Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

Mostly simple overlapping changes.

For example, David Ahern's adjacency list revamp in 'net-next'
conflicted with an adjacency list traversal bug fix in 'net'.

Signed-off-by: David S. Miller <davem@davemloft.net>

kernel/cgroup.c

@@ -64,6 +64,9 @@
 #include <linux/file.h>
 #include <net/sock.h>
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/cgroup.h>
+
 /*
  * pidlists linger the following amount before being destroyed.  The goal
  * is avoiding frequent destruction in the middle of consecutive read calls
@@ -1176,6 +1179,8 @@ static void cgroup_destroy_root(struct cgroup_root *root)
 	struct cgroup *cgrp = &root->cgrp;
 	struct cgrp_cset_link *link, *tmp_link;
 
+	trace_cgroup_destroy_root(root);
+
 	cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp);
 
 	BUG_ON(atomic_read(&root->nr_cgrps));
@@ -1874,6 +1879,9 @@ static int cgroup_remount(struct kernfs_root *kf_root, int *flags, char *data)
 		strcpy(root->release_agent_path, opts.release_agent);
 		spin_unlock(&release_agent_path_lock);
 	}
+
+	trace_cgroup_remount(root);
+
  out_unlock:
 	kfree(opts.release_agent);
 	kfree(opts.name);
@@ -2031,6 +2039,8 @@ static int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask)
 	if (ret)
 		goto destroy_root;
 
+	trace_cgroup_setup_root(root);
+
 	/*
 	 * There must be no failure case after here, since rebinding takes
 	 * care of subsystems' refcounts, which are explicitly dropped in
@@ -2315,22 +2325,18 @@ static struct file_system_type cgroup2_fs_type = {
 	.fs_flags = FS_USERNS_MOUNT,
 };
 
-static char *cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen,
-				   struct cgroup_namespace *ns)
+static int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen,
+				 struct cgroup_namespace *ns)
 {
 	struct cgroup *root = cset_cgroup_from_root(ns->root_cset, cgrp->root);
-	int ret;
 
-	ret = kernfs_path_from_node(cgrp->kn, root->kn, buf, buflen);
-	if (ret < 0 || ret >= buflen)
-		return NULL;
-	return buf;
+	return kernfs_path_from_node(cgrp->kn, root->kn, buf, buflen);
 }
 
-char *cgroup_path_ns(struct cgroup *cgrp, char *buf, size_t buflen,
-		     struct cgroup_namespace *ns)
+int cgroup_path_ns(struct cgroup *cgrp, char *buf, size_t buflen,
+		   struct cgroup_namespace *ns)
 {
-	char *ret;
+	int ret;
 
 	mutex_lock(&cgroup_mutex);
 	spin_lock_irq(&css_set_lock);
@@ -2357,12 +2363,12 @@ EXPORT_SYMBOL_GPL(cgroup_path_ns);
  *
  * Return value is the same as kernfs_path().
  */
-char *task_cgroup_path(struct task_struct *task, char *buf, size_t buflen)
+int task_cgroup_path(struct task_struct *task, char *buf, size_t buflen)
 {
 	struct cgroup_root *root;
 	struct cgroup *cgrp;
 	int hierarchy_id = 1;
-	char *path = NULL;
+	int ret;
 
 	mutex_lock(&cgroup_mutex);
 	spin_lock_irq(&css_set_lock);
@@ -2371,16 +2377,15 @@ char *task_cgroup_path(struct task_struct *task, char *buf, size_t buflen)
 
 	if (root) {
 		cgrp = task_cgroup_from_root(task, root);
-		path = cgroup_path_ns_locked(cgrp, buf, buflen, &init_cgroup_ns);
+		ret = cgroup_path_ns_locked(cgrp, buf, buflen, &init_cgroup_ns);
 	} else {
 		/* if no hierarchy exists, everyone is in "/" */
-		if (strlcpy(buf, "/", buflen) < buflen)
-			path = buf;
+		ret = strlcpy(buf, "/", buflen);
 	}
 
 	spin_unlock_irq(&css_set_lock);
 	mutex_unlock(&cgroup_mutex);
-	return path;
+	return ret;
 }
 EXPORT_SYMBOL_GPL(task_cgroup_path);
 
