h index 1c95131ef760..123c9b7c3979 100644 --- a/kernel/audit.h +++ b/kernel/audit.h @@ -85,6 +85,7 @@ struct audit_names { struct filename *name; int name_len; /* number of chars to log */ + bool hidden; /* don't log this record */ bool name_put; /* call __putname()? */ unsigned long ino; diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 6bd4a90d1991..4d230eafd5e8 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -423,7 +423,7 @@ static struct audit_entry *audit_data_to_entry(struct audit_rule_data *data, f->lsm_rule = NULL; /* Support legacy tests for a valid loginuid */ - if ((f->type == AUDIT_LOGINUID) && (f->val == 4294967295)) { + if ((f->type == AUDIT_LOGINUID) && (f->val == ~0U)) { f->type = AUDIT_LOGINUID_SET; f->val = 0; } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 3c8a601324a2..03a3af8538bd 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -733,6 +733,22 @@ static enum audit_state audit_filter_task(struct task_struct *tsk, char **key) return AUDIT_BUILD_CONTEXT; } +static int audit_in_mask(const struct audit_krule *rule, unsigned long val) +{ + int word, bit; + + if (val > 0xffffffff) + return false; + + word = AUDIT_WORD(val); + if (word >= AUDIT_BITMASK_SIZE) + return false; + + bit = AUDIT_BIT(val); + + return rule->mask[word] & bit; +} + /* At syscall entry and exit time, this filter is called if the * audit_state is not low enough that auditing cannot take place, but is * also not high enough that we already know we have to write an audit @@ -750,11 +766,8 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, rcu_read_lock(); if (!list_empty(list)) { - int word = AUDIT_WORD(ctx->major); - int bit = AUDIT_BIT(ctx->major); - list_for_each_entry_rcu(e, list, list) { - if ((e->rule.mask[word] & bit) == bit && + if (audit_in_mask(&e->rule, ctx->major) && audit_filter_rules(tsk, &e->rule, ctx, NULL, &state, false)) { rcu_read_unlock(); @@ -774,20 +787,16 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, static int audit_filter_inode_name(struct task_struct *tsk, struct audit_names *n, struct audit_context *ctx) { - int word, bit; int h = audit_hash_ino((u32)n->ino); struct list_head *list = &audit_inode_hash[h]; struct audit_entry *e; enum audit_state state; - word = AUDIT_WORD(ctx->major); - bit = AUDIT_BIT(ctx->major); - if (list_empty(list)) return 0; list_for_each_entry_rcu(e, list, list) { - if ((e->rule.mask[word] & bit) == bit && + if (audit_in_mask(&e->rule, ctx->major) && audit_filter_rules(tsk, &e->rule, ctx, n, &state, false)) { ctx->current_state = state; return 1; @@ -1399,8 +1408,11 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts } i = 0; - list_for_each_entry(n, &context->names_list, list) + list_for_each_entry(n, &context->names_list, list) { + if (n->hidden) + continue; audit_log_name(context, n, NULL, i++, &call_panic); + } /* Send end of event record to help user space know we are finished */ ab = audit_log_start(context, GFP_KERNEL, AUDIT_EOE); @@ -1769,14 +1781,15 @@ void audit_putname(struct filename *name) * __audit_inode - store the inode and device from a lookup * @name: name being audited * @dentry: dentry being audited - * @parent: does this dentry represent the parent? + * @flags: attributes for this particular entry */ void __audit_inode(struct filename *name, const struct dentry *dentry, - unsigned int parent) + unsigned int flags) { struct audit_context *context = current->audit_context; const struct inode *inode = dentry->d_inode; struct audit_names *n; + bool parent = flags & AUDIT_INODE_PARENT; if (!context->in_syscall) return; @@ -1831,6 +1844,8 @@ out: if (parent) { n->name_len = n->name ? parent_len(n->name->name) : AUDIT_NAME_FULL; n->type = AUDIT_TYPE_PARENT; + if (flags & AUDIT_INODE_HIDDEN) + n->hidden = true; } else { n->name_len = AUDIT_NAME_FULL; n->type = AUDIT_TYPE_NORMAL; diff --git a/kernel/capability.c b/kernel/capability.c index f6c2ce5701e1..d52eecc0942b 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -445,22 +445,18 @@ bool nsown_capable(int cap) } /** - * inode_capable - Check superior capability over inode + * capable_wrt_inode_uidgid - Check nsown_capable and uid and gid mapped * @inode: The inode in question * @cap: The capability in question * - * Return true if the current task has the given superior capability - * targeted at it's own user namespace and that the given inode is owned - * by the current user namespace or a child namespace. - * - * Currently we check to see if an inode is owned by the current - * user namespace by seeing if the inode's owner maps into the - * current user namespace. - * + * Return true if the current task has the given capability targeted at + * its own user namespace and that the given inode's uid and gid are + * mapped into the current user namespace. */ -bool inode_capable(const struct inode *inode, int cap) +bool capable_wrt_inode_uidgid(const struct inode *inode, int cap) { struct user_namespace *ns = current_user_ns(); - return ns_capable(ns, cap) && kuid_has_mapping(ns, inode->i_uid); + return ns_capable(ns, cap) && kuid_has_mapping(ns, inode->i_uid) && + kgid_has_mapping(ns, inode->i_gid); } diff --git a/kernel/cgroup.c b/kernel/cgroup.c index a7c9e6ddb979..d0def7fc2848 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -92,6 +92,14 @@ static DEFINE_MUTEX(cgroup_mutex); static DEFINE_MUTEX(cgroup_root_mutex); /* + * cgroup destruction makes heavy use of work items and there can be a lot + * of concurrent destructions. Use a separate workqueue so that cgroup + * destruction work items don't end up filling up max_active of system_wq + * which may lead to deadlock. + */ +static struct workqueue_struct *cgroup_destroy_wq; + +/* * Generate an array of cgroup subsystem pointers. At boot time, this is * populated with the built in subsystems, and modular subsystems are * registered after that. The mutable section of this array is protected by @@ -873,7 +881,7 @@ static void cgroup_free_rcu(struct rcu_head *head) { struct cgroup *cgrp = container_of(head, struct cgroup, rcu_head); - schedule_work(&cgrp->free_work); + queue_work(cgroup_destroy_wq, &cgrp->free_work); } static void cgroup_diput(struct dentry *dentry, struct inode *inode) @@ -1995,7 +2003,7 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, /* @tsk either already exited or can't exit until the end */ if (tsk->flags & PF_EXITING) - continue; + goto next; /* as per above, nr_threads may decrease, but not increase. */ BUG_ON(i >= group_size); @@ -2003,7 +2011,7 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, ent.cgrp = task_cgroup_from_root(tsk, root); /* nothing to do if this task is already in the cgroup */ if (ent.cgrp == cgrp) - continue; + goto next; /* * saying GFP_ATOMIC has no effect here because we did prealloc * earlier, but it's good form to communicate our expectations. @@ -2011,7 +2019,7 @@ static int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk, retval = flex_array_put(group, i, &ent, GFP_ATOMIC); BUG_ON(retval != 0); i++; - + next: if (!threadgroup) break; } while_each_thread(leader, tsk); @@ -2769,13 +2777,17 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, { LIST_HEAD(pending); struct cgroup *cgrp, *n; + struct super_block *sb = ss->root->sb; /* %NULL @cfts indicates abort and don't bother if @ss isn't attached */ - if (cfts && ss->root != &rootnode) { + if (cfts && ss->root != &rootnode && + atomic_inc_not_zero(&sb->s_active)) { list_for_each_entry(cgrp, &ss->root->allcg_list, allcg_node) { dget(cgrp->dentry); list_add_tail(&cgrp->cft_q_node, &pending); } + } else { + sb = NULL; } mutex_unlock(&cgroup_mutex); @@ -2798,6 +2810,9 @@ static void cgroup_cfts_commit(struct cgroup_subsys *ss, dput(cgrp->dentry); } + if (sb) + deactivate_super(sb); + mutex_unlock(&cgroup_cft_mutex); } @@ -3727,6 +3742,23 @@ static int cgroup_write_notify_on_release(struct cgroup *cgrp, } /* + * When dput() is called asynchronously, if umount has been done and + * then deactivate_super() in cgroup_free_fn() kills the superblock, + * there's a small window that vfs will see the root dentry with non-zero + * refcnt and trigger BUG(). + * + * That's why we hold a reference before dput() and drop it right after. + */ +static void cgroup_dput(struct cgroup *cgrp) +{ + struct super_block *sb = cgrp->root->sb; + + atomic_inc(&sb->s_active); + dput(cgrp->dentry); + deactivate_super(sb); +} + +/* * Unregister event and free resources. * * Gets called from workqueue. @@ -3746,7 +3778,7 @@ static void cgroup_event_remove(struct work_struct *work) eventfd_ctx_put(event->eventfd); kfree(event); - dput(cgrp->dentry); + cgroup_dput(cgrp); } /* @@ -4031,12 +4063,8 @@ static void css_dput_fn(struct work_struct *work) { struct cgroup_subsys_state *css = container_of(work, struct cgroup_subsys_state, dput_work); - struct dentry *dentry = css->cgroup->dentry; - struct super_block *sb = dentry->d_sb; - atomic_inc(&sb->s_active); - dput(dentry); - deactivate_super(sb); + cgroup_dput(css->cgroup); } static void init_cgroup_css(struct cgroup_subsys_state *css, @@ -4666,6 +4694,22 @@ out: return err; } +static int __init cgroup_wq_init(void) +{ + /* + * There isn't much point in executing destruction path in + * parallel. Good chunk is serialized with cgroup_mutex anyway. + * Use 1 for @max_active. + * + * We would prefer to do this in cgroup_init() above, but that + * is called before init_workqueues(): so leave this until after. + */ + cgroup_destroy_wq = alloc_workqueue("cgroup_destroy", 0, 1); + BUG_ON(!cgroup_destroy_wq); + return 0; +} +core_initcall(cgroup_wq_init); + /* * proc_cgroup_show() * - Print task's cgroup paths into seq_file, one line for each hierarchy @@ -4976,7 +5020,7 @@ void __css_put(struct cgroup_subsys_state *css) v = css_unbias_refcnt(atomic_dec_return(&css->refcnt)); if (v == 0) - schedule_work(&css->dput_work); + queue_work(cgroup_destroy_wq, &css->dput_work); } EXPORT_SYMBOL_GPL(__css_put); diff --git a/kernel/cpu.c b/kernel/cpu.c index 198a38883e64..bc255e25d5dd 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -698,10 +698,12 @@ void set_cpu_present(unsigned int cpu, bool present) void set_cpu_online(unsigned int cpu, bool online) { - if (online) + if (online) { cpumask_set_cpu(cpu, to_cpumask(cpu_online_bits)); - else + cpumask_set_cpu(cpu, to_cpumask(cpu_active_bits)); + } else { cpumask_clear_cpu(cpu, to_cpumask(cpu_online_bits)); + } } void set_cpu_active(unsigned int cpu, bool active) diff --git a/kernel/cpu/idle.c b/kernel/cpu/idle.c index e695c0a0bcb5..c261409500e4 100644 --- a/kernel/cpu/idle.c +++ b/kernel/cpu/idle.c @@ -44,7 +44,7 @@ static inline int cpu_idle_poll(void) rcu_idle_enter(); trace_cpu_idle_rcuidle(0, smp_processor_id()); local_irq_enable(); - while (!need_resched()) + while (!tif_need_resched()) cpu_relax(); trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, smp_processor_id()); rcu_idle_exit(); @@ -92,8 +92,7 @@ static void cpu_idle_loop(void) if (cpu_idle_force_poll || tick_check_broadcast_expired()) { cpu_idle_poll(); } else { - current_clr_polling(); - if (!need_resched()) { + if (!current_clr_polling_and_test()) { stop_critical_timings(); rcu_idle_enter(); arch_cpu_idle(); @@ -103,7 +102,7 @@ static void cpu_idle_loop(void) } else { local_irq_enable(); } - current_set_polling(); + __current_set_polling(); } arch_cpu_idle_exit(); } @@ -129,7 +128,7 @@ void cpu_startup_entry(enum cpuhp_state state) */ boot_init_stack_canary(); #endif - current_set_polling(); + __current_set_polling(); arch_cpu_idle_prepare(); cpu_idle_loop(); } diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 64b3f791bbe5..d9dd521ddd6b 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -984,8 +984,10 @@ static void cpuset_change_task_nodemask(struct task_struct *tsk, need_loop = task_has_mempolicy(tsk) || !nodes_intersects(*newmems, tsk->mems_allowed); - if (need_loop) + if (need_loop) { + local_irq_disable(); write_seqcount_begin(&tsk->mems_allowed_seq); + } nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); @@ -993,8 +995,10 @@ static void cpuset_change_task_nodemask(struct task_struct *tsk, mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP2); tsk->mems_allowed = *newmems; - if (need_loop) + if (need_loop) { write_seqcount_end(&tsk->mems_allowed_seq); + local_irq_enable(); + } task_unlock(tsk); } @@ -1502,11 +1506,13 @@ static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) { struct cpuset *cs = cgroup_cs(cgrp); cpuset_filetype_t type = cft->private; - int retval = -ENODEV; + int retval = 0; mutex_lock(&cpuset_mutex); - if (!is_cpuset_online(cs)) + if (!is_cpuset_online(cs)) { + retval = -ENODEV; goto out_unlock; + } switch (type) { case FILE_CPU_EXCLUSIVE: @@ -2416,9 +2422,9 @@ int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask) task_lock(current); cs = nearest_hardwall_ancestor(task_cs(current)); + allowed = node_isset(node, cs->mems_allowed); task_unlock(current); - allowed = node_isset(node, cs->mems_allowed); mutex_unlock(&callback_mutex); return allowed; } diff --git a/kernel/events/core.c b/kernel/events/core.c index b391907d5352..459b94c94721 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -165,25 +165,109 @@ int sysctl_perf_event_mlock __read_mostly = 512 + (PAGE_SIZE / 1024); /* 'free' /* * max perf event sample rate */ -#define DEFAULT_MAX_SAMPLE_RATE 100000 -int sysctl_perf_event_sample_rate __read_mostly = DEFAULT_MAX_SAMPLE_RATE; -static int max_samples_per_tick __read_mostly = - DIV_ROUND_UP(DEFAULT_MAX_SAMPLE_RATE, HZ); +#define DEFAULT_MAX_SAMPLE_RATE 100000 +#define DEFAULT_SAMPLE_PERIOD_NS (NSEC_PER_SEC / DEFAULT_MAX_SAMPLE_RATE) +#define DEFAULT_CPU_TIME_MAX_PERCENT 25 + +int sysctl_perf_event_sample_rate __read_mostly = DEFAULT_MAX_SAMPLE_RATE; + +static int max_samples_per_tick __read_mostly = DIV_ROUND_UP(DEFAULT_MAX_SAMPLE_RATE, HZ); +static int perf_sample_period_ns __read_mostly = DEFAULT_SAMPLE_PERIOD_NS; + +static atomic_t perf_sample_allowed_ns __read_mostly = + ATOMIC_INIT( DEFAULT_SAMPLE_PERIOD_NS * DEFAULT_CPU_TIME_MAX_PERCENT / 100); + +void update_perf_cpu_limits(void) +{ + u64 tmp = perf_sample_period_ns; + + tmp *= sysctl_perf_cpu_time_max_percent; + do_div(tmp, 100); + atomic_set(&perf_sample_allowed_ns, tmp); +} int perf_proc_update_handler(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - int ret = proc_dointvec(table, write, buffer, lenp, ppos); + int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (ret || !write) return ret; max_samples_per_tick = DIV_ROUND_UP(sysctl_perf_event_sample_rate, HZ); + perf_sample_period_ns = NSEC_PER_SEC / sysctl_perf_event_sample_rate; + update_perf_cpu_limits(); + + return 0; +} + +int sysctl_perf_cpu_time_max_percent __read_mostly = DEFAULT_CPU_TIME_MAX_PERCENT; + +int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, + loff_t *ppos) +{ + int ret = proc_dointvec(table, write, buffer, lenp, ppos); + + if (ret || !write) + return ret; + + update_perf_cpu_limits(); return 0; } +/* + * perf samples are done in some very critical code paths (NMIs). + * If they take too much CPU time, the system can lock up and not + * get any real work done. This will drop the sample rate when + * we detect that events are taking too long. + */ +#define NR_ACCUMULATED_SAMPLES 128 +DEFINE_PER_CPU(u64, running_sample_length); + +void perf_sample_event_took(u64 sample_len_ns) +{ + u64 avg_local_sample_len; + u64 local_samples_len; + + if (atomic_read(&perf_sample_allowed_ns) == 0) + return; + + /* decay the counter by 1 average sample */ + local_samples_len = __get_cpu_var(running_sample_length); + local_samples_len -= local_samples_len/NR_ACCUMULATED_SAMPLES; + local_samples_len += sample_len_ns; + __get_cpu_var(running_sample_length) = local_samples_len; + + /* + * note: this will be biased artifically low until we have + * seen NR_ACCUMULATED_SAMPLES. Doing it this way keeps us + * from having to maintain a count. + */ + avg_local_sample_len = local_samples_len/NR_ACCUMULATED_SAMPLES; + + if (avg_local_sample_len <= atomic_read(&perf_sample_allowed_ns)) + return; + + if (max_samples_per_tick <= 1) + return; + + max_samples_per_tick = DIV_ROUND_UP(max_samples_per_tick, 2); + sysctl_perf_event_sample_rate = max_samples_per_tick * HZ; + perf_sample_period_ns = NSEC_PER_SEC / sysctl_perf_event_sample_rate; + + printk_ratelimited(KERN_WARNING + "perf samples too long (%lld > %d), lowering " + "kernel.perf_event_max_sample_rate to %d\n", + avg_local_sample_len, + atomic_read(&perf_sample_allowed_ns), + sysctl_perf_event_sample_rate); + + update_perf_cpu_limits(); +} + static atomic64_t perf_event_id; static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx, @@ -761,8 +845,18 @@ perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags) { struct perf_event_context *ctx; - rcu_read_lock(); retry: + /* + * One of the few rules of preemptible RCU is that one cannot do + * rcu_read_unlock() while holding a scheduler (or nested) lock when + * part of the read side critical section was preemptible -- see + * rcu_read_unlock_special(). + * + * Since ctx->lock nests under rq->lock we must ensure the entire read + * side critical section is non-preemptible. + */ + preempt_disable(); + rcu_read_lock(); ctx = rcu_dereference(task->perf_event_ctxp[ctxn]); if (ctx) { /* @@ -778,6 +872,8 @@ retry: raw_spin_lock_irqsave(&ctx->lock, *flags); if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) { raw_spin_unlock_irqrestore(&ctx->lock, *flags); + rcu_read_unlock(); + preempt_enable(); goto retry; } @@ -787,6 +883,7 @@ retry: } } rcu_read_unlock(); + preempt_enable(); return ctx; } @@ -1224,6 +1321,11 @@ group_sched_out(struct perf_event *group_event, cpuctx->exclusive = 0; } +struct remove_event { + struct perf_event *event; + bool detach_group; +}; + /* * Cross CPU call to remove a performance event * @@ -1232,12 +1334,15 @@ group_sched_out(struct perf_event *group_event, */ static int __perf_remove_from_context(void *info) { - struct perf_event *event = info; + struct remove_event *re = info; + struct perf_event *event = re->event; struct perf_event_context *ctx = event->ctx; struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); raw_spin_lock(&ctx->lock); event_sched_out(event, cpuctx, ctx); + if (re->detach_group) + perf_group_detach(event); list_del_event(event, ctx); if (!ctx->nr_events && cpuctx->task_ctx == ctx) { ctx->is_active = 0; @@ -1262,10 +1367,14 @@ static int __perf_remove_from_context(void *info) * When called from perf_event_exit_task, it's OK because the * context has been detached from its task. */ -static void perf_remove_from_context(struct perf_event *event) +static void perf_remove_from_context(struct perf_event *event, bool detach_group) { struct perf_event_context *ctx = event->ctx; struct task_struct *task = ctx->task; + struct remove_event re = { + .event = event, + .detach_group = detach_group, + }; lockdep_assert_held(&ctx->mutex); @@ -1274,12 +1383,12 @@ static void perf_remove_from_context(struct perf_event *event) * Per cpu events are removed via an smp call and * the removal is always successful. */ - cpu_function_call(event->cpu, __perf_remove_from_context, event); + cpu_function_call(event->cpu, __perf_remove_from_context, &re); return; } retry: - if (!task_function_call(task, __perf_remove_from_context, event)) + if (!task_function_call(task, __perf_remove_from_context, &re)) return; raw_spin_lock_irq(&ctx->lock); @@ -1296,6 +1405,8 @@ retry: * Since the task isn't running, its safe to remove the event, us * holding the ctx->lock ensures the task won't get scheduled in. */ + if (detach_group) + perf_group_detach(event); list_del_event(event, ctx); raw_spin_unlock_irq(&ctx->lock); } @@ -1761,7 +1872,16 @@ static int __perf_event_enable(void *info) struct perf_cpu_context *cpuctx = __get_cpu_context(ctx); int err; - if (WARN_ON_ONCE(!ctx->is_active)) + /* + * There's a time window between 'ctx->is_active' check + * in perf_event_enable function and this place having: + * - IRQs on + * - ctx->lock unlocked + * + * where the task could be killed and 'ctx' deactivated + * by perf_event_exit_task. + */ + if (!ctx->is_active) return -EINVAL; raw_spin_lock(&ctx->lock); @@ -1994,9 +2114,6 @@ static void __perf_event_sync_stat(struct perf_event *event, perf_event_update_userpage(next_event); } -#define list_next_entry(pos, member) \ - list_entry(pos->member.next, typeof(*pos), member) - static void perf_event_sync_stat(struct perf_event_context *ctx, struct perf_event_context *next_ctx) { @@ -2996,10 +3113,7 @@ int perf_event_release_kernel(struct perf_event *event) * to trigger the AB-BA case. */ mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING); - raw_spin_lock_irq(&ctx->lock); - perf_group_detach(event); - raw_spin_unlock_irq(&ctx->lock); - perf_remove_from_context(event); + perf_remove_from_context(event, true); mutex_unlock(&ctx->mutex); free_event(event); @@ -5025,6 +5139,9 @@ struct swevent_htable { /* Recursion avoidance in each contexts */ int recursion[PERF_NR_CONTEXTS]; + + /* Keeps track of cpu being initialized/exited */ + bool online; }; static DEFINE_PER_CPU(struct swevent_htable, swevent_htable); @@ -5271,8 +5388,14 @@ static int perf_swevent_add(struct perf_event *event, int flags) hwc->state = !(flags & PERF_EF_START); head = find_swevent_head(swhash, event); - if (WARN_ON_ONCE(!head)) + if (!head) { + /* + * We can race with cpu hotplug code. Do not + * WARN if the cpu just got unplugged. + */ + WARN_ON_ONCE(swhash->online); return -EINVAL; + } hlist_add_head_rcu(&event->hlist_entry, head); @@ -6562,6 +6685,9 @@ SYSCALL_DEFINE5(perf_event_open, if (attr.freq) { if (attr.sample_freq > sysctl_perf_event_sample_rate) return -EINVAL; + } else { + if (attr.sample_period & (1ULL << 63)) + return -EINVAL; } /* @@ -6708,7 +6834,7 @@ SYSCALL_DEFINE5(perf_event_open, struct perf_event_context *gctx = group_leader->ctx; mutex_lock(&gctx->mutex); - perf_remove_from_context(group_leader); + perf_remove_from_context(group_leader, false); /* * Removing from the context ends up with disabled @@ -6718,7 +6844,7 @@ SYSCALL_DEFINE5(perf_event_open, perf_event__state_init(group_leader); list_for_each_entry(sibling, &group_leader->sibling_list, group_entry) { - perf_remove_from_context(sibling); + perf_remove_from_context(sibling, false); perf_event__state_init(sibling); put_ctx(gctx); } @@ -6848,7 +6974,7 @@ void perf_pmu_migrate_context(struct pmu *pmu, int src_cpu, int dst_cpu) mutex_lock(&src_ctx->mutex); list_for_each_entry_safe(event, tmp, &src_ctx->event_list, event_entry) { - perf_remove_from_context(event); + perf_remove_from_context(event, false); put_ctx(src_ctx); list_add(&event->event_entry, &events); } @@ -6908,13 +7034,7 @@ __perf_event_exit_task(struct perf_event *child_event, struct perf_event_context *child_ctx, struct task_struct *child) { - if (child_event->parent) { - raw_spin_lock_irq(&child_ctx->lock); - perf_group_detach(child_event); - raw_spin_unlock_irq(&child_ctx->lock); - } - - perf_remove_from_context(child_event); + perf_remove_from_context(child_event, !!child_event->parent); /* * It can happen that the parent exits first, and has events @@ -7228,7 +7348,7 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent, * child. */ - child_ctx = alloc_perf_context(event->pmu, child); + child_ctx = alloc_perf_context(parent_ctx->pmu, child); if (!child_ctx) return -ENOMEM; @@ -7376,6 +7496,7 @@ static void __cpuinit perf_event_init_cpu(int cpu) struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); mutex_lock(&swhash->hlist_mutex); + swhash->online = true; if (swhash->hlist_refcount > 0) { struct swevent_hlist *hlist; @@ -7398,15 +7519,15 @@ static void perf_pmu_rotate_stop(struct pmu *pmu) static void __perf_event_exit_context(void *__info) { + struct remove_event re = { .detach_group = false }; struct perf_event_context *ctx = __info; - struct perf_event *event, *tmp; perf_pmu_rotate_stop(ctx->pmu); - list_for_each_entry_safe(event, tmp, &ctx->pinned_groups, group_entry) - __perf_remove_from_context(event); - list_for_each_entry_safe(event, tmp, &ctx->flexible_groups, group_entry) - __perf_remove_from_context(event); + rcu_read_lock(); + list_for_each_entry_rcu(re.event, &ctx->event_list, event_entry) + __perf_remove_from_context(&re); + rcu_read_unlock(); } static void perf_event_exit_cpu_context(int cpu) @@ -7430,11 +7551,12 @@ static void perf_event_exit_cpu(int cpu) { struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu); + perf_event_exit_cpu_context(cpu); + mutex_lock(&swhash->hlist_mutex); + swhash->online = false; swevent_hlist_release(swhash); mutex_unlock(&swhash->hlist_mutex); - - perf_event_exit_cpu_context(cpu); } #else static inline void perf_event_exit_cpu(int cpu) { } diff --git a/kernel/events/ring_buffer.c b/kernel/events/ring_buffer.c index cd55144270b5..9c2ddfbf4525 100644 --- a/kernel/events/ring_buffer.c +++ b/kernel/events/ring_buffer.c @@ -87,10 +87,31 @@ again: goto out; /* - * Publish the known good head. Rely on the full barrier implied - * by atomic_dec_and_test() order the rb->head read and this - * write. + * Since the mmap() consumer (userspace) can run on a different CPU: + * + * kernel user + * + * READ ->data_tail READ ->data_head + * smp_mb() (A) smp_rmb() (C) + * WRITE $data READ $data + * smp_wmb() (B) smp_mb() (D) + * STORE ->data_head WRITE ->data_tail + * + * Where A pairs with D, and B pairs with C. + * + * I don't think A needs to be a full barrier because we won't in fact + * write data until we see the store from userspace. So we simply don't + * issue the data WRITE until we observe it. Be conservative for now. + * + * OTOH, D needs to be a full barrier since it separates the data READ + * from the tail WRITE. + * + * For B a WMB is sufficient since it separates two WRITEs, and for C + * an RMB is sufficient since it separates two READs. + * + * See perf_output_begin(). */ + smp_wmb(); rb->user_page->data_head = head; /* @@ -154,9 +175,11 @@ int perf_output_begin(struct perf_output_handle *handle, * Userspace could choose to issue a mb() before updating the * tail pointer. So that all reads will be completed before the * write is issued. + * + * See perf_output_put_handle(). */ tail = ACCESS_ONCE(rb->user_page->data_tail); - smp_rmb(); + smp_mb(); offset = head = local_read(&rb->head); head += size; if (unlikely(!perf_output_space(rb, tail, offset, head))) diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index f3569747d629..ad8e1bdca70e 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1682,12 +1682,10 @@ static bool handle_trampoline(struct pt_regs *regs) tmp = ri; ri = ri->next; kfree(tmp); + utask->depth--; if (!chained) break; - - utask->depth--; - BUG_ON(!ri); } diff --git a/kernel/exit.c b/kernel/exit.c index 7bb73f9d09db..6682b2ea5b11 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -570,9 +570,6 @@ static void reparent_leader(struct task_struct *father, struct task_struct *p, struct list_head *dead) { list_move_tail(&p->sibling, &p->real_parent->children); - - if (p->exit_state == EXIT_DEAD) - return; /* * If this is a threaded reparent there is no need to * notify anyone anything has happened. @@ -580,9 +577,19 @@ static void reparent_leader(struct task_struct *father, struct task_struct *p, if (same_thread_group(p->real_parent, father)) return; - /* We don't want people slaying init. */ + /* + * We don't want people slaying init. + * + * Note: we do this even if it is EXIT_DEAD, wait_task_zombie() + * can change ->exit_state to EXIT_ZOMBIE. If this is the final + * state, do_notify_parent() was already called and ->exit_signal + * doesn't matter. + */ p->exit_signal = SIGCHLD; + if (p->exit_state == EXIT_DEAD) + return; + /* If it has exited notify the new parent about this child's death. */ if (!p->ptrace && p->exit_state == EXIT_ZOMBIE && thread_group_empty(p)) { @@ -794,6 +801,8 @@ void do_exit(long code) exit_shm(tsk); exit_files(tsk); exit_fs(tsk); + if (group_dead) + disassociate_ctty(1); exit_task_namespaces(tsk); exit_task_work(tsk); check_stack_usage(); @@ -809,13 +818,9 @@ void do_exit(long code) cgroup_exit(tsk, 1); - if (group_dead) - disassociate_ctty(1); - module_put(task_thread_info(tsk)->exec_domain->module); proc_exit_connector(tsk); - /* * FIXME: do that only when needed, using sched_exit tracepoint */ diff --git a/kernel/fork.c b/kernel/fork.c index 987b28a1f01b..ff7be9dac4c1 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -544,6 +544,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p) mm->cached_hole_size = ~0UL; mm_init_aio(mm); mm_init_owner(mm, p); + clear_tlb_flush_pending(mm); if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; @@ -1171,10 +1172,11 @@ static struct task_struct *copy_process(unsigned long clone_flags, return ERR_PTR(-EINVAL); /* - * If the new process will be in a different pid namespace - * don't allow the creation of threads. + * If the new process will be in a different pid namespace don't + * allow it to share a thread group or signal handlers with the + * forking task. */ - if ((clone_flags & (CLONE_VM|CLONE_NEWPID)) && + if ((clone_flags & (CLONE_SIGHAND | CLONE_NEWPID)) && (task_active_pid_ns(current) != current->nsproxy->pid_ns)) return ERR_PTR(-EINVAL); @@ -1675,6 +1677,12 @@ SYSCALL_DEFINE5(clone, unsigned long, newsp, unsigned long, clone_flags, int __user *, parent_tidptr, int __user *, child_tidptr, int, tls_val) +#elif defined(CONFIG_CLONE_BACKWARDS3) +SYSCALL_DEFINE6(clone, unsigned long, clone_flags, unsigned long, newsp, + int, stack_size, + int __user *, parent_tidptr, + int __user *, child_tidptr, + int, tls_val) #else SYSCALL_DEFINE5(clone, unsigned long, clone_flags, unsigned long, newsp, int __user *, parent_tidptr, diff --git a/kernel/freezer.c b/kernel/freezer.c index c38893b0efba..78758512b1e1 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -19,6 +19,12 @@ EXPORT_SYMBOL(system_freezing_cnt); bool pm_freezing; bool pm_nosig_freezing; +/* + * Temporary export for the deadlock workaround in ata_scsi_hotplug(). + * Remove once the hack becomes unnecessary. + */ +EXPORT_SYMBOL_GPL(pm_freezing); + /* protects freezing and frozen transitions */ static DEFINE_SPINLOCK(freezer_lock); diff --git a/kernel/futex.c b/kernel/futex.c index b26dcfc02c94..625a4e659e7a 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -61,12 +61,15 @@ #include <linux/nsproxy.h> #include <linux/ptrace.h> #include <linux/sched/rt.h> +#include <linux/hugetlb.h> #include <asm/futex.h> #include "rtmutex_common.h" +#ifndef CONFIG_HAVE_FUTEX_CMPXCHG int __read_mostly futex_cmpxchg_enabled; +#endif #define FUTEX_HASHBITS (CONFIG_BASE_SMALL ? 4 : 8) @@ -286,7 +289,7 @@ again: put_page(page); /* serialize against __split_huge_page_splitting() */ local_irq_disable(); - if (likely(__get_user_pages_fast(address, 1, 1, &page) == 1)) { + if (likely(__get_user_pages_fast(address, 1, !ro, &page) == 1)) { page_head = compound_head(page); /* * page_head is valid pointer but we must pin @@ -365,7 +368,7 @@ again: } else { key->both.offset |= FUT_OFF_INODE; /* inode-based key */ key->shared.inode = page_head->mapping->host; - key->shared.pgoff = page_head->index; + key->shared.pgoff = basepage_index(page); } get_futex_key_refs(key); @@ -589,6 +592,55 @@ void exit_pi_state_list(struct task_struct *curr) raw_spin_unlock_irq(&curr->pi_lock); } +/* + * We need to check the following states: + * + * Waiter | pi_state | pi->owner | uTID | uODIED | ? + * + * [1] NULL | --- | --- | 0 | 0/1 | Valid + * [2] NULL | --- | --- | >0 | 0/1 | Valid + * + * [3] Found | NULL | -- | Any | 0/1 | Invalid + * + * [4] Found | Found | NULL | 0 | 1 | Valid + * [5] Found | Found | NULL | >0 | 1 | Invalid + * + * [6] Found | Found | task | 0 | 1 | Valid + * + * [7] Found | Found | NULL | Any | 0 | Invalid + * + * [8] Found | Found | task | ==taskTID | 0/1 | Valid + * [9] Found | Found | task | 0 | 0 | Invalid + * [10] Found | Found | task | !