@@ -2830,6 +2835,10 @@ static int cgroup_attach_task(struct cgroup *dst_cgrp,
 		ret = cgroup_migrate(leader, threadgroup, dst_cgrp->root);
 
 	cgroup_migrate_finish(&preloaded_csets);
+
+	if (!ret)
+		trace_cgroup_attach_task(dst_cgrp, leader, threadgroup);
+
 	return ret;
 }
 
@@ -3611,6 +3620,8 @@ static int cgroup_rename(struct kernfs_node *kn, struct kernfs_node *new_parent,
 	mutex_lock(&cgroup_mutex);
 
 	ret = kernfs_rename(kn, new_parent, new_name_str);
+	if (!ret)
+		trace_cgroup_rename(cgrp);
 
 	mutex_unlock(&cgroup_mutex);
 
@@ -4381,6 +4392,8 @@ int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from)
 
 		if (task) {
 			ret = cgroup_migrate(task, false, to->root);
+			if (!ret)
+				trace_cgroup_transfer_tasks(to, task, false);
 			put_task_struct(task);
 		}
 	} while (task && !ret);
@@ -5046,6 +5059,8 @@ static void css_release_work_fn(struct work_struct *work)
 			ss->css_released(css);
 	} else {
 		/* cgroup release path */
+		trace_cgroup_release(cgrp);
+
 		cgroup_idr_remove(&cgrp->root->cgroup_idr, cgrp->id);
 		cgrp->id = -1;
 
@@ -5332,6 +5347,8 @@ static int cgroup_mkdir(struct kernfs_node *parent_kn, const char *name,
 	if (ret)
 		goto out_destroy;
 
+	trace_cgroup_mkdir(cgrp);
+
 	/* let's create and online css's */
 	kernfs_activate(kn);
 
@@ -5507,6 +5524,9 @@ static int cgroup_rmdir(struct kernfs_node *kn)
 
 	ret = cgroup_destroy_locked(cgrp);
 
+	if (!ret)
+		trace_cgroup_rmdir(cgrp);
+
 	cgroup_kn_unlock(kn);
 	return ret;
 }
@@ -5743,7 +5763,7 @@ core_initcall(cgroup_wq_init);
 int proc_cgroup_show(struct seq_file *m, struct pid_namespace *ns,
 		     struct pid *pid, struct task_struct *tsk)
 {
-	char *buf, *path;
+	char *buf;
 	int retval;
 	struct cgroup_root *root;
 
@@ -5786,17 +5806,17 @@ int proc_cgroup_show(struct seq_file *m, struct pid_namespace *ns,
 		 * " (deleted)" is appended to the cgroup path.
 		 */
 		if (cgroup_on_dfl(cgrp) || !(tsk->flags & PF_EXITING)) {
-			path = cgroup_path_ns_locked(cgrp, buf, PATH_MAX,
+			retval = cgroup_path_ns_locked(cgrp, buf, PATH_MAX,
 						current->nsproxy->cgroup_ns);
-			if (!path) {
+			if (retval >= PATH_MAX)
 				retval = -ENAMETOOLONG;
+			if (retval < 0)
 				goto out_unlock;
-			}
-		} else {
-			path = "/";
-		}
 
-		seq_puts(m, path);
+			seq_puts(m, buf);
+		} else {
+			seq_puts(m, "/");
+		}
 
 		if (cgroup_on_dfl(cgrp) && cgroup_is_dead(cgrp))
 			seq_puts(m, " (deleted)\n");
@@ -6062,8 +6082,9 @@ static void cgroup_release_agent(struct work_struct *work)
 {
 	struct cgroup *cgrp =
 		container_of(work, struct cgroup, release_agent_work);
-	char *pathbuf = NULL, *agentbuf = NULL, *path;
+	char *pathbuf = NULL, *agentbuf = NULL;
 	char *argv[3], *envp[3];
+	int ret;
 
 	mutex_lock(&cgroup_mutex);
 