=taskTID | 0/1 | Invalid + * + * [1] Indicates that the kernel can acquire the futex atomically. We + * came came here due to a stale FUTEX_WAITERS/FUTEX_OWNER_DIED bit. + * + * [2] Valid, if TID does not belong to a kernel thread. If no matching + * thread is found then it indicates that the owner TID has died. + * + * [3] Invalid. The waiter is queued on a non PI futex + * + * [4] Valid state after exit_robust_list(), which sets the user space + * value to FUTEX_WAITERS | FUTEX_OWNER_DIED. + * + * [5] The user space value got manipulated between exit_robust_list() + * and exit_pi_state_list() + * + * [6] Valid state after exit_pi_state_list() which sets the new owner in + * the pi_state but cannot access the user space value. + * + * [7] pi_state->owner can only be NULL when the OWNER_DIED bit is set. + * + * [8] Owner and user space value match + * + * [9] There is no transient state which sets the user space TID to 0 + * except exit_robust_list(), but this is indicated by the + * FUTEX_OWNER_DIED bit. See [4] + * + * [10] There is no transient state which leaves owner and user space + * TID out of sync. + */ static int lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, union futex_key *key, struct futex_pi_state **ps) @@ -604,12 +656,13 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, plist_for_each_entry_safe(this, next, head, list) { if (match_futex(&this->key, key)) { /* - * Another waiter already exists - bump up - * the refcount and return its pi_state: + * Sanity check the waiter before increasing + * the refcount and attaching to it. */ pi_state = this->pi_state; /* - * Userspace might have messed up non-PI and PI futexes + * Userspace might have messed up non-PI and + * PI futexes [3] */ if (unlikely(!pi_state)) return -EINVAL; @@ -617,34 +670,70 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, WARN_ON(!atomic_read(&pi_state->refcount)); /* - * When pi_state->owner is NULL then the owner died - * and another waiter is on the fly. pi_state->owner - * is fixed up by the task which acquires - * pi_state->rt_mutex. - * - * We do not check for pid == 0 which can happen when - * the owner died and robust_list_exit() cleared the - * TID. + * Handle the owner died case: */ - if (pid && pi_state->owner) { + if (uval & FUTEX_OWNER_DIED) { /* - * Bail out if user space manipulated the - * futex value. + * exit_pi_state_list sets owner to NULL and + * wakes the topmost waiter. The task which + * acquires the pi_state->rt_mutex will fixup + * owner. */ - if (pid != task_pid_vnr(pi_state->owner)) + if (!pi_state->owner) { + /* + * No pi state owner, but the user + * space TID is not 0. Inconsistent + * state. [5] + */ + if (pid) + return -EINVAL; + /* + * Take a ref on the state and + * return. [4] + */ + goto out_state; + } + + /* + * If TID is 0, then either the dying owner + * has not yet executed exit_pi_state_list() + * or some waiter acquired the rtmutex in the + * pi state, but did not yet fixup the TID in + * user space. + * + * Take a ref on the state and return. [6] + */ + if (!pid) + goto out_state; + } else { + /* + * If the owner died bit is not set, + * then the pi_state must have an + * owner. [7] + */ + if (!pi_state->owner) return -EINVAL; } + /* + * Bail out if user space manipulated the + * futex value. If pi state exists then the + * owner TID must be the same as the user + * space TID. [9/10] + */ + if (pid != task_pid_vnr(pi_state->owner)) + return -EINVAL; + + out_state: atomic_inc(&pi_state->refcount); *ps = pi_state; - return 0; } } /* * We are the first waiter - try to look up the real owner and attach - * the new pi_state to it, but bail out when TID = 0 + * the new pi_state to it, but bail out when TID = 0 [1] */ if (!pid) return -ESRCH; @@ -652,6 +741,11 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, if (!p) return -ESRCH; + if (!p->mm) { + put_task_struct(p); + return -EPERM; + } + /* * We need to look at the task state flags to figure out, * whether the task is exiting. To protect against the do_exit @@ -672,6 +766,9 @@ lookup_pi_state(u32 uval, struct futex_hash_bucket *hb, return ret; } + /* + * No existing pi state. First waiter. [2] + */ pi_state = alloc_pi_state(); /* @@ -743,10 +840,18 @@ retry: return -EDEADLK; /* - * Surprise - we got the lock. Just return to userspace: + * Surprise - we got the lock, but we do not trust user space at all. */ - if (unlikely(!curval)) - return 1; + if (unlikely(!curval)) { + /* + * We verify whether there is kernel state for this + * futex. If not, we can safely assume, that the 0 -> + * TID transition is correct. If state exists, we do + * not bother to fixup the user space state as it was + * corrupted already. + */ + return futex_top_waiter(hb, key) ? -EINVAL : 1; + } uval = curval; @@ -876,6 +981,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) struct task_struct *new_owner; struct futex_pi_state *pi_state = this->pi_state; u32 uninitialized_var(curval), newval; + int ret = 0; if (!pi_state) return -EINVAL; @@ -899,23 +1005,19 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) new_owner = this->task; /* - * We pass it to the next owner. (The WAITERS bit is always - * kept enabled while there is PI state around. We must also - * preserve the owner died bit.) + * We pass it to the next owner. The WAITERS bit is always + * kept enabled while there is PI state around. We cleanup the + * owner died bit, because we are the owner. */ - if (!(uval & FUTEX_OWNER_DIED)) { - int ret = 0; - - newval = FUTEX_WAITERS | task_pid_vnr(new_owner); + newval = FUTEX_WAITERS | task_pid_vnr(new_owner); - if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) - ret = -EFAULT; - else if (curval != uval) - ret = -EINVAL; - if (ret) { - raw_spin_unlock(&pi_state->pi_mutex.wait_lock); - return ret; - } + if (cmpxchg_futex_value_locked(&curval, uaddr, uval, newval)) + ret = -EFAULT; + else if (curval != uval) + ret = -EINVAL; + if (ret) { + raw_spin_unlock(&pi_state->pi_mutex.wait_lock); + return ret; } raw_spin_lock_irq(&pi_state->owner->pi_lock); @@ -1194,7 +1296,7 @@ void requeue_pi_wake_futex(struct futex_q *q, union futex_key *key, * * Return: * 0 - failed to acquire the lock atomically; - * 1 - acquired the lock; + * >0 - acquired the lock, return value is vpid of the top_waiter * <0 - error */ static int futex_proxy_trylock_atomic(u32 __user *pifutex, @@ -1205,7 +1307,7 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, { struct futex_q *top_waiter = NULL; u32 curval; - int ret; + int ret, vpid; if (get_futex_value_locked(&curval, pifutex)) return -EFAULT; @@ -1233,11 +1335,13 @@ static int futex_proxy_trylock_atomic(u32 __user *pifutex, * the contended case or if set_waiters is 1. The pi_state is returned * in ps in contended cases. */ + vpid = task_pid_vnr(top_waiter->task); ret = futex_lock_pi_atomic(pifutex, hb2, key2, ps, top_waiter->task, set_waiters); - if (ret == 1) + if (ret == 1) { requeue_pi_wake_futex(top_waiter, key2, hb2); - + return vpid; + } return ret; } @@ -1269,10 +1373,16 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags, struct futex_hash_bucket *hb1, *hb2; struct plist_head *head1; struct futex_q *this, *next; - u32 curval2; if (requeue_pi) { /* + * Requeue PI only works on two distinct uaddrs. This + * check is only valid for private futexes. See below. + */ + if (uaddr1 == uaddr2) + return -EINVAL; + + /* * requeue_pi requires a pi_state, try to allocate it now * without any locks in case it fails. */ @@ -1310,6 +1420,15 @@ retry: if (unlikely(ret != 0)) goto out_put_key1; + /* + * The check above which compares uaddrs is not sufficient for + * shared futexes. We need to compare the keys: + */ + if (requeue_pi && match_futex(&key1, &key2)) { + ret = -EINVAL; + goto out_put_keys; + } + hb1 = hash_futex(&key1); hb2 = hash_futex(&key2); @@ -1355,16 +1474,25 @@ retry_private: * At this point the top_waiter has either taken uaddr2 or is * waiting on it. If the former, then the pi_state will not * exist yet, look it up one more time to ensure we have a - * reference to it. + * reference to it. If the lock was taken, ret contains the + * vpid of the top waiter task. */ - if (ret == 1) { + if (ret > 0) { WARN_ON(pi_state); drop_count++; task_count++; - ret = get_futex_value_locked(&curval2, uaddr2); - if (!ret) - ret = lookup_pi_state(curval2, hb2, &key2, - &pi_state); + /* + * If we acquired the lock, then the user + * space value of uaddr2 should be vpid. It + * cannot be changed by the top waiter as it + * is blocked on hb2 lock if it tries to do + * so. If something fiddled with it behind our + * back the pi state lookup might unearth + * it. So we rather use the known value than + * rereading and handing potential crap to + * lookup_pi_state. + */ + ret = lookup_pi_state(ret, hb2, &key2, &pi_state); } switch (ret) { @@ -2134,9 +2262,10 @@ retry: /* * To avoid races, try to do the TID -> 0 atomic transition * again. If it succeeds then we can return without waking - * anyone else up: + * anyone else up. We only try this if neither the waiters nor + * the owner died bit are set. */ - if (!(uval & FUTEX_OWNER_DIED) && + if (!(uval & ~FUTEX_TID_MASK) && cmpxchg_futex_value_locked(&uval, uaddr, vpid, 0)) goto pi_faulted; /* @@ -2168,11 +2297,9 @@ retry: /* * No waiters - kernel unlocks the futex: */ - if (!(uval & FUTEX_OWNER_DIED)) { - ret = unlock_futex_pi(uaddr, uval); - if (ret == -EFAULT) - goto pi_faulted; - } + ret = unlock_futex_pi(uaddr, uval); + if (ret == -EFAULT) + goto pi_faulted; out_unlock: spin_unlock(&hb->lock); @@ -2331,6 +2458,15 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags, if (ret) goto out_key2; + /* + * The check above which compares uaddrs is not sufficient for + * shared futexes. We need to compare the keys: + */ + if (match_futex(&q.key, &key2)) { + ret = -EINVAL; + goto out_put_keys; + } + /* Queue the futex_q, drop the hb lock, wait for wakeup. */ futex_wait_queue_me(hb, &q, to); @@ -2728,10 +2864,10 @@ SYSCALL_DEFINE6(futex, u32 __user *, uaddr, int, op, u32, val, return do_futex(uaddr, op, val, tp, uaddr2, val2, val3); } -static int __init futex_init(void) +static void __init futex_detect_cmpxchg(void) { +#ifndef CONFIG_HAVE_FUTEX_CMPXCHG u32 curval; - int i; /* * This will fail and we want it. Some arch implementations do @@ -2745,6 +2881,14 @@ static int __init futex_init(void) */ if (cmpxchg_futex_value_locked(&curval, NULL, 0, 0) == -EFAULT) futex_cmpxchg_enabled = 1; +#endif +} + +static int __init futex_init(void) +{ + int i; + + futex_detect_cmpxchg(); for (i = 0; i < ARRAY_SIZE(futex_queues); i++) { plist_head_init(&futex_queues[i].chain); diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index fd4b13b131f8..aadf4b7a607c 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -245,6 +245,11 @@ again: goto again; } timer->base = new_base; + } else { + if (cpu != this_cpu && hrtimer_check_target(timer, new_base)) { + cpu = this_cpu; + goto again; + } } return new_base; } @@ -580,6 +585,23 @@ hrtimer_force_reprogram(struct hrtimer_cpu_base *cpu_base, int skip_equal) cpu_base->expires_next.tv64 = expires_next.tv64; + /* + * If a hang was detected in the last timer interrupt then we + * leave the hang delay active in the hardware. We want the + * system to make progress. That also prevents the following + * scenario: + * T1 expires 50ms from now + * T2 expires 5s from now + * + * T1 is removed, so this code is called and would reprogram + * the hardware to 5s from now. Any hrtimer_start after that + * will not reprogram the hardware due to hang_detected being + * set. So we'd effectivly block all timers until the T2 event + * fires. + */ + if (cpu_base->hang_detected) + return; + if (cpu_base->expires_next.tv64 != KTIME_MAX) tick_program_event(cpu_base->expires_next, 1); } @@ -721,17 +743,20 @@ static int hrtimer_switch_to_hres(void) return 1; } +static void clock_was_set_work(struct work_struct *work) +{ + clock_was_set(); +} + +static DECLARE_WORK(hrtimer_work, clock_was_set_work); + /* - * Called from timekeeping code to reprogramm the hrtimer interrupt - * device. If called from the timer interrupt context we defer it to - * softirq context. + * Called from timekeeping and resume code to reprogramm the hrtimer + * interrupt device on all cpus. */ void clock_was_set_delayed(void) { - struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases); - - cpu_base->clock_was_set = 1; - __raise_softirq_irqoff(HRTIMER_SOFTIRQ); + schedule_work(&hrtimer_work); } #else @@ -780,8 +805,10 @@ void hrtimers_resume(void) WARN_ONCE(!irqs_disabled(), KERN_INFO "hrtimers_resume() called with IRQs enabled!"); + /* Retrigger on the local CPU */ retrigger_next_event(NULL); - timerfd_clock_was_set(); + /* And schedule a retrigger for all others */ + clock_was_set_delayed(); } static inline void timer_stats_hrtimer_set_start_info(struct hrtimer *timer) @@ -972,11 +999,8 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, /* Remove an active timer from the queue: */ ret = remove_hrtimer(timer, base); - /* Switch the timer base, if necessary: */ - new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); - if (mode & HRTIMER_MODE_REL) { - tim = ktime_add_safe(tim, new_base->get_time()); + tim = ktime_add_safe(tim, base->get_time()); /* * CONFIG_TIME_LOW_RES is a temporary way for architectures * to signal that they simply return xtime in @@ -991,6 +1015,9 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim, hrtimer_set_expires_range_ns(timer, tim, delta_ns); + /* Switch the timer base, if necessary: */ + new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED); + timer_stats_hrtimer_set_start_info(timer); leftmost = enqueue_hrtimer(timer, new_base); @@ -1432,13 +1459,6 @@ void hrtimer_peek_ahead_timers(void) static void run_hrtimer_softirq(struct softirq_action *h) { - struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases); - - if (cpu_base->clock_was_set) { - cpu_base->clock_was_set = 0; - clock_was_set(); - } - hrtimer_peek_ahead_timers(); } diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index 192a302d6cfd..3fcb6faa5fa6 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -23,10 +23,35 @@ static struct lock_class_key irq_desc_lock_class; #if defined(CONFIG_SMP) +static int __init irq_affinity_setup(char *str) +{ + zalloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); + cpulist_parse(str, irq_default_affinity); + /* + * Set at least the boot cpu. We don't want to end up with + * bugreports caused by random comandline masks + */ + cpumask_set_cpu(smp_processor_id(), irq_default_affinity); + return 1; +} +__setup("irqaffinity=", irq_affinity_setup); + +extern struct cpumask hmp_slow_cpu_mask; + static void __init init_irq_default_affinity(void) { - alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); - cpumask_setall(irq_default_affinity); +#ifdef CONFIG_CPUMASK_OFFSTACK + if (!irq_default_affinity) + zalloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT); +#endif +#ifdef CONFIG_SCHED_HMP + if (!cpumask_empty(&hmp_slow_cpu_mask)) { + cpumask_copy(irq_default_affinity, &hmp_slow_cpu_mask); + return; + } +#endif + if (cpumask_empty(irq_default_affinity)) + cpumask_setall(irq_default_affinity); } #else static void __init init_irq_default_affinity(void) @@ -274,6 +299,7 @@ struct irq_desc *irq_to_desc(unsigned int irq) { return (irq < NR_IRQS) ? irq_desc + irq : NULL; } +EXPORT_SYMBOL(irq_to_desc); static void free_desc(unsigned int irq) { diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index fa17855ca65a..8815abfdf2cb 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -150,7 +150,7 @@ int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask, struct irq_chip *chip = irq_data_get_irq_chip(data); int ret; - ret = chip->irq_set_affinity(data, mask, false); + ret = chip->irq_set_affinity(data, mask, force); switch (ret) { case IRQ_SET_MASK_OK: cpumask_copy(data->affinity, mask); @@ -162,7 +162,8 @@ int irq_do_set_affinity(struct irq_data *data, const struct cpumask *mask, return ret; } -int __irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask) +int irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask, + bool force) { struct irq_chip *chip = irq_data_get_irq_chip(data); struct irq_desc *desc = irq_data_to_desc(data); @@ -172,7 +173,7 @@ int __irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask) return -EINVAL; if (irq_can_move_pcntxt(data)) { - ret = irq_do_set_affinity(data, mask, false); + ret = irq_do_set_affinity(data, mask, force); } else { irqd_set_move_pending(data); irq_copy_pending(desc, mask); @@ -187,13 +188,7 @@ int __irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask) return ret; } -/** - * irq_set_affinity - Set the irq affinity of a given irq - * @irq: Interrupt to set affinity - * @mask: cpumask - * - */ -int irq_set_affinity(unsigned int irq, const struct cpumask *mask) +int __irq_set_affinity(unsigned int irq, const struct cpumask *mask, bool force) { struct irq_desc *desc = irq_to_desc(irq); unsigned long flags; @@ -203,7 +198,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *mask) return -EINVAL; raw_spin_lock_irqsave(&desc->lock, flags); - ret = __irq_set_affinity_locked(irq_desc_get_irq_data(desc), mask); + ret = irq_set_affinity_locked(irq_desc_get_irq_data(desc), mask, force); raw_spin_unlock_irqrestore(&desc->lock, flags); return ret; } @@ -555,9 +550,9 @@ int can_request_irq(unsigned int irq, unsigned long irqflags) return 0; if (irq_settings_can_request(desc)) { - if (desc->action) - if (irqflags & desc->action->flags & IRQF_SHARED) - canrequest =1; + if (!desc->action || + irqflags & desc->action->flags & IRQF_SHARED) + canrequest = 1; } irq_put_desc_unlock(desc, flags); return canrequest; @@ -802,8 +797,7 @@ static irqreturn_t irq_thread_fn(struct irq_desc *desc, static void wake_threads_waitq(struct irq_desc *desc) { - if (atomic_dec_and_test(&desc->threads_active) && - waitqueue_active(&desc->wait_for_threads)) + if (atomic_dec_and_test(&desc->threads_active)) wake_up(&desc->wait_for_threads); } diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index cb228bf21760..abcd6ca86cb7 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -50,7 +50,7 @@ static void resume_irqs(bool want_early) bool is_early = desc->action && desc->action->flags & IRQF_EARLY_RESUME; - if (is_early != want_early) + if (!is_early && want_early) continue; raw_spin_lock_irqsave(&desc->lock, flags); diff --git a/kernel/kexec.c b/kernel/kexec.c index 59f7b55ba745..1f8d9382dbac 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -47,6 +47,9 @@ u32 vmcoreinfo_note[VMCOREINFO_NOTE_SIZE/4]; size_t vmcoreinfo_size; size_t vmcoreinfo_max_size = sizeof(vmcoreinfo_data); +/* Flag to indicate we are going to kexec a new kernel */ +bool kexec_in_progress = false; + /* Location of the reserved area for the crash kernel */ struct resource crashk_res = { .name = "Crash kernel", @@ -1678,6 +1681,7 @@ int kernel_kexec(void) } else #endif { + kexec_in_progress = true; kernel_restart_prepare(NULL); printk(KERN_EMERG "Starting new kernel\n"); machine_shutdown(); diff --git a/kernel/module.c b/kernel/module.c index cab4bce49c23..10a3af821d28 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2927,7 +2927,6 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) { /* Module within temporary copy. */ struct module *mod; - Elf_Shdr *pcpusec; int err; mod = setup_load_info(info, flags); @@ -2942,17 +2941,10 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) err = module_frob_arch_sections(info->hdr, info->sechdrs, info->secstrings, mod); if (err < 0) - goto out; + return ERR_PTR(err); - pcpusec = &info->sechdrs[info->index.pcpu]; - if (pcpusec->sh_size) { - /* We have a special allocation for this section. */ - err = percpu_modalloc(mod, - pcpusec->sh_size, pcpusec->sh_addralign); - if (err) - goto out; - pcpusec->sh_flags &= ~(unsigned long)SHF_ALLOC; - } + /* We will do a special allocation for per-cpu sections later. */ + info->sechdrs[info->index.pcpu].sh_flags &= ~(unsigned long)SHF_ALLOC; /* Determine total sizes, and put offsets in sh_entsize. For now this is done generically; there doesn't appear to be any @@ -2963,17 +2955,22 @@ static struct module *layout_and_allocate(struct load_info *info, int flags) /* Allocate and move to the final place */ err = move_module(mod, info); if (err) - goto free_percpu; + return ERR_PTR(err); /* Module has been copied to its final place now: return it. */ mod = (void *)info->sechdrs[info->index.mod].sh_addr; kmemleak_load_module(mod, info); return mod; +} -free_percpu: - percpu_modfree(mod); -out: - return ERR_PTR(err); +static int alloc_module_percpu(struct module *mod, struct load_info *info) +{ + Elf_Shdr *pcpusec = &info->sechdrs[info->index.pcpu]; + if (!pcpusec->sh_size) + return 0; + + /* We have a special allocation for this section. */ + return percpu_modalloc(mod, pcpusec->sh_size, pcpusec->sh_addralign); } /* mod is no longer valid after this! */ @@ -3237,6 +3234,11 @@ static int load_module(struct load_info *info, const char __user *uargs, } #endif + /* To avoid stressing percpu allocator, do this once we're unique. */ + err = alloc_module_percpu(mod, info); + if (err) + goto unlink_mod; + /* Now module is in final location, initialize linked lists, etc. */ err = module_unload_init(mod); if (err) @@ -3277,6 +3279,9 @@ static int load_module(struct load_info *info, const char __user *uargs, dynamic_debug_setup(info->debug, info->num_debug); + /* Ftrace init must be called in the MODULE_STATE_UNFORMED state */ + ftrace_module_init(mod); + /* Finally it's fully formed, ready to start executing. */ err = complete_formation(mod, info); if (err) diff --git a/kernel/pid.c b/kernel/pid.c index 0db3e791a06d..0eb6d8e8b1da 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -264,6 +264,7 @@ void free_pid(struct pid *pid) struct pid_namespace *ns = upid->ns; hlist_del_rcu(&upid->pid_chain); switch(--ns->nr_hashed) { + case 2: case 1: /* When all that is left in the pid namespace * is the reaper wake up the reaper. The reaper diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index 6917e8edb48e..e32703d5e0ab 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -312,7 +312,9 @@ static void *pidns_get(struct task_struct *task) struct pid_namespace *ns; rcu_read_lock(); - ns = get_pid_ns(task_active_pid_ns(task)); + ns = task_active_pid_ns(task); + if (ns) + get_pid_ns(ns); rcu_read_unlock(); return ns; diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 5dfdc9ea180b..46455961a88f 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -263,6 +263,26 @@ config PM_GENERIC_DOMAINS bool depends on PM +config WQ_POWER_EFFICIENT_DEFAULT + bool "Enable workqueue power-efficient mode by default" + depends on PM + default n + help + Per-cpu workqueues are generally preferred because they show + better performance thanks to cache locality; unfortunately, + per-cpu workqueues tend to be more power hungry than unbound + workqueues. + + Enabling workqueue.power_efficient kernel parameter makes the + per-cpu workqueues which were observed to contribute + significantly to power consumption unbound, leading to measurably + lower power usage at the cost of small performance overhead. + + This config option determines whether workqueue.power_efficient + is enabled by default. + + If in doubt, say N. + config PM_GENERIC_DOMAINS_SLEEP def_bool y depends on PM_SLEEP && PM_GENERIC_DOMAINS diff --git a/kernel/power/autosleep.c b/kernel/power/autosleep.c index c6422ffeda9a..9012ecf7b814 100644 --- a/kernel/power/autosleep.c +++ b/kernel/power/autosleep.c @@ -32,7 +32,8 @@ static void try_to_suspend(struct work_struct *work) mutex_lock(&autosleep_lock); - if (!pm_save_wakeup_count(initial_count)) { + if (!pm_save_wakeup_count(initial_count) || + system_state != SYSTEM_RUNNING) { mutex_unlock(&autosleep_lock); goto out; } diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 587dddeebf15..25cf89bc659e 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -293,6 +293,15 @@ int pm_qos_request_active(struct pm_qos_request *req) } EXPORT_SYMBOL_GPL(pm_qos_request_active); +static void __pm_qos_update_request(struct pm_qos_request *req, + s32 new_value) +{ + if (new_value != req->node.prio) + pm_qos_update_target( + pm_qos_array[req->pm_qos_class]->constraints, + &req->node, PM_QOS_UPDATE_REQ, new_value); +} + /** * pm_qos_work_fn - the timeout handler of pm_qos_update_request_timeout * @work: work struct for the delayed work (timeout) @@ -305,7 +314,7 @@ static void pm_qos_work_fn(struct work_struct *work) struct pm_qos_request, work); - pm_qos_update_request(req, PM_QOS_DEFAULT_VALUE); + __pm_qos_update_request(req, PM_QOS_DEFAULT_VALUE); } /** @@ -365,6 +374,8 @@ void pm_qos_update_request(struct pm_qos_request *req, pm_qos_update_target( pm_qos_array[req->pm_qos_class]->constraints, &req->node, PM_QOS_UPDATE_REQ, new_value); + + __pm_qos_update_request(req, new_value); } EXPORT_SYMBOL_GPL(pm_qos_update_request); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 0de28576807d..91c04f16e79c 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1398,7 +1398,11 @@ int hibernate_preallocate_memory(void) * highmem and non-highmem zones separately. */ pages_highmem = preallocate_image_highmem(highmem / 2); - alloc = (count - max_size) - pages_highmem; + alloc = count - max_size; + if (alloc > pages_highmem) + alloc -= pages_highmem; + else + alloc = 0; pages = preallocate_image_memory(alloc, avail_normal); if (pages < alloc) { /* We have exhausted non-highmem pages, try highmem. */ diff --git a/kernel/printk.c b/kernel/printk.c index 8212c1aef125..d37d45c90ae6 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1369,9 +1369,9 @@ static int console_trylock_for_printk(unsigned int cpu) } } logbuf_cpu = UINT_MAX; + raw_spin_unlock(&logbuf_lock); if (wake) up(&console_sem); - raw_spin_unlock(&logbuf_lock); return retval; } diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 335a7ae697f5..afadcf7b4a22 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -257,7 +257,8 @@ ok: if (task->mm) dumpable = get_dumpable(task->mm); rcu_read_lock(); - if (!dumpable && !ptrace_has_cap(__task_cred(task)->user_ns, mode)) { + if (dumpable != SUID_DUMP_USER && + !ptrace_has_cap(__task_cred(task)->user_ns, mode)) { rcu_read_unlock(); return -EPERM; } diff --git a/kernel/sched/auto_group.c b/kernel/sched/auto_group.c index 64de5f8b0c9e..4a073539c58e 100644 --- a/kernel/sched/auto_group.c +++ b/kernel/sched/auto_group.c @@ -77,8 +77,6 @@ static inline struct autogroup *autogroup_create(void) if (IS_ERR(tg)) goto out_free; - sched_online_group(tg, &root_task_group); - kref_init(&ag->kref); init_rwsem(&ag->lock); ag->id = atomic_inc_return(&autogroup_seq_nr); @@ -98,6 +96,7 @@ static inline struct autogroup *autogroup_create(void) #endif tg->autogroup = ag; + sched_online_group(tg, &root_task_group); return ag; out_free: diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e8b335016c52..6299bb47de74 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1407,7 +1407,11 @@ void scheduler_ipi(void) { if (llist_empty(&this_rq()->wake_list) && !tick_nohz_full_cpu(smp_processor_id()) - && !got_nohz_idle_kick()) + && !got_nohz_idle_kick() +#ifdef CONFIG_SCHED_HMP + && !this_rq()->wake_for_idle_pull +#endif + ) return; /* @@ -1434,6 +1438,11 @@ void scheduler_ipi(void) this_rq()->idle_balance = 1; raise_softirq_irqoff(SCHED_SOFTIRQ); } +#ifdef CONFIG_SCHED_HMP + else if (unlikely(this_rq()->wake_for_idle_pull)) + raise_softirq_irqoff(SCHED_SOFTIRQ); +#endif + irq_exit(); } @@ -1487,7 +1496,13 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) unsigned long flags; int cpu, success = 0; - smp_wmb(); + /* + * If we are going to wake up a thread waiting for CONDITION we + * need to ensure that CONDITION=1 done by the caller can not be + * reordered with p->state check below. This pairs with mb() in + * set_current_state() the waiting thread does. + */ + smp_mb__before_spinlock(); raw_spin_lock_irqsave(&p->pi_lock, flags); if (!(p->state & state)) goto out; @@ -1617,6 +1632,20 @@ static void __sched_fork(struct task_struct *p) #if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED) p->se.avg.runnable_avg_period = 0; p->se.avg.runnable_avg_sum = 0; +#ifdef CONFIG_SCHED_HMP + /* keep LOAD_AVG_MAX in sync with fair.c if load avg series is changed */ +#define LOAD_AVG_MAX 47742 + if (p->mm) { + p->se.avg.hmp_last_up_migration = 0; + p->se.avg.hmp_last_down_migration = 0; + p->se.avg.load_avg_ratio = 1023; + p->se.avg.load_avg_contrib = + (1023 * scale_load_down(p->se.load.weight)); + p->se.avg.runnable_avg_period = LOAD_AVG_MAX; + p->se.avg.runnable_avg_sum = LOAD_AVG_MAX; + p->se.avg.usage_avg_sum = LOAD_AVG_MAX; + } +#endif #endif #ifdef CONFIG_SCHEDSTATS memset(&p->se.statistics, 0, sizeof(p->se.statistics)); @@ -2966,6 +2995,12 @@ need_resched: if (sched_feat(HRTICK)) hrtick_clear(rq); + /* + * Make sure that signal_pending_state()->signal_pending() below + * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) + * done by the caller to avoid the race with signal_wake_up(). + */ + smp_mb__before_spinlock(); raw_spin_lock_irq(&rq->lock); switch_count = &prev->nivcsw; @@ -3813,6 +3848,8 @@ static struct task_struct *find_process_by_pid(pid_t pid) return pid ? find_task_by_vpid(pid) : current; } +extern struct cpumask hmp_slow_cpu_mask; + /* Actually do priority change: must hold rq lock. */ static void __setscheduler(struct rq *rq, struct task_struct *p, int policy, int prio) @@ -3822,8 +3859,17 @@ __setscheduler(struct rq *rq, struct task_struct *p, int policy, int prio) p->normal_prio = normal_prio(p); /* we are holding p->pi_lock already */ p->prio = rt_mutex_getprio(p); - if (rt_prio(p->prio)) + if (rt_prio(p->prio)) { p->sched_class = &rt_sched_class; +#ifdef CONFIG_SCHED_HMP + if (!cpumask_empty(&hmp_slow_cpu_mask)) + if (cpumask_equal(&p->cpus_allowed, cpu_all_mask)) { + p->nr_cpus_allowed = + cpumask_weight(&hmp_slow_cpu_mask); + do_set_cpus_allowed(p, &hmp_slow_cpu_mask); + } +#endif + } else p->sched_class = &fair_sched_class; set_load_weight(p); @@ -5258,7 +5304,6 @@ static int __cpuinit sched_cpu_active(struct notifier_block *nfb, unsigned long action, void *hcpu) { switch (action & ~CPU_TASKS_FROZEN) { - case CPU_STARTING: case CPU_DOWN_FAILED: set_cpu_active((long)hcpu, true); return NOTIFY_OK; @@ -7800,7 +7845,12 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) runtime_enabled = quota != RUNTIME_INF; runtime_was_enabled = cfs_b->quota != RUNTIME_INF; - account_cfs_bandwidth_used(runtime_enabled, runtime_was_enabled); + /* + * If we need to toggle cfs_bandwidth_used, off->on must occur + * before making related changes, and on->off must occur afterwards + */ + if (runtime_enabled && !runtime_was_enabled) + cfs_bandwidth_usage_inc(); raw_spin_lock_irq(&cfs_b->lock); cfs_b->period = ns_to_ktime(period); cfs_b->quota = quota; @@ -7826,6 +7876,8 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) unthrottle_cfs_rq(cfs_rq); raw_spin_unlock_irq(&rq->lock); } + if (runtime_was_enabled && !runtime_enabled) + cfs_bandwidth_usage_dec(); out_unlock: mutex_unlock(&cfs_constraints_mutex); diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index 1095e878a46f..b3f0a2783369 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -70,8 +70,7 @@ int cpupri_find(struct cpupri *cp, struct task_struct *p, int idx = 0; int task_pri = convert_prio(p->prio); - if (task_pri >= MAX_RT_PRIO) - return 0; + BUG_ON(task_pri >= CPUPRI_NR_PRIORITIES); for (idx = 0; idx < task_pri; idx++) { struct cpupri_vec *vec = &cp->pri_to_cpu[idx]; diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index b5ccba22603b..c23a8fd36149 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -326,50 +326,50 @@ out: * softirq as those do not count in task exec_runtime any more. */ static void irqtime_account_process_tick(struct task_struct *p, int user_tick, - struct rq *rq) + struct rq *rq, int ticks) { - cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy); + cputime_t scaled = cputime_to_scaled(cputime_one_jiffy); + u64 cputime = (__force u64) cputime_one_jiffy; u64 *cpustat = kcpustat_this_cpu->cpustat; if (steal_account_process_tick()) return; + cputime *= ticks; + scaled *= ticks; + if (irqtime_account_hi_update()) { - cpustat[CPUTIME_IRQ] += (__force u64) cputime_one_jiffy; + cpustat[CPUTIME_IRQ] += cputime; } else if (irqtime_account_si_update()) { - cpustat[CPUTIME_SOFTIRQ] += (__force u64) cputime_one_jiffy; + cpustat[CPUTIME_SOFTIRQ] += cputime; } else if (this_cpu_ksoftirqd() == p) { /* * ksoftirqd time do not get accounted in cpu_softirq_time. * So, we have to handle it separately here. * Also, p->stime needs to be updated for ksoftirqd. */ - __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, - CPUTIME_SOFTIRQ); + __account_system_time(p, cputime, scaled, CPUTIME_SOFTIRQ); } else if (user_tick) { - account_user_time(p, cputime_one_jiffy, one_jiffy_scaled); + account_user_time(p, cputime, scaled); } else if (p == rq->idle) { - account_idle_time(cputime_one_jiffy); + account_idle_time(cputime); } else if (p->flags & PF_VCPU) { /* System time or guest time */ - account_guest_time(p, cputime_one_jiffy, one_jiffy_scaled); + account_guest_time(p, cputime, scaled); } else { - __account_system_time(p, cputime_one_jiffy, one_jiffy_scaled, - CPUTIME_SYSTEM); + __account_system_time(p, cputime, scaled, CPUTIME_SYSTEM); } } static void irqtime_account_idle_ticks(int ticks) { - int i; struct rq *rq = this_rq(); - for (i = 0; i < ticks; i++) - irqtime_account_process_tick(current, 0, rq); + irqtime_account_process_tick(current, 0, rq, ticks); } #else /* CONFIG_IRQ_TIME_ACCOUNTING */ static inline void irqtime_account_idle_ticks(int ticks) {} static inline void irqtime_account_process_tick(struct task_struct *p, int user_tick, - struct rq *rq) {} + struct rq *rq, int nr_ticks) {} #endif /* CONFIG_IRQ_TIME_ACCOUNTING */ /* @@ -464,7 +464,7 @@ void account_process_tick(struct task_struct *p, int user_tick) return; if (sched_clock_irqtime) { - irqtime_account_process_tick(p, user_tick, rq); + irqtime_account_process_tick(p, user_tick, rq, 1); return; } @@ -558,7 +558,7 @@ static void cputime_adjust(struct task_cputime *curr, struct cputime *prev, cputime_t *ut, cputime_t *st) { - cputime_t rtime, stime, utime, total; + cputime_t rtime, stime, utime; if (vtime_accounting_enabled()) { *ut = curr->utime; @@ -566,9 +566,6 @@ static void cputime_adjust(struct task_cputime *curr, return; } - stime = curr->stime; - total = stime + curr->utime; - /* * Tick based cputime accounting depend on random scheduling * timeslices of a task to be interrupted or not by the timer. @@ -589,13 +586,19 @@ static void cputime_adjust(struct task_cputime *curr, if (prev->stime + prev->utime >= rtime) goto out; - if (total) { + stime = curr->stime; + utime = curr->utime; + + if (utime == 0) { + stime = rtime; + } else if (stime == 0) { + utime = rtime; + } else { + cputime_t total = stime + utime; + stime = scale_stime((__force u64)stime, (__force u64)rtime, (__force u64)total); utime = rtime - stime; - } else { - stime = rtime; - utime = 0; } /* diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 75024a673520..c2665cd29594 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -94,6 +94,7 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group #ifdef CONFIG_SMP P(se->avg.runnable_avg_sum); P(se->avg.runnable_avg_period); + P(se->avg.usage_avg_sum); P(se->avg.load_avg_contrib); P(se->avg.decay_count); #endif @@ -223,6 +224,16 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) cfs_rq->tg_runnable_contrib); SEQ_printf(m, " .