@@ -6073,13 +6094,13 @@ static void cgroup_release_agent(struct work_struct *work)
 		goto out;
 
 	spin_lock_irq(&css_set_lock);
-	path = cgroup_path_ns_locked(cgrp, pathbuf, PATH_MAX, &init_cgroup_ns);
+	ret = cgroup_path_ns_locked(cgrp, pathbuf, PATH_MAX, &init_cgroup_ns);
 	spin_unlock_irq(&css_set_lock);
-	if (!path)
+	if (ret < 0 || ret >= PATH_MAX)
 		goto out;
 
 	argv[0] = agentbuf;
-	argv[1] = path;
+	argv[1] = pathbuf;
 	argv[2] = NULL;
 
 	/* minimal command environment */

kernel/cpu.c

@@ -228,7 +228,7 @@ static struct {
 	.wq = __WAIT_QUEUE_HEAD_INITIALIZER(cpu_hotplug.wq),
 	.lock = __MUTEX_INITIALIZER(cpu_hotplug.lock),
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
-	.dep_map = {.name = "cpu_hotplug.lock" },
+	.dep_map = STATIC_LOCKDEP_MAP_INIT("cpu_hotplug.dep_map", &cpu_hotplug.dep_map),
 #endif
 };
 

kernel/cpuset.c

@@ -2715,7 +2715,7 @@ void __cpuset_memory_pressure_bump(void)
 int proc_cpuset_show(struct seq_file *m, struct pid_namespace *ns,
 		     struct pid *pid, struct task_struct *tsk)
 {
-	char *buf, *p;
+	char *buf;
 	struct cgroup_subsys_state *css;
 	int retval;
 
@@ -2724,14 +2724,15 @@ int proc_cpuset_show(struct seq_file *m, struct pid_namespace *ns,
 	if (!buf)
 		goto out;
 
-	retval = -ENAMETOOLONG;
 	css = task_get_css(tsk, cpuset_cgrp_id);
-	p = cgroup_path_ns(css->cgroup, buf, PATH_MAX,
-			   current->nsproxy->cgroup_ns);
+	retval = cgroup_path_ns(css->cgroup, buf, PATH_MAX,
+				current->nsproxy->cgroup_ns);
 	css_put(css);
-	if (!p)
+	if (retval >= PATH_MAX)
+		retval = -ENAMETOOLONG;
+	if (retval < 0)
 		goto out_free;
-	seq_puts(m, p);
+	seq_puts(m, buf);
 	seq_putc(m, '\n');
 	retval = 0;
 out_free:

kernel/events/core.c

@@ -1960,6 +1960,12 @@ void perf_event_disable(struct perf_event *event)
 }
 EXPORT_SYMBOL_GPL(perf_event_disable);
 
+void perf_event_disable_inatomic(struct perf_event *event)
+{
+	event->pending_disable = 1;
+	irq_work_queue(&event->pending);
+}
+
 static void perf_set_shadow_time(struct perf_event *event,
 				 struct perf_event_context *ctx,
 				 u64 tstamp)
@@ -7075,8 +7081,8 @@ static int __perf_event_overflow(struct perf_event *event,
 	if (events && atomic_dec_and_test(&event->event_limit)) {
 		ret = 1;
 		event->pending_kill = POLL_HUP;
-		event->pending_disable = 1;
-		irq_work_queue(&event->pending);
+
+		perf_event_disable_inatomic(event);
 	}
 
 	READ_ONCE(event->overflow_handler)(event, data, regs);
@@ -8855,7 +8861,10 @@ EXPORT_SYMBOL_GPL(perf_pmu_register);
 
 void perf_pmu_unregister(struct pmu *pmu)
 {
+	int remove_device;
+
 	mutex_lock(&pmus_lock);
+	remove_device = pmu_bus_running;
 	list_del_rcu(&pmu->entry);
 	mutex_unlock(&pmus_lock);
 