%-30s: %d\n", "tg->runnable_avg", atomic_read(&cfs_rq->tg->runnable_avg)); + SEQ_printf(m, " .%-30s: %d\n", "tg->usage_avg", + atomic_read(&cfs_rq->tg->usage_avg)); +#endif +#ifdef CONFIG_CFS_BANDWIDTH + SEQ_printf(m, " .%-30s: %d\n", "tg->cfs_bandwidth.timer_active", + cfs_rq->tg->cfs_bandwidth.timer_active); + SEQ_printf(m, " .%-30s: %d\n", "throttled", + cfs_rq->throttled); + SEQ_printf(m, " .%-30s: %d\n", "throttle_count", + cfs_rq->throttle_count); #endif print_cfs_group_stats(m, cpu, cfs_rq->tg); @@ -566,6 +577,12 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) "nr_involuntary_switches", (long long)p->nivcsw); P(se.load.weight); +#if defined(CONFIG_SMP) && defined(CONFIG_FAIR_GROUP_SCHED) + P(se.avg.runnable_avg_sum); + P(se.avg.runnable_avg_period); + P(se.avg.load_avg_contrib); + P(se.avg.decay_count); +#endif P(policy); P(prio); #undef PN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c61a614465c8..71da724449ee 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -31,9 +31,21 @@ #include <linux/task_work.h> #include <trace/events/sched.h> +#include <linux/sysfs.h> +#include <linux/vmalloc.h> +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE +/* Include cpufreq header to add a notifier so that cpu frequency + * scaling can track the current CPU frequency + */ +#include <linux/cpufreq.h> +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ +#ifdef CONFIG_SCHED_HMP +#include <linux/cpuidle.h> +#endif #include "sched.h" + /* * Targeted preemption latency for CPU-bound tasks: * (default: 6ms * (1 + ilog(ncpus)), units: nanoseconds) @@ -936,6 +948,13 @@ void task_numa_work(struct callback_head *work) if (vma->vm_end - vma->vm_start < HPAGE_SIZE) continue; + /* + * Skip inaccessible VMAs to avoid any confusion between + * PROT_NONE and NUMA hinting ptes + */ + if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))) + continue; + do { start = max(start, vma->vm_start); end = ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE); @@ -1201,8 +1220,91 @@ static u32 __compute_runnable_contrib(u64 n) return contrib + runnable_avg_yN_sum[n]; } -/* - * We can represent the historical contribution to runnable average as the +#ifdef CONFIG_SCHED_HMP +#define HMP_VARIABLE_SCALE_SHIFT 16ULL +struct hmp_global_attr { + struct attribute attr; + ssize_t (*show)(struct kobject *kobj, + struct attribute *attr, char *buf); + ssize_t (*store)(struct kobject *a, struct attribute *b, + const char *c, size_t count); + int *value; + int (*to_sysfs)(int); + int (*from_sysfs)(int); + ssize_t (*to_sysfs_text)(char *buf, int buf_size); +}; + +#define HMP_DATA_SYSFS_MAX 8 + +struct hmp_data_struct { +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + int freqinvar_load_scale_enabled; +#endif + int multiplier; /* used to scale the time delta */ + struct attribute_group attr_group; + struct attribute *attributes[HMP_DATA_SYSFS_MAX + 1]; + struct hmp_global_attr attr[HMP_DATA_SYSFS_MAX]; +} hmp_data; + +static u64 hmp_variable_scale_convert(u64 delta); +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE +/* Frequency-Invariant Load Modification: + * Loads are calculated as in PJT's patch however we also scale the current + * contribution in line with the frequency of the CPU that the task was + * executed on. + * In this version, we use a simple linear scale derived from the maximum + * frequency reported by CPUFreq. As an example: + * + * Consider that we ran a task for 100% of the previous interval. + * + * Our CPU was under asynchronous frequency control through one of the + * CPUFreq governors. + * + * The CPUFreq governor reports that it is able to scale the CPU between + * 500MHz and 1GHz. + * + * During the period, the CPU was running at 1GHz. + * + * In this case, our load contribution for that period is calculated as + * 1 * (number_of_active_microseconds) + * + * This results in our task being able to accumulate maximum load as normal. + * + * + * Consider now that our CPU was executing at 500MHz. + * + * We now scale the load contribution such that it is calculated as + * 0.5 * (number_of_active_microseconds) + * + * Our task can only record 50% maximum load during this period. + * + * This represents the task consuming 50% of the CPU's *possible* compute + * capacity. However the task did consume 100% of the CPU's *available* + * compute capacity which is the value seen by the CPUFreq governor and + * user-side CPU Utilization tools. + * + * Restricting tracked load to be scaled by the CPU's frequency accurately + * represents the consumption of possible compute capacity and allows the + * HMP migration's simple threshold migration strategy to interact more + * predictably with CPUFreq's asynchronous compute capacity changes. + */ +#define SCHED_FREQSCALE_SHIFT 10 +struct cpufreq_extents { + u32 curr_scale; + u32 min; + u32 max; + u32 flags; +}; +/* Flag set when the governor in use only allows one frequency. + * Disables scaling. + */ +#define SCHED_LOAD_FREQINVAR_SINGLEFREQ 0x01 + +static struct cpufreq_extents freq_scale[CONFIG_NR_CPUS]; +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ +#endif /* CONFIG_SCHED_HMP */ + +/* We can represent the historical contribution to runnable average as the * coefficients of a geometric series. To do this we sub-divide our runnable * history into segments of approximately 1ms (1024us); label the segment that * occurred N-ms ago p_N, with p_0 corresponding to the current period, e.g. @@ -1231,13 +1333,24 @@ static u32 __compute_runnable_contrib(u64 n) */ static __always_inline int __update_entity_runnable_avg(u64 now, struct sched_avg *sa, - int runnable) + int runnable, + int running, + int cpu) { u64 delta, periods; u32 runnable_contrib; int delta_w, decayed = 0; +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + u64 scaled_delta; + u32 scaled_runnable_contrib; + int scaled_delta_w; + u32 curr_scale = 1024; +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ delta = now - sa->last_runnable_update; +#ifdef CONFIG_SCHED_HMP + delta = hmp_variable_scale_convert(delta); +#endif /* * This should only happen when time goes backwards, which it * unfortunately does during sched clock init when we swap over to TSC. @@ -1256,6 +1369,12 @@ static __always_inline int __update_entity_runnable_avg(u64 now, return 0; sa->last_runnable_update = now; +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + /* retrieve scale factor for load */ + if (hmp_data.freqinvar_load_scale_enabled) + curr_scale = freq_scale[cpu].curr_scale; +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ + /* delta_w is the amount already accumulated against our next period */ delta_w = sa->runnable_avg_period % 1024; if (delta + delta_w >= 1024) { @@ -1268,8 +1387,20 @@ static __always_inline int __update_entity_runnable_avg(u64 now, * period and accrue it. */ delta_w = 1024 - delta_w; + /* scale runnable time if necessary */ +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + scaled_delta_w = (delta_w * curr_scale) + >> SCHED_FREQSCALE_SHIFT; + if (runnable) + sa->runnable_avg_sum += scaled_delta_w; + if (running) + sa->usage_avg_sum += scaled_delta_w; +#else if (runnable) sa->runnable_avg_sum += delta_w; + if (running) + sa->usage_avg_sum += delta_w; +#endif /* #ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ sa->runnable_avg_period += delta_w; delta -= delta_w; @@ -1277,22 +1408,49 @@ static __always_inline int __update_entity_runnable_avg(u64 now, /* Figure out how many additional periods this update spans */ periods = delta / 1024; delta %= 1024; - + /* decay the load we have accumulated so far */ sa->runnable_avg_sum = decay_load(sa->runnable_avg_sum, periods + 1); sa->runnable_avg_period = decay_load(sa->runnable_avg_period, periods + 1); - + sa->usage_avg_sum = decay_load(sa->usage_avg_sum, periods + 1); + /* add the contribution from this period */ /* Efficiently calculate \sum (1..n_period) 1024*y^i */ runnable_contrib = __compute_runnable_contrib(periods); + /* Apply load scaling if necessary. + * Note that multiplying the whole series is same as + * multiplying all terms + */ +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + scaled_runnable_contrib = (runnable_contrib * curr_scale) + >> SCHED_FREQSCALE_SHIFT; + if (runnable) + sa->runnable_avg_sum += scaled_runnable_contrib; + if (running) + sa->usage_avg_sum += scaled_runnable_contrib; +#else if (runnable) sa->runnable_avg_sum += runnable_contrib; + if (running) + sa->usage_avg_sum += runnable_contrib; +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ sa->runnable_avg_period += runnable_contrib; } /* Remainder of delta accrued against u_0` */ + /* scale if necessary */ +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + scaled_delta = ((delta * curr_scale) >> SCHED_FREQSCALE_SHIFT); + if (runnable) + sa->runnable_avg_sum += scaled_delta; + if (running) + sa->usage_avg_sum += scaled_delta; +#else if (runnable) sa->runnable_avg_sum += delta; + if (running) + sa->usage_avg_sum += delta; +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ sa->runnable_avg_period += delta; return decayed; @@ -1305,12 +1463,9 @@ static inline u64 __synchronize_entity_decay(struct sched_entity *se) u64 decays = atomic64_read(&cfs_rq->decay_counter); decays -= se->avg.decay_count; - if (!decays) - return 0; - - se->avg.load_avg_contrib = decay_load(se->avg.load_avg_contrib, decays); + if (decays) + se->avg.load_avg_contrib = decay_load(se->avg.load_avg_contrib, decays); se->avg.decay_count = 0; - return decays; } @@ -1338,16 +1493,28 @@ static inline void __update_tg_runnable_avg(struct sched_avg *sa, struct cfs_rq *cfs_rq) { struct task_group *tg = cfs_rq->tg; - long contrib; + long contrib, usage_contrib; /* The fraction of a cpu used by this cfs_rq */ contrib = div_u64(sa->runnable_avg_sum << NICE_0_SHIFT, sa->runnable_avg_period + 1); contrib -= cfs_rq->tg_runnable_contrib; - if (abs(contrib) > cfs_rq->tg_runnable_contrib / 64) { + usage_contrib = div_u64(sa->usage_avg_sum << NICE_0_SHIFT, + sa->runnable_avg_period + 1); + usage_contrib -= cfs_rq->tg_usage_contrib; + + /* + * contrib/usage at this point represent deltas, only update if they + * are substantive. + */ + if ((abs(contrib) > cfs_rq->tg_runnable_contrib / 64) || + (abs(usage_contrib) > cfs_rq->tg_usage_contrib / 64)) { atomic_add(contrib, &tg->runnable_avg); cfs_rq->tg_runnable_contrib += contrib; + + atomic_add(usage_contrib, &tg->usage_avg); + cfs_rq->tg_usage_contrib += usage_contrib; } } @@ -1408,12 +1575,18 @@ static inline void __update_task_entity_contrib(struct sched_entity *se) contrib = se->avg.runnable_avg_sum * scale_load_down(se->load.weight); contrib /= (se->avg.runnable_avg_period + 1); se->avg.load_avg_contrib = scale_load(contrib); + trace_sched_task_load_contrib(task_of(se), se->avg.load_avg_contrib); + contrib = se->avg.runnable_avg_sum * scale_load_down(NICE_0_LOAD); + contrib /= (se->avg.runnable_avg_period + 1); + se->avg.load_avg_ratio = scale_load(contrib); + trace_sched_task_runnable_ratio(task_of(se), se->avg.load_avg_ratio); } /* Compute the current contribution to load_avg by se, return any delta */ -static long __update_entity_load_avg_contrib(struct sched_entity *se) +static long __update_entity_load_avg_contrib(struct sched_entity *se, long *ratio) { long old_contrib = se->avg.load_avg_contrib; + long old_ratio = se->avg.load_avg_ratio; if (entity_is_task(se)) { __update_task_entity_contrib(se); @@ -1422,6 +1595,8 @@ static long __update_entity_load_avg_contrib(struct sched_entity *se) __update_group_entity_contrib(se); } + if (ratio) + *ratio = se->avg.load_avg_ratio - old_ratio; return se->avg.load_avg_contrib - old_contrib; } @@ -1441,9 +1616,13 @@ static inline void update_entity_load_avg(struct sched_entity *se, int update_cfs_rq) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - long contrib_delta; + long contrib_delta, ratio_delta; u64 now; + int cpu = -1; /* not used in normal case */ +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + cpu = cfs_rq->rq->cpu; +#endif /* * For a group entity we need to use their owned cfs_rq_clock_task() in * case they are the parent of a throttled hierarchy. @@ -1453,18 +1632,21 @@ static inline void update_entity_load_avg(struct sched_entity *se, else now = cfs_rq_clock_task(group_cfs_rq(se)); - if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq)) + if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq, + cfs_rq->curr == se, cpu)) return; - contrib_delta = __update_entity_load_avg_contrib(se); + contrib_delta = __update_entity_load_avg_contrib(se, &ratio_delta); if (!update_cfs_rq) return; - if (se->on_rq) + if (se->on_rq) { cfs_rq->runnable_load_avg += contrib_delta; - else + rq_of(cfs_rq)->avg.load_avg_ratio += ratio_delta; + } else { subtract_blocked_load_contrib(cfs_rq, -contrib_delta); + } } /* @@ -1497,8 +1679,17 @@ static void update_cfs_rq_blocked_load(struct cfs_rq *cfs_rq, int force_update) static inline void update_rq_runnable_avg(struct rq *rq, int runnable) { - __update_entity_runnable_avg(rq->clock_task, &rq->avg, runnable); + int cpu = -1; /* not used in normal case */ + +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + cpu = rq->cpu; +#endif + __update_entity_runnable_avg(rq->clock_task, &rq->avg, runnable, + runnable, cpu); __update_tg_runnable_avg(&rq->avg, &rq->cfs); + trace_sched_rq_runnable_ratio(cpu_of(rq), rq->avg.load_avg_ratio); + trace_sched_rq_runnable_load(cpu_of(rq), rq->cfs.runnable_load_avg); + trace_sched_rq_nr_running(cpu_of(rq), rq->nr_running, rq->nr_iowait.counter); } /* Add the load generated by se into cfs_rq's child load-average */ @@ -1540,6 +1731,8 @@ static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq, } cfs_rq->runnable_load_avg += se->avg.load_avg_contrib; + rq_of(cfs_rq)->avg.load_avg_ratio += se->avg.load_avg_ratio; + /* we force update consideration on load-balancer moves */ update_cfs_rq_blocked_load(cfs_rq, !wakeup); } @@ -1558,6 +1751,8 @@ static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq, update_cfs_rq_blocked_load(cfs_rq, !sleep); cfs_rq->runnable_load_avg -= se->avg.load_avg_contrib; + rq_of(cfs_rq)->avg.load_avg_ratio -= se->avg.load_avg_ratio; + if (sleep) { cfs_rq->blocked_load_avg += se->avg.load_avg_contrib; se->avg.decay_count = atomic64_read(&cfs_rq->decay_counter); @@ -1886,6 +2081,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) */ update_stats_wait_end(cfs_rq, se); __dequeue_entity(cfs_rq, se); + update_entity_load_avg(se, 1); } update_stats_curr_start(cfs_rq, se); @@ -1984,6 +2180,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) */ update_entity_load_avg(curr, 1); update_cfs_rq_blocked_load(cfs_rq, 1); + update_cfs_shares(cfs_rq); #ifdef CONFIG_SCHED_HRTICK /* @@ -2021,13 +2218,14 @@ static inline bool cfs_bandwidth_used(void) return static_key_false(&__cfs_bandwidth_used); } -void account_cfs_bandwidth_used(int enabled, int was_enabled) +void cfs_bandwidth_usage_inc(void) +{ + static_key_slow_inc(&__cfs_bandwidth_used); +} + +void cfs_bandwidth_usage_dec(void) { - /* only need to count groups transitioning between enabled/!enabled */ - if (enabled && !was_enabled) - static_key_slow_inc(&__cfs_bandwidth_used); - else if (!enabled && was_enabled) - static_key_slow_dec(&__cfs_bandwidth_used); + static_key_slow_dec(&__cfs_bandwidth_used); } #else /* HAVE_JUMP_LABEL */ static bool cfs_bandwidth_used(void) @@ -2035,7 +2233,8 @@ static bool cfs_bandwidth_used(void) return true; } -void account_cfs_bandwidth_used(int enabled, int was_enabled) {} +void cfs_bandwidth_usage_inc(void) {} +void cfs_bandwidth_usage_dec(void) {} #endif /* HAVE_JUMP_LABEL */ /* @@ -2287,6 +2486,8 @@ static void throttle_cfs_rq(struct cfs_rq *cfs_rq) cfs_rq->throttled_clock = rq->clock; raw_spin_lock(&cfs_b->lock); list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); + if (!cfs_b->timer_active) + __start_cfs_bandwidth(cfs_b); raw_spin_unlock(&cfs_b->lock); } @@ -2398,6 +2599,13 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) if (idle) goto out_unlock; + /* + * if we have relooped after returning idle once, we need to update our + * status as actually running, so that other cpus doing + * __start_cfs_bandwidth will stop trying to cancel us. + */ + cfs_b->timer_active = 1; + __refill_cfs_bandwidth_runtime(cfs_b); if (!throttled) { @@ -2458,7 +2666,13 @@ static const u64 min_bandwidth_expiration = 2 * NSEC_PER_MSEC; /* how long we wait to gather additional slack before distributing */ static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC; -/* are we near the end of the current quota period? */ +/* + * Are we near the end of the current quota period? + * + * Requires cfs_b->lock for hrtimer_expires_remaining to be safe against the + * hrtimer base being cleared by __hrtimer_start_range_ns. In the case of + * migrate_hrtimers, base is never cleared, so we are fine. + */ static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) { struct hrtimer *refresh_timer = &cfs_b->period_timer; @@ -2534,10 +2748,12 @@ static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) u64 expires; /* confirm we're still not at a refresh boundary */ - if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) + raw_spin_lock(&cfs_b->lock); + if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) { + raw_spin_unlock(&cfs_b->lock); return; + } - raw_spin_lock(&cfs_b->lock); if (cfs_b->quota != RUNTIME_INF && cfs_b->runtime > slice) { runtime = cfs_b->runtime; cfs_b->runtime = 0; @@ -2662,11 +2878,11 @@ void __start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) * (timer_active==0 becomes visible before the hrtimer call-back * terminates). In either case we ensure that it's re-programmed */ - while (unlikely(hrtimer_active(&cfs_b->period_timer))) { + while (unlikely(hrtimer_active(&cfs_b->period_timer)) && + hrtimer_try_to_cancel(&cfs_b->period_timer) < 0) { + /* bounce the lock to allow do_sched_cfs_period_timer to run */ raw_spin_unlock(&cfs_b->lock); - /* ensure cfs_b->lock is available while we wait */ - hrtimer_cancel(&cfs_b->period_timer); - + cpu_relax(); raw_spin_lock(&cfs_b->lock); /* if someone else restarted the timer then we're done */ if (cfs_b->timer_active) @@ -3314,6 +3530,834 @@ done: return target; } +#ifdef CONFIG_SCHED_HMP +/* + * Heterogenous multiprocessor (HMP) optimizations + * + * The cpu types are distinguished using a list of hmp_domains + * which each represent one cpu type using a cpumask. + * The list is assumed ordered by compute capacity with the + * fastest domain first. + */ +DEFINE_PER_CPU(struct hmp_domain *, hmp_cpu_domain); +static const int hmp_max_tasks = 5; + +extern void __init arch_get_hmp_domains(struct list_head *hmp_domains_list); + +#ifdef CONFIG_CPU_IDLE +/* + * hmp_idle_pull: + * + * In this version we have stopped using forced up migrations when we + * detect that a task running on a little CPU should be moved to a bigger + * CPU. In most cases, the bigger CPU is in a deep sleep state and a forced + * migration means we stop the task immediately but need to wait for the + * target CPU to wake up before we can restart the task which is being + * moved. Instead, we now wake a big CPU with an IPI and ask it to pull + * a task when ready. This allows the task to continue executing on its + * current CPU, reducing the amount of time that the task is stalled for. + * + * keepalive timers: + * + * The keepalive timer is used as a way to keep a CPU engaged in an + * idle pull operation out of idle while waiting for the source + * CPU to stop and move the task. Ideally this would not be necessary + * and we could impose a temporary zero-latency requirement on the + * current CPU, but in the current QoS framework this will result in + * all CPUs in the system being unable to enter idle states which is + * not desirable. The timer does not perform any work when it expires. + */ +struct hmp_keepalive { + bool init; + ktime_t delay; /* if zero, no need for timer */ + struct hrtimer timer; +}; +DEFINE_PER_CPU(struct hmp_keepalive, hmp_cpu_keepalive); + +/* setup per-cpu keepalive timers */ +static enum hrtimer_restart hmp_cpu_keepalive_notify(struct hrtimer *hrtimer) +{ + return HRTIMER_NORESTART; +} + +/* + * Work out if any of the idle states have an exit latency too high for us. + * ns_delay is passed in containing the max we are willing to tolerate. + * If there are none, set ns_delay to zero. + * If there are any, set ns_delay to + * ('target_residency of state with shortest too-big latency' - 1) * 1000. + */ +static void hmp_keepalive_delay(unsigned int *ns_delay) +{ + struct cpuidle_driver *drv; + drv = cpuidle_driver_ref(); + if (drv) { + unsigned int us_delay = UINT_MAX; + unsigned int us_max_delay = *ns_delay / 1000; + int idx; + /* if cpuidle states are guaranteed to be sorted we + * could stop at the first match. + */ + for (idx = 0; idx < drv->state_count; idx++) { + if (drv->states[idx].exit_latency > us_max_delay && + drv->states[idx].target_residency < us_delay) { + us_delay = drv->states[idx].target_residency; + } + } + if (us_delay == UINT_MAX) + *ns_delay = 0; /* no timer required */ + else + *ns_delay = 1000 * (us_delay - 1); + } + cpuidle_driver_unref(); +} + +static void hmp_cpu_keepalive_trigger(void) +{ + int cpu = smp_processor_id(); + struct hmp_keepalive *keepalive = &per_cpu(hmp_cpu_keepalive, cpu); + if (!keepalive->init) { + unsigned int ns_delay = 100000; /* tolerate 100usec delay */ + + hrtimer_init(&keepalive->timer, + CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED); + keepalive->timer.function = hmp_cpu_keepalive_notify; + + hmp_keepalive_delay(&ns_delay); + keepalive->delay = ns_to_ktime(ns_delay); + keepalive->init = true; + } + if (ktime_to_ns(keepalive->delay)) + hrtimer_start(&keepalive->timer, + keepalive->delay, HRTIMER_MODE_REL_PINNED); +} + +static void hmp_cpu_keepalive_cancel(int cpu) +{ + struct hmp_keepalive *keepalive = &per_cpu(hmp_cpu_keepalive, cpu); + if (keepalive->init) + hrtimer_cancel(&keepalive->timer); +} +#else /* !CONFIG_CPU_IDLE */ +static void hmp_cpu_keepalive_trigger(void) +{ +} + +static void hmp_cpu_keepalive_cancel(int cpu) +{ +} +#endif + +/* Setup hmp_domains */ +static int __init hmp_cpu_mask_setup(void) +{ + char buf[64]; + struct hmp_domain *domain; + struct list_head *pos; + int dc, cpu; + + pr_debug("Initializing HMP scheduler:\n"); + + /* Initialize hmp_domains using platform code */ + arch_get_hmp_domains(&hmp_domains); + if (list_empty(&hmp_domains)) { + pr_debug("HMP domain list is empty!\n"); + return 0; + } + + /* Print hmp_domains */ + dc = 0; + list_for_each(pos, &hmp_domains) { + domain = list_entry(pos, struct hmp_domain, hmp_domains); + cpulist_scnprintf(buf, 64, &domain->possible_cpus); + pr_debug(" HMP domain %d: %s\n", dc, buf); + + for_each_cpu_mask(cpu, domain->possible_cpus) { + per_cpu(hmp_cpu_domain, cpu) = domain; + } + dc++; + } + + return 1; +} + +static struct hmp_domain *hmp_get_hmp_domain_for_cpu(int cpu) +{ + struct hmp_domain *domain; + struct list_head *pos; + + list_for_each(pos, &hmp_domains) { + domain = list_entry(pos, struct hmp_domain, hmp_domains); + if(cpumask_test_cpu(cpu, &domain->possible_cpus)) + return domain; + } + return NULL; +} + +static void hmp_online_cpu(int cpu) +{ + struct hmp_domain *domain = hmp_get_hmp_domain_for_cpu(cpu); + + if(domain) + cpumask_set_cpu(cpu, &domain->cpus); +} + +static void hmp_offline_cpu(int cpu) +{ + struct hmp_domain *domain = hmp_get_hmp_domain_for_cpu(cpu); + + if(domain) + cpumask_clear_cpu(cpu, &domain->cpus); + + hmp_cpu_keepalive_cancel(cpu); +} +/* + * Needed to determine heaviest tasks etc. + */ +static inline unsigned int hmp_cpu_is_fastest(int cpu); +static inline unsigned int hmp_cpu_is_slowest(int cpu); +static inline struct hmp_domain *hmp_slower_domain(int cpu); +static inline struct hmp_domain *hmp_faster_domain(int cpu); + +/* must hold runqueue lock for queue se is currently on */ +static struct sched_entity *hmp_get_heaviest_task( + struct sched_entity *se, int target_cpu) +{ + int num_tasks = hmp_max_tasks; + struct sched_entity *max_se = se; + unsigned long int max_ratio = se->avg.load_avg_ratio; + const struct cpumask *hmp_target_mask = NULL; + struct hmp_domain *hmp; + + if (hmp_cpu_is_fastest(cpu_of(se->cfs_rq->rq))) + return max_se; + + hmp = hmp_faster_domain(cpu_of(se->cfs_rq->rq)); + hmp_target_mask = &hmp->cpus; + if (target_cpu >= 0) { + /* idle_balance gets run on a CPU while + * it is in the middle of being hotplugged + * out. Bail early in that case. + */ + if(!cpumask_test_cpu(target_cpu, hmp_target_mask)) + return NULL; + hmp_target_mask = cpumask_of(target_cpu); + } + /* The currently running task is not on the runqueue */ + se = __pick_first_entity(cfs_rq_of(se)); + + while (num_tasks && se) { + if (entity_is_task(se) && + se->avg.load_avg_ratio > max_ratio && + cpumask_intersects(hmp_target_mask, + tsk_cpus_allowed(task_of(se)))) { + max_se = se; + max_ratio = se->avg.load_avg_ratio; + } + se = __pick_next_entity(se); + num_tasks--; + } + return max_se; +} + +static struct sched_entity *hmp_get_lightest_task( + struct sched_entity *se, int migrate_down) +{ + int num_tasks = hmp_max_tasks; + struct sched_entity *min_se = se; + unsigned long int min_ratio = se->avg.load_avg_ratio; + const struct cpumask *hmp_target_mask = NULL; + + if (migrate_down) { + struct hmp_domain *hmp; + if (hmp_cpu_is_slowest(cpu_of(se->cfs_rq->rq))) + return min_se; + hmp = hmp_slower_domain(cpu_of(se->cfs_rq->rq)); + hmp_target_mask = &hmp->cpus; + } + /* The currently running task is not on the runqueue */ + se = __pick_first_entity(cfs_rq_of(se)); + + while (num_tasks && se) { + if (entity_is_task(se) && + (se->avg.load_avg_ratio < min_ratio && + hmp_target_mask && + cpumask_intersects(hmp_target_mask, + tsk_cpus_allowed(task_of(se))))) { + min_se = se; + min_ratio = se->avg.load_avg_ratio; + } + se = __pick_next_entity(se); + num_tasks--; + } + return min_se; +} + +/* + * Migration thresholds should be in the range [0..1023] + * hmp_up_threshold: min. load required for migrating tasks to a faster cpu + * hmp_down_threshold: max. load allowed for tasks migrating to a slower cpu + * + * hmp_up_prio: Only up migrate task with high priority (<hmp_up_prio) + * hmp_next_up_threshold: Delay before next up migration (1024 ~= 1 ms) + * hmp_next_down_threshold: Delay before next down migration (1024 ~= 1 ms) + * + * Small Task Packing: + * We can choose to fill the littlest CPUs in an HMP system rather than + * the typical spreading mechanic. This behavior is controllable using + * two variables. + * hmp_packing_enabled: runtime control over pack/spread + * hmp_full_threshold: Consider a CPU with this much unweighted load full + */ +unsigned int hmp_up_threshold = 700; +unsigned int hmp_down_threshold = 512; +#ifdef CONFIG_SCHED_HMP_PRIO_FILTER +unsigned int hmp_up_prio = NICE_TO_PRIO(CONFIG_SCHED_HMP_PRIO_FILTER_VAL); +#endif +unsigned int hmp_next_up_threshold = 4096; +unsigned int hmp_next_down_threshold = 4096; + +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING +/* + * Set the default packing threshold to try to keep little + * CPUs at no more than 80% of their maximum frequency if only + * packing a small number of small tasks. Bigger tasks will + * raise frequency as normal. + * In order to pack a task onto a CPU, the sum of the + * unweighted runnable_avg load of existing tasks plus the + * load of the new task must be less than hmp_full_threshold. + * + * This works in conjunction with frequency-invariant load + * and DVFS governors. Since most DVFS governors aim for 80% + * utilisation, we arrive at (0.8*0.8*(max_load=1024))=655 + * and use a value slightly lower to give a little headroom + * in the decision. + * Note that the most efficient frequency is different for + * each system so /sys/kernel/hmp/packing_limit should be + * configured at runtime for any given platform to achieve + * optimal energy usage. Some systems may not benefit from + * packing, so this feature can also be disabled at runtime + * with /sys/kernel/hmp/packing_enable + */ +unsigned int hmp_packing_enabled = 1; +unsigned int hmp_full_threshold = 650; +#endif + +static unsigned int hmp_up_migration(int cpu, int *target_cpu, struct sched_entity *se); +static unsigned int hmp_down_migration(int cpu, struct sched_entity *se); +static inline unsigned int hmp_domain_min_load(struct hmp_domain *hmpd, + int *min_cpu, struct cpumask *affinity); + +static inline struct hmp_domain *hmp_smallest_domain(void) +{ + return list_entry(hmp_domains.prev, struct hmp_domain, hmp_domains); +} + +/* Check if cpu is in fastest hmp_domain */ +static inline unsigned int hmp_cpu_is_fastest(int cpu) +{ + struct list_head *pos; + + pos = &hmp_cpu_domain(cpu)->hmp_domains; + return pos == hmp_domains.next; +} + +/* Check if cpu is in slowest hmp_domain */ +static inline unsigned int hmp_cpu_is_slowest(int cpu) +{ + struct list_head *pos; + + pos = &hmp_cpu_domain(cpu)->hmp_domains; + return list_is_last(pos, &hmp_domains); +} + +/* Next (slower) hmp_domain relative to cpu */ +static inline struct hmp_domain *hmp_slower_domain(int cpu) +{ + struct list_head *pos; + + pos = &hmp_cpu_domain(cpu)->hmp_domains; + return list_entry(pos->next, struct hmp_domain, hmp_domains); +} + +/* Previous (faster) hmp_domain relative to cpu */ +static inline struct hmp_domain *hmp_faster_domain(int cpu) +{ + struct list_head *pos; + + pos = &hmp_cpu_domain(cpu)->hmp_domains; + return list_entry(pos->prev, struct hmp_domain, hmp_domains); +} + +/* + * Selects a cpu in previous (faster) hmp_domain + */ +static inline unsigned int hmp_select_faster_cpu(struct task_struct *tsk, + int cpu) +{ + int lowest_cpu=NR_CPUS; + __always_unused int lowest_ratio; + struct hmp_domain *hmp; + + if (hmp_cpu_is_fastest(cpu)) + hmp = hmp_cpu_domain(cpu); + else + hmp = hmp_faster_domain(cpu); + + lowest_ratio = hmp_domain_min_load(hmp, &lowest_cpu, + tsk_cpus_allowed(tsk)); + + return lowest_cpu; +} + +/* + * Selects a cpu in next (slower) hmp_domain + * Note that cpumask_any_and() returns the first cpu in the cpumask + */ +static inline unsigned int hmp_select_slower_cpu(struct task_struct *tsk, + int cpu) +{ + int lowest_cpu=NR_CPUS; + struct hmp_domain *hmp; + __always_unused int lowest_ratio; + + if (hmp_cpu_is_slowest(cpu)) + hmp = hmp_cpu_domain(cpu); + else + hmp = hmp_slower_domain(cpu); + + lowest_ratio = hmp_domain_min_load(hmp, &lowest_cpu, + tsk_cpus_allowed(tsk)); + + return lowest_cpu; +} +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING +/* + * Select the 'best' candidate little CPU to wake up on. + * Implements a packing strategy which examines CPU in + * logical CPU order, and selects the first which will + * be loaded less than hmp_full_threshold according to + * the sum of the tracked load of the runqueue and the task. + */ +static inline unsigned int hmp_best_little_cpu(struct task_struct *tsk, + int cpu) { + int tmp_cpu; + unsigned long estimated_load; + struct hmp_domain *hmp; + struct sched_avg *avg; + struct cpumask allowed_hmp_cpus; + + if(!hmp_packing_enabled || + tsk->se.avg.load_avg_ratio > ((NICE_0_LOAD * 90)/100)) + return hmp_select_slower_cpu(tsk, cpu); + + if (hmp_cpu_is_slowest(cpu)) + hmp = hmp_cpu_domain(cpu); + else + hmp = hmp_slower_domain(cpu); + + /* respect affinity */ + cpumask_and(&allowed_hmp_cpus, &hmp->cpus, + tsk_cpus_allowed(tsk)); + + for_each_cpu_mask(tmp_cpu, allowed_hmp_cpus) { + avg = &cpu_rq(tmp_cpu)->avg; + /* estimate new rq load if we add this task */ + estimated_load = avg->load_avg_ratio + + tsk->se.avg.load_avg_ratio; + if (estimated_load <= hmp_full_threshold) { + cpu = tmp_cpu; + break; + } + } + /* if no match was found, the task uses the initial value */ + return cpu; +} +#endif +static inline void hmp_next_up_delay(struct sched_entity *se, int cpu) +{ + /* hack - always use clock from first online CPU */ + u64 now = cpu_rq(cpumask_first(cpu_online_mask))->clock_task; + se->avg.hmp_last_up_migration = now; + se->avg.hmp_last_down_migration = 0; + cpu_rq(cpu)->avg.hmp_last_up_migration = now; + cpu_rq(cpu)->avg.hmp_last_down_migration = 0; +} + +static inline void hmp_next_down_delay(struct sched_entity *se, int cpu) +{ + /* hack - always use clock from first online CPU */ + u64 now = cpu_rq(cpumask_first(cpu_online_mask))->clock_task; + se->avg.hmp_last_down_migration = now; + se->avg.hmp_last_up_migration = 0; + cpu_rq(cpu)->avg.hmp_last_down_migration = now; + cpu_rq(cpu)->avg.hmp_last_up_migration = 0; +} + +/* + * Heterogenous multiprocessor (HMP) optimizations + * + * These functions allow to change the growing speed of the load_avg_ratio + * by default it goes from 0 to 0.5 in LOAD_AVG_PERIOD = 32ms + * This can now be changed with /sys/kernel/hmp/load_avg_period_ms. + * + * These functions also allow to change the up and down threshold of HMP + * using /sys/kernel/hmp/{up,down}_threshold. + * Both must be between 0 and 1023. The threshold that is compared + * to the load_avg_ratio is up_threshold/1024 and down_threshold/1024. + * + * For instance, if load_avg_period = 64 and up_threshold = 512, an idle + * task with a load of 0 will reach the threshold after 64ms of busy loop. + * + * Changing load_avg_periods_ms has the same effect than changing the + * default scaling factor Y=1002/1024 in the load_avg_ratio computation to + * (1002/1024.0)^(LOAD_AVG_PERIOD/load_avg_period_ms), but the last one + * could trigger overflows. + * For instance, with Y = 1023/1024 in __update_task_entity_contrib() + * "contrib = se->avg.runnable_avg_sum * scale_load_down(se->load.weight);" + * could be overflowed for a weight > 2^12 even is the load_avg_contrib + * should still be a 32bits result. This would not happen by multiplicating + * delta time by 1/22 and setting load_avg_period_ms = 706. + */ + +/* + * By scaling the delta time it end-up increasing or decrease the + * growing speed of the per entity load_avg_ratio + * The scale factor hmp_data.multiplier is a fixed point + * number: (32-HMP_VARIABLE_SCALE_SHIFT).HMP_VARIABLE_SCALE_SHIFT + */ +static inline u64 hmp_variable_scale_convert(u64 delta) +{ +#ifdef CONFIG_HMP_VARIABLE_SCALE + u64 high = delta >> 32ULL; + u64 low = delta & 0xffffffffULL; + low *= hmp_data.multiplier; + high *= hmp_data.multiplier; + return (low >> HMP_VARIABLE_SCALE_SHIFT) + + (high << (32ULL - HMP_VARIABLE_SCALE_SHIFT)); +#else + return delta; +#endif +} + +static ssize_t hmp_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct hmp_global_attr *hmp_attr = + container_of(attr, struct hmp_global_attr, attr); + int temp; + + if (hmp_attr->to_sysfs_text != NULL) + return hmp_attr->to_sysfs_text(buf, PAGE_SIZE); + + temp = *(hmp_attr->value); + if (hmp_attr->to_sysfs != NULL) + temp = hmp_attr->to_sysfs(temp); + + return (ssize_t)sprintf(buf, "%d\n", temp); +} + +static ssize_t hmp_store(struct kobject *a, struct attribute *attr, + const char *buf, size_t count) +{ + int temp; + ssize_t ret = count; + struct hmp_global_attr *hmp_attr = + container_of(attr, struct hmp_global_attr, attr); + char *str = vmalloc(count + 1); + if (str == NULL) + return -ENOMEM; + memcpy(str, buf, count); + str[count] = 0; + if (sscanf(str, "%d", &temp) < 1) + ret = -EINVAL; + else { + if (hmp_attr->from_sysfs != NULL) + temp = hmp_attr->from_sysfs(temp); + if (temp < 0) + ret = -EINVAL; + else + *(hmp_attr->value) = temp; + } + vfree(str); + return ret; +} + +static ssize_t hmp_print_domains(char *outbuf, int outbufsize) +{ + char buf[64]; + const char nospace[] = "%s", space[] = " %s"; + const char *fmt = nospace; + struct hmp_domain *domain; + struct list_head *pos; + int outpos = 0; + list_for_each(pos, &hmp_domains) { + domain = list_entry(pos, struct hmp_domain, hmp_domains); + if (cpumask_scnprintf(buf, 64, &domain->possible_cpus)) { + outpos += sprintf(outbuf+outpos, fmt, buf); + fmt = space; + } + } + strcat(outbuf, "\n"); + return outpos+1; +} + +#ifdef CONFIG_HMP_VARIABLE_SCALE +static int hmp_period_tofrom_sysfs(int value) +{ + return (LOAD_AVG_PERIOD << HMP_VARIABLE_SCALE_SHIFT) / value; +} +#endif +/* max value for threshold is 1024 */ +static int hmp_theshold_from_sysfs(int value) +{ + if (value > 1024) + return -1; + return value; +} +#if defined(CONFIG_SCHED_HMP_LITTLE_PACKING) || \ + defined(CONFIG_HMP_FREQUENCY_INVARIANT_SCALE) +/* toggle control is only 0,1 off/on */ +static int hmp_toggle_from_sysfs(int value) +{ + if (value < 0 || value > 1) + return -1; + return value; +} +#endif +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING +/* packing value must be non-negative */ +static int hmp_packing_from_sysfs(int value) +{ + if (value < 0) + return -1; + return value; +} +#endif +static void hmp_attr_add( + const char *name, + int *value, + int (*to_sysfs)(int), + int (*from_sysfs)(int), + ssize_t (*to_sysfs_text)(char *, int), + umode_t mode) +{ + int i = 0; + while (hmp_data.attributes[i] != NULL) { + i++; + if (i >= HMP_DATA_SYSFS_MAX) + return; + } + if (mode) + hmp_data.attr[i].attr.mode = mode; + else + hmp_data.attr[i].attr.mode = 0644; + hmp_data.attr[i].show = hmp_show; + hmp_data.attr[i].store = hmp_store; + hmp_data.attr[i].attr.name = name; + hmp_data.attr[i].value = value; + hmp_data.attr[i].to_sysfs = to_sysfs; + hmp_data.attr[i].from_sysfs = from_sysfs; + hmp_data.attr[i].to_sysfs_text = to_sysfs_text; + hmp_data.attributes[i] = &hmp_data.attr[i].attr; + hmp_data.attributes[i + 1] = NULL; +} + +static int hmp_attr_init(void) +{ + int ret; + memset(&hmp_data, sizeof(hmp_data), 0); + hmp_attr_add("hmp_domains", + NULL, + NULL, + NULL, + hmp_print_domains, + 0444); + hmp_attr_add("up_threshold", + &hmp_up_threshold, + NULL, + hmp_theshold_from_sysfs, + NULL, + 0); + hmp_attr_add("down_threshold", + &hmp_down_threshold, + NULL, + hmp_theshold_from_sysfs, + NULL, + 0); +#ifdef CONFIG_HMP_VARIABLE_SCALE + /* by default load_avg_period_ms == LOAD_AVG_PERIOD + * meaning no change + */ + hmp_data.multiplier = hmp_period_tofrom_sysfs(LOAD_AVG_PERIOD); + hmp_attr_add("load_avg_period_ms", + &hmp_data.multiplier, + hmp_period_tofrom_sysfs, + hmp_period_tofrom_sysfs, + NULL, + 0); +#endif +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE + /* default frequency-invariant scaling ON */ + hmp_data.freqinvar_load_scale_enabled = 1; + hmp_attr_add("frequency_invariant_load_scale", + &hmp_data.freqinvar_load_scale_enabled, + NULL, + hmp_toggle_from_sysfs, + NULL, + 0); +#endif +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING + hmp_attr_add("packing_enable", + &hmp_packing_enabled, + NULL, + hmp_toggle_from_sysfs, + NULL, + 0); + hmp_attr_add("packing_limit", + &hmp_full_threshold, + NULL, + hmp_packing_from_sysfs, + NULL, + 0); +#endif + hmp_data.attr_group.name = "hmp"; + hmp_data.attr_group.attrs = hmp_data.attributes; + ret = sysfs_create_group(kernel_kobj, + &hmp_data.attr_group); + return 0; +} +late_initcall(hmp_attr_init); +/* + * return the load of the lowest-loaded CPU in a given HMP domain + * min_cpu optionally points to an int to receive the CPU. + * affinity optionally points to a cpumask containing the + * CPUs to be considered. note: + * + min_cpu = NR_CPUS only if no CPUs are in the set of + * affinity && hmp_domain cpus + * + min_cpu will always otherwise equal one of the CPUs in + * the hmp domain + * + when more than one CPU has the same load, the one which + * is least-recently-disturbed by an HMP migration will be + * selected + * + if all CPUs are equally loaded or idle and the times are + * all the same, the first in the set will be used + * + if affinity is not set, cpu_online_mask is used + */ +static inline unsigned int hmp_domain_min_load(struct hmp_domain *hmpd, + int *min_cpu, struct cpumask *affinity) +{ + int cpu; + int min_cpu_runnable_temp = NR_CPUS; + u64 min_target_last_migration = ULLONG_MAX; + u64 curr_last_migration; + unsigned long min_runnable_load = INT_MAX; + unsigned long contrib; + struct sched_avg *avg; + struct cpumask temp_cpumask; + /* + * only look at CPUs allowed if specified, + * otherwise look at all online CPUs in the + * right HMP domain + */ + cpumask_and(&temp_cpumask, &hmpd->cpus, affinity ? affinity : cpu_online_mask); + + for_each_cpu_mask(cpu, temp_cpumask) { + avg = &cpu_rq(cpu)->avg; + /* used for both up and down migration */ + curr_last_migration = avg->hmp_last_up_migration ? + avg->hmp_last_up_migration : avg->hmp_last_down_migration; + + contrib = avg->load_avg_ratio; + /* + * Consider a runqueue completely busy if there is any load + * on it. Definitely not the best for overall fairness, but + * does well in typical Android use cases. + */ + if (contrib) + contrib = 1023; + + if ((contrib < min_runnable_load) || + (contrib == min_runnable_load && + curr_last_migration < min_target_last_migration)) { + /* + * if the load is the same target the CPU with + * the longest time since a migration. + * This is to spread migration load between + * members of a domain more evenly when the + * domain is fully loaded + */ + min_runnable_load = contrib; + min_cpu_runnable_temp = cpu; + min_target_last_migration = curr_last_migration; + } + } + + if (min_cpu) + *min_cpu = min_cpu_runnable_temp; + + return min_runnable_load; +} + +/* + * Calculate the task starvation + * This is the ratio of actually running time vs. runnable time. + * If the two are equal the task is getting the cpu time it needs or + * it is alone on the cpu and the cpu is fully utilized. + */ +static inline unsigned int hmp_task_starvation(struct sched_entity *se) +{ + u32 starvation; + + starvation = se->avg.usage_avg_sum * scale_load_down(NICE_0_LOAD); + starvation /= (se->avg.runnable_avg_sum + 1); + + return scale_load(starvation); +} + +static inline unsigned int hmp_offload_down(int cpu, struct sched_entity *se) +{ + int min_usage; + int dest_cpu = NR_CPUS; + + if (hmp_cpu_is_slowest(cpu)) + return NR_CPUS; + + /* Is there an idle CPU in the current domain */ + min_usage = hmp_domain_min_load(hmp_cpu_domain(cpu), NULL, NULL); + if (min_usage == 0) { + trace_sched_hmp_offload_abort(cpu, min_usage, "load"); + return NR_CPUS; + } + + /* Is the task alone on the cpu? */ + if (cpu_rq(cpu)->cfs.h_nr_running < 2) { + trace_sched_hmp_offload_abort(cpu, + cpu_rq(cpu)->cfs.h_nr_running, "nr_running"); + return NR_CPUS; + } + + /* Is the task actually starving? */ + /* >=25% ratio running/runnable = starving */ + if (hmp_task_starvation(se) > 768) { + trace_sched_hmp_offload_abort(cpu, hmp_task_starvation(se), + "starvation"); + return NR_CPUS; + } + + /* Does the slower domain have any idle CPUs? */ + min_usage = hmp_domain_min_load(hmp_slower_domain(cpu), &dest_cpu, + tsk_cpus_allowed(task_of(se))); + + if (min_usage == 0) { + trace_sched_hmp_offload_succeed(cpu, dest_cpu); + return dest_cpu; + } else + trace_sched_hmp_offload_abort(cpu,min_usage,"slowdomain"); + return NR_CPUS; +} +#endif /* CONFIG_SCHED_HMP */ + /* * sched_balance_self: balance the current task (running on cpu) in domains * that have the 'flag' flag set. In practice, this is SD_BALANCE_FORK and @@ -3338,6 +4382,19 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) if (p->nr_cpus_allowed == 1) return prev_cpu; +#ifdef CONFIG_SCHED_HMP + /* always put non-kernel forking tasks on a big domain */ + if (p->mm && (sd_flag & SD_BALANCE_FORK)) { + new_cpu = hmp_select_faster_cpu(p, prev_cpu); + if (new_cpu != NR_CPUS) { + hmp_next_up_delay(&p->se, new_cpu); + return new_cpu; + } + /* failed to perform HMP fork balance, use normal balance */ + new_cpu = cpu; + } +#endif + if (sd_flag & SD_BALANCE_WAKE) { if (cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) want_affine = 1; @@ -3412,6 +4469,31 @@ select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags) unlock: rcu_read_unlock(); +#ifdef CONFIG_SCHED_HMP + prev_cpu = task_cpu(p); + + if (hmp_up_migration(prev_cpu, &new_cpu, &p->se)) { + hmp_next_up_delay(&p->se, new_cpu); + trace_sched_hmp_migrate(p, new_cpu, HMP_MIGRATE_WAKEUP); + return new_cpu; + } + if (hmp_down_migration(prev_cpu, &p->se)) { +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING + new_cpu = hmp_best_little_cpu(p, prev_cpu); +#else + new_cpu = hmp_select_slower_cpu(p, prev_cpu); +#endif + if (new_cpu != prev_cpu) { + hmp_next_down_delay(&p->se, new_cpu); + trace_sched_hmp_migrate(p, new_cpu, HMP_MIGRATE_WAKEUP); + return new_cpu; + } + } + /* Make sure that the task stays in its previous hmp domain */ + if (!cpumask_test_cpu(new_cpu, &hmp_cpu_domain(prev_cpu)->cpus)) + return prev_cpu; +#endif + return new_cpu; } @@ -3421,6 +4503,16 @@ unlock: * load-balance). */ #ifdef CONFIG_FAIR_GROUP_SCHED + +#ifdef CONFIG_NO_HZ_COMMON +static int nohz_test_cpu(int cpu); +#else +static inline int nohz_test_cpu(int cpu) +{ + return 0; +} +#endif + /* * Called immediately before a task is migrated to a new cpu; task_cpu(p) and * cfs_rq_of(p) references at time of call are still valid and identify the @@ -3440,6 +4532,25 @@ migrate_task_rq_fair(struct task_struct *p, int next_cpu) * be negative here since on-rq tasks have decay-count == 0. */ if (se->avg.decay_count) { + /* + * If we migrate a sleeping task away from a CPU + * which has the tick stopped, then both the clock_task + * and decay_counter will be out of date for that CPU + * and we will not decay load correctly. + */ + if (!se->on_rq && nohz_test_cpu(task_cpu(p))) { + struct rq *rq = cpu_rq(task_cpu(p)); + unsigned long flags; + /* + * Current CPU cannot be holding rq->lock in this + * circumstance, but another might be. We must hold + * rq->lock before we go poking around in its clocks + */ + raw_spin_lock_irqsave(&rq->lock, flags); + update_rq_clock(rq); + update_cfs_rq_blocked_load(cfs_rq, 0); + raw_spin_unlock_irqrestore(&rq->lock, flags); + } se->avg.decay_count = -__synchronize_entity_decay(se); atomic64_add(se->avg.load_avg_contrib, &cfs_rq->removed_load); } @@ -3945,7 +5056,6 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) * 1) task is cache cold, or * 2) too many balance attempts have failed. */ - tsk_cache_hot = task_hot(p, env->src_rq->clock_task, env->sd); if (!tsk_cache_hot || env->sd->nr_balance_failed > env->sd->cache_nice_tries) { @@ -5231,6 +6341,16 @@ out: return ld_moved; } +#ifdef CONFIG_SCHED_HMP +static unsigned int hmp_idle_pull(int this_cpu); +static int move_specific_task(struct lb_env *env, struct task_struct *pm); +#else +static int move_specific_task(struct lb_env *env, struct task_struct *pm) +{ + return 0; +} +#endif + /* * idle_balance is called by schedule() if this_cpu is about to become * idle. Attempts to pull tasks from other CPUs. @@ -5275,7 +6395,10 @@ void idle_balance(int this_cpu, struct rq *this_rq) } } rcu_read_unlock(); - +#ifdef CONFIG_SCHED_HMP + if (!pulled_task) + pulled_task = hmp_idle_pull(this_cpu); +#endif raw_spin_lock(&this_rq->lock); if (pulled_task || time_after(jiffies, this_rq->next_balance)) { @@ -5287,22 +6410,19 @@ void idle_balance(int this_cpu, struct rq *this_rq) } } -/* - * active_load_balance_cpu_stop is run by cpu stopper. It pushes - * running tasks off the busiest CPU onto idle CPUs. It requires at - * least 1 task to be running on each physical CPU where possible, and - * avoids physical / logical imbalances. - */ -static int active_load_balance_cpu_stop(void *data) +static int __do_active_load_balance_cpu_stop(void *data, bool check_sd_lb_flag) { struct rq *busiest_rq = data; int busiest_cpu = cpu_of(busiest_rq); int target_cpu = busiest_rq->push_cpu; struct rq *target_rq = cpu_rq(target_cpu); struct sched_domain *sd; + struct task_struct *p = NULL; raw_spin_lock_irq(&busiest_rq->lock); - +#ifdef CONFIG_SCHED_HMP + p = busiest_rq->migrate_task; +#endif /* make sure the requested cpu hasn't gone down in the meantime */ if (unlikely(busiest_cpu != smp_processor_id() || !busiest_rq->active_balance)) @@ -5312,6 +6432,11 @@ static int active_load_balance_cpu_stop(void *data) if (busiest_rq->nr_running <= 1) goto out_unlock; + if (!check_sd_lb_flag) { + /* Task has migrated meanwhile, abort forced migration */ + if (task_rq(p) != busiest_rq) + goto out_unlock; + } /* * This condition is "impossible", if it occurs * we need to fix it. Originally reported by @@ -5325,12 +6450,14 @@ static int active_load_balance_cpu_stop(void *data) /* Search for an sd spanning us and the target CPU. */ rcu_read_lock(); for_each_domain(target_cpu, sd) { - if ((sd->flags & SD_LOAD_BALANCE) && - cpumask_test_cpu(busiest_cpu, sched_domain_span(sd))) + if (((check_sd_lb_flag && sd->flags & SD_LOAD_BALANCE) || + !check_sd_lb_flag) && + cpumask_test_cpu(busiest_cpu, sched_domain_span(sd))) break; } if (likely(sd)) { + bool success = false; struct lb_env env = { .sd = sd, .dst_cpu = target_cpu, @@ -5342,7 +6469,14 @@ static int active_load_balance_cpu_stop(void *data) schedstat_inc(sd, alb_count); - if (move_one_task(&env)) + if (check_sd_lb_flag) { + if (move_one_task(&env)) + success = true; + } else { + if (move_specific_task(&env, p)) + success = true; + } + if (success) schedstat_inc(sd, alb_pushed); else schedstat_inc(sd, alb_failed); @@ -5350,11 +6484,24 @@ static int active_load_balance_cpu_stop(void *data) rcu_read_unlock(); double_unlock_balance(busiest_rq, target_rq); out_unlock: + if (!check_sd_lb_flag) + put_task_struct(p); busiest_rq->active_balance = 0; raw_spin_unlock_irq(&busiest_rq->lock); return 0; } +/* + * active_load_balance_cpu_stop is run by cpu stopper. It pushes + * running tasks off the busiest CPU onto idle CPUs. It requires at + * least 1 task to be running on each physical CPU where possible, and + * avoids physical / logical imbalances. + */ +static int active_load_balance_cpu_stop(void *data) +{ + return __do_active_load_balance_cpu_stop(data, true); +} + #ifdef CONFIG_NO_HZ_COMMON /* * idle load balancing details @@ -5368,12 +6515,65 @@ static struct { unsigned long next_balance; /* in jiffy units */ } nohz ____cacheline_aligned; +/* + * nohz_test_cpu used when load tracking is enabled. FAIR_GROUP_SCHED + * dependency below may be removed when load tracking guards are + * removed. + */ +#ifdef CONFIG_FAIR_GROUP_SCHED +static int nohz_test_cpu(int cpu) +{ + return cpumask_test_cpu(cpu, nohz.idle_cpus_mask); +} +#endif + +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING +/* + * Decide if the tasks on the busy CPUs in the + * littlest domain would benefit from an idle balance + */ +static int hmp_packing_ilb_needed(int cpu) +{ + struct hmp_domain *hmp; + /* always allow ilb on non-slowest domain */ + if (!hmp_cpu_is_slowest(cpu)) + return 1; + + /* if disabled, use normal ILB behaviour */ + if (!hmp_packing_enabled) + return 1; + + hmp = hmp_cpu_domain(cpu); + for_each_cpu_and(cpu, &hmp->cpus, nohz.idle_cpus_mask) { + /* only idle balance if a CPU is loaded over threshold */ + if (cpu_rq(cpu)->avg.load_avg_ratio > hmp_full_threshold) + return 1; + } + return 0; +} +#endif + static inline int find_new_ilb(int call_cpu) { int ilb = cpumask_first(nohz.idle_cpus_mask); +#ifdef CONFIG_SCHED_HMP + int ilb_needed = 1; + + /* restrict nohz balancing to occur in the same hmp domain */ + ilb = cpumask_first_and(nohz.idle_cpus_mask, + &((struct hmp_domain *)hmp_cpu_domain(call_cpu))->cpus); +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING + if (ilb < nr_cpu_ids) + ilb_needed = hmp_packing_ilb_needed(ilb); +#endif + + if (ilb_needed && ilb < nr_cpu_ids && idle_cpu(ilb)) + return ilb; +#else if (ilb < nr_cpu_ids && idle_cpu(ilb)) return ilb; +#endif return nr_cpu_ids; } @@ -5650,6 +6850,18 @@ static inline int nohz_kick_needed(struct rq *rq, int cpu) if (time_before(now, nohz.next_balance)) return 0; +#ifdef CONFIG_SCHED_HMP + /* + * Bail out if there are no nohz CPUs in our + * HMP domain, since we will move tasks between + * domains through wakeup and force balancing + * as necessary based upon task load. + */ + if (cpumask_first_and(nohz.idle_cpus_mask, + &((struct hmp_domain *)hmp_cpu_domain(cpu))->cpus) >= nr_cpu_ids) + return 0; +#endif + if (rq->nr_running >= 2) goto need_kick; @@ -5682,6 +6894,442 @@ need_kick: static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) { } #endif +#ifdef CONFIG_SCHED_HMP +static unsigned int hmp_task_eligible_for_up_migration(struct sched_entity *se) +{ + /* below hmp_up_threshold, never eligible */ + if (se->avg.load_avg_ratio < hmp_up_threshold) + return 0; + return 1; +} + +/* Check if task should migrate to a faster cpu */ +static unsigned int hmp_up_migration(int cpu, int *target_cpu, struct sched_entity *se) +{ + struct task_struct *p = task_of(se); + int temp_target_cpu; + u64 now; + + if (hmp_cpu_is_fastest(cpu)) + return 0; + +#ifdef CONFIG_SCHED_HMP_PRIO_FILTER + /* Filter by task priority */ + if (p->prio >= hmp_up_prio) + return 0; +#endif + if (!hmp_task_eligible_for_up_migration(se)) + return 0; + + /* Let the task load settle before doing another up migration */ + /* hack - always use clock from first online CPU */ + now = cpu_rq(cpumask_first(cpu_online_mask))->clock_task; + if (((now - se->avg.hmp_last_up_migration) >> 10) + < hmp_next_up_threshold) + return 0; + + /* hmp_domain_min_load only returns 0 for an + * idle CPU or 1023 for any partly-busy one. + * Be explicit about requirement for an idle CPU. + */ + if (hmp_domain_min_load(hmp_faster_domain(cpu), &temp_target_cpu, + tsk_cpus_allowed(p)) == 0 && temp_target_cpu != NR_CPUS) { + if(target_cpu) + *target_cpu = temp_target_cpu; + return 1; + } + return 0; +} + +/* Check if task should migrate to a slower cpu */ +static unsigned int hmp_down_migration(int cpu, struct sched_entity *se) +{ + struct task_struct *p = task_of(se); + u64 now; + + if (hmp_cpu_is_slowest(cpu)) { +#ifdef CONFIG_SCHED_HMP_LITTLE_PACKING + if(hmp_packing_enabled) + return 1; + else +#endif + return 0; + } + +#ifdef CONFIG_SCHED_HMP_PRIO_FILTER + /* Filter by task priority */ + if ((p->prio >= hmp_up_prio) && + cpumask_intersects(&hmp_slower_domain(cpu)->cpus, + tsk_cpus_allowed(p))) { + return 1; + } +#endif + + /* Let the task load settle before doing another down migration */ + /* hack - always use clock from first online CPU */ + now = cpu_rq(cpumask_first(cpu_online_mask))->clock_task; + if (((now - se->avg.hmp_last_down_migration) >> 10) + < hmp_next_down_threshold) + return 0; + + if (cpumask_intersects(&hmp_slower_domain(cpu)->cpus, + tsk_cpus_allowed(p)) + && se->avg.load_avg_ratio < hmp_down_threshold) { + return 1; + } + return 0; +} + +/* + * hmp_can_migrate_task - may task p from runqueue rq be migrated to this_cpu? + * Ideally this function should be merged with can_migrate_task() to avoid + * redundant code. + */ +static int hmp_can_migrate_task(struct task_struct *p, struct lb_env *env) +{ + int tsk_cache_hot = 0; + + /* + * We do not migrate tasks that are: + * 1) running (obviously), or + * 2) cannot be migrated to this CPU due to cpus_allowed + */ + if (!cpumask_test_cpu(env->dst_cpu, tsk_cpus_allowed(p))) { + schedstat_inc(p, se.statistics.nr_failed_migrations_affine); + return 0; + } + env->flags &= ~LBF_ALL_PINNED; + + if (task_running(env->src_rq, p)) { + schedstat_inc(p, se.statistics.nr_failed_migrations_running); + return 0; + } + + /* + * Aggressive migration if: + * 1) task is cache cold, or + * 2) too many balance attempts have failed. + */ + + tsk_cache_hot = task_hot(p, env->src_rq->clock_task, env->sd); + if (!tsk_cache_hot || + env->sd->nr_balance_failed > env->sd->cache_nice_tries) { +#ifdef CONFIG_SCHEDSTATS + if (tsk_cache_hot) { + schedstat_inc(env->sd, lb_hot_gained[env->idle]); + schedstat_inc(p, se.statistics.nr_forced_migrations); + } +#endif + return 1; + } + + return 1; +} + +/* + * move_specific_task tries to move a specific task. + * Returns 1 if successful and 0 otherwise. + * Called with both runqueues locked. + */ +static int move_specific_task(struct lb_env *env, struct task_struct *pm) +{ + struct task_struct *p, *n; + + list_for_each_entry_safe(p, n, &env->src_rq->cfs_tasks, se.group_node) { + if (throttled_lb_pair(task_group(p), env->src_rq->cpu, + env->dst_cpu)) + continue; + + if (!hmp_can_migrate_task(p, env)) + continue; + /* Check if we found the right task */ + if (p != pm) + continue; + + move_task(p, env); + /* + * Right now, this is only the third place move_task() + * is called, so we can safely collect move_task() + * stats here rather than inside move_task(). + */ + schedstat_inc(env->sd, lb_gained[env->idle]); + return 1; + } + return 0; +} + +/* + * hmp_active_task_migration_cpu_stop is run by cpu stopper and used to + * migrate a specific task from one runqueue to another. + * hmp_force_up_migration uses this to push a currently running task + * off a runqueue. hmp_idle_pull uses this to pull a currently + * running task to an idle runqueue. + * Reuses __do_active_load_balance_cpu_stop to actually do the work. + */ +static int hmp_active_task_migration_cpu_stop(void *data) +{ + return __do_active_load_balance_cpu_stop(data, false); +} + +/* + * Move task in a runnable state to another CPU. + * + * Tailored on 'active_load_balance_cpu_stop' with slight + * modification to locking and pre-transfer checks. Note + * rq->lock must be held before calling. + */ +static void hmp_migrate_runnable_task(struct rq *rq) +{ + struct sched_domain *sd; + int src_cpu = cpu_of(rq); + struct rq *src_rq = rq; + int dst_cpu = rq->push_cpu; + struct rq *dst_rq = cpu_rq(dst_cpu); + struct task_struct *p = rq->migrate_task; + /* + * One last check to make sure nobody else is playing + * with the source rq. + */ + if (src_rq->active_balance) + goto out; + + if (src_rq->nr_running <= 1) + goto out; + + if (task_rq(p) != src_rq) + goto out; + /* + * Not sure if this applies here but one can never + * be too cautious + */ + BUG_ON(src_rq == dst_rq); + + double_lock_balance(src_rq, dst_rq); + + rcu_read_lock(); + for_each_domain(dst_cpu, sd) { + if (cpumask_test_cpu(src_cpu, sched_domain_span(sd))) + break; + } + + if (likely(sd)) { + struct lb_env env = { + .sd = sd, + .dst_cpu = dst_cpu, + .dst_rq = dst_rq, + .src_cpu = src_cpu, + .src_rq = src_rq, + .idle = CPU_IDLE, + }; + + schedstat_inc(sd, alb_count); + + if (move_specific_task(&env, p)) + schedstat_inc(sd, alb_pushed); + else + schedstat_inc(sd, alb_failed); + } + + rcu_read_unlock(); + double_unlock_balance(src_rq, dst_rq); +out: + put_task_struct(p); +} + +static DEFINE_SPINLOCK(hmp_force_migration); + +/* + * hmp_force_up_migration checks runqueues for tasks that need to + * be actively migrated to a faster cpu. + */ +static void hmp_force_up_migration(int this_cpu) +{ + int cpu, target_cpu; + struct sched_entity *curr, *orig; + struct rq *target; + unsigned long flags; + unsigned int force, got_target; + struct task_struct *p; + + if (!spin_trylock(&hmp_force_migration)) + return; + for_each_online_cpu(cpu) { + force = 0; + got_target = 0; + target = cpu_rq(cpu); + raw_spin_lock_irqsave(&target->lock, flags); + curr = target->cfs.curr; + if (!curr || target->active_balance) { + raw_spin_unlock_irqrestore(&target->lock, flags); + continue; + } + if (!entity_is_task(curr)) { + struct cfs_rq *cfs_rq; + + cfs_rq = group_cfs_rq(curr); + while (cfs_rq) { + curr = cfs_rq->curr; + cfs_rq = group_cfs_rq(curr); + } + } + orig = curr; + curr = hmp_get_heaviest_task(curr, -1); + if (!curr) { + raw_spin_unlock_irqrestore(&target->lock, flags); + continue; + } + p = task_of(curr); + if (hmp_up_migration(cpu, &target_cpu, curr)) { + cpu_rq(target_cpu)->wake_for_idle_pull = 1; + raw_spin_unlock_irqrestore(&target->lock, flags); + spin_unlock(&hmp_force_migration); + smp_send_reschedule(target_cpu); + return; + } + if (!got_target) { + /* + * For now we just check the currently running task. + * Selecting the lightest task for offloading will + * require extensive book keeping. + */ + curr = hmp_get_lightest_task(orig, 1); + p = task_of(curr); + target->push_cpu = hmp_offload_down(cpu, curr); + if (target->push_cpu < NR_CPUS) { + get_task_struct(p); + target->migrate_task = p; + got_target = 1; + trace_sched_hmp_migrate(p, target->push_cpu, HMP_MIGRATE_OFFLOAD); + hmp_next_down_delay(&p->se, target->push_cpu); + } + } + /* + * We have a target with no active_balance. If the task + * is not currently running move it, otherwise let the + * CPU stopper take care of it. + */ + if (got_target) { + if (!task_running(target, p)) { + trace_sched_hmp_migrate_force_running(p, 0); + hmp_migrate_runnable_task(target); + } else { + target->active_balance = 1; + force = 1; + } + } + + raw_spin_unlock_irqrestore(&target->lock, flags); + + if (force) + stop_one_cpu_nowait(cpu_of(target), + hmp_active_task_migration_cpu_stop, + target, &target->active_balance_work); + } + spin_unlock(&hmp_force_migration); +} +/* + * hmp_idle_pull looks at little domain runqueues to see + * if a task should be pulled. + * + * Reuses hmp_force_migration spinlock. + * + */ +static unsigned int hmp_idle_pull(int this_cpu) +{ + int cpu; + struct sched_entity *curr, *orig; + struct hmp_domain *hmp_domain = NULL; + struct rq *target = NULL, *rq; + unsigned long flags, ratio = 0; + unsigned int force = 0; + struct task_struct *p = NULL; + + if (!hmp_cpu_is_slowest(this_cpu)) + hmp_domain = hmp_slower_domain(this_cpu); + if (!hmp_domain) + return 0; + + if (!spin_trylock(&hmp_force_migration)) + return 0; + + /* first select a task */ + for_each_cpu(cpu, &hmp_domain->cpus) { + rq = cpu_rq(cpu); + raw_spin_lock_irqsave(&rq->lock, flags); + curr = rq->cfs.curr; + if (!curr) { + raw_spin_unlock_irqrestore(&rq->lock, flags); + continue; + } + if (!entity_is_task(curr)) { + struct cfs_rq *cfs_rq; + + cfs_rq = group_cfs_rq(curr); + while (cfs_rq) { + curr = cfs_rq->curr; + if (!entity_is_task(curr)) + cfs_rq = group_cfs_rq(curr); + else + cfs_rq = NULL; + } + } + orig = curr; + curr = hmp_get_heaviest_task(curr, this_cpu); + /* check if heaviest eligible task on this + * CPU is heavier than previous task + */ + if (curr && hmp_task_eligible_for_up_migration(curr) && + curr->avg.load_avg_ratio > ratio && + cpumask_test_cpu(this_cpu, + tsk_cpus_allowed(task_of(curr)))) { + p = task_of(curr); + target = rq; + ratio = curr->avg.load_avg_ratio; + } + raw_spin_unlock_irqrestore(&rq->lock, flags); + } + + if (!p) + goto done; + + /* now we have a candidate */ + raw_spin_lock_irqsave(&target->lock, flags); + if (!target->active_balance && task_rq(p) == target) { + get_task_struct(p); + target->push_cpu = this_cpu; + target->migrate_task = p; + trace_sched_hmp_migrate(p, target->push_cpu, HMP_MIGRATE_IDLE_PULL); + hmp_next_up_delay(&p->se, target->push_cpu); + /* + * if the task isn't running move it right away. + * Otherwise setup the active_balance mechanic and let + * the CPU stopper do its job. + */ + if (!task_running(target, p)) { + trace_sched_hmp_migrate_idle_running(p, 0); + hmp_migrate_runnable_task(target); + } else { + target->active_balance = 1; + force = 1; + } + } + raw_spin_unlock_irqrestore(&target->lock, flags); + + if (force) { + /* start timer to keep us awake */ + hmp_cpu_keepalive_trigger(); + stop_one_cpu_nowait(cpu_of(target), + hmp_active_task_migration_cpu_stop, + target, &target->active_balance_work); + } +done: + spin_unlock(&hmp_force_migration); + return force; +} +#else +static void hmp_force_up_migration(int this_cpu) { } +#endif /* CONFIG_SCHED_HMP */ + /* * run_rebalance_domains is triggered when needed from the scheduler tick. * Also triggered for nohz idle balancing (with nohz_balancing_kick set). @@ -5693,6 +7341,20 @@ static void run_rebalance_domains(struct softirq_action *h) enum cpu_idle_type idle = this_rq->idle_balance ? CPU_IDLE : CPU_NOT_IDLE; +#ifdef CONFIG_SCHED_HMP + /* shortcut for hmp idle pull wakeups */ + if (unlikely(this_rq->wake_for_idle_pull)) { + this_rq->wake_for_idle_pull = 0; + if (hmp_idle_pull(this_cpu)) { + /* break out unless running nohz idle as well */ + if (idle != CPU_IDLE) + return; + } + } +#endif + + hmp_force_up_migration(this_cpu); + rebalance_domains(this_cpu, idle); /* @@ -5725,11 +7387,17 @@ void trigger_load_balance(struct rq *rq, int cpu) static void rq_online_fair(struct rq *rq) { +#ifdef CONFIG_SCHED_HMP + hmp_online_cpu(rq->cpu); +#endif update_sysctl(); } static void rq_offline_fair(struct rq *rq) { +#ifdef CONFIG_SCHED_HMP + hmp_offline_cpu(rq->cpu); +#endif update_sysctl(); /* Ensure any throttled groups are reachable by pick_next_task */ @@ -5777,11 +7445,15 @@ static void task_fork_fair(struct task_struct *p) cfs_rq = task_cfs_rq(current); curr = cfs_rq->curr; - if (unlikely(task_cpu(p) != this_cpu)) { - rcu_read_lock(); - __set_task_cpu(p, this_cpu); - rcu_read_unlock(); - } + /* + * Not only the cpu but also the task_group of the parent might have + * been changed after parent->se.parent,cfs_rq were copied to + * child->se.parent,cfs_rq. So call __set_task_cpu() to make those + * of child point to valid ones. + */ + rcu_read_lock(); + __set_task_cpu(p, this_cpu); + rcu_read_unlock(); update_curr(cfs_rq); @@ -5831,15 +7503,15 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p) struct cfs_rq *cfs_rq = cfs_rq_of(se); /* - * Ensure the task's vruntime is normalized, so that when its + * Ensure the task's vruntime is normalized, so that when it's * switched back to the fair class the enqueue_entity(.flags=0) will * do the right thing. * - * If it was on_rq, then the dequeue_entity(.flags=0) will already - * have normalized the vruntime, if it was !on_rq, then only when + * If it's on_rq, then the dequeue_entity(.flags=0) will already + * have normalized the vruntime, if it's !on_rq, then only when * the task is sleeping will it still have non-normalized vruntime. */ - if (!se->on_rq && p->state != TASK_RUNNING) { + if (!p->on_rq && p->state != TASK_RUNNING) { /* * Fix up our vruntime so that the current sleep doesn't * cause 'unlimited' sleep bonus. @@ -6060,7 +7732,8 @@ void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, se->cfs_rq = parent->my_q; se->my_q = cfs_rq; - update_load_set(&se->load, 0); + /* guarantee group entities always have weight */ + update_load_set(&se->load, NICE_0_LOAD); se->parent = parent; } @@ -6192,6 +7865,139 @@ __init void init_sched_fair_class(void) zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT); cpu_notifier(sched_ilb_notifier, 0); #endif + +#ifdef CONFIG_SCHED_HMP + hmp_cpu_mask_setup(); +#endif #endif /* SMP */ } + +#ifdef CONFIG_HMP_FREQUENCY_INVARIANT_SCALE +static u32 cpufreq_calc_scale(u32 min, u32 max, u32 curr) +{ + u32 result = curr / max; + return result; +} + +/* Called when the CPU Frequency is changed. + * Once for each CPU. + */ +static int cpufreq_callback(struct notifier_block *nb, + unsigned long val, void *data) +{ + struct cpufreq_freqs *freq = data; + int cpu = freq->cpu; + struct cpufreq_extents *extents; + + if (freq->flags & CPUFREQ_CONST_LOOPS) + return NOTIFY_OK; + + if (val != CPUFREQ_POSTCHANGE) + return NOTIFY_OK; + + /* if dynamic load scale is disabled, set the load scale to 1.0 */ + if (!hmp_data.freqinvar_load_scale_enabled) { + freq_scale[cpu].curr_scale = 1024; + return NOTIFY_OK; + } + + extents = &freq_scale[cpu]; + if (extents->flags & SCHED_LOAD_FREQINVAR_SINGLEFREQ) { + /* If our governor was recognised as a single-freq governor, + * use 1.0 + */ + extents->curr_scale = 1024; + } else { + extents->curr_scale = cpufreq_calc_scale(extents->min, + extents->max, freq->new); + } + + return NOTIFY_OK; +} + +/* Called when the CPUFreq governor is changed. + * Only called for the CPUs which are actually changed by the + * userspace. + */ +static int cpufreq_policy_callback(struct notifier_block *nb, + unsigned long event, void *data) +{ + struct cpufreq_policy *policy = data; + struct cpufreq_extents *extents; + int cpu, singleFreq = 0; + static const char performance_governor[] = "performance"; + static const char powersave_governor[] = "powersave"; + + if (event == CPUFREQ_START) + return 0; + + if (event != CPUFREQ_INCOMPATIBLE) + return 0; + + /* CPUFreq governors do not accurately report the range of + * CPU Frequencies they will choose from. + * We recognise performance and powersave governors as + * single-frequency only. + */ + if (!strncmp(policy->governor->name, performance_governor, + strlen(performance_governor)) || + !strncmp(policy->governor->name, powersave_governor, + strlen(powersave_governor))) + singleFreq = 1; + + /* Make sure that all CPUs impacted by this policy are + * updated since we will only get a notification when the + * user explicitly changes the policy on a CPU. + */ + for_each_cpu(cpu, policy->cpus) { + extents = &freq_scale[cpu]; + extents->max = policy->max >> SCHED_FREQSCALE_SHIFT; + extents->min = policy->min >> SCHED_FREQSCALE_SHIFT; + if (!hmp_data.freqinvar_load_scale_enabled) { + extents->curr_scale = 1024; + } else if (singleFreq) { + extents->flags |= SCHED_LOAD_FREQINVAR_SINGLEFREQ; + extents->curr_scale = 1024; + } else { + extents->flags &= ~SCHED_LOAD_FREQINVAR_SINGLEFREQ; + extents->curr_scale = cpufreq_calc_scale(extents->min, + extents->max, policy->cur); + } + } + + return 0; +} + +static struct notifier_block cpufreq_notifier = { + .notifier_call = cpufreq_callback, +}; +static struct notifier_block cpufreq_policy_notifier = { + .notifier_call = cpufreq_policy_callback, +}; + +static int __init register_sched_cpufreq_notifier(void) +{ + int ret = 0; + + /* init safe defaults since there are no policies at registration */ + for (ret = 0; ret < CONFIG_NR_CPUS; ret++) { + /* safe defaults */ + freq_scale[ret].max = 1024; + freq_scale[ret].min = 1024; + freq_scale[ret].curr_scale = 1024; + } + + pr_info("sched: registering cpufreq notifiers for scale-invariant loads\n"); + ret = cpufreq_register_notifier(&cpufreq_policy_notifier, + CPUFREQ_POLICY_NOTIFIER); + + if (ret != -EINVAL) + ret = cpufreq_register_notifier(&cpufreq_notifier, + CPUFREQ_TRANSITION_NOTIFIER); + + return ret; +} + +core_initcall(register_sched_cpufreq_notifier); +#endif /* CONFIG_HMP_FREQUENCY_INVARIANT_SCALE */ diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 127a2c4cf4ab..15334e6de832 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -964,6 +964,13 @@ inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); +#ifdef CONFIG_RT_GROUP_SCHED + /* + * Change rq's cpupri only if rt_rq is the top queue. + */ + if (&rq->rt != rt_rq) + return; +#endif if (rq->online && prio < prev_prio) cpupri_set(&rq->rd->cpupri, rq->cpu, prio); } @@ -973,6 +980,13 @@ dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); +#ifdef CONFIG_RT_GROUP_SCHED + /* + * Change rq's cpupri only if rt_rq is the top queue. + */ + if (&rq->rt != rt_rq) + return; +#endif if (rq->online && rt_rq->highest_prio.curr != prev_prio) cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio.curr); } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index ce39224d6155..0d19ede6849e 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -142,7 +142,7 @@ struct task_group { atomic_t load_weight; atomic64_t load_avg; - atomic_t runnable_avg; + atomic_t runnable_avg, usage_avg; #endif #ifdef CONFIG_RT_GROUP_SCHED @@ -279,7 +279,7 @@ struct cfs_rq { #endif /* CONFIG_FAIR_GROUP_SCHED */ /* These always depend on CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_FAIR_GROUP_SCHED - u32 tg_runnable_contrib; + u32 tg_runnable_contrib, tg_usage_contrib; u64 tg_load_contrib; #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -464,6 +464,10 @@ struct rq { int active_balance; int push_cpu; struct cpu_stop_work active_balance_work; +#ifdef CONFIG_SCHED_HMP + struct task_struct *migrate_task; + int wake_for_idle_pull; +#endif /* cpu of this runqueue: */ int cpu; int online; @@ -642,6 +646,12 @@ static inline unsigned int group_first_cpu(struct sched_group *group) extern int group_balance_cpu(struct sched_group *sg); +#ifdef CONFIG_SCHED_HMP +static LIST_HEAD(hmp_domains); +DECLARE_PER_CPU(struct hmp_domain *, hmp_cpu_domain); +#define hmp_cpu_domain(cpu) (per_cpu(hmp_cpu_domain, (cpu))) +#endif /* CONFIG_SCHED_HMP */ + #endif /* CONFIG_SMP */ #include "stats.h" @@ -1318,7 +1328,8 @@ extern void print_rt_stats(struct seq_file *m, int cpu); extern void init_cfs_rq(struct cfs_rq *cfs_rq); extern void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq); -extern void account_cfs_bandwidth_used(int enabled, int was_enabled); +extern void cfs_bandwidth_usage_inc(void); +extern void cfs_bandwidth_usage_dec(void); #ifdef CONFIG_NO_HZ_COMMON enum rq_nohz_flag_bits { diff --git a/kernel/smp.c b/kernel/smp.c index 4dba0f7b72ad..23ccc67dcbb2 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -12,6 +12,8 @@ #include <linux/gfp.h> #include <linux/smp.h> #include <linux/cpu.h> +#define CREATE_TRACE_POINTS +#include <trace/events/smp.h> #include "smpboot.h" @@ -159,8 +161,10 @@ void generic_exec_single(int cpu, struct call_single_data *csd, int wait) * locking and barrier primitives. Generic code isn't really * equipped to do the right thing... */ - if (ipi) + if (ipi) { + trace_smp_call_func_send(csd->func, cpu); arch_send_call_function_single_ipi(cpu); + } if (wait) csd_lock_wait(csd); @@ -197,8 +201,9 @@ void generic_smp_call_function_single_interrupt(void) * so save them away before making the call: */ csd_flags = csd->flags; - + trace_smp_call_func_entry(csd->func); csd->func(csd->info); + trace_smp_call_func_exit(csd->func); /* * Unlocked CSDs are valid through generic_exec_single(): @@ -228,6 +233,7 @@ int smp_call_function_single(int cpu, smp_call_func_t func, void *info, int this_cpu; int err = 0; + trace_smp_call_func_send(func, cpu); /* * prevent preemption and reschedule on another processor, * as well as CPU removal @@ -245,7 +251,9 @@ int smp_call_function_single(int cpu, smp_call_func_t func, void *info, if (cpu == this_cpu) { local_irq_save(flags); + trace_smp_call_func_entry(func); func(info); + trace_smp_call_func_exit(func); local_irq_restore(flags); } else { if ((unsigned)cpu < nr_cpu_ids && cpu_online(cpu)) { diff --git a/kernel/softirq.c b/kernel/softirq.c index 3d6833f125d3..787b3a032429 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -330,10 +330,19 @@ void irq_enter(void) static inline void invoke_softirq(void) { - if (!force_irqthreads) - __do_softirq(); - else + if (!force_irqthreads) { + /* + * We can safely execute softirq on the current stack if + * it is the irq stack, because it should be near empty + * at this stage. But we have no way to know if the arch + * calls irq_exit() on the irq stack. So call softirq + * in its own stack to prevent from any overrun on top + * of a potentially deep task stack. + */ + do_softirq(); + } else { wakeup_softirqd(); + } } static inline void tick_irq_exit(void) diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 9edcf456e0fc..9469f4c61a30 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -144,6 +144,11 @@ static int min_percpu_pagelist_fract = 8; static int ngroups_max = NGROUPS_MAX; static const int cap_last_cap = CAP_LAST_CAP; +/*this is needed for proc_doulongvec_minmax of sysctl_hung_task_timeout_secs */ +#ifdef CONFIG_DETECT_HUNG_TASK +static unsigned long hung_task_timeout_max = (LONG_MAX/HZ); +#endif + #ifdef CONFIG_INOTIFY_USER #include <linux/inotify.h> #endif @@ -966,6 +971,7 @@ static struct ctl_table kern_table[] = { .maxlen = sizeof(unsigned long), .mode = 0644, .proc_handler = proc_dohung_task_timeout_secs, + .extra2 = &hung_task_timeout_max, }, { .procname = "hung_task_warnings", @@ -1043,6 +1049,16 @@ static struct ctl_table kern_table[] = { .maxlen = sizeof(sysctl_perf_event_sample_rate), .mode = 0644, .proc_handler = perf_proc_update_handler, + .extra1 = &one, + }, + { + .procname = "perf_cpu_time_max_percent", + .data = &sysctl_perf_cpu_time_max_percent, + .maxlen = sizeof(sysctl_perf_cpu_time_max_percent), + .mode = 0644, + .proc_handler = perf_cpu_time_max_percent_handler, + .extra1 = &zero, + .extra2 = &one_hundred, }, #endif #ifdef CONFIG_KMEMCHECK diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index f11d83b12949..a8f5084dcde7 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -445,7 +445,7 @@ static int alarm_clock_getres(const clockid_t which_clock, struct timespec *tp) clockid_t baseid = alarm_bases[clock2alarm(which_clock)].base_clockid; if (!alarmtimer_get_rtcdev()) - return -ENOTSUPP; + return -EINVAL; return hrtimer_get_res(baseid, tp); } @@ -462,7 +462,7 @@ static int alarm_clock_get(clockid_t which_clock, struct timespec *tp) struct alarm_base *base = &alarm_bases[clock2alarm(which_clock)]; if (!alarmtimer_get_rtcdev()) - return -ENOTSUPP; + return -EINVAL; *tp = ktime_to_timespec(base->gettime()); return 0; diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index c6d6400ee137..9df0e3b19f09 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -15,7 +15,6 @@ #include <linux/hrtimer.h> #include <linux/init.h> #include <linux/module.h> -#include <linux/notifier.h> #include <linux/smp.h> #include "tick-internal.h" @@ -23,36 +22,67 @@ /* The registered clock event devices */ static LIST_HEAD(clockevent_devices); static LIST_HEAD(clockevents_released); - -/* Notification for clock events */ -static RAW_NOTIFIER_HEAD(clockevents_chain); - /* Protection for the above */ static DEFINE_RAW_SPINLOCK(clockevents_lock); -/** - * clockevents_delta2ns - Convert a latch value (device ticks) to nanoseconds - * @latch: value to convert - * @evt: pointer to clock event device descriptor - * - * Math helper, returns latch value converted to nanoseconds (bound checked) - */ -u64 clockevent_delta2ns(unsigned long latch, struct clock_event_device *evt) +static u64 cev_delta2ns(unsigned long latch, struct clock_event_device *evt, + bool ismax) { u64 clc = (u64) latch << evt->shift; + u64 rnd; if (unlikely(!evt->mult)) { evt->mult = 1; WARN_ON(1); } + rnd = (u64) evt->mult - 1; + + /* + * Upper bound sanity check. If the backwards conversion is + * not equal latch, we know that the above shift overflowed. + */ + if ((clc >> evt->shift) != (u64)latch) + clc = ~0ULL; + + /* + * Scaled math oddities: + * + * For mult <= (1 << shift) we can safely add mult - 1 to + * prevent integer rounding loss. So the backwards conversion + * from nsec to device ticks will be correct. + * + * For mult > (1 << shift), i.e. device frequency is > 1GHz we + * need to be careful. Adding mult - 1 will result in a value + * which when converted back to device ticks can be larger + * than latch by up to (mult - 1) >> shift. For the min_delta + * calculation we still want to apply this in order to stay + * above the minimum device ticks limit. For the upper limit + * we would end up with a latch value larger than the upper + * limit of the device, so we omit the add to stay below the + * device upper boundary. + * + * Also omit the add if it would overflow the u64 boundary. + */ + if ((~0ULL - clc > rnd) && + (!ismax || evt->mult <= (1U << evt->shift))) + clc += rnd; do_div(clc, evt->mult); - if (clc < 1000) - clc = 1000; - if (clc > KTIME_MAX) - clc = KTIME_MAX; - return clc; + /* Deltas less than 1usec are pointless noise */ + return clc > 1000 ? clc : 1000; +} + +/** + * clockevents_delta2ns - Convert a latch value (device ticks) to nanoseconds + * @latch: value to convert + * @evt: pointer to clock event device descriptor + * + * Math helper, returns latch value converted to nanoseconds (bound checked) + */ +u64 clockevent_delta2ns(unsigned long latch, struct clock_event_device *evt) +{ + return cev_delta2ns(latch, evt, false); } EXPORT_SYMBOL_GPL(clockevent_delta2ns); @@ -232,30 +262,6 @@ int clockevents_program_event(struct clock_event_device *dev, ktime_t expires, return (rc && force) ? clockevents_program_min_delta(dev) : rc; } -/** - * clockevents_register_notifier - register a clock events change listener - */ -int clockevents_register_notifier(struct notifier_block *nb) -{ - unsigned long flags; - int ret; - - raw_spin_lock_irqsave(&clockevents_lock, flags); - ret = raw_notifier_chain_register(&clockevents_chain, nb); - raw_spin_unlock_irqrestore(&clockevents_lock, flags); - - return ret; -} - -/* - * Notify about a clock event change. Called with clockevents_lock - * held. - */ -static void clockevents_do_notify(unsigned long reason, void *dev) -{ - raw_notifier_call_chain(&clockevents_chain, reason, dev); -} - /* * Called after a notify add to make devices available which were * released from the notifier call. @@ -269,7 +275,7 @@ static void clockevents_notify_released(void) struct clock_event_device, list); list_del(&dev->list); list_add(&dev->list, &clockevent_devices); - clockevents_do_notify(CLOCK_EVT_NOTIFY_ADD, dev); + tick_check_new_device(dev); } } @@ -290,7 +296,7 @@ void clockevents_register_device(struct clock_event_device *dev) raw_spin_lock_irqsave(&clockevents_lock, flags); list_add(&dev->list, &clockevent_devices); - clockevents_do_notify(CLOCK_EVT_NOTIFY_ADD, dev); + tick_check_new_device(dev); clockevents_notify_released(); raw_spin_unlock_irqrestore(&clockevents_lock, flags); @@ -317,8 +323,8 @@ void clockevents_config(struct clock_event_device *dev, u32 freq) sec = 600; clockevents_calc_mult_shift(dev, freq, sec); - dev->min_delta_ns = clockevent_delta2ns(dev->min_delta_ticks, dev); - dev->max_delta_ns = clockevent_delta2ns(dev->max_delta_ticks, dev); + dev->min_delta_ns = cev_delta2ns(dev->min_delta_ticks, dev, false); + dev->max_delta_ns = cev_delta2ns(dev->max_delta_ticks, dev, true); } /** @@ -386,6 +392,7 @@ void clockevents_exchange_device(struct clock_event_device *old, * released list and do a notify add later. */ if (old) { + module_put(old->owner); clockevents_set_mode(old, CLOCK_EVT_MODE_UNUSED); list_del(&old->list); list_add(&old->list, &clockevents_released); @@ -433,7 +440,7 @@ void clockevents_notify(unsigned long reason, void *arg) int cpu; raw_spin_lock_irqsave(&clockevents_lock, flags); - clockevents_do_notify(reason, arg); + tick_notify(reason, arg); switch (reason) { case CLOCK_EVT_NOTIFY_CPU_DEAD: diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c index 7a925ba456fb..a6a5bf53e86d 100644 --- a/kernel/time/jiffies.c +++ b/kernel/time/jiffies.c @@ -51,7 +51,13 @@ * HZ shrinks, so values greater than 8 overflow 32bits when * HZ=100. */ +#if HZ < 34 +#define JIFFIES_SHIFT 6 +#elif HZ < 67 +#define JIFFIES_SHIFT 7 +#else #define JIFFIES_SHIFT 8 +#endif static cycle_t jiffies_read(struct clocksource *cs) { diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index 8f5b3b98577b..af8d1d4f3d55 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -475,6 +475,7 @@ static void sync_cmos_clock(struct work_struct *work) * called as close as possible to 500 ms before the new second starts. * This code is run on a timer. If the clock is set, that timer * may not expire at the correct time. Thus, we adjust... + * We want the clock to be within a couple of ticks from the target. */ if (!ntp_synced()) { /* @@ -485,7 +486,7 @@ static void sync_cmos_clock(struct work_struct *work) } getnstimeofday(&now); - if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec / 2) { + if (abs(now.tv_nsec - (NSEC_PER_SEC / 2)) <= tick_nsec * 5) { struct timespec adjust = now; fail = -ENODEV; @@ -516,13 +517,13 @@ static void sync_cmos_clock(struct work_struct *work) schedule_delayed_work(&sync_cmos_work, timespec_to_jiffies(&next)); } -static void notify_cmos_timer(void) +void ntp_notify_cmos_timer(void) { schedule_delayed_work(&sync_cmos_work, 0); } #else -static inline void notify_cmos_timer(void) { } +void ntp_notify_cmos_timer(void) { } #endif @@ -687,8 +688,6 @@ int __do_adjtimex(struct timex *txc, struct timespec *ts, s32 *time_tai) if (!(time_status & STA_NANO)) txc->time.tv_usec /= NSEC_PER_USEC; - notify_cmos_timer(); - return result; } diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 20d6fba70652..19ee339a1d0d 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -19,6 +19,7 @@ #include <linux/profile.h> #include <linux/sched.h> #include <linux/smp.h> +#include <linux/module.h> #include "tick-internal.h" @@ -29,6 +30,7 @@ static struct tick_device tick_broadcast_device; static cpumask_var_t tick_broadcast_mask; +static cpumask_var_t tick_broadcast_on; static cpumask_var_t tmpmask; static DEFINE_RAW_SPINLOCK(tick_broadcast_lock); static int tick_broadcast_force; @@ -64,17 +66,34 @@ static void tick_broadcast_start_periodic(struct clock_event_device *bc) /* * Check, if the device can be utilized as broadcast device: */ -int tick_check_broadcast_device(struct clock_event_device *dev) +static bool tick_check_broadcast_device(struct clock_event_device *curdev, + struct clock_event_device *newdev) +{ + if ((newdev->features & CLOCK_EVT_FEAT_DUMMY) || + (newdev->features & CLOCK_EVT_FEAT_C3STOP)) + return false; + + if (tick_broadcast_device.mode == TICKDEV_MODE_ONESHOT && + !(newdev->features & CLOCK_EVT_FEAT_ONESHOT)) + return false; + + return !curdev || newdev->rating > curdev->rating; +} + +/* + * Conditionally install/replace broadcast device + */ +void tick_install_broadcast_device(struct clock_event_device *dev) { struct clock_event_device *cur = tick_broadcast_device.evtdev; - if ((dev->features & CLOCK_EVT_FEAT_DUMMY) || - (tick_broadcast_device.evtdev && - tick_broadcast_device.evtdev->rating >= dev->rating) || - (dev->features & CLOCK_EVT_FEAT_C3STOP)) - return 0; + if (!tick_check_broadcast_device(cur, dev)) + return; - clockevents_exchange_device(tick_broadcast_device.evtdev, dev); + if (!try_module_get(dev->owner)) + return; + + clockevents_exchange_device(cur, dev); if (cur) cur->event_handler = clockevents_handle_noop; tick_broadcast_device.evtdev = dev; @@ -90,7 +109,6 @@ int tick_check_broadcast_device(struct clock_event_device *dev) */ if (dev->features & CLOCK_EVT_FEAT_ONESHOT) tick_clock_notify(); - return 1; } /* @@ -123,8 +141,9 @@ static void tick_device_setup_broadcast_func(struct clock_event_device *dev) */ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu) { + struct clock_event_device *bc = tick_broadcast_device.evtdev; unsigned long flags; - int ret = 0; + int ret; raw_spin_lock_irqsave(&tick_broadcast_lock, flags); @@ -138,20 +157,59 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu) dev->event_handler = tick_handle_periodic; tick_device_setup_broadcast_func(dev); cpumask_set_cpu(cpu, tick_broadcast_mask); - tick_broadcast_start_periodic(tick_broadcast_device.evtdev); + tick_broadcast_start_periodic(bc); ret = 1; } else { /* - * When the new device is not affected by the stop - * feature and the cpu is marked in the broadcast mask - * then clear the broadcast bit. + * Clear the broadcast bit for this cpu if the + * device is not power state affected. */ - if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) { - int cpu = smp_processor_id(); + if (!(dev->features & CLOCK_EVT_FEAT_C3STOP)) cpumask_clear_cpu(cpu, tick_broadcast_mask); - tick_broadcast_clear_oneshot(cpu); - } else { + else tick_device_setup_broadcast_func(dev); + + /* + * Clear the broadcast bit if the CPU is not in + * periodic broadcast on state. + */ + if (!cpumask_test_cpu(cpu, tick_broadcast_on)) + cpumask_clear_cpu(cpu, tick_broadcast_mask); + + switch (tick_broadcast_device.mode) { + case TICKDEV_MODE_ONESHOT: + /* + * If the system is in oneshot mode we can + * unconditionally clear the oneshot mask bit, + * because the CPU is running and therefore + * not in an idle state which causes the power + * state affected device to stop. Let the + * caller initialize the device. + */ + tick_broadcast_clear_oneshot(cpu); + ret = 0; + break; + + case TICKDEV_MODE_PERIODIC: + /* + * If the system is in periodic mode, check + * whether the broadcast device can be + * switched off now. + */ + if (cpumask_empty(tick_broadcast_mask) && bc) + clockevents_shutdown(bc); + /* + * If we kept the cpu in the broadcast mask, + * tell the caller to leave the per cpu device + * in shutdown state. The periodic interrupt + * is delivered by the broadcast device. + */ + ret = cpumask_test_cpu(cpu, tick_broadcast_mask); + break; + default: + /* Nothing to do */ + ret = 0; + break; } } raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); @@ -281,6 +339,7 @@ static void tick_do_broadcast_on_off(unsigned long *reason) switch (*reason) { case CLOCK_EVT_NOTIFY_BROADCAST_ON: case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: + cpumask_set_cpu(cpu, tick_broadcast_on); if (!cpumask_test_and_set_cpu(cpu, tick_broadcast_mask)) { if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) @@ -290,8 +349,12 @@ static void tick_do_broadcast_on_off(unsigned long *reason) tick_broadcast_force = 1; break; case CLOCK_EVT_NOTIFY_BROADCAST_OFF: - if (!tick_broadcast_force && - cpumask_test_and_clear_cpu(cpu, tick_broadcast_mask)) { + if (tick_broadcast_force) + break; + cpumask_clear_cpu(cpu, tick_broadcast_on); + if (!tick_device_is_functional(dev)) + break; + if (cpumask_test_and_clear_cpu(cpu, tick_broadcast_mask)) { if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) tick_setup_periodic(dev, 0); @@ -349,6 +412,7 @@ void tick_shutdown_broadcast(unsigned int *cpup) bc = tick_broadcast_device.evtdev; cpumask_clear_cpu(cpu, tick_broadcast_mask); + cpumask_clear_cpu(cpu, tick_broadcast_on); if (tick_broadcast_device.mode == TICKDEV_MODE_PERIODIC) { if (bc && cpumask_empty(tick_broadcast_mask)) @@ -475,7 +539,15 @@ void tick_check_oneshot_broadcast(int cpu) if (cpumask_test_cpu(cpu, tick_broadcast_oneshot_mask)) { struct tick_device *td = &per_cpu(tick_cpu_device, cpu); - clockevents_set_mode(td->evtdev, CLOCK_EVT_MODE_ONESHOT); + /* + * We might be in the middle of switching over from + * periodic to oneshot. If the CPU has not yet + * switched over, leave the device alone. + */ + if (td->mode == TICKDEV_MODE_ONESHOT) { + clockevents_set_mode(td->evtdev, + CLOCK_EVT_MODE_ONESHOT); + } } } @@ -522,6 +594,13 @@ again: cpumask_clear(tick_broadcast_force_mask); /* + * Sanity check. Catch the case where we try to broadcast to + * offline cpus. + */ + if (WARN_ON_ONCE(!cpumask_subset(tmpmask, cpu_online_mask))) + cpumask_and(tmpmask, tmpmask, cpu_online_mask); + + /* * Wakeup the cpus which have an expired event. */ tick_do_broadcast(tmpmask); @@ -673,6 +752,7 @@ out: static void tick_broadcast_clear_oneshot(int cpu) { cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask); + cpumask_clear_cpu(cpu, tick_broadcast_pending_mask); } static void tick_broadcast_init_next_event(struct cpumask *mask, @@ -761,10 +841,12 @@ void tick_shutdown_broadcast_oneshot(unsigned int *cpup) raw_spin_lock_irqsave(&tick_broadcast_lock, flags); /* - * Clear the broadcast mask flag for the dead cpu, but do not - * stop the broadcast device! + * Clear the broadcast masks for the dead cpu, but do not stop + * the broadcast device! */ cpumask_clear_cpu(cpu, tick_broadcast_oneshot_mask); + cpumask_clear_cpu(cpu, tick_broadcast_pending_mask); + cpumask_clear_cpu(cpu, tick_broadcast_force_mask); raw_spin_unlock_irqrestore(&tick_broadcast_lock, flags); } @@ -792,6 +874,7 @@ bool tick_broadcast_oneshot_available(void) void __init tick_broadcast_init(void) { zalloc_cpumask_var(&tick_broadcast_mask, GFP_NOWAIT); + zalloc_cpumask_var(&tick_broadcast_on, GFP_NOWAIT); zalloc_cpumask_var(&tmpmask, GFP_NOWAIT); #ifdef CONFIG_TICK_ONESHOT zalloc_cpumask_var(&tick_broadcast_oneshot_mask, GFP_NOWAIT); diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index 5d3fb100bc06..086216c433fa 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -18,6 +18,7 @@ #include <linux/percpu.h> #include <linux/profile.h> #include <linux/sched.h> +#include <linux/module.h> #include <asm/irq_regs.h> @@ -194,7 +195,8 @@ static void tick_setup_device(struct tick_device *td, * When global broadcasting is active, check if the current * device is registered as a placeholder for broadcast mode. * This allows us to handle this x86 misfeature in a generic - * way. + * way. This function also returns !=0 when we keep the + * current active broadcast state for this CPU. */ if (tick_device_uses_broadcast(newdev, cpu)) return; @@ -205,14 +207,50 @@ static void tick_setup_device(struct tick_device *td, tick_setup_oneshot(newdev, handler, next_event); } +static bool tick_check_percpu(struct clock_event_device *curdev, + struct clock_event_device *newdev, int cpu) +{ + if (!cpumask_test_cpu(cpu, newdev->cpumask)) + return false; + if (cpumask_equal(newdev->cpumask, cpumask_of(cpu))) + return true; + /* Check if irq affinity can be set */ + if (newdev->irq >= 0 && !irq_can_set_affinity(newdev->irq)) + return false; + /* Prefer an existing cpu local device */ + if (curdev && cpumask_equal(curdev->cpumask, cpumask_of(cpu))) + return false; + return true; +} + +static bool tick_check_preferred(struct clock_event_device *curdev, + struct clock_event_device *newdev) +{ + /* Prefer oneshot capable device */ + if (!(newdev->features & CLOCK_EVT_FEAT_ONESHOT)) { + if (curdev && (curdev->features & CLOCK_EVT_FEAT_ONESHOT)) + return false; + if (tick_oneshot_mode_active()) + return false; + } + + /* + * Use the higher rated one, but prefer a CPU local device with a lower + * rating than a non-CPU local device + */ + return !curdev || + newdev->rating > curdev->rating || + !cpumask_equal(curdev->cpumask, newdev->cpumask); +} + /* * Check, if the new registered device should be used. */ -static int tick_check_new_device(struct clock_event_device *newdev) +void tick_check_new_device(struct clock_event_device *newdev) { struct clock_event_device *curdev; struct tick_device *td; - int cpu, ret = NOTIFY_OK; + int cpu; unsigned long flags; raw_spin_lock_irqsave(&tick_device_lock, flags); @@ -225,40 +263,15 @@ static int tick_check_new_device(struct clock_event_device *newdev) curdev = td->evtdev; /* cpu local device ? */ - if (!cpumask_equal(newdev->cpumask, cpumask_of(cpu))) { - - /* - * If the cpu affinity of the device interrupt can not - * be set, ignore it. - */ - if (!irq_can_set_affinity(newdev->irq)) - goto out_bc; + if (!tick_check_percpu(curdev, newdev, cpu)) + goto out_bc; - /* - * If we have a cpu local device already, do not replace it - * by a non cpu local device - */ - if (curdev && cpumask_equal(curdev->cpumask, cpumask_of(cpu))) - goto out_bc; - } + /* Preference decision */ + if (!tick_check_preferred(curdev, newdev)) + goto out_bc; - /* - * If we have an active device, then check the rating and the oneshot - * feature. - */ - if (curdev) { - /* - * Prefer one shot capable devices ! - */ - if ((curdev->features & CLOCK_EVT_FEAT_ONESHOT) && - !