@@ -8869,10 +8878,12 @@ void perf_pmu_unregister(struct pmu *pmu)
 	free_percpu(pmu->pmu_disable_count);
 	if (pmu->type >= PERF_TYPE_MAX)
 		idr_remove(&pmu_idr, pmu->type);
-	if (pmu->nr_addr_filters)
-		device_remove_file(pmu->dev, &dev_attr_nr_addr_filters);
-	device_del(pmu->dev);
-	put_device(pmu->dev);
+	if (remove_device) {
+		if (pmu->nr_addr_filters)
+			device_remove_file(pmu->dev, &dev_attr_nr_addr_filters);
+		device_del(pmu->dev);
+		put_device(pmu->dev);
+	}
 	free_pmu_context(pmu);
 }
 EXPORT_SYMBOL_GPL(perf_pmu_unregister);

kernel/events/uprobes.c

@@ -300,7 +300,8 @@ int uprobe_write_opcode(struct mm_struct *mm, unsigned long vaddr,
 
 retry:
 	/* Read the page with vaddr into memory */
-	ret = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &old_page, &vma);
+	ret = get_user_pages_remote(NULL, mm, vaddr, 1, FOLL_FORCE, &old_page,
+			&vma);
 	if (ret <= 0)
 		return ret;
 
@@ -1710,7 +1711,8 @@ static int is_trap_at_addr(struct mm_struct *mm, unsigned long vaddr)
 	 * but we treat this as a 'remote' access since it is
 	 * essentially a kernel access to the memory.
 	 */
-	result = get_user_pages_remote(NULL, mm, vaddr, 1, 0, 1, &page, NULL);
+	result = get_user_pages_remote(NULL, mm, vaddr, 1, FOLL_FORCE, &page,
+			NULL);
 	if (result < 0)
 		return result;
 

kernel/fork.c

@@ -547,7 +547,8 @@ free_tsk:
 }
 
 #ifdef CONFIG_MMU
-static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
+static __latent_entropy int dup_mmap(struct mm_struct *mm,
+					struct mm_struct *oldmm)
 {
 	struct vm_area_struct *mpnt, *tmp, *prev, **pprev;
 	struct rb_node **rb_link, *rb_parent;
@@ -1441,7 +1442,8 @@ init_task_pid(struct task_struct *task, enum pid_type type, struct pid *pid)
  * parts of the process environment (as per the clone
  * flags). The actual kick-off is left to the caller.
  */
-static struct task_struct *copy_process(unsigned long clone_flags,
+static __latent_entropy struct task_struct *copy_process(
+					unsigned long clone_flags,
 					unsigned long stack_start,
 					unsigned long stack_size,
 					int __user *child_tidptr,
@@ -1926,6 +1928,7 @@ long _do_fork(unsigned long clone_flags,
 
 	p = copy_process(clone_flags, stack_start, stack_size,
 			 child_tidptr, NULL, trace, tls, NUMA_NO_NODE);
+	add_latent_entropy();
 	/*
 	 * Do this prior waking up the new thread - the thread pointer
 	 * might get invalid after that point, if the thread exits quickly.

kernel/irq/manage.c

@@ -721,6 +721,7 @@ int irq_set_parent(int irq, int parent_irq)
 	irq_put_desc_unlock(desc, flags);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(irq_set_parent);
 #endif
 
 /*

kernel/kcov.c

@@ -53,8 +53,15 @@ void notrace __sanitizer_cov_trace_pc(void)
 	/*
 	 * We are interested in code coverage as a function of a syscall inputs,
 	 * so we ignore code executed in interrupts.
+	 * The checks for whether we are in an interrupt are open-coded, because
+	 * 1. We can't use in_interrupt() here, since it also returns true
+	 *    when we are inside local_bh_disable() section.
+	 * 2. We don't want to use (in_irq() | in_serving_softirq() | in_nmi()),
+	 *    since that leads to slower generated code (three separate tests,
+	 *    one for each of the flags).
 	 */
-	if (!t || in_interrupt())
+	if (!t || (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_OFFSET
+							| NMI_MASK)))
 		return;
 	mode = READ_ONCE(t->kcov_mode);
 	if (mode == KCOV_MODE_TRACE) {

kernel/power/suspend.c

@@ -498,9 +498,9 @@ static int enter_state(suspend_state_t state)
 
 #ifndef CONFIG_SUSPEND_SKIP_SYNC
 	trace_suspend_resume(TPS("sync_filesystems"), 0, true);
-	printk(KERN_INFO "PM: Syncing filesystems ... ");
+	pr_info("PM: Syncing filesystems ... ");
 	sys_sync();
-	printk("done.\n");
+	pr_cont("done.\n");
 	trace_suspend_resume(TPS("sync_filesystems"), 0, false);
 #endif
 

kernel/printk/printk.c

@@ -1769,6 +1769,10 @@ static size_t log_output(int facility, int level, enum log_flags lflags, const c
 		cont_flush();
 	}
 
+	/* Skip empty continuation lines that couldn't be added - they just flush */
+	if (!text_len && (lflags & LOG_CONT))
+		return 0;
+
 	/* If it doesn't end in a newline, try to buffer the current line */
 	if (!(lflags & LOG_NEWLINE)) {
 		if (cont_add(facility, level, lflags, text, text_len))

kernel/ptrace.c

@@ -537,7 +537,7 @@ int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst
 		int this_len, retval;
 
 		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
-		retval = access_process_vm(tsk, src, buf, this_len, 0);
+		retval = access_process_vm(tsk, src, buf, this_len, FOLL_FORCE);
 		if (!retval) {
 			if (copied)
 				break;
@@ -564,7 +564,8 @@ int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long ds
 		this_len = (len > sizeof(buf)) ? sizeof(buf) : len;
 		if (copy_from_user(buf, src, this_len))
 			return -EFAULT;
-		retval = access_process_vm(tsk, dst, buf, this_len, 1);
+		retval = access_process_vm(tsk, dst, buf, this_len,
+				FOLL_FORCE | FOLL_WRITE);
 		if (!retval) {
 			if (copied)
 				break;
@@ -1127,7 +1128,7 @@ int generic_ptrace_peekdata(struct task_struct *tsk, unsigned long addr,
 	unsigned long tmp;
 	int copied;
 
-	copied = access_process_vm(tsk, addr, &tmp, sizeof(tmp), 0);
+	copied = access_process_vm(tsk, addr, &tmp, sizeof(tmp), FOLL_FORCE);
 	if (copied != sizeof(tmp))
 		return -EIO;
 	return put_user(tmp, (unsigned long __user *)data);
@@ -1138,7 +1139,8 @@ int generic_ptrace_pokedata(struct task_struct *tsk, unsigned long addr,
 {
 	int copied;
 
-	copied = access_process_vm(tsk, addr, &data, sizeof(data), 1);
+	copied = access_process_vm(tsk, addr, &data, sizeof(data),
+			FOLL_FORCE | FOLL_WRITE);
 	return (copied == sizeof(data)) ? 0 : -EIO;
 }
 
@@ -1155,7 +1157,8 @@ int compat_ptrace_request(struct task_struct *child, compat_long_t request,
 	switch (request) {
 	case PTRACE_PEEKTEXT:
 	case PTRACE_PEEKDATA:
-		ret = access_process_vm(child, addr, &word, sizeof(word), 0);
+		ret = access_process_vm(child, addr, &word, sizeof(word),
+				FOLL_FORCE);
 		if (ret != sizeof(word))
 			ret = -EIO;
 		else
@@ -1164,7 +1167,8 @@ int compat_ptrace_request(struct task_struct *child, compat_long_t request,
 
 	case PTRACE_POKETEXT:
 	case PTRACE_POKEDATA:
-		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
+		ret = access_process_vm(child, addr, &data, sizeof(data),
+				FOLL_FORCE | FOLL_WRITE);
 		ret = (ret != sizeof(data) ? -EIO : 0);
 		break;
 

kernel/rcu/tiny.c

@@ -170,7 +170,7 @@ static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
 				      false));
 }
 
-static void rcu_process_callbacks(struct softirq_action *unused)
+static __latent_entropy void rcu_process_callbacks(struct softirq_action *unused)
 {
 	__rcu_process_callbacks(&rcu_sched_ctrlblk);
 	__rcu_process_callbacks(&rcu_bh_ctrlblk);

kernel/rcu/tree.c

@@ -3013,7 +3013,7 @@ __rcu_process_callbacks(struct rcu_state *rsp)
 /*
  * Do RCU core processing for the current CPU.
  */
-static void rcu_process_callbacks(struct softirq_action *unused)
+static __latent_entropy void rcu_process_callbacks(struct softirq_action *unused)
 {
 	struct rcu_state *rsp;
 

kernel/sched/core.c

@@ -7515,11 +7515,27 @@ static struct kmem_cache *task_group_cache __read_mostly;
 DECLARE_PER_CPU(cpumask_var_t, load_balance_mask);
 DECLARE_PER_CPU(cpumask_var_t, select_idle_mask);
 