(newdev->features & CLOCK_EVT_FEAT_ONESHOT)) - goto out_bc; - /* - * Check the rating - */ - if (curdev->rating >= newdev->rating) - goto out_bc; - } + if (!try_module_get(newdev->owner)) + return; /* * Replace the eventually existing device by the new @@ -275,18 +288,14 @@ static int tick_check_new_device(struct clock_event_device *newdev) tick_oneshot_notify(); raw_spin_unlock_irqrestore(&tick_device_lock, flags); - return NOTIFY_STOP; + return; out_bc: /* * Can the new device be used as a broadcast device ? */ - if (tick_check_broadcast_device(newdev)) - ret = NOTIFY_STOP; - + tick_install_broadcast_device(newdev); raw_spin_unlock_irqrestore(&tick_device_lock, flags); - - return ret; } /* @@ -360,17 +369,10 @@ static void tick_resume(void) raw_spin_unlock_irqrestore(&tick_device_lock, flags); } -/* - * Notification about clock event devices - */ -static int tick_notify(struct notifier_block *nb, unsigned long reason, - void *dev) +void tick_notify(unsigned long reason, void *dev) { switch (reason) { - case CLOCK_EVT_NOTIFY_ADD: - return tick_check_new_device(dev); - case CLOCK_EVT_NOTIFY_BROADCAST_ON: case CLOCK_EVT_NOTIFY_BROADCAST_OFF: case CLOCK_EVT_NOTIFY_BROADCAST_FORCE: @@ -404,21 +406,12 @@ static int tick_notify(struct notifier_block *nb, unsigned long reason, default: break; } - - return NOTIFY_OK; } -static struct notifier_block tick_notifier = { - .notifier_call = tick_notify, -}; - /** * tick_init - initialize the tick control - * - * Register the notifier with the clockevents framework */ void __init tick_init(void) { - clockevents_register_notifier(&tick_notifier); tick_broadcast_init(); } diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h index f0299eae4602..60742fe6f63d 100644 --- a/kernel/time/tick-internal.h +++ b/kernel/time/tick-internal.h @@ -18,6 +18,8 @@ extern int tick_do_timer_cpu __read_mostly; extern void tick_setup_periodic(struct clock_event_device *dev, int broadcast); extern void tick_handle_periodic(struct clock_event_device *dev); +extern void tick_notify(unsigned long reason, void *dev); +extern void tick_check_new_device(struct clock_event_device *dev); extern void clockevents_shutdown(struct clock_event_device *dev); @@ -90,7 +92,7 @@ static inline bool tick_broadcast_oneshot_available(void) { return false; } */ #ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST extern int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu); -extern int tick_check_broadcast_device(struct clock_event_device *dev); +extern void tick_install_broadcast_device(struct clock_event_device *dev); extern int tick_is_broadcast_device(struct clock_event_device *dev); extern void tick_broadcast_on_off(unsigned long reason, int *oncpu); extern void tick_shutdown_broadcast(unsigned int *cpup); @@ -102,9 +104,8 @@ tick_set_periodic_handler(struct clock_event_device *dev, int broadcast); #else /* !BROADCAST */ -static inline int tick_check_broadcast_device(struct clock_event_device *dev) +static inline void tick_install_broadcast_device(struct clock_event_device *dev) { - return 0; } static inline int tick_is_broadcast_device(struct clock_event_device *dev) diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 0cf1c1453181..4251374578bc 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -832,13 +832,10 @@ void tick_nohz_irq_exit(void) { struct tick_sched *ts = &__get_cpu_var(tick_cpu_sched); - if (ts->inidle) { - /* Cancel the timer because CPU already waken up from the C-states*/ - menu_hrtimer_cancel(); + if (ts->inidle) __tick_nohz_idle_enter(ts); - } else { + else tick_nohz_full_stop_tick(ts); - } } /** @@ -936,8 +933,6 @@ void tick_nohz_idle_exit(void) ts->inidle = 0; - /* Cancel the timer because CPU already waken up from the C-states*/ - menu_hrtimer_cancel(); if (ts->idle_active || ts->tick_stopped) now = ktime_get(); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index baeeb5c87cf1..76fefb1613b2 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -72,7 +72,7 @@ static void tk_set_wall_to_mono(struct timekeeper *tk, struct timespec wtm) tk->wall_to_monotonic = wtm; set_normalized_timespec(&tmp, -wtm.tv_sec, -wtm.tv_nsec); tk->offs_real = timespec_to_ktime(tmp); - tk->offs_tai = ktime_sub(tk->offs_real, ktime_set(tk->tai_offset, 0)); + tk->offs_tai = ktime_add(tk->offs_real, ktime_set(tk->tai_offset, 0)); } static void tk_set_sleep_time(struct timekeeper *tk, struct timespec t) @@ -590,7 +590,7 @@ s32 timekeeping_get_tai_offset(void) static void __timekeeping_set_tai_offset(struct timekeeper *tk, s32 tai_offset) { tk->tai_offset = tai_offset; - tk->offs_tai = ktime_sub(tk->offs_real, ktime_set(tai_offset, 0)); + tk->offs_tai = ktime_add(tk->offs_real, ktime_set(tai_offset, 0)); } /** @@ -605,6 +605,7 @@ void timekeeping_set_tai_offset(s32 tai_offset) raw_spin_lock_irqsave(&timekeeper_lock, flags); write_seqcount_begin(&timekeeper_seq); __timekeeping_set_tai_offset(tk, tai_offset); + timekeeping_update(tk, false, true); write_seqcount_end(&timekeeper_seq); raw_spin_unlock_irqrestore(&timekeeper_lock, flags); clock_was_set(); @@ -1007,6 +1008,8 @@ static int timekeeping_suspend(void) timekeeping_suspend_time = timespec_add(timekeeping_suspend_time, delta_delta); } + + timekeeping_update(tk, false, true); write_seqcount_end(&timekeeper_seq); raw_spin_unlock_irqrestore(&timekeeper_lock, flags); @@ -1236,9 +1239,10 @@ out_adjust: * It also calls into the NTP code to handle leapsecond processing. * */ -static inline void accumulate_nsecs_to_secs(struct timekeeper *tk) +static inline unsigned int accumulate_nsecs_to_secs(struct timekeeper *tk) { u64 nsecps = (u64)NSEC_PER_SEC << tk->shift; + unsigned int clock_set = 0; while (tk->xtime_nsec >= nsecps) { int leap; @@ -1260,9 +1264,10 @@ static inline void accumulate_nsecs_to_secs(struct timekeeper *tk) __timekeeping_set_tai_offset(tk, tk->tai_offset - leap); - clock_was_set_delayed(); + clock_set = 1; } } + return clock_set; } /** @@ -1275,7 +1280,8 @@ static inline void accumulate_nsecs_to_secs(struct timekeeper *tk) * Returns the unconsumed cycles. */ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset, - u32 shift) + u32 shift, + unsigned int *clock_set) { cycle_t interval = tk->cycle_interval << shift; u64 raw_nsecs; @@ -1289,7 +1295,7 @@ static cycle_t logarithmic_accumulation(struct timekeeper *tk, cycle_t offset, tk->cycle_last += interval; tk->xtime_nsec += tk->xtime_interval << shift; - accumulate_nsecs_to_secs(tk); + *clock_set |= accumulate_nsecs_to_secs(tk); /* Accumulate raw time */ raw_nsecs = (u64)tk->raw_interval << shift; @@ -1328,7 +1334,7 @@ static inline void old_vsyscall_fixup(struct timekeeper *tk) tk->xtime_nsec -= remainder; tk->xtime_nsec += 1ULL << tk->shift; tk->ntp_error += remainder << tk->ntp_error_shift; - + tk->ntp_error -= (1ULL << tk->shift) << tk->ntp_error_shift; } #else #define old_vsyscall_fixup(tk) @@ -1347,6 +1353,7 @@ static void update_wall_time(void) struct timekeeper *tk = &shadow_timekeeper; cycle_t offset; int shift = 0, maxshift; + unsigned int clock_set = 0; unsigned long flags; raw_spin_lock_irqsave(&timekeeper_lock, flags); @@ -1381,7 +1388,8 @@ static void update_wall_time(void) maxshift = (64 - (ilog2(ntp_tick_length())+1)) - 1; shift = min(shift, maxshift); while (offset >= tk->cycle_interval) { - offset = logarithmic_accumulation(tk, offset, shift); + offset = logarithmic_accumulation(tk, offset, shift, + &clock_set); if (offset < tk->cycle_interval<<shift) shift--; } @@ -1399,7 +1407,7 @@ static void update_wall_time(void) * Finally, make sure that after the rounding * xtime_nsec isn't larger than NSEC_PER_SEC */ - accumulate_nsecs_to_secs(tk); + clock_set |= accumulate_nsecs_to_secs(tk); write_seqcount_begin(&timekeeper_seq); /* Update clock->cycle_last with the new value */ @@ -1419,6 +1427,10 @@ static void update_wall_time(void) write_seqcount_end(&timekeeper_seq); out: raw_spin_unlock_irqrestore(&timekeeper_lock, flags); + if (clock_set) + /* have to call outside the timekeeper_seq */ + clock_was_set_delayed(); + } /** @@ -1677,11 +1689,16 @@ int do_adjtimex(struct timex *txc) if (tai != orig_tai) { __timekeeping_set_tai_offset(tk, tai); - clock_was_set_delayed(); + timekeeping_update(tk, false, true); } write_seqcount_end(&timekeeper_seq); raw_spin_unlock_irqrestore(&timekeeper_lock, flags); + if (tai != orig_tai) + clock_was_set(); + + ntp_notify_cmos_timer(); + return ret; } diff --git a/kernel/time/timer_list.c b/kernel/time/timer_list.c index 3bdf28323012..61ed862cdd37 100644 --- a/kernel/time/timer_list.c +++ b/kernel/time/timer_list.c @@ -265,10 +265,9 @@ static inline void timer_list_header(struct seq_file *m, u64 now) static int timer_list_show(struct seq_file *m, void *v) { struct timer_list_iter *iter = v; - u64 now = ktime_to_ns(ktime_get()); if (iter->cpu == -1 && !iter->second_pass) - timer_list_header(m, now); + timer_list_header(m, iter->now); else if (!iter->second_pass) print_cpu(m, iter->cpu, iter->now); #ifdef CONFIG_GENERIC_CLOCKEVENTS @@ -298,33 +297,41 @@ void sysrq_timer_list_show(void) return; } -static void *timer_list_start(struct seq_file *file, loff_t *offset) +static void *move_iter(struct timer_list_iter *iter, loff_t offset) { - struct timer_list_iter *iter = file->private; - - if (!*offset) { - iter->cpu = -1; - iter->now = ktime_to_ns(ktime_get()); - } else if (iter->cpu >= nr_cpu_ids) { + for (; offset; offset--) { + iter->cpu = cpumask_next(iter->cpu, cpu_online_mask); + if (iter->cpu >= nr_cpu_ids) { #ifdef CONFIG_GENERIC_CLOCKEVENTS - if (!iter->second_pass) { - iter->cpu = -1; - iter->second_pass = true; - } else - return NULL; + if (!iter->second_pass) { + iter->cpu = -1; + iter->second_pass = true; + } else + return NULL; #else - return NULL; + return NULL; #endif + } } return iter; } +static void *timer_list_start(struct seq_file *file, loff_t *offset) +{ + struct timer_list_iter *iter = file->private; + + if (!*offset) + iter->now = ktime_to_ns(ktime_get()); + iter->cpu = -1; + iter->second_pass = false; + return move_iter(iter, *offset); +} + static void *timer_list_next(struct seq_file *file, void *v, loff_t *offset) { struct timer_list_iter *iter = file->private; - iter->cpu = cpumask_next(iter->cpu, cpu_online_mask); ++*offset; - return timer_list_start(file, offset); + return move_iter(iter, 1); } static void timer_list_stop(struct seq_file *seq, void *v) diff --git a/kernel/timer.c b/kernel/timer.c index 15ffdb3f1948..20f45ea6f5a4 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -149,9 +149,11 @@ static unsigned long round_jiffies_common(unsigned long j, int cpu, /* now that we have rounded, subtract the extra skew again */ j -= cpu * 3; - if (j <= jiffies) /* rounding ate our timeout entirely; */ - return original; - return j; + /* + * Make sure j is still in the future. Otherwise return the + * unmodified value. + */ + return time_is_after_jiffies(j) ? j : original; } /** @@ -820,7 +822,7 @@ unsigned long apply_slack(struct timer_list *timer, unsigned long expires) bit = find_last_bit(&mask, BITS_PER_LONG); - mask = (1 << bit) - 1; + mask = (1UL << bit) - 1; expires_limit = expires_limit & ~(mask); diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index b8b8560bfb95..686417ba5cd1 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -685,6 +685,7 @@ void blk_trace_shutdown(struct request_queue *q) * blk_add_trace_rq - Add a trace for a request oriented action * @q: queue the io is for * @rq: the source request + * @nr_bytes: number of completed bytes * @what: the action * * Description: @@ -692,7 +693,7 @@ void blk_trace_shutdown(struct request_queue *q) * **/ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, - u32 what) + unsigned int nr_bytes, u32 what) { struct blk_trace *bt = q->blk_trace; @@ -701,11 +702,11 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, if (rq->cmd_type == REQ_TYPE_BLOCK_PC) { what |= BLK_TC_ACT(BLK_TC_PC); - __blk_add_trace(bt, 0, blk_rq_bytes(rq), rq->cmd_flags, + __blk_add_trace(bt, 0, nr_bytes, rq->cmd_flags, what, rq->errors, rq->cmd_len, rq->cmd); } else { what |= BLK_TC_ACT(BLK_TC_FS); - __blk_add_trace(bt, blk_rq_pos(rq), blk_rq_bytes(rq), + __blk_add_trace(bt, blk_rq_pos(rq), nr_bytes, rq->cmd_flags, what, rq->errors, 0, NULL); } } @@ -713,33 +714,34 @@ static void blk_add_trace_rq(struct request_queue *q, struct request *rq, static void blk_add_trace_rq_abort(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, BLK_TA_ABORT); + blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_ABORT); } static void blk_add_trace_rq_insert(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, BLK_TA_INSERT); + blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_INSERT); } static void blk_add_trace_rq_issue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, BLK_TA_ISSUE); + blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_ISSUE); } static void blk_add_trace_rq_requeue(void *ignore, struct request_queue *q, struct request *rq) { - blk_add_trace_rq(q, rq, BLK_TA_REQUEUE); + blk_add_trace_rq(q, rq, blk_rq_bytes(rq), BLK_TA_REQUEUE); } static void blk_add_trace_rq_complete(void *ignore, struct request_queue *q, - struct request *rq) + struct request *rq, + unsigned int nr_bytes) { - blk_add_trace_rq(q, rq, BLK_TA_COMPLETE); + blk_add_trace_rq(q, rq, nr_bytes, BLK_TA_COMPLETE); } /** diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 6c508ff33c62..797d3b91a30b 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -85,6 +85,8 @@ int function_trace_stop __read_mostly; /* Current function tracing op */ struct ftrace_ops *function_trace_op __read_mostly = &ftrace_list_end; +/* What to set function_trace_op to */ +static struct ftrace_ops *set_function_trace_op; /* List for set_ftrace_pid's pids. */ LIST_HEAD(ftrace_pids); @@ -278,6 +280,29 @@ static void update_global_ops(void) global_ops.func = func; } +static void ftrace_sync(struct work_struct *work) +{ + /* + * This function is just a stub to implement a hard force + * of synchronize_sched(). This requires synchronizing + * tasks even in userspace and idle. + * + * Yes, function tracing is rude. + */ +} + +static void ftrace_sync_ipi(void *data) +{ + /* Probably not needed, but do it anyway */ + smp_rmb(); +} + +#ifdef CONFIG_FUNCTION_GRAPH_TRACER +static void update_function_graph_func(void); +#else +static inline void update_function_graph_func(void) { } +#endif + static void update_ftrace_function(void) { ftrace_func_t func; @@ -296,16 +321,61 @@ static void update_ftrace_function(void) !FTRACE_FORCE_LIST_FUNC)) { /* Set the ftrace_ops that the arch callback uses */ if (ftrace_ops_list == &global_ops) - function_trace_op = ftrace_global_list; + set_function_trace_op = ftrace_global_list; else - function_trace_op = ftrace_ops_list; + set_function_trace_op = ftrace_ops_list; func = ftrace_ops_list->func; } else { /* Just use the default ftrace_ops */ - function_trace_op = &ftrace_list_end; + set_function_trace_op = &ftrace_list_end; func = ftrace_ops_list_func; } + /* If there's no change, then do nothing more here */ + if (ftrace_trace_function == func) + return; + + update_function_graph_func(); + + /* + * If we are using the list function, it doesn't care + * about the function_trace_ops. + */ + if (func == ftrace_ops_list_func) { + ftrace_trace_function = func; + /* + * Don't even bother setting function_trace_ops, + * it would be racy to do so anyway. + */ + return; + } + +#ifndef CONFIG_DYNAMIC_FTRACE + /* + * For static tracing, we need to be a bit more careful. + * The function change takes affect immediately. Thus, + * we need to coorditate the setting of the function_trace_ops + * with the setting of the ftrace_trace_function. + * + * Set the function to the list ops, which will call the + * function we want, albeit indirectly, but it handles the + * ftrace_ops and doesn't depend on function_trace_op. + */ + ftrace_trace_function = ftrace_ops_list_func; + /* + * Make sure all CPUs see this. Yes this is slow, but static + * tracing is slow and nasty to have enabled. + */ + schedule_on_each_cpu(ftrace_sync); + /* Now all cpus are using the list ops. */ + function_trace_op = set_function_trace_op; + /* Make sure the function_trace_op is visible on all CPUs */ + smp_wmb(); + /* Nasty way to force a rmb on all cpus */ + smp_call_function(ftrace_sync_ipi, NULL, 1); + /* OK, we are all set to update the ftrace_trace_function now! */ +#endif /* !CONFIG_DYNAMIC_FTRACE */ + ftrace_trace_function = func; } @@ -367,9 +437,6 @@ static int remove_ftrace_list_ops(struct ftrace_ops **list, static int __register_ftrace_function(struct ftrace_ops *ops) { - if (unlikely(ftrace_disabled)) - return -ENODEV; - if (FTRACE_WARN_ON(ops == &global_ops)) return -EINVAL; @@ -417,9 +484,6 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) { int ret; - if (ftrace_disabled) - return -ENODEV; - if (WARN_ON(!(ops->flags & FTRACE_OPS_FL_ENABLED))) return -EBUSY; @@ -434,16 +498,6 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) } else if (ops->flags & FTRACE_OPS_FL_CONTROL) { ret = remove_ftrace_list_ops(&ftrace_control_list, &control_ops, ops); - if (!ret) { - /* - * The ftrace_ops is now removed from the list, - * so there'll be no new users. We must ensure - * all current users are done before we free - * the control data. - */ - synchronize_sched(); - control_ops_free(ops); - } } else ret = remove_ftrace_ops(&ftrace_ops_list, ops); @@ -453,13 +507,6 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_enabled) update_ftrace_function(); - /* - * Dynamic ops may be freed, we must make sure that all - * callers are done before leaving this function. - */ - if (ops->flags & FTRACE_OPS_FL_DYNAMIC) - synchronize_sched(); - return 0; } @@ -756,7 +803,7 @@ static int ftrace_profile_init(void) int cpu; int ret = 0; - for_each_online_cpu(cpu) { + for_each_possible_cpu(cpu) { ret = ftrace_profile_init_cpu(cpu); if (ret) break; @@ -1416,12 +1463,22 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, * the hashes are freed with call_rcu_sched(). */ static int -ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip) +ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { struct ftrace_hash *filter_hash; struct ftrace_hash *notrace_hash; int ret; +#ifdef CONFIG_DYNAMIC_FTRACE_WITH_REGS + /* + * There's a small race when adding ops that the ftrace handler + * that wants regs, may be called without them. We can not + * allow that handler to be called if regs is NULL. + */ + if (regs == NULL && (ops->flags & FTRACE_OPS_FL_SAVE_REGS)) + return 0; +#endif + filter_hash = rcu_dereference_raw_notrace(ops->filter_hash); notrace_hash = rcu_dereference_raw_notrace(ops->notrace_hash); @@ -1948,8 +2005,14 @@ void ftrace_modify_all_code(int command) else if (command & FTRACE_DISABLE_CALLS) ftrace_replace_code(0); - if (command & FTRACE_UPDATE_TRACE_FUNC) + if (command & FTRACE_UPDATE_TRACE_FUNC) { + function_trace_op = set_function_trace_op; + smp_wmb(); + /* If irqs are disabled, we are in stop machine */ + if (!irqs_disabled()) + smp_call_function(ftrace_sync_ipi, NULL, 1); ftrace_update_ftrace_func(ftrace_trace_function); + } if (command & FTRACE_START_FUNC_RET) ftrace_enable_ftrace_graph_caller(); @@ -2038,10 +2101,15 @@ static void ftrace_startup_enable(int command) static int ftrace_startup(struct ftrace_ops *ops, int command) { bool hash_enable = true; + int ret; if (unlikely(ftrace_disabled)) return -ENODEV; + ret = __register_ftrace_function(ops); + if (ret) + return ret; + ftrace_start_up++; command |= FTRACE_UPDATE_CALLS; @@ -2063,12 +2131,17 @@ static int ftrace_startup(struct ftrace_ops *ops, int command) return 0; } -static void ftrace_shutdown(struct ftrace_ops *ops, int command) +static int ftrace_shutdown(struct ftrace_ops *ops, int command) { bool hash_disable = true; + int ret; if (unlikely(ftrace_disabled)) - return; + return -ENODEV; + + ret = __unregister_ftrace_function(ops); + if (ret) + return ret; ftrace_start_up--; /* @@ -2102,10 +2175,42 @@ static void ftrace_shutdown(struct ftrace_ops *ops, int command) command |= FTRACE_UPDATE_TRACE_FUNC; } - if (!command || !ftrace_enabled) - return; + if (!command || !ftrace_enabled) { + /* + * If these are control ops, they still need their + * per_cpu field freed. Since, function tracing is + * not currently active, we can just free them + * without synchronizing all CPUs. + */ + if (ops->flags & FTRACE_OPS_FL_CONTROL) + control_ops_free(ops); + return 0; + } ftrace_run_update_code(command); + + /* + * Dynamic ops may be freed, we must make sure that all + * callers are done before leaving this function. + * The same goes for freeing the per_cpu data of the control + * ops. + * + * Again, normal synchronize_sched() is not good enough. + * We need to do a hard force of sched synchronization. + * This is because we use preempt_disable() to do RCU, but + * the function tracers can be called where RCU is not watching + * (like before user_exit()). We can not rely on the RCU + * infrastructure to do the synchronization, thus we must do it + * ourselves. + */ + if (ops->flags & (FTRACE_OPS_FL_DYNAMIC | FTRACE_OPS_FL_CONTROL)) { + schedule_on_each_cpu(ftrace_sync); + + if (ops->flags & FTRACE_OPS_FL_CONTROL) + control_ops_free(ops); + } + + return 0; } static void ftrace_startup_sysctl(void) @@ -2134,12 +2239,57 @@ static cycle_t ftrace_update_time; static unsigned long ftrace_update_cnt; unsigned long ftrace_update_tot_cnt; -static int ops_traces_mod(struct ftrace_ops *ops) +static inline int ops_traces_mod(struct ftrace_ops *ops) { - struct ftrace_hash *hash; + /* + * Filter_hash being empty will default to trace module. + * But notrace hash requires a test of individual module functions. + */ + return ftrace_hash_empty(ops->filter_hash) && + ftrace_hash_empty(ops->notrace_hash); +} - hash = ops->filter_hash; - return ftrace_hash_empty(hash); +/* + * Check if the current ops references the record. + * + * If the ops traces all functions, then it was already accounted for. + * If the ops does not trace the current record function, skip it. + * If the ops ignores the function via notrace filter, skip it. + */ +static inline bool +ops_references_rec(struct ftrace_ops *ops, struct dyn_ftrace *rec) +{ + /* If ops isn't enabled, ignore it */ + if (!(ops->flags & FTRACE_OPS_FL_ENABLED)) + return 0; + + /* If ops traces all mods, we already accounted for it */ + if (ops_traces_mod(ops)) + return 0; + + /* The function must be in the filter */ + if (!ftrace_hash_empty(ops->filter_hash) && + !ftrace_lookup_ip(ops->filter_hash, rec->ip)) + return 0; + + /* If in notrace hash, we ignore it too */ + if (ftrace_lookup_ip(ops->notrace_hash, rec->ip)) + return 0; + + return 1; +} + +static int referenced_filters(struct dyn_ftrace *rec) +{ + struct ftrace_ops *ops; + int cnt = 0; + + for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) { + if (ops_references_rec(ops, rec)) + cnt++; + } + + return cnt; } static int ftrace_update_code(struct module *mod) @@ -2148,6 +2298,7 @@ static int ftrace_update_code(struct module *mod) struct dyn_ftrace *p; cycle_t start, stop; unsigned long ref = 0; + bool test = false; int i; /* @@ -2161,9 +2312,12 @@ static int ftrace_update_code(struct module *mod) for (ops = ftrace_ops_list; ops != &ftrace_list_end; ops = ops->next) { - if (ops->flags & FTRACE_OPS_FL_ENABLED && - ops_traces_mod(ops)) - ref++; + if (ops->flags & FTRACE_OPS_FL_ENABLED) { + if (ops_traces_mod(ops)) + ref++; + else + test = true; + } } } @@ -2173,12 +2327,16 @@ static int ftrace_update_code(struct module *mod) for (pg = ftrace_new_pgs; pg; pg = pg->next) { for (i = 0; i < pg->index; i++) { + int cnt = ref; + /* If something went wrong, bail without enabling anything */ if (unlikely(ftrace_disabled)) return -1; p = &pg->records[i]; - p->flags = ref; + if (test) + cnt += referenced_filters(p); + p->flags = cnt; /* * Do the initial record conversion from mcount jump @@ -2198,7 +2356,7 @@ static int ftrace_update_code(struct module *mod) * conversion puts the module to the correct state, thus * passing the ftrace_make_call check. */ - if (ftrace_start_up && ref) { + if (ftrace_start_up && cnt) { int failed = __ftrace_replace_code(p, 1); if (failed) ftrace_bug(failed, p->ip); @@ -2957,16 +3115,13 @@ static void __enable_ftrace_function_probe(void) if (i == FTRACE_FUNC_HASHSIZE) return; - ret = __register_ftrace_function(&trace_probe_ops); - if (!ret) - ret = ftrace_startup(&trace_probe_ops, 0); + ret = ftrace_startup(&trace_probe_ops, 0); ftrace_probe_registered = 1; } static void __disable_ftrace_function_probe(void) { - int ret; int i; if (!ftrace_probe_registered) @@ -2979,9 +3134,7 @@ static void __disable_ftrace_function_probe(void) } /* no more funcs left */ - ret = __unregister_ftrace_function(&trace_probe_ops); - if (!ret) - ftrace_shutdown(&trace_probe_ops, 0); + ftrace_shutdown(&trace_probe_ops, 0); ftrace_probe_registered = 0; } @@ -4069,16 +4222,11 @@ static void ftrace_init_module(struct module *mod, ftrace_process_locs(mod, start, end); } -static int ftrace_module_notify_enter(struct notifier_block *self, - unsigned long val, void *data) +void ftrace_module_init(struct module *mod) { - struct module *mod = data; - - if (val == MODULE_STATE_COMING) - ftrace_init_module(mod, mod->ftrace_callsites, - mod->ftrace_callsites + - mod->num_ftrace_callsites); - return 0; + ftrace_init_module(mod, mod->ftrace_callsites, + mod->ftrace_callsites + + mod->num_ftrace_callsites); } static int ftrace_module_notify_exit(struct notifier_block *self, @@ -4092,11 +4240,6 @@ static int ftrace_module_notify_exit(struct notifier_block *self, return 0; } #else -static int ftrace_module_notify_enter(struct notifier_block *self, - unsigned long val, void *data) -{ - return 0; -} static int ftrace_module_notify_exit(struct notifier_block *self, unsigned long val, void *data) { @@ -4104,11 +4247,6 @@ static int ftrace_module_notify_exit(struct notifier_block *self, } #endif /* CONFIG_MODULES */ -struct notifier_block ftrace_module_enter_nb = { - .notifier_call = ftrace_module_notify_enter, - .priority = INT_MAX, /* Run before anything that can use kprobes */ -}; - struct notifier_block ftrace_module_exit_nb = { .notifier_call = ftrace_module_notify_exit, .priority = INT_MIN, /* Run after anything that can remove kprobes */ @@ -4145,10 +4283,6 @@ void __init ftrace_init(void) __start_mcount_loc, __stop_mcount_loc); - ret = register_module_notifier(&ftrace_module_enter_nb); - if (ret) - pr_warning("Failed to register trace ftrace module enter notifier\n"); - ret = register_module_notifier(&ftrace_module_exit_nb); if (ret) pr_warning("Failed to register trace ftrace module exit notifier\n"); @@ -4178,17 +4312,20 @@ core_initcall(ftrace_nodyn_init); static inline int ftrace_init_dyn_debugfs(struct dentry *d_tracer) { return 0; } static inline void ftrace_startup_enable(int command) { } /* Keep as macros so we do not need to define the commands */ -# define ftrace_startup(ops, command) \ - ({ \ - (ops)->flags |= FTRACE_OPS_FL_ENABLED; \ - 0; \ +# define ftrace_startup(ops, command) \ + ({ \ + int ___ret = __register_ftrace_function(ops); \ + if (!___ret) \ + (ops)->flags |= FTRACE_OPS_FL_ENABLED; \ + ___ret; \ }) -# define ftrace_shutdown(ops, command) do { } while (0) +# define ftrace_shutdown(ops, command) __unregister_ftrace_function(ops) + # define ftrace_startup_sysctl() do { } while (0) # define ftrace_shutdown_sysctl() do { } while (0) static inline int -ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip) +ftrace_ops_test(struct ftrace_ops *ops, unsigned long ip, void *regs) { return 1; } @@ -4211,7 +4348,7 @@ ftrace_ops_control_func(unsigned long ip, unsigned long parent_ip, do_for_each_ftrace_op(op, ftrace_control_list) { if (!(op->flags & FTRACE_OPS_FL_STUB) && !ftrace_function_local_disabled(op) && - ftrace_ops_test(op, ip)) + ftrace_ops_test(op, ip, regs)) op->func(ip, parent_ip, op, regs); } while_for_each_ftrace_op(op); trace_recursion_clear(TRACE_CONTROL_BIT); @@ -4244,7 +4381,7 @@ __ftrace_ops_list_func(unsigned long ip, unsigned long parent_ip, */ preempt_disable_notrace(); do_for_each_ftrace_op(op, ftrace_ops_list) { - if (ftrace_ops_test(op, ip)) + if (ftrace_ops_test(op, ip, regs)) op->func(ip, parent_ip, op, regs); } while_for_each_ftrace_op(op); preempt_enable_notrace(); @@ -4583,9 +4720,7 @@ int register_ftrace_function(struct ftrace_ops *ops) mutex_lock(&ftrace_lock); - ret = __register_ftrace_function(ops); - if (!ret) - ret = ftrace_startup(ops, 0); + ret = ftrace_startup(ops, 0); mutex_unlock(&ftrace_lock); @@ -4604,9 +4739,7 @@ int unregister_ftrace_function(struct ftrace_ops *ops) int ret; mutex_lock(&ftrace_lock); - ret = __unregister_ftrace_function(ops); - if (!ret) - ftrace_shutdown(ops, 0); + ret = ftrace_shutdown(ops, 0); mutex_unlock(&ftrace_lock); return ret; @@ -4666,6 +4799,7 @@ int ftrace_graph_entry_stub(struct ftrace_graph_ent *trace) trace_func_graph_ret_t ftrace_graph_return = (trace_func_graph_ret_t)ftrace_stub; trace_func_graph_ent_t ftrace_graph_entry = ftrace_graph_entry_stub; +static trace_func_graph_ent_t __ftrace_graph_entry = ftrace_graph_entry_stub; /* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */ static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) @@ -4800,6 +4934,37 @@ ftrace_suspend_notifier_call(struct notifier_block *bl, unsigned long state, return NOTIFY_DONE; } +/* Just a place holder for function graph */ +static struct ftrace_ops fgraph_ops __read_mostly = { + .func = ftrace_stub, + .flags = FTRACE_OPS_FL_STUB | FTRACE_OPS_FL_GLOBAL | + FTRACE_OPS_FL_RECURSION_SAFE, +}; + +static int ftrace_graph_entry_test(struct ftrace_graph_ent *trace) +{ + if (!ftrace_ops_test(&global_ops, trace->func, NULL)) + return 0; + return __ftrace_graph_entry(trace); +} + +/* + * The function graph tracer should only trace the functions defined + * by set_ftrace_filter and set_ftrace_notrace. If another function + * tracer ops is registered, the graph tracer requires testing the + * function against the global ops, and not just trace any function + * that any ftrace_ops registered. + */ +static void update_function_graph_func(void) +{ + if (ftrace_ops_list == &ftrace_list_end || + (ftrace_ops_list == &global_ops && + global_ops.next == &ftrace_list_end)) + ftrace_graph_entry = __ftrace_graph_entry; + else + ftrace_graph_entry = ftrace_graph_entry_test; +} + int register_ftrace_graph(trace_func_graph_ret_t retfunc, trace_func_graph_ent_t entryfunc) { @@ -4824,9 +4989,18 @@ int register_ftrace_graph(trace_func_graph_ret_t retfunc, } ftrace_graph_return = retfunc; - ftrace_graph_entry = entryfunc; - ret = ftrace_startup(&global_ops, FTRACE_START_FUNC_RET); + /* + * Update the indirect function to the entryfunc, and the + * function that gets called to the entry_test first. Then + * call the update fgraph entry function to determine if + * the entryfunc should be called directly or not. + */ + __ftrace_graph_entry = entryfunc; + ftrace_graph_entry = ftrace_graph_entry_test; + update_function_graph_func(); + + ret = ftrace_startup(&fgraph_ops, FTRACE_START_FUNC_RET); out: mutex_unlock(&ftrace_lock); @@ -4843,7 +5017,8 @@ void unregister_ftrace_graph(void) ftrace_graph_active--; ftrace_graph_return = (trace_func_graph_ret_t)ftrace_stub; ftrace_graph_entry = ftrace_graph_entry_stub; - ftrace_shutdown(&global_ops, FTRACE_STOP_FUNC_RET); + __ftrace_graph_entry = ftrace_graph_entry_stub; + ftrace_shutdown(&fgraph_ops, FTRACE_STOP_FUNC_RET); unregister_pm_notifier(&ftrace_suspend_notifier); unregister_trace_sched_switch(ftrace_graph_probe_sched_switch, NULL); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index e444ff88f0a4..fd12cc56371f 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2396,6 +2396,13 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, write &= RB_WRITE_MASK; tail = write - length; + /* + * If this is the first commit on the page, then it has the same + * timestamp as the page itself. + */ + if (!tail) + delta = 0; + /* See if we shot pass the end of this buffer page */ if (unlikely(write > BUF_PAGE_SIZE)) return rb_move_tail(cpu_buffer, length, tail, diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index e71a8be4a6ee..6dbdf277c8fe 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -193,6 +193,37 @@ static struct trace_array global_trace; LIST_HEAD(ftrace_trace_arrays); +int trace_array_get(struct trace_array *this_tr) +{ + struct trace_array *tr; + int ret = -ENODEV; + + mutex_lock(&trace_types_lock); + list_for_each_entry(tr, &ftrace_trace_arrays, list) { + if (tr == this_tr) { + tr->ref++; + ret = 0; + break; + } + } + mutex_unlock(&trace_types_lock); + + return ret; +} + +static void __trace_array_put(struct trace_array *this_tr) +{ + WARN_ON(!this_tr->ref); + this_tr->ref--; +} + +void trace_array_put(struct trace_array *this_tr) +{ + mutex_lock(&trace_types_lock); + __trace_array_put(this_tr); + mutex_unlock(&trace_types_lock); +} + int filter_current_check_discard(struct ring_buffer *buffer, struct ftrace_event_call *call, void *rec, struct ring_buffer_event *event) @@ -201,23 +232,43 @@ int filter_current_check_discard(struct ring_buffer *buffer, } EXPORT_SYMBOL_GPL(filter_current_check_discard); -cycle_t ftrace_now(int cpu) +cycle_t buffer_ftrace_now(struct trace_buffer *buf, int cpu) { u64 ts; /* Early boot up does not have a buffer yet */ - if (!global_trace.trace_buffer.buffer) + if (!buf->buffer) return trace_clock_local(); - ts = ring_buffer_time_stamp(global_trace.trace_buffer.buffer, cpu); - ring_buffer_normalize_time_stamp(global_trace.trace_buffer.buffer, cpu, &ts); + ts = ring_buffer_time_stamp(buf->buffer, cpu); + ring_buffer_normalize_time_stamp(buf->buffer, cpu, &ts); return ts; } +cycle_t ftrace_now(int cpu) +{ + return buffer_ftrace_now(&global_trace.trace_buffer, cpu); +} + +/** + * tracing_is_enabled - Show if global_trace has been disabled + * + * Shows if the global trace has been enabled or not. It uses the + * mirror flag "buffer_disabled" to be used in fast paths such as for + * the irqsoff tracer. But it may be inaccurate due to races. If you + * need to know the accurate state, use tracing_is_on() which is a little + * slower, but accurate. + */ int tracing_is_enabled(void) { - return tracing_is_on(); + /* + * For quick access (irqsoff uses this in fast path), just + * return the mirror variable of the state of the ring buffer. + * It's a little racy, but we don't really care. + */ + smp_rmb(); + return !global_trace.buffer_disabled; } /* @@ -240,7 +291,7 @@ static struct tracer *trace_types __read_mostly; /* * trace_types_lock is used to protect the trace_types list. */ -static DEFINE_MUTEX(trace_types_lock); +DEFINE_MUTEX(trace_types_lock); /* * serialize the access of the ring buffer @@ -330,6 +381,23 @@ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | TRACE_ITER_GRAPH_TIME | TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE | TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS | TRACE_ITER_FUNCTION; +void tracer_tracing_on(struct trace_array *tr) +{ + if (tr->trace_buffer.buffer) + ring_buffer_record_on(tr->trace_buffer.buffer); + /* + * This flag is looked at when buffers haven't been allocated + * yet, or by some tracers (like irqsoff), that just want to + * know if the ring buffer has been disabled, but it can handle + * races of where it gets disabled but we still do a record. + * As the check is in the fast path of the tracers, it is more + * important to be fast than accurate. + */ + tr->buffer_disabled = 0; + /* Make the flag seen by readers */ + smp_wmb(); +} + /** * tracing_on - enable tracing buffers * @@ -338,15 +406,7 @@ unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | */ void tracing_on(void) { - if (global_trace.trace_buffer.buffer) - ring_buffer_record_on(global_trace.trace_buffer.buffer); - /* - * This flag is only looked at when buffers haven't been - * allocated yet. We don't really care about the race - * between setting this flag and actually turning - * on the buffer. - */ - global_trace.buffer_disabled = 