+#define WAIT_TABLE_BITS 8
+#define WAIT_TABLE_SIZE (1 << WAIT_TABLE_BITS)
+static wait_queue_head_t bit_wait_table[WAIT_TABLE_SIZE] __cacheline_aligned;
+
+wait_queue_head_t *bit_waitqueue(void *word, int bit)
+{
+	const int shift = BITS_PER_LONG == 32 ? 5 : 6;
+	unsigned long val = (unsigned long)word << shift | bit;
+
+	return bit_wait_table + hash_long(val, WAIT_TABLE_BITS);
+}
+EXPORT_SYMBOL(bit_waitqueue);
+
 void __init sched_init(void)
 {
 	int i, j;
 	unsigned long alloc_size = 0, ptr;
 
+	for (i = 0; i < WAIT_TABLE_SIZE; i++)
+		init_waitqueue_head(bit_wait_table + i);
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	alloc_size += 2 * nr_cpu_ids * sizeof(void **);
 #endif

kernel/sched/debug.c

@@ -415,7 +415,8 @@ static char *task_group_path(struct task_group *tg)
 	if (autogroup_path(tg, group_path, PATH_MAX))
 		return group_path;
 
-	return cgroup_path(tg->css.cgroup, group_path, PATH_MAX);
+	cgroup_path(tg->css.cgroup, group_path, PATH_MAX);
+	return group_path;
 }
 #endif
 

kernel/sched/fair.c

@@ -690,7 +690,14 @@ void init_entity_runnable_average(struct sched_entity *se)
 	 * will definitely be update (after enqueue).
 	 */
 	sa->period_contrib = 1023;
-	sa->load_avg = scale_load_down(se->load.weight);
+	/*
+	 * Tasks are intialized with full load to be seen as heavy tasks until
+	 * they get a chance to stabilize to their real load level.
+	 * Group entities are intialized with zero load to reflect the fact that
+	 * nothing has been attached to the task group yet.
+	 */
+	if (entity_is_task(se))
+		sa->load_avg = scale_load_down(se->load.weight);
 	sa->load_sum = sa->load_avg * LOAD_AVG_MAX;
 	/*
 	 * At this point, util_avg won't be used in select_task_rq_fair anyway
@@ -5471,13 +5478,18 @@ static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd
  */
 static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
 {
-	struct sched_domain *this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc));
-	u64 avg_idle = this_rq()->avg_idle;
-	u64 avg_cost = this_sd->avg_scan_cost;
+	struct sched_domain *this_sd;
+	u64 avg_cost, avg_idle = this_rq()->avg_idle;
 	u64 time, cost;
 	s64 delta;
 	int cpu, wrap;
 
+	this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc));
+	if (!this_sd)
+		return -1;
+
+	avg_cost = this_sd->avg_scan_cost;
+
 	/*
 	 * Due to large variance we need a large fuzz factor; hackbench in
 	 * particularly is sensitive here.
@@ -8522,7 +8534,7 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { }
  * run_rebalance_domains is triggered when needed from the scheduler tick.
  * Also triggered for nohz idle balancing (with nohz_balancing_kick set).
  */
-static void run_rebalance_domains(struct softirq_action *h)
+static __latent_entropy void run_rebalance_domains(struct softirq_action *h)
 {
 	struct rq *this_rq = this_rq();
 	enum cpu_idle_type idle = this_rq->idle_balance ?
@@ -8827,7 +8839,6 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 {
 	struct sched_entity *se;
 	struct cfs_rq *cfs_rq;
-	struct rq *rq;
 	int i;
 
 	tg->cfs_rq = kzalloc(sizeof(cfs_rq) * nr_cpu_ids, GFP_KERNEL);
@@ -8842,8 +8853,6 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 	init_cfs_bandwidth(tg_cfs_bandwidth(tg));
 
 	for_each_possible_cpu(i) {
-		rq = cpu_rq(i);
-
 		cfs_rq = kzalloc_node(sizeof(struct cfs_rq),
 				      GFP_KERNEL, cpu_to_node(i));
 		if (!cfs_rq)

kernel/sched/wait.c

@@ -480,16 +480,6 @@ void wake_up_bit(void *word, int bit)
 }
 EXPORT_SYMBOL(wake_up_bit);
 
-wait_queue_head_t *bit_waitqueue(void *word, int bit)
-{
-	const int shift = BITS_PER_LONG == 32 ? 5 : 6;
-	const struct zone *zone = page_zone(virt_to_page(word));
-	unsigned long val = (unsigned long)word << shift | bit;
-
-	return &zone->wait_table[hash_long(val, zone->wait_table_bits)];
-}
-EXPORT_SYMBOL(bit_waitqueue);
-
 /*
  * Manipulate the atomic_t address to produce a better bit waitqueue table hash
  * index (we're keying off bit -1, but that would produce a horrible hash

kernel/softirq.c

@@ -58,7 +58,7 @@ static struct softirq_action softirq_vec[NR_SOFTIRQS] __cacheline_aligned_in_smp
 DEFINE_PER_CPU(struct task_struct *, ksoftirqd);
 
 const char * const softirq_to_name[NR_SOFTIRQS] = {
-	"HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "BLOCK_IOPOLL",
+	"HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "IRQ_POLL",
 	"TASKLET", "SCHED", "HRTIMER", "RCU"
 };
 
@@ -496,7 +496,7 @@ void __tasklet_hi_schedule_first(struct tasklet_struct *t)
 }
 EXPORT_SYMBOL(__tasklet_hi_schedule_first);
 
-static void tasklet_action(struct softirq_action *a)
+static __latent_entropy void tasklet_action(struct softirq_action *a)
 {
 	struct tasklet_struct *list;
 
@@ -532,7 +532,7 @@ static void tasklet_action(struct softirq_action *a)
 	}
 }
 
-static void tasklet_hi_action(struct softirq_action *a)
+static __latent_entropy void tasklet_hi_action(struct softirq_action *a)
 {
 	struct tasklet_struct *list;
 

kernel/time/alarmtimer.c

@@ -542,7 +542,6 @@ static int alarm_clock_get(clockid_t which_clock, struct timespec *tp)
 static int alarm_timer_create(struct k_itimer *new_timer)
 {
 	enum  alarmtimer_type type;
-	struct alarm_base *base;
 
 	if (!alarmtimer_get_rtcdev())
 		return -ENOTSUPP;
@@ -551,7 +550,6 @@ static int alarm_timer_create(struct k_itimer *new_timer)
 		return -EPERM;
 
 	type = clock2alarm(new_timer->it_clock);
-	base = &alarm_bases[type];
 	alarm_init(&new_timer->it.alarm.alarmtimer, type, alarm_handle_timer);
 	return 0;
 }

kernel/time/timer.c

@@ -878,7 +878,7 @@ static inline struct timer_base *get_timer_base(u32 tflags)
 
 #ifdef CONFIG_NO_HZ_COMMON
 static inline struct timer_base *
-__get_target_base(struct timer_base *base, unsigned tflags)
+get_target_base(struct timer_base *base, unsigned tflags)
 {
 #ifdef CONFIG_SMP
 	if ((tflags & TIMER_PINNED) || !base->migration_enabled)
@@ -891,25 +891,27 @@ __get_target_base(struct timer_base *base, unsigned tflags)
 
 static inline void forward_timer_base(struct timer_base *base)
 {
+	unsigned long jnow = READ_ONCE(jiffies);
+
 	/*
 	 * We only forward the base when it's idle and we have a delta between
 	 * base clock and jiffies.
 	 */
-	if (!base->is_idle || (long) (jiffies - base->clk) < 2)
+	if (!base->is_idle || (long) (jnow - base->clk) < 2)
 		return;
 