0; + tracer_tracing_on(&global_trace); } EXPORT_SYMBOL_GPL(tracing_on); @@ -364,6 +424,9 @@ int __trace_puts(unsigned long ip, const char *str, int size) unsigned long irq_flags; int alloc; + if (unlikely(tracing_selftest_running || tracing_disabled)) + return 0; + alloc = sizeof(*entry) + size + 2; /* possible \n added */ local_save_flags(irq_flags); @@ -404,6 +467,9 @@ int __trace_bputs(unsigned long ip, const char *str) unsigned long irq_flags; int size = sizeof(struct bputs_entry); + if (unlikely(tracing_selftest_running || tracing_disabled)) + return 0; + local_save_flags(irq_flags); buffer = global_trace.trace_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_BPUTS, size, @@ -540,6 +606,23 @@ void tracing_snapshot_alloc(void) EXPORT_SYMBOL_GPL(tracing_snapshot_alloc); #endif /* CONFIG_TRACER_SNAPSHOT */ +void tracer_tracing_off(struct trace_array *tr) +{ + if (tr->trace_buffer.buffer) + ring_buffer_record_off(tr->trace_buffer.buffer); + /* + * This flag is looked at when buffers haven't been allocated + * yet, or by some tracers (like irqsoff), that just want to + * know if the ring buffer has been disabled, but it can handle + * races of where it gets disabled but we still do a record. + * As the check is in the fast path of the tracers, it is more + * important to be fast than accurate. + */ + tr->buffer_disabled = 1; + /* Make the flag seen by readers */ + smp_wmb(); +} + /** * tracing_off - turn off tracing buffers * @@ -550,26 +633,29 @@ EXPORT_SYMBOL_GPL(tracing_snapshot_alloc); */ void tracing_off(void) { - if (global_trace.trace_buffer.buffer) - ring_buffer_record_off(global_trace.trace_buffer.buffer); - /* - * This flag is only looked at when buffers haven't been - * allocated yet. We don't really care about the race - * between setting this flag and actually turning - * on the buffer. - */ - global_trace.buffer_disabled = 1; + tracer_tracing_off(&global_trace); } EXPORT_SYMBOL_GPL(tracing_off); /** + * tracer_tracing_is_on - show real state of ring buffer enabled + * @tr : the trace array to know if ring buffer is enabled + * + * Shows real state of the ring buffer if it is enabled or not. + */ +int tracer_tracing_is_on(struct trace_array *tr) +{ + if (tr->trace_buffer.buffer) + return ring_buffer_record_is_on(tr->trace_buffer.buffer); + return !tr->buffer_disabled; +} + +/** * tracing_is_on - show state of ring buffers enabled */ int tracing_is_on(void) { - if (global_trace.trace_buffer.buffer) - return ring_buffer_record_is_on(global_trace.trace_buffer.buffer); - return !global_trace.buffer_disabled; + return tracer_tracing_is_on(&global_trace); } EXPORT_SYMBOL_GPL(tracing_is_on); @@ -746,9 +832,12 @@ int trace_get_user(struct trace_parser *parser, const char __user *ubuf, if (isspace(ch)) { parser->buffer[parser->idx] = 0; parser->cont = false; - } else { + } else if (parser->idx < parser->size - 1) { parser->cont = true; parser->buffer[parser->idx++] = ch; + } else { + ret = -EINVAL; + goto out; } *ppos += read; @@ -1119,7 +1208,7 @@ void tracing_reset_online_cpus(struct trace_buffer *buf) /* Make sure all commits have finished */ synchronize_sched(); - buf->time_start = ftrace_now(buf->cpu); + buf->time_start = buffer_ftrace_now(buf, buf->cpu); for_each_online_cpu(cpu) ring_buffer_reset_cpu(buffer, cpu); @@ -1127,23 +1216,17 @@ void tracing_reset_online_cpus(struct trace_buffer *buf) ring_buffer_record_enable(buffer); } -void tracing_reset_current(int cpu) -{ - tracing_reset(&global_trace.trace_buffer, cpu); -} - +/* Must have trace_types_lock held */ void tracing_reset_all_online_cpus(void) { struct trace_array *tr; - mutex_lock(&trace_types_lock); list_for_each_entry(tr, &ftrace_trace_arrays, list) { tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE tracing_reset_online_cpus(&tr->max_buffer); #endif } - mutex_unlock(&trace_types_lock); } #define SAVED_CMDLINES 128 @@ -2760,6 +2843,17 @@ static int s_show(struct seq_file *m, void *v) return 0; } +/* + * Should be used after trace_array_get(), trace_types_lock + * ensures that i_cdev was already initialized. + */ +static inline int tracing_get_cpu(struct inode *inode) +{ + if (inode->i_cdev) /* See trace_create_cpu_file() */ + return (long)inode->i_cdev - 1; + return RING_BUFFER_ALL_CPUS; +} + static const struct seq_operations tracer_seq_ops = { .start = s_start, .next = s_next, @@ -2770,8 +2864,7 @@ static const struct seq_operations tracer_seq_ops = { static struct trace_iterator * __tracing_open(struct inode *inode, struct file *file, bool snapshot) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int cpu; @@ -2812,8 +2905,8 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) iter->trace_buffer = &tr->trace_buffer; iter->snapshot = snapshot; iter->pos = -1; + iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); - iter->cpu_file = tc->cpu; /* Notify the tracer early; before we stop tracing. */ if (iter->trace && iter->trace->open) @@ -2850,8 +2943,6 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) tracing_iter_reset(iter, cpu); } - tr->ref++; - mutex_unlock(&trace_types_lock); return iter; @@ -2874,24 +2965,41 @@ int tracing_open_generic(struct inode *inode, struct file *filp) return 0; } +/* + * Open and update trace_array ref count. + * Must have the current trace_array passed to it. + */ +int tracing_open_generic_tr(struct inode *inode, struct file *filp) +{ + struct trace_array *tr = inode->i_private; + + if (tracing_disabled) + return -ENODEV; + + if (trace_array_get(tr) < 0) + return -ENODEV; + + filp->private_data = inode->i_private; + + return 0; +} + static int tracing_release(struct inode *inode, struct file *file) { + struct trace_array *tr = inode->i_private; struct seq_file *m = file->private_data; struct trace_iterator *iter; - struct trace_array *tr; int cpu; - if (!(file->f_mode & FMODE_READ)) + if (!(file->f_mode & FMODE_READ)) { + trace_array_put(tr); return 0; + } + /* Writes do not use seq_file */ iter = m->private; - tr = iter->tr; - mutex_lock(&trace_types_lock); - WARN_ON(!tr->ref); - tr->ref--; - for_each_tracing_cpu(cpu) { if (iter->buffer_iter[cpu]) ring_buffer_read_finish(iter->buffer_iter[cpu]); @@ -2903,6 +3011,9 @@ static int tracing_release(struct inode *inode, struct file *file) if (!iter->snapshot) /* reenable tracing if it was previously enabled */ tracing_start_tr(tr); + + __trace_array_put(tr); + mutex_unlock(&trace_types_lock); mutex_destroy(&iter->mutex); @@ -2910,24 +3021,44 @@ static int tracing_release(struct inode *inode, struct file *file) kfree(iter->trace); kfree(iter->buffer_iter); seq_release_private(inode, file); + return 0; } +static int tracing_release_generic_tr(struct inode *inode, struct file *file) +{ + struct trace_array *tr = inode->i_private; + + trace_array_put(tr); + return 0; +} + +static int tracing_single_release_tr(struct inode *inode, struct file *file) +{ + struct trace_array *tr = inode->i_private; + + trace_array_put(tr); + + return single_release(inode, file); +} + static int tracing_open(struct inode *inode, struct file *file) { + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int ret = 0; + if (trace_array_get(tr) < 0) + return -ENODEV; + /* If this file was open for write, then erase contents */ - if ((file->f_mode & FMODE_WRITE) && - (file->f_flags & O_TRUNC)) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { + int cpu = tracing_get_cpu(inode); - if (tc->cpu == RING_BUFFER_ALL_CPUS) + if (cpu == RING_BUFFER_ALL_CPUS) tracing_reset_online_cpus(&tr->trace_buffer); else - tracing_reset(&tr->trace_buffer, tc->cpu); + tracing_reset(&tr->trace_buffer, cpu); } if (file->f_mode & FMODE_READ) { @@ -2937,6 +3068,10 @@ static int tracing_open(struct inode *inode, struct file *file) else if (trace_flags & TRACE_ITER_LATENCY_FMT) iter->iter_flags |= TRACE_FILE_LAT_FMT; } + + if (ret < 0) + trace_array_put(tr); + return ret; } @@ -3293,17 +3428,27 @@ tracing_trace_options_write(struct file *filp, const char __user *ubuf, static int tracing_trace_options_open(struct inode *inode, struct file *file) { + struct trace_array *tr = inode->i_private; + int ret; + if (tracing_disabled) return -ENODEV; - return single_open(file, tracing_trace_options_show, inode->i_private); + if (trace_array_get(tr) < 0) + return -ENODEV; + + ret = single_open(file, tracing_trace_options_show, inode->i_private); + if (ret < 0) + trace_array_put(tr); + + return ret; } static const struct file_operations tracing_iter_fops = { .open = tracing_trace_options_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = tracing_single_release_tr, .write = tracing_trace_options_write, }; @@ -3783,20 +3928,23 @@ tracing_max_lat_write(struct file *filp, const char __user *ubuf, static int tracing_open_pipe(struct inode *inode, struct file *filp) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; int ret = 0; if (tracing_disabled) return -ENODEV; + if (trace_array_get(tr) < 0) + return -ENODEV; + mutex_lock(&trace_types_lock); /* create a buffer to store the information to pass to userspace */ iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { ret = -ENOMEM; + __trace_array_put(tr); goto out; } @@ -3826,9 +3974,9 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) if (trace_clocks[tr->clock_id].in_ns) iter->iter_flags |= TRACE_FILE_TIME_IN_NS; - iter->cpu_file = tc->cpu; - iter->tr = tc->tr; - iter->trace_buffer = &tc->tr->trace_buffer; + iter->tr = tr; + iter->trace_buffer = &tr->trace_buffer; + iter->cpu_file = tracing_get_cpu(inode); mutex_init(&iter->mutex); filp->private_data = iter; @@ -3843,6 +3991,7 @@ out: fail: kfree(iter->trace); kfree(iter); + __trace_array_put(tr); mutex_unlock(&trace_types_lock); return ret; } @@ -3850,6 +3999,7 @@ fail: static int tracing_release_pipe(struct inode *inode, struct file *file) { struct trace_iterator *iter = file->private_data; + struct trace_array *tr = inode->i_private; mutex_lock(&trace_types_lock); @@ -3863,6 +4013,8 @@ static int tracing_release_pipe(struct inode *inode, struct file *file) kfree(iter->trace); kfree(iter); + trace_array_put(tr); + return 0; } @@ -3939,7 +4091,7 @@ static int tracing_wait_pipe(struct file *filp) * * iter->pos will be 0 if we haven't read anything. */ - if (!tracing_is_enabled() && iter->pos) + if (!tracing_is_on() && iter->pos) break; } @@ -4000,6 +4152,7 @@ waitagain: memset(&iter->seq, 0, sizeof(struct trace_iterator) - offsetof(struct trace_iterator, seq)); + cpumask_clear(iter->started); iter->pos = -1; trace_event_read_lock(); @@ -4200,15 +4353,16 @@ static ssize_t tracing_entries_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; - struct trace_array *tr = tc->tr; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; + int cpu = tracing_get_cpu(inode); char buf[64]; int r = 0; ssize_t ret; mutex_lock(&trace_types_lock); - if (tc->cpu == RING_BUFFER_ALL_CPUS) { + if (cpu == RING_BUFFER_ALL_CPUS) { int cpu, buf_size_same; unsigned long size; @@ -4235,7 +4389,7 @@ tracing_entries_read(struct file *filp, char __user *ubuf, } else r = sprintf(buf, "X\n"); } else - r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->trace_buffer.data, tc->cpu)->entries >> 10); + r = sprintf(buf, "%lu\n", per_cpu_ptr(tr->trace_buffer.data, cpu)->entries >> 10); mutex_unlock(&trace_types_lock); @@ -4247,7 +4401,8 @@ static ssize_t tracing_entries_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; unsigned long val; int ret; @@ -4261,8 +4416,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, /* value is in KB */ val <<= 10; - - ret = tracing_resize_ring_buffer(tc->tr, val, tc->cpu); + ret = tracing_resize_ring_buffer(tr, val, tracing_get_cpu(inode)); if (ret < 0) return ret; @@ -4316,10 +4470,12 @@ tracing_free_buffer_release(struct inode *inode, struct file *filp) /* disable tracing ? */ if (trace_flags & TRACE_ITER_STOP_ON_FREE) - tracing_off(); + tracer_tracing_off(tr); /* resize the ring buffer to 0 */ tracing_resize_ring_buffer(tr, 0, RING_BUFFER_ALL_CPUS); + trace_array_put(tr); + return 0; } @@ -4328,6 +4484,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *fpos) { unsigned long addr = (unsigned long)ubuf; + struct trace_array *tr = filp->private_data; struct ring_buffer_event *event; struct ring_buffer *buffer; struct print_entry *entry; @@ -4387,7 +4544,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, local_save_flags(irq_flags); size = sizeof(*entry) + cnt + 2; /* possible \n added */ - buffer = global_trace.trace_buffer.buffer; + buffer = tr->trace_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, irq_flags, preempt_count()); if (!event) { @@ -4478,12 +4635,12 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, * New clock may not be consistent with the previous clock. * Reset the buffer so that it doesn't have incomparable timestamps. */ - tracing_reset_online_cpus(&global_trace.trace_buffer); + tracing_reset_online_cpus(&tr->trace_buffer); #ifdef CONFIG_TRACER_MAX_TRACE if (tr->flags & TRACE_ARRAY_FL_GLOBAL && tr->max_buffer.buffer) ring_buffer_set_clock(tr->max_buffer.buffer, trace_clocks[i].func); - tracing_reset_online_cpus(&global_trace.max_buffer); + tracing_reset_online_cpus(&tr->max_buffer); #endif mutex_unlock(&trace_types_lock); @@ -4495,10 +4652,20 @@ static ssize_t tracing_clock_write(struct file *filp, const char __user *ubuf, static int tracing_clock_open(struct inode *inode, struct file *file) { + struct trace_array *tr = inode->i_private; + int ret; + if (tracing_disabled) return -ENODEV; - return single_open(file, tracing_clock_show, inode->i_private); + if (trace_array_get(tr)) + return -ENODEV; + + ret = single_open(file, tracing_clock_show, inode->i_private); + if (ret < 0) + trace_array_put(tr); + + return ret; } struct ftrace_buffer_info { @@ -4510,31 +4677,40 @@ struct ftrace_buffer_info { #ifdef CONFIG_TRACER_SNAPSHOT static int tracing_snapshot_open(struct inode *inode, struct file *file) { - struct trace_cpu *tc = inode->i_private; + struct trace_array *tr = inode->i_private; struct trace_iterator *iter; struct seq_file *m; int ret = 0; + if (trace_array_get(tr) < 0) + return -ENODEV; + if (file->f_mode & FMODE_READ) { iter = __tracing_open(inode, file, true); if (IS_ERR(iter)) ret = PTR_ERR(iter); } else { /* Writes still need the seq_file to hold the private data */ + ret = -ENOMEM; m = kzalloc(sizeof(*m), GFP_KERNEL); if (!m) - return -ENOMEM; + goto out; iter = kzalloc(sizeof(*iter), GFP_KERNEL); if (!iter) { kfree(m); - return -ENOMEM; + goto out; } - iter->tr = tc->tr; - iter->trace_buffer = &tc->tr->max_buffer; - iter->cpu_file = tc->cpu; + ret = 0; + + iter->tr = tr; + iter->trace_buffer = &tr->max_buffer; + iter->cpu_file = tracing_get_cpu(inode); m->private = iter; file->private_data = m; } +out: + if (ret < 0) + trace_array_put(tr); return ret; } @@ -4616,9 +4792,12 @@ out: static int tracing_snapshot_release(struct inode *inode, struct file *file) { struct seq_file *m = file->private_data; + int ret; + + ret = tracing_release(inode, file); if (file->f_mode & FMODE_READ) - return tracing_release(inode, file); + return ret; /* If write only, the seq_file is just a stub */ if (m) @@ -4684,34 +4863,38 @@ static const struct file_operations tracing_pipe_fops = { }; static const struct file_operations tracing_entries_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_entries_read, .write = tracing_entries_write, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; static const struct file_operations tracing_total_entries_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_total_entries_read, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; static const struct file_operations tracing_free_buffer_fops = { + .open = tracing_open_generic_tr, .write = tracing_free_buffer_write, .release = tracing_free_buffer_release, }; static const struct file_operations tracing_mark_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .write = tracing_mark_write, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; static const struct file_operations trace_clock_fops = { .open = tracing_clock_open, .read = seq_read, .llseek = seq_lseek, - .release = single_release, + .release = tracing_single_release_tr, .write = tracing_clock_write, }; @@ -4736,23 +4919,26 @@ static const struct file_operations snapshot_raw_fops = { static int tracing_buffers_open(struct inode *inode, struct file *filp) { - struct trace_cpu *tc = inode->i_private; - struct trace_array *tr = tc->tr; + struct trace_array *tr = inode->i_private; struct ftrace_buffer_info *info; + int ret; if (tracing_disabled) return -ENODEV; + if (trace_array_get(tr) < 0) + return -ENODEV; + info = kzalloc(sizeof(*info), GFP_KERNEL); - if (!info) + if (!info) { + trace_array_put(tr); return -ENOMEM; + } mutex_lock(&trace_types_lock); - tr->ref++; - info->iter.tr = tr; - info->iter.cpu_file = tc->cpu; + info->iter.cpu_file = tracing_get_cpu(inode); info->iter.trace = tr->current_trace; info->iter.trace_buffer = &tr->trace_buffer; info->spare = NULL; @@ -4763,7 +4949,11 @@ static int tracing_buffers_open(struct inode *inode, struct file *filp) mutex_unlock(&trace_types_lock); - return nonseekable_open(inode, filp); + ret = nonseekable_open(inode, filp); + if (ret < 0) + trace_array_put(tr); + + return ret; } static unsigned int @@ -4863,8 +5053,7 @@ static int tracing_buffers_release(struct inode *inode, struct file *file) mutex_lock(&trace_types_lock); - WARN_ON(!iter->tr->ref); - iter->tr->ref--; + __trace_array_put(iter->tr); if (info->spare) ring_buffer_free_read_page(iter->trace_buffer->buffer, info->spare); @@ -5066,14 +5255,14 @@ static ssize_t tracing_stats_read(struct file *filp, char __user *ubuf, size_t count, loff_t *ppos) { - struct trace_cpu *tc = filp->private_data; - struct trace_array *tr = tc->tr; + struct inode *inode = file_inode(filp); + struct trace_array *tr = inode->i_private; struct trace_buffer *trace_buf = &tr->trace_buffer; + int cpu = tracing_get_cpu(inode); struct trace_seq *s; unsigned long cnt; unsigned long long t; unsigned long usec_rem; - int cpu = tc->cpu; s = kmalloc(sizeof(*s), GFP_KERNEL); if (!s) @@ -5126,9 +5315,10 @@ tracing_stats_read(struct file *filp, char __user *ubuf, } static const struct file_operations tracing_stats_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = tracing_stats_read, .llseek = generic_file_llseek, + .release = tracing_release_generic_tr, }; #ifdef CONFIG_DYNAMIC_FTRACE @@ -5317,10 +5507,20 @@ static struct dentry *tracing_dentry_percpu(struct trace_array *tr, int cpu) return tr->percpu_dir; } +static struct dentry * +trace_create_cpu_file(const char *name, umode_t mode, struct dentry *parent, + void *data, long cpu, const struct file_operations *fops) +{ + struct dentry *ret = trace_create_file(name, mode, parent, data, fops); + + if (ret) /* See tracing_get_cpu() */ + ret->d_inode->i_cdev = (void *)(cpu + 1); + return ret; +} + static void tracing_init_debugfs_percpu(struct trace_array *tr, long cpu) { - struct trace_array_cpu *data = per_cpu_ptr(tr->trace_buffer.data, cpu); struct dentry *d_percpu = tracing_dentry_percpu(tr, cpu); struct dentry *d_cpu; char cpu_dir[30]; /* 30 characters should be more than enough */ @@ -5336,28 +5536,28 @@ tracing_init_debugfs_percpu(struct trace_array *tr, long cpu) } /* per cpu trace_pipe */ - trace_create_file("trace_pipe", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_pipe_fops); + trace_create_cpu_file("trace_pipe", 0444, d_cpu, + tr, cpu, &tracing_pipe_fops); /* per cpu trace */ - trace_create_file("trace", 0644, d_cpu, - (void *)&data->trace_cpu, &tracing_fops); + trace_create_cpu_file("trace", 0644, d_cpu, + tr, cpu, &tracing_fops); - trace_create_file("trace_pipe_raw", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_buffers_fops); + trace_create_cpu_file("trace_pipe_raw", 0444, d_cpu, + tr, cpu, &tracing_buffers_fops); - trace_create_file("stats", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_stats_fops); + trace_create_cpu_file("stats", 0444, d_cpu, + tr, cpu, &tracing_stats_fops); - trace_create_file("buffer_size_kb", 0444, d_cpu, - (void *)&data->trace_cpu, &tracing_entries_fops); + trace_create_cpu_file("buffer_size_kb", 0444, d_cpu, + tr, cpu, &tracing_entries_fops); #ifdef CONFIG_TRACER_SNAPSHOT - trace_create_file("snapshot", 0644, d_cpu, - (void *)&data->trace_cpu, &snapshot_fops); + trace_create_cpu_file("snapshot", 0644, d_cpu, + tr, cpu, &snapshot_fops); - trace_create_file("snapshot_raw", 0444, d_cpu, - (void *)&data->trace_cpu, &snapshot_raw_fops); + trace_create_cpu_file("snapshot_raw", 0444, d_cpu, + tr, cpu, &snapshot_raw_fops); #endif } @@ -5612,15 +5812,10 @@ rb_simple_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { struct trace_array *tr = filp->private_data; - struct ring_buffer *buffer = tr->trace_buffer.buffer; char buf[64]; int r; - if (buffer) - r = ring_buffer_record_is_on(buffer); - else - r = 0; - + r = tracer_tracing_is_on(tr); r = sprintf(buf, "%d\n", r); return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); @@ -5642,11 +5837,11 @@ rb_simple_write(struct file *filp, const char __user *ubuf, if (buffer) { mutex_lock(&trace_types_lock); if (val) { - ring_buffer_record_on(buffer); + tracer_tracing_on(tr); if (tr->current_trace->start) tr->current_trace->start(tr); } else { - ring_buffer_record_off(buffer); + tracer_tracing_off(tr); if (tr->current_trace->stop) tr->current_trace->stop(tr); } @@ -5659,9 +5854,10 @@ rb_simple_write(struct file *filp, const char __user *ubuf, } static const struct file_operations rb_simple_fops = { - .open = tracing_open_generic, + .open = tracing_open_generic_tr, .read = rb_simple_read, .write = rb_simple_write, + .release = tracing_release_generic_tr, .llseek = default_llseek, }; @@ -5688,6 +5884,8 @@ allocate_trace_buffer(struct trace_array *tr, struct trace_buffer *buf, int size rb_flags = trace_flags & TRACE_ITER_OVERWRITE ? RB_FL_OVERWRITE : 0; + buf->tr = tr; + buf->buffer = ring_buffer_alloc(size, rb_flags); if (!buf->buffer) return -ENOMEM; @@ -5775,8 +5973,10 @@ static int new_instance_create(const char *name) goto out_free_tr; ret = event_trace_add_tracer(tr->dir, tr); - if (ret) + if (ret) { + debugfs_remove_recursive(tr->dir); goto out_free_tr; + } init_tracer_debugfs(tr, tr->dir); @@ -5922,13 +6122,13 @@ init_tracer_debugfs(struct trace_array *tr, struct dentry *d_tracer) tr, &tracing_iter_fops); trace_create_file("trace", 0644, d_tracer, - (void *)&tr->trace_cpu, &tracing_fops); + tr, &tracing_fops); trace_create_file("trace_pipe", 0444, d_tracer, - (void *)&tr->trace_cpu, &tracing_pipe_fops); + tr, &tracing_pipe_fops); trace_create_file("buffer_size_kb", 0644, d_tracer, - (void *)&tr->trace_cpu, &tracing_entries_fops); + tr, &tracing_entries_fops); trace_create_file("buffer_total_size_kb", 0444, d_tracer, tr, &tracing_total_entries_fops); @@ -5943,11 +6143,11 @@ init_tracer_debugfs(struct trace_array *tr, struct dentry *d_tracer) &trace_clock_fops); trace_create_file("tracing_on", 0644, d_tracer, - tr, &rb_simple_fops); + tr, &rb_simple_fops); #ifdef CONFIG_TRACER_SNAPSHOT trace_create_file("snapshot", 0644, d_tracer, - (void *)&tr->trace_cpu, &snapshot_fops); + tr, &snapshot_fops); #endif for_each_tracing_cpu(cpu) diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 20572ed88c5c..51b44483eb78 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -224,6 +224,11 @@ enum { extern struct list_head ftrace_trace_arrays; +extern struct mutex trace_types_lock; + +extern int trace_array_get(struct trace_array *tr); +extern void trace_array_put(struct trace_array *tr); + /* * The global tracer (top) should be the first trace array added, * but we check the flag anyway. diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 84b1e045faba..8354dc81ae64 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -26,7 +26,7 @@ static int perf_trace_event_perm(struct ftrace_event_call *tp_event, { /* The ftrace function trace is allowed only for root. */ if (ftrace_event_is_function(tp_event) && - perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN)) + perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN)) return -EPERM; /* No tracing, just counting, so no obvious leak */ diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 27963e2bf4bf..001b349af939 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -27,12 +27,6 @@ DEFINE_MUTEX(event_mutex); -DEFINE_MUTEX(event_storage_mutex); -EXPORT_SYMBOL_GPL(event_storage_mutex); - -char event_storage[EVENT_STORAGE_SIZE]; -EXPORT_SYMBOL_GPL(event_storage); - LIST_HEAD(ftrace_events); static LIST_HEAD(ftrace_common_fields); @@ -41,6 +35,23 @@ static LIST_HEAD(ftrace_common_fields); static struct kmem_cache *field_cachep; static struct kmem_cache *file_cachep; +#define SYSTEM_FL_FREE_NAME (1 << 31) + +static inline int system_refcount(struct event_subsystem *system) +{ + return system->ref_count & ~SYSTEM_FL_FREE_NAME; +} + +static int system_refcount_inc(struct event_subsystem *system) +{ + return (system->ref_count++) & ~SYSTEM_FL_FREE_NAME; +} + +static int system_refcount_dec(struct event_subsystem *system) +{ + return (--system->ref_count) & ~SYSTEM_FL_FREE_NAME; +} + /* Double loops, do not use break, only goto's work */ #define do_for_each_event_file(tr, file) \ list_for_each_entry(tr, &ftrace_trace_arrays, list) { \ @@ -97,7 +108,7 @@ static int __trace_define_field(struct list_head *head, const char *type, field = kmem_cache_alloc(field_cachep, GFP_TRACE); if (!field) - goto err; + return -ENOMEM; field->name = name; field->type = type; @@ -114,11 +125,6 @@ static int __trace_define_field(struct list_head *head, const char *type, list_add(&field->link, head); return 0; - -err: - kmem_cache_free(field_cachep, field); - - return -ENOMEM; } int trace_define_field(struct ftrace_event_call *call, const char *type, @@ -349,8 +355,8 @@ static void __put_system(struct event_subsystem *system) { struct event_filter *filter = system->filter; - WARN_ON_ONCE(system->ref_count == 0); - if (--system->ref_count) + WARN_ON_ONCE(system_refcount(system) == 0); + if (system_refcount_dec(system)) return; list_del(&system->list); @@ -359,13 +365,15 @@ static void __put_system(struct event_subsystem *system) kfree(filter->filter_string); kfree(filter); } + if (system->ref_count & SYSTEM_FL_FREE_NAME) + kfree(system->name); kfree(system); } static void __get_system(struct event_subsystem *system) { - WARN_ON_ONCE(system->ref_count == 0); - system->ref_count++; + WARN_ON_ONCE(system_refcount(system) == 0); + system_refcount_inc(system); } static void __get_system_dir(struct ftrace_subsystem_dir *dir) @@ -379,7 +387,7 @@ static void __put_system_dir(struct ftrace_subsystem_dir *dir) { WARN_ON_ONCE(dir->ref_count == 0); /* If the subsystem is about to be freed, the dir must be too */ - WARN_ON_ONCE(dir->subsystem->ref_count == 1 && dir->ref_count != 1); + WARN_ON_ONCE(system_refcount(dir->subsystem) == 1 && dir->ref_count != 1); __put_system(dir->subsystem); if (!--dir->ref_count) @@ -393,17 +401,55 @@ static void put_system(struct ftrace_subsystem_dir *dir) mutex_unlock(&event_mutex); } +static void remove_subsystem(struct ftrace_subsystem_dir *dir) +{ + if (!dir) + return; + + if (!--dir->nr_events) { + debugfs_remove_recursive(dir->entry); + list_del(&dir->list); + __put_system_dir(dir); + } +} + +static void *event_file_data(struct file *filp) +{ + return ACCESS_ONCE(file_inode(filp)->i_private); +} + +static void remove_event_file_dir(struct ftrace_event_file *file) +{ + struct dentry *dir = file->dir; + struct dentry *child; + + if (dir) { + spin_lock(&dir->d_lock); /* probably unneeded */ + list_for_each_entry(child, &dir->d_subdirs, d_u.d_child) { + if (child->d_inode) /* probably unneeded */ + child->d_inode->i_private = NULL; + } + spin_unlock(&dir->d_lock); + + debugfs_remove_recursive(dir); + } + + list_del(&file->list); + remove_subsystem(file->system); + kmem_cache_free(file_cachep, file); +} + /* * __ftrace_set_clr_event(NULL, NULL, NULL, set) will set/unset all events. */ -static int __ftrace_set_clr_event(struct trace_array *tr, const char *match, - const char *sub, const char *event, int set) +static int +__ftrace_set_clr_event_nolock(struct trace_array *tr, const char *match, + const char *sub, const char *event, int set) { struct ftrace_event_file *file; struct ftrace_event_call *call; int ret = -EINVAL; - mutex_lock(&event_mutex); list_for_each_entry(file, &tr->events, list) { call = file->event_call; @@ -429,6 +475,17 @@ static int __ftrace_set_clr_event(struct trace_array *tr, const char *match, ret = 0; } + + return ret; +} + +static int __ftrace_set_clr_event(struct trace_array *tr, const char *match, + const char *sub, const char *event, int set) +{ + int ret; + + mutex_lock(&event_mutex); + ret = __ftrace_set_clr_event_nolock(tr, match, sub, event, set); mutex_unlock(&event_mutex); return ret; @@ -623,13 +680,23 @@ static ssize_t event_enable_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_file *file = filp->private_data; + struct ftrace_event_file *file; + unsigned long flags; char *buf; - if (file->flags & FTRACE_EVENT_FL_ENABLED) { - if (file->flags & FTRACE_EVENT_FL_SOFT_DISABLED) + mutex_lock(&event_mutex); + file = event_file_data(filp); + if (likely(file)) + flags = file->flags; + mutex_unlock(&event_mutex); + + if (!file) + return -ENODEV; + + if (flags & FTRACE_EVENT_FL_ENABLED) { + if (flags & FTRACE_EVENT_FL_SOFT_DISABLED) buf = "0*\n"; - else if (file->flags & FTRACE_EVENT_FL_SOFT_MODE) + else if (flags & FTRACE_EVENT_FL_SOFT_MODE) buf = "1*\n"; else buf = "1\n"; @@ -643,13 +710,10 @@ static ssize_t event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_file *file = filp->private_data; + struct ftrace_event_file *file; unsigned long val; int ret; - if (!file) - return -EINVAL; - ret = kstrtoul_from_user(ubuf, cnt, 10, &val); if (ret) return ret; @@ -661,8 +725,11 @@ event_enable_write(struct file *filp, const char __user *ubuf, size_t cnt, switch (val) { case 0: case 1: + ret = -ENODEV; mutex_lock(&event_mutex); - ret = ftrace_event_enable_disable(file, val); + file = event_file_data(filp); + if (likely(file)) + ret = ftrace_event_enable_disable(file, val); mutex_unlock(&event_mutex); break; @@ -769,7 +836,7 @@ enum { static void *f_next(struct seq_file *m, void *v, loff_t *pos) { - struct ftrace_event_call *call = m->private; + struct ftrace_event_call *call = event_file_data(m->private); struct ftrace_event_field *field; struct list_head *common_head = &ftrace_common_fields; struct list_head *head = trace_get_fields(call); @@ -813,6 +880,11 @@ static void *f_start(struct seq_file *m, loff_t *pos) loff_t l = 0; void *p; + /* ->stop() is called even if ->start() fails */ + mutex_lock(&event_mutex); + if (!event_file_data(m->private)) + return ERR_PTR(-ENODEV); + /* Start by showing the header */ if (!*pos) return (void *)FORMAT_HEADER; @@ -827,7 +899,7 @@ static void *f_start(struct seq_file *m, loff_t *pos) static int f_show(struct seq_file *m, void *v) { - struct ftrace_event_call *call = m->private; + struct ftrace_event_call *call = event_file_data(m->private); struct ftrace_event_field *field; const char *array_descriptor; @@ -878,6 +950,7 @@ static int f_show(struct seq_file *m, void *v) static void f_stop(struct seq_file *m, void *p) { + mutex_unlock(&event_mutex); } static const struct seq_operations trace_format_seq_ops = { @@ -889,7 +962,6 @@ static const struct seq_operations trace_format_seq_ops = { static int trace_format_open(struct inode *inode, struct file *file) { - struct ftrace_event_call *call = inode->i_private; struct seq_file *m; int ret; @@ -898,7 +970,7 @@ static int trace_format_open(struct inode *inode, struct file *file) return ret; m = file->private_data; - m->private = call; + m->private = file; return 0; } @@ -906,19 +978,22 @@ static int trace_format_open(struct inode *inode, struct file *file) static ssize_t event_id_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; + int id = (long)event_file_data(filp); struct trace_seq *s; int r; if (*ppos) return 0; + if (unlikely(!id)) + return -ENODEV; + s = kmalloc(sizeof(*s), GFP_KERNEL); if (!s) return -ENOMEM; trace_seq_init(s); - trace_seq_printf(s, "%d\n", call->event.type); + trace_seq_printf(s, "%d\n", id); r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); @@ -930,21 +1005,28 @@ static ssize_t event_filter_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call; struct trace_seq *s; - int r; + int r = -ENODEV; if (*ppos) return 0; s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) return -ENOMEM; trace_seq_init(s); - print_event_filter(call, s); - r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); + mutex_lock(&event_mutex); + call = event_file_data(filp); + if (call) + print_event_filter(call, s); + mutex_unlock(&event_mutex); + + if (call) + r = simple_read_from_buffer(ubuf, cnt, ppos, s->buffer, s->len); kfree(s); @@ -955,9 +1037,9 @@ static ssize_t event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - struct ftrace_event_call *call = filp->private_data; + struct ftrace_event_call *call; char *buf; - int err; + int err = -ENODEV; if (cnt >= PAGE_SIZE) return -EINVAL; @@ -972,7 +1054,12 @@ event_filter_write(struct file *filp, const char __user *ubuf, size_t cnt, } buf[cnt] = '\0'; - err = apply_event_filter(call, buf); + mutex_lock(&event_mutex); + call = event_file_data(filp); + if (call) + err = apply_event_filter(call, buf); + mutex_unlock(&event_mutex); + free_page((unsigned long) buf); if (err < 0) return err; @@ -992,6 +1079,7 @@ static int subsystem_open(struct inode *inode, struct file *filp) int ret; /* Make sure the system still exists */ + mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); list_for_each_entry(tr, &ftrace_trace_arrays, list) { list_for_each_entry(dir, &tr->systems, list) { @@ -1007,6 +1095,7 @@ static int subsystem_open(struct inode *inode, struct file *filp) } exit_loop: mutex_unlock(&event_mutex); + mutex_unlock(&trace_types_lock); if (!system) return -ENODEV; @@ -1014,9 +1103,17 @@ static int subsystem_open(struct inode *inode, struct file *filp) /* Some versions of gcc think dir can be uninitialized here */ WARN_ON(!dir); + /* Still need to increment the ref count of the system */ + if (trace_array_get(tr) < 0) { + put_system(dir); + return -ENODEV; + } + ret = tracing_open_generic(inode, filp); - if (ret < 0) + if (ret < 0) { + trace_array_put(tr); put_system(dir); + } return ret; } @@ -1027,16 +1124,23 @@ static int system_tr_open(struct inode *inode, struct file *filp) struct trace_array *tr = inode->i_private; int ret; + if (trace_array_get(tr) < 0) + return -ENODEV; + /* Make a temporary dir that has no system but points to tr */ dir = kzalloc(sizeof(*dir), GFP_KERNEL); - if (!dir) + if (!dir) { + trace_array_put(tr); return -ENOMEM; + } dir->tr = tr; ret = tracing_open_generic(inode, filp); - if (ret < 0) + if (ret < 0) { + trace_array_put(tr); kfree(dir); + } filp->private_data = dir; @@ -1047,6 +1151,8 @@ static int subsystem_release(struct inode *inode, struct file *file) { struct ftrace_subsystem_dir *dir = file->private_data; + trace_array_put(dir->tr); + /* * If dir->subsystem is NULL, then this is a temporary * descriptor that was made for a trace_array to enable @@ -1143,6 +1249,7 @@ show_header(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) static int ftrace_event_avail_open(struct inode *inode, struct file *file); static int ftrace_event_set_open(struct inode *inode, struct file *file); +static int ftrace_event_release(struct inode *inode, struct file *file); static const struct seq_operations show_event_seq_ops = { .start = t_start, @@ -1170,7 +1277,7 @@ static const struct file_operations ftrace_set_event_fops = { .read = seq_read, .write = ftrace_event_write, .llseek = seq_lseek, - .release = seq_release, + .release = ftrace_event_release, }; static const struct file_operations ftrace_enable_fops = { @@ -1188,7 +1295,6 @@ static const struct file_operations ftrace_event_format_fops = { }; static const struct file_operations ftrace_event_id_fops = { - .open = tracing_open_generic, .read = event_id_read, .llseek = default_llseek, }; @@ -1247,6 +1353,15 @@ ftrace_event_open(struct inode *inode, struct file *file, return ret; } +static int ftrace_event_release(struct inode *inode, struct file *file) +{ + struct trace_array *tr = inode->i_private; + + trace_array_put(tr); + + return seq_release(inode, file); +} + static int ftrace_event_avail_open(struct inode *inode, struct file *file) { @@ -1260,12 +1375,19 @@ ftrace_event_set_open(struct inode *inode, struct file *file) { const struct seq_operations *seq_ops = &show_set_event_seq_ops; struct trace_array *tr = inode->i_private; + int ret; + + if (trace_array_get(tr) < 0) + return -ENODEV; if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) ftrace_clear_events(tr); - return ftrace_event_open(inode, file, seq_ops); + ret = ftrace_event_open(inode, file, seq_ops); + if (ret < 0) + trace_array_put(tr); + return ret; } static struct event_subsystem * @@ -1279,7 +1401,15 @@ create_new_subsystem(const char *name) return NULL; system->ref_count = 1; - system->name = name; + + /* Only allocate if dynamic (kprobes and modules) */ + if (!core_kernel_data((unsigned long)name)) { + system->ref_count |= SYSTEM_FL_FREE_NAME; + system->name = kstrdup(name, GFP_KERNEL); + if (!system->name) + goto out_free; + } else + system->name = name; system->filter = NULL; @@ -1292,6 +1422,8 @@ create_new_subsystem(const char *name) return system; out_free: + if (system->ref_count & SYSTEM_FL_FREE_NAME) + kfree(system->name); kfree(system); return NULL; } @@ -1410,8 +1542,8 @@ event_create_dir(struct dentry *parent, #ifdef CONFIG_PERF_EVENTS if (call->event.type && call->class->reg) - trace_create_file("id", 0444, file->dir, call, - id); + trace_create_file("id", 0444, file->dir, + (void *)(long)call->event.type, id); #endif /* @@ -1436,33 +1568,16 @@ event_create_dir(struct dentry *parent, return 0; } -static void remove_subsystem(struct ftrace_subsystem_dir *dir) -{ - if (!dir) - return; - - if (!--dir->nr_events) { - debugfs_remove_recursive(dir->entry); - list_del(&dir->list); - __put_system_dir(dir); - } -} - static void remove_event_from_tracers(struct ftrace_event_call *call) { struct ftrace_event_file *file; struct trace_array *tr; do_for_each_event_file_safe(tr, file) { - if (file->event_call != call) continue; - list_del(&file->list); - debugfs_remove_recursive(file->dir); - remove_subsystem(file->system); - kmem_cache_free(file_cachep, file); - + remove_event_file_dir(file); /* * The do_for_each_event_file_safe() is * a double loop. After finding the call for this @@ -1591,6 +1706,7 @@ static void __add_event_to_tracers(struct ftrace_event_call *call, int trace_add_event_call(struct ftrace_event_call *call) { int ret; + mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); ret = __register_event(call, NULL); @@ -1598,11 +1714,13 @@ int trace_add_event_call(struct ftrace_event_call *call) __add_event_to_tracers(call, NULL); mutex_unlock(&event_mutex); + mutex_unlock(&trace_types_lock); return ret; } /* - * Must be called under locking both of event_mutex and trace_event_sem. + * Must be called under locking of trace_types_lock, event_mutex and + * trace_event_sem. */ static void __trace_remove_event_call(struct ftrace_event_call *call) { @@ -1611,14 +1729,47 @@ static void __trace_remove_event_call(struct ftrace_event_call *call) destroy_preds(call); } +static int probe_remove_event_call(struct ftrace_event_call *call) +{ + struct trace_array *tr; + struct ftrace_event_file *file; + +#ifdef CONFIG_PERF_EVENTS + if (call->perf_refcount) + return -EBUSY; +#endif + do_for_each_event_file(tr, file) { + if (file->event_call != call) + continue; + /* + * We can't rely on ftrace_event_enable_disable(enable => 0) + * we are going to do, FTRACE_EVENT_FL_SOFT_MODE can suppress + * TRACE_REG_UNREGISTER. + */ + if (file->flags & FTRACE_EVENT_FL_ENABLED) + return -EBUSY; + break; + } while_for_each_event_file(); + + __trace_remove_event_call(call); + + return 0; +} + /* Remove an event_call */ -void trace_remove_event_call(struct ftrace_event_call *call) +int trace_remove_event_call(struct ftrace_event_call *call) { + int ret; + + mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); down_write(&trace_event_sem); - __trace_remove_event_call(call); + ret = probe_remove_event_call(call); up_write(&trace_event_sem); mutex_unlock(&event_mutex); + mutex_unlock(&trace_types_lock); + + return ret; } #define for_each_event(event, start, end) \ @@ -1703,6 +1854,16 @@ static void trace_module_add_events(struct module *mod) struct ftrace_module_file_ops *file_ops = NULL; struct ftrace_event_call **call, **start, **end; + if (!mod->num_trace_events) + return; + + /* Don't add infrastructure for mods without tracepoints */ + if (trace_module_has_bad_taint(mod)) { + pr_err("%s: module has bad taint, not creating trace events\n", + mod->name); + return; + } + start = mod->trace_events; end = mod->trace_events + mod->num_trace_events; @@ -1762,6 +1923,7 @@ static int trace_module_notify(struct notifier_block *self, { struct module *mod = data; + mutex_lock(&trace_types_lock); mutex_lock(&event_mutex); switch (val) { case MODULE_STATE_COMING: @@ -1772,6 +1934,7 @@ static int trace_module_notify(struct notifier_block *self, break; } mutex_unlock(&event_mutex); + mutex_unlock(&trace_types_lock); return 0; } @@ -2188,12 +2351,8 @@ __trace_remove_event_dirs(struct trace_array *tr) { struct ftrace_event_file *file, *next; - list_for_each_entry_safe(file, next, &tr->events, list) { - list_del(&file->list); - debugfs_remove_recursive(file->dir); - remove_subsystem(file->system); - kmem_cache_free(file_cachep, file); - } + list_for_each_entry_safe(file, next, &tr->events, list) + remove_event_file_dir(file); } static void @@ -2329,11 +2488,11 @@ early_event_add_tracer(struct dentry *parent, struct trace_array *tr) int event_trace_del_tracer(struct trace_array *tr) { - /* Disable any running events */ - __ftrace_set_clr_event(tr, NULL, NULL, NULL, 0); - mutex_lock(&event_mutex); + /* Disable any running events */ + __ftrace_set_clr_event_nolock(tr, NULL, NULL, NULL, 0); + down_write(&trace_event_sem); __trace_remove_event_dirs(tr); debugfs_remove_recursive(tr->event_dir); diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index e1b653f7e1ca..0a1edc694d67 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -631,17 +631,15 @@ static void append_filter_err(struct filter_parse_state *ps, free_page((unsigned long) buf); } +/* caller must hold event_mutex */ void print_event_filter(struct ftrace_event_call *call, struct trace_seq *s) { - struct event_filter *filter; + struct event_filter *filter = call->filter; - mutex_lock(&event_mutex); - filter = call->filter; if (filter && filter->filter_string) trace_seq_printf(s, "%s\n", filter->filter_string); else trace_seq_printf(s, "none\n"); - mutex_unlock(&event_mutex); } void print_subsystem_event_filter(struct event_subsystem *system, @@ -1835,23 +1833,22 @@ static int create_system_filter(struct event_subsystem *system, return err; } +/* caller must hold event_mutex */ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) { struct event_filter *filter; - int err = 0; - - mutex_lock(&event_mutex); + int err; if (!strcmp(strstrip(filter_string), "0")) { filter_disable(call); filter = call->filter; if (!filter) - goto out_unlock; + return 0; RCU_INIT_POINTER(call->filter, NULL); /* Make sure the filter is not being used */ synchronize_sched(); __free_filter(filter); - goto out_unlock; + return 0; } err = create_filter(call, filter_string, true, &filter); @@ -1878,8 +1875,6 @@ int apply_event_filter(struct ftrace_event_call *call, char *filter_string) __free_filter(tmp); } } -out_unlock: - mutex_unlock(&event_mutex); return err; } diff --git a/kernel/trace/trace_export.c b/kernel/trace/trace_export.c index d21a74670088..d7d0b50b1b70 100644 --- a/kernel/trace/trace_export.c +++ b/kernel/trace/trace_export.c @@ -95,15 +95,12 @@ static void __always_unused ____ftrace_check_##name(void) \ #undef __array #define __array(type, item, len) \ do { \ + char *type_str = #type"["__stringify(len)"]"; \ BUILD_BUG_ON(len > MAX_FILTER_STR_VAL); \ - mutex_lock(&event_storage_mutex); \ - snprintf(event_storage, sizeof(event_storage), \ - "%s[%d]", #type, len); \ - ret = trace_define_field(event_call, event_storage, #item, \ + ret = trace_define_field(event_call, type_str, #item, \ offsetof(typeof(field), item), \ sizeof(field.item), \ is_signed_type(type), filter_type); \ - mutex_unlock(&event_storage_mutex); \ if (ret) \ return ret; \ } while (0); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index b19d065a28cb..2aefbee93a6d 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -373,7 +373,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip) struct trace_array_cpu *data; unsigned long flags; - if (likely(!tracer_enabled)) + if (!tracer_enabled || !tracing_is_enabled()) return; cpu = raw_smp_processor_id(); @@ -416,7 +416,7 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip) else return; - if (!tracer_enabled) + if (!tracer_enabled || !tracing_is_enabled()) return; data = per_cpu_ptr(tr->trace_buffer.data, cpu); diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 9f46e98ba8f2..64abc8ca928b 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -90,7 +90,7 @@ static __kprobes bool trace_probe_is_on_module(struct trace_probe *tp) } static int register_probe_event(struct trace_probe *tp); -static void unregister_probe_event(struct trace_probe *tp); +static int unregister_probe_event(struct trace_probe *tp); static DEFINE_MUTEX(probe_lock); static LIST_HEAD(probe_list); @@ -281,6 +281,8 @@ trace_probe_file_index(struct trace_probe *tp, struct ftrace_event_file *file) static int disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) { + struct ftrace_event_file **old = NULL; + int wait = 0; int ret = 0; mutex_lock(&probe_enable_lock); @@ -314,10 +316,7 @@ disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) } rcu_assign_pointer(tp->files, new); - - /* Make sure the probe is done with old files */ - synchronize_sched(); - kfree(old); + wait = 1; } else tp->flags &= ~TP_FLAG_PROFILE; @@ -326,11 +325,25 @@ disable_trace_probe(struct trace_probe *tp, struct ftrace_event_file *file) disable_kretprobe(&tp->rp); else disable_kprobe(&tp->rp.kp); + wait = 1; } out_unlock: mutex_unlock(&probe_enable_lock); + if (wait) { + /* + * Synchronize with kprobe_trace_func/kretprobe_trace_func + * to ensure disabled (all running handlers are finished). + * This is not only for kfree(), but also the caller, + * trace_remove_event_call() supposes it for releasing + * event_call related objects, which will be accessed in + * the kprobe_trace_func/kretprobe_trace_func. + */ + synchronize_sched(); + kfree(old); /* Ignored if link == NULL */ + } + return ret; } @@ -398,9 +411,12 @@ static int unregister_trace_probe(struct trace_probe *tp) if (trace_probe_is_enabled(tp)) return -EBUSY; + /* Will fail if probe is being used by ftrace or perf */ + if (unregister_probe_event(tp)) + return -EBUSY; + __unregister_trace_probe(tp); list_del(&tp->list); - unregister_probe_event(tp); return 0; } @@ -679,7 +695,9 @@ static int release_all_trace_probes(void) /* TODO: Use batch unregistration */ while (!list_empty(&probe_list)) { tp = list_entry(probe_list.next, struct trace_probe, list); - unregister_trace_probe(tp); + ret = unregister_trace_probe(tp); + if (ret) + goto end; free_trace_probe(tp); } @@ -1312,11 +1330,15 @@ static int register_probe_event(struct trace_probe *tp) return ret; } -static void unregister_probe_event(struct trace_probe *tp) +static int unregister_probe_event(struct trace_probe *tp) { + int ret; + /* tp->event is unregistered in trace_remove_event_call() */ - trace_remove_event_call(&tp->call); - kfree(tp->call.print_fmt); + ret = trace_remove_event_call(&tp->call); + if (!ret) + kfree(tp->call.print_fmt); + return ret; } /* Make a debugfs interface for controlling probe points */ diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index 8f2ac73c7a5f..322e16461072 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -306,6 +306,8 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) struct syscall_metadata *sys_data; struct ring_buffer_event *event; struct ring_buffer *buffer; + unsigned long irq_flags; + int pc; int syscall_nr; int size; @@ -321,9 +323,12 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) size = sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args; + local_save_flags(irq_flags); + pc = preempt_count(); + buffer = tr->trace_buffer.buffer; event = trace_buffer_lock_reserve(buffer, - sys_data->enter_event->event.type, size, 0, 0); + sys_data->enter_event->event.type, size, irq_flags, pc); if (!event) return; @@ -333,7 +338,8 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) if (!filter_current_check_discard(buffer, sys_data->enter_event, entry, event)) - trace_current_buffer_unlock_commit(buffer, event, 0, 0); + trace_current_buffer_unlock_commit(buffer, event, + irq_flags, pc); } static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) @@ -343,6 +349,8 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) struct syscall_metadata *sys_data; struct ring_buffer_event *event; struct ring_buffer *buffer; + unsigned long irq_flags; + int pc; int syscall_nr; syscall_nr = trace_get_syscall_nr(current, regs); @@ -355,9 +363,13 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) if (!sys_data) return; + local_save_flags(irq_flags); + pc = preempt_count(); + buffer = tr->trace_buffer.buffer; event = trace_buffer_lock_reserve(buffer, - sys_data->exit_event->event.type, sizeof(*entry), 0, 0); + sys_data->exit_event->event.type, sizeof(*entry), + irq_flags, pc); if (!event) return; @@ -367,7 +379,8 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) if (!filter_current_check_discard(buffer, sys_data->exit_event, entry, event)) - trace_current_buffer_unlock_commit(buffer, event, 0, 0); + trace_current_buffer_unlock_commit(buffer, event, + irq_flags, pc); } static int reg_event_syscall_enter(struct ftrace_event_file *file, diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 32494fb0ee64..6fd72b768522 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -70,7 +70,7 @@ struct trace_uprobe { (sizeof(struct probe_arg) * (n))) static int register_uprobe_event(struct trace_uprobe *tu); -static void unregister_uprobe_event(struct trace_uprobe *tu); +static int unregister_uprobe_event(struct trace_uprobe *tu); static DEFINE_MUTEX(uprobe_lock); static LIST_HEAD(uprobe_list); @@ -164,11 +164,17 @@ static struct trace_uprobe *find_probe_event(const char *event, const char *grou } /* Unregister a trace_uprobe and probe_event: call with locking uprobe_lock */ -static void unregister_trace_uprobe(struct trace_uprobe *tu) +static int unregister_trace_uprobe(struct trace_uprobe *tu) { + int ret; + + ret = unregister_uprobe_event(tu); + if (ret) + return ret; + list_del(&tu->list); - unregister_uprobe_event(tu); free_trace_uprobe(tu); + return 0; } /* Register a trace_uprobe and probe_event */ @@ -181,9 +187,12 @@ static int register_trace_uprobe(struct trace_uprobe *tu) /* register as an event */ old_tp = find_probe_event(tu->call.name, tu->call.class->system); - if (old_tp) + if (old_tp) { /* delete old event */ - unregister_trace_uprobe(old_tp); + ret = unregister_trace_uprobe(old_tp); + if (ret) + goto end; + } ret = register_uprobe_event(tu); if (ret) { @@ -256,6 +265,8 @@ static int create_trace_uprobe(int argc, char **argv) group = UPROBE_EVENT_SYSTEM; if (is_delete) { + int ret; + if (!event) { pr_info("Delete command needs an event name.\n"); return -EINVAL; @@ -269,9 +280,9 @@ static int create_trace_uprobe(int argc, char **argv) return -ENOENT; } /* delete an event */ - unregister_trace_uprobe(tu); + ret = unregister_trace_uprobe(tu); mutex_unlock(&uprobe_lock); - return 0; + return ret; } if (argc < 2) { @@ -283,8 +294,10 @@ static int create_trace_uprobe(int argc, char **argv) return -EINVAL; } arg = strchr(argv[1], ':'); - if (!arg) + if (!arg) { + ret = -EINVAL; goto fail_address_parse; + } *arg++ = '\0'; filename = argv[1]; @@ -406,16 +419,20 @@ fail_address_parse: return ret; } -static void cleanup_all_probes(void) +static int cleanup_all_probes(void) { struct trace_uprobe *tu; + int ret = 0; mutex_lock(&uprobe_lock); while (!list_empty(&uprobe_list)) { tu = list_entry(uprobe_list.next, struct trace_uprobe, list); - unregister_trace_uprobe(tu); + ret = unregister_trace_uprobe(tu); + if (ret) + break; } mutex_unlock(&uprobe_lock); + return ret; } /* Probes listing interfaces */ @@ -460,8 +477,13 @@ static const struct seq_operations probes_seq_op = { static int probes_open(struct inode *inode, struct file *file) { - if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) - cleanup_all_probes(); + int ret; + + if ((file->f_mode & FMODE_WRITE) && (file->f_flags & O_TRUNC)) { + ret = cleanup_all_probes(); + if (ret) + return ret; + } return seq_open(file, &probes_seq_op); } @@ -968,12 +990,17 @@ static int register_uprobe_event(struct trace_uprobe *tu) return ret; } -static void unregister_uprobe_event(struct trace_uprobe *tu) +static int unregister_uprobe_event(struct trace_uprobe *tu) { + int ret; + /* tu->event is unregistered in trace_remove_event_call() */ - trace_remove_event_call(&tu->call); + ret = trace_remove_event_call(&tu->call); + if (ret) + return ret; kfree(tu->call.print_fmt); tu->call.print_fmt = NULL; + return 0; } /* Make a trace interface for controling probe points */ diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 29f26540e9c9..63630aef3bd3 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -631,17 +631,25 @@ void tracepoint_iter_reset(struct tracepoint_iter *iter) EXPORT_SYMBOL_GPL(tracepoint_iter_reset); #ifdef CONFIG_MODULES +bool trace_module_has_bad_taint(struct module *mod) +{ + return mod->taints & ~((1 << TAINT_OOT_MODULE) | (1 << TAINT_CRAP)); +} + static int tracepoint_module_coming(struct module *mod) { struct tp_module *tp_mod, *iter; int ret = 0; + if (!mod->num_tracepoints) + return 0; + /* * We skip modules that taint the kernel, especially those with different * module headers (for forced load), to make sure we don't cause a crash. * Staging and out-of-tree GPL modules are fine. */ - if (mod->taints & ~((1 << TAINT_OOT_MODULE) | (1 << TAINT_CRAP))) + if (trace_module_has_bad_taint(mod)) return 0; mutex_lock(&tracepoints_mutex); tp_mod = kmalloc(sizeof(struct tp_module), GFP_KERNEL); @@ -679,6 +687,9 @@ static int tracepoint_module_going(struct module *mod) { struct tp_module *pos; + if (!mod->num_tracepoints) + return 0; + mutex_lock(&tracepoints_mutex); tracepoint_update_probe_range(mod->tracepoints_ptrs, mod->tracepoints_ptrs + mod->num_tracepoints); diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c index d8c30db06c5b..9bea1d7dd21f 100644 --- a/kernel/user_namespace.c +++ b/kernel/user_namespace.c @@ -62,6 +62,9 @@ int create_user_ns(struct cred *new) kgid_t group = new->egid; int ret; + if (parent_ns->level > 32) + return -EUSERS; + /* * Verify that we can not violate the policy of which files * may be accessed that is specified by the root directory, @@ -92,6 +95,7 @@ int create_user_ns(struct cred *new) atomic_set(&ns->count, 1); /* Leave the new->user_ns reference with the new user namespace. */ ns->parent = parent_ns; + ns->level = parent_ns->level + 1; ns->owner = owner; ns->group = group; @@ -105,16 +109,21 @@ int create_user_ns(struct cred *new) int unshare_userns(unsigned long unshare_flags, struct cred **new_cred) { struct cred *cred; + int err = -ENOMEM; if (!(unshare_flags & CLONE_NEWUSER)) return 0; cred = prepare_creds(); - if (!cred) - return -ENOMEM; + if (cred) { + err = create_user_ns(cred); + if (err) + put_cred(cred); + else + *new_cred = cred; + } - *new_cred = cred; - return create_user_ns(cred); + return err; } void free_user_ns(struct user_namespace *ns) @@ -139,7 +148,7 @@ static u32 map_id_range_down(struct uid_gid_map *map, u32 id, u32 count) /* Find the matching extent */ extents = map->nr_extents; - smp_read_barrier_depends(); + smp_rmb(); for (idx = 0; idx < extents; idx++) { first = map->extent[idx].first; last = first + map->extent[idx].count - 1; @@ -163,7 +172,7 @@ static u32 map_id_down(struct uid_gid_map *map, u32 id) /* Find the matching extent */ extents = map->nr_extents; - smp_read_barrier_depends(); + smp_rmb(); for (idx = 0; idx < extents; idx++) { first = map->extent[idx].first; last = first + map->extent[idx].count - 1; @@ -186,7 +195,7 @@ static u32 map_id_up(struct uid_gid_map *map, u32 id) /* Find the matching extent */ extents = map->nr_extents; - smp_read_barrier_depends(); + smp_rmb(); for (idx = 0; idx < extents; idx++) { first = map->extent[idx].lower_first; last = first + map->extent[idx].count - 1; @@ -602,9 +611,8 @@ static ssize_t map_write(struct file *file, const char __user *buf, * were written before the count of the extents. * * To achieve this smp_wmb() is used on guarantee the write - * order and smp_read_barrier_depends() is guaranteed that we - * don't have crazy architectures returning stale data. - * + * order and smp_rmb() is guaranteed that we don't have crazy + * architectures returning stale data. */ mutex_lock(&id_map_mutex); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index ee8e29a2320c..28c7fcde35af 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -272,6 +272,15 @@ static cpumask_var_t *wq_numa_possible_cpumask; static bool wq_disable_numa; module_param_named(disable_numa, wq_disable_numa, bool, 0444); +/* see the comment above the definition of WQ_POWER_EFFICIENT */ +#ifdef CONFIG_WQ_POWER_EFFICIENT_DEFAULT +static bool wq_power_efficient = true; +#else +static bool wq_power_efficient; +#endif + +module_param_named(power_efficient, wq_power_efficient, bool, 0444); + static bool wq_numa_enabled; /* unbound NUMA affinity enabled */ /* buf for wq_update_unbound_numa_attrs(), protected by CPU hotplug exclusion */ @@ -295,6 +304,9 @@ static DEFINE_HASHTABLE(unbound_pool_hash, UNBOUND_POOL_HASH_ORDER); /* I: attributes used when instantiating standard unbound pools on demand */ static struct workqueue_attrs *unbound_std_wq_attrs[NR_STD_WORKER_POOLS]; +/* I: attributes used when instantiating ordered pools on demand */ +static struct workqueue_attrs *ordered_wq_attrs[NR_STD_WORKER_POOLS]; + struct workqueue_struct *system_wq __read_mostly; EXPORT_SYMBOL(system_wq); struct workqueue_struct *system_highpri_wq __read_mostly; @@ -305,6 +317,10 @@ struct workqueue_struct *system_unbound_wq __read_mostly; EXPORT_SYMBOL_GPL(system_unbound_wq); struct workqueue_struct *system_freezable_wq __read_mostly; EXPORT_SYMBOL_GPL(system_freezable_wq); +struct workqueue_struct *system_power_efficient_wq __read_mostly; +EXPORT_SYMBOL_GPL(system_power_efficient_wq); +struct workqueue_struct *system_freezable_power_efficient_wq __read_mostly; +EXPORT_SYMBOL_GPL(system_freezable_power_efficient_wq); static int worker_thread(void *__worker); static void copy_workqueue_attrs(struct workqueue_attrs *to, @@ -1820,6 +1836,12 @@ static void destroy_worker(struct worker *worker) if (worker->flags & WORKER_IDLE) pool->nr_idle--; + /* + * Once WORKER_DIE is set, the kworker may destroy itself at any + * point. Pin to ensure the task stays until we're done with it. + */ + get_task_struct(worker->task); + list_del_init(&worker->entry); worker->flags |= WORKER_DIE; @@ -1828,6 +1850,7 @@ static void destroy_worker(struct worker *worker) spin_unlock_irq(&pool->lock); kthread_stop(worker->task); + put_task_struct(worker->task); kfree(worker); spin_lock_irq(&pool->lock); @@ -1871,6 +1894,12 @@ static void send_mayday(struct work_struct *work) /* mayday mayday mayday */ if (list_empty(&pwq->mayday_node)) { + /* + * If @pwq is for an unbound wq, its base ref may be put at + * any time due to an attribute change. Pin @pwq until the + * rescuer is done with it. + */ + get_pwq(pwq); list_add_tail(&pwq->mayday_node, &wq->maydays); wake_up_process(wq->rescuer->task); } @@ -2188,6 +2217,15 @@ __acquires(&pool->lock) dump_stack(); } + /* + * The following prevents a kworker from hogging CPU on !PREEMPT + * kernels, where a requeueing work item waiting for something to + * happen could deadlock with stop_machine as such work item could + * indefinitely requeue itself while all other CPUs are trapped in + * stop_machine. + */ + cond_resched(); + spin_lock_irq(&pool->lock); /* clear cpu intensive status */ @@ -2337,6 +2375,7 @@ static int rescuer_thread(void *__rescuer) struct worker *rescuer = __rescuer; struct workqueue_struct *wq = rescuer->rescue_wq; struct list_head *scheduled = &rescuer->scheduled; + bool should_stop; set_user_nice(current, RESCUER_NICE_LEVEL); @@ -2348,11 +2387,15 @@ static int rescuer_thread(void *__rescuer) repeat: set_current_state(TASK_INTERRUPTIBLE); - if (kthread_should_stop()) { - __set_current_state(TASK_RUNNING); - rescuer->task->flags &= ~PF_WQ_WORKER; - return 0; - } + /* + * By the time the rescuer is requested to stop, the workqueue + * shouldn't have any work pending, but @wq->maydays may still have + * pwq(s) queued. This can happen by non-rescuer workers consuming + * all the work items before the rescuer got to them. Go through + * @wq->maydays processing before acting on should_stop so that the + * list is always empty on exit. + */ + should_stop = kthread_should_stop(); /* see whether any pwq is asking for help */ spin_lock_irq(&wq_mayday_lock); @@ -2384,6 +2427,12 @@ repeat: process_scheduled_works(rescuer); /* + * Put the reference grabbed by send_mayday(). @pool won't + * go away while we're holding its lock. + */ + put_pwq(pwq); + + /* * Leave this pool. If keep_working() is %true, notify a * regular worker; otherwise, we end up with 0 concurrency * and stalling the execution. @@ -2398,6 +2447,12 @@ repeat: spin_unlock_irq(&wq_mayday_lock); + if (should_stop) { + __set_current_state(TASK_RUNNING); + rescuer->task->flags &= ~PF_WQ_WORKER; + return 0; + } + /* rescuers should never participate in concurrency management */ WARN_ON_ONCE(!(rescuer->flags & WORKER_NOT_RUNNING)); schedule(); @@ -3398,6 +3453,12 @@ static void copy_workqueue_attrs(struct workqueue_attrs *to, { to->nice = from->nice; cpumask_copy(to->cpumask, from->cpumask); + /* + * Unlike hash and equality test, this function doesn't ignore + * ->no_numa as it is used for both pool and wq attrs. Instead, + * get_unbound_pool() explicitly clears ->no_numa after copying. + */ + to->no_numa = from->no_numa; } /* hash value of the content of @attr */ @@ -3565,6 +3626,12 @@ static struct worker_pool *get_unbound_pool(const struct workqueue_attrs *attrs) lockdep_set_subclass(&pool->lock, 1); /* see put_pwq() */ copy_workqueue_attrs(pool->attrs, attrs); + /* + * no_numa isn't a worker_pool attribute, always clear it. See + * 'struct workqueue_attrs' comments for detail. + */ + pool->attrs->no_numa = false; + /* if cpumask is contained inside a NUMA node, we belong to that node */ if (wq_numa_enabled) { for_each_node(node) { @@ -4012,7 +4079,8 @@ static void wq_update_unbound_numa(struct workqueue_struct *wq, int cpu, if (!pwq) { pr_warning("workqueue: allocation failed while updating NUMA affinity of \"%s\"\n", wq->name); - goto out_unlock; + mutex_lock(&wq->mutex); + goto use_dfl_pwq; } /* @@ -4038,7 +4106,7 @@ out_unlock: static int alloc_and_link_pwqs(struct workqueue_struct *wq) { bool highpri = wq->flags & WQ_HIGHPRI; - int cpu; + int cpu, ret; if (!(wq->flags & WQ_UNBOUND)) { wq->cpu_pwqs = alloc_percpu(struct pool_workqueue); @@ -4058,6 +4126,13 @@ static int alloc_and_link_pwqs(struct workqueue_struct *wq) mutex_unlock(&wq->mutex); } return 0; + } else if (wq->flags & __WQ_ORDERED) { + ret = apply_workqueue_attrs(wq, ordered_wq_attrs[highpri]); + /* there should only be single pwq for ordering guarantee */ + WARN(!ret && (wq->pwqs.next != &wq->dfl_pwq->pwqs_node || + wq->pwqs.prev != &wq->dfl_pwq->pwqs_node), + "ordering guarantee broken for workqueue %s\n", wq->name); + return ret; } else { return apply_workqueue_attrs(wq, unbound_std_wq_attrs[highpri]); } @@ -4086,6 +4161,10 @@ struct workqueue_struct *__alloc_workqueue_key(const char *fmt, struct workqueue_struct *wq; struct pool_workqueue *pwq; + /* see the comment above the definition of WQ_POWER_EFFICIENT */ + if ((flags & WQ_POWER_EFFICIENT) && wq_power_efficient) + flags |= WQ_UNBOUND; + /* allocate wq and format name */ if (flags & WQ_UNBOUND) tbl_size = wq_numa_tbl_len * sizeof(wq->numa_pwq_tbl[0]); @@ -4969,13 +5048,23 @@ static int __init init_workqueues(void) } } - /* create default unbound wq attrs */ + /* create default unbound and ordered wq attrs */ for (i = 0; i < NR_STD_WORKER_POOLS; i++) { struct workqueue_attrs *attrs; BUG_ON(!(attrs = alloc_workqueue_attrs(GFP_KERNEL))); attrs->nice = std_nice[i]; unbound_std_wq_attrs[i] = attrs; + + /* + * An ordered wq should have only one pwq as ordering is + * guaranteed by max_active which is enforced by pwqs. + * Turn off NUMA so that dfl_pwq is used for all nodes. + */ + BUG_ON(!(attrs = alloc_workqueue_attrs(GFP_KERNEL))); + attrs->nice = std_nice[i]; + attrs->no_numa = true; + ordered_wq_attrs[i] = attrs; } system_wq = alloc_workqueue("events", 0, 0); @@ -4985,8 +5074,15 @@ static int __init init_workqueues(void) WQ_UNBOUND_MAX_ACTIVE); system_freezable_wq = alloc_workqueue("events_freezable", WQ_FREEZABLE, 0); + system_power_efficient_wq = alloc_workqueue("events_power_efficient", + WQ_POWER_EFFICIENT, 0); + system_freezable_power_efficient_wq = alloc_workqueue("events_freezable_power_efficient", + WQ_FREEZABLE | WQ_POWER_EFFICIENT, + 0); BUG_ON(!system_wq || !system_highpri_wq || !system_long_wq || - !system_unbound_wq || !system_freezable_wq); + !system_unbound_wq || !system_freezable_wq || + !system_power_efficient_wq || + !system_freezable_power_efficient_wq); return 0; } early_initcall(init_workqueues); |