 	/*
 	 * If the next expiry value is > jiffies, then we fast forward to
 	 * jiffies otherwise we forward to the next expiry value.
 	 */
-	if (time_after(base->next_expiry, jiffies))
-		base->clk = jiffies;
+	if (time_after(base->next_expiry, jnow))
+		base->clk = jnow;
 	else
 		base->clk = base->next_expiry;
 }
 #else
 static inline struct timer_base *
-__get_target_base(struct timer_base *base, unsigned tflags)
+get_target_base(struct timer_base *base, unsigned tflags)
 {
 	return get_timer_this_cpu_base(tflags);
 }
@@ -917,14 +919,6 @@ __get_target_base(struct timer_base *base, unsigned tflags)
 static inline void forward_timer_base(struct timer_base *base) { }
 #endif
 
-static inline struct timer_base *
-get_target_base(struct timer_base *base, unsigned tflags)
-{
-	struct timer_base *target = __get_target_base(base, tflags);
-
-	forward_timer_base(target);
-	return target;
-}
 
 /*
  * We are using hashed locking: Holding per_cpu(timer_bases[x]).lock means
@@ -943,7 +937,14 @@ static struct timer_base *lock_timer_base(struct timer_list *timer,
 {
 	for (;;) {
 		struct timer_base *base;
-		u32 tf = timer->flags;
+		u32 tf;
+
+		/*
+		 * We need to use READ_ONCE() here, otherwise the compiler
+		 * might re-read @tf between the check for TIMER_MIGRATING
+		 * and spin_lock().
+		 */
+		tf = READ_ONCE(timer->flags);
 
 		if (!(tf & TIMER_MIGRATING)) {
 			base = get_timer_base(tf);
@@ -964,6 +965,8 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 	unsigned long clk = 0, flags;
 	int ret = 0;
 
+	BUG_ON(!timer->function);
+
 	/*
 	 * This is a common optimization triggered by the networking code - if
 	 * the timer is re-modified to have the same timeout or ends up in the
@@ -972,13 +975,16 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 	if (timer_pending(timer)) {
 		if (timer->expires == expires)
 			return 1;
-		/*
-		 * Take the current timer_jiffies of base, but without holding
-		 * the lock!
-		 */
-		base = get_timer_base(timer->flags);
-		clk = base->clk;
 
+		/*
+		 * We lock timer base and calculate the bucket index right
+		 * here. If the timer ends up in the same bucket, then we
+		 * just update the expiry time and avoid the whole
+		 * dequeue/enqueue dance.
+		 */
+		base = lock_timer_base(timer, &flags);
+
+		clk = base->clk;
 		idx = calc_wheel_index(expires, clk);
 
 		/*
@@ -988,14 +994,14 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		 */
 		if (idx == timer_get_idx(timer)) {
 			timer->expires = expires;
-			return 1;
+			ret = 1;
+			goto out_unlock;
 		}
+	} else {
+		base = lock_timer_base(timer, &flags);
 	}
 
 	timer_stats_timer_set_start_info(timer);
-	BUG_ON(!timer->function);
-
-	base = lock_timer_base(timer, &flags);
 
 	ret = detach_if_pending(timer, base, false);
 	if (!ret && pending_only)
@@ -1025,12 +1031,16 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
 		}
 	}
 
+	/* Try to forward a stale timer base clock */
+	forward_timer_base(base);
+
 	timer->expires = expires;
 	/*
 	 * If 'idx' was calculated above and the base time did not advance
-	 * between calculating 'idx' and taking the lock, only enqueue_timer()
-	 * and trigger_dyntick_cpu() is required. Otherwise we need to
-	 * (re)calculate the wheel index via internal_add_timer().
+	 * between calculating 'idx' and possibly switching the base, only
+	 * enqueue_timer() and trigger_dyntick_cpu() is required. Otherwise
+	 * we need to (re)calculate the wheel index via
+	 * internal_add_timer().
 	 */
 	if (idx != UINT_MAX && clk == base->clk) {
 		enqueue_timer(base, timer, idx);
@@ -1510,12 +1520,16 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
 	is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA);
 	base->next_expiry = nextevt;
 	/*
-	 * We have a fresh next event. Check whether we can forward the base:
+	 * We have a fresh next event. Check whether we can forward the
+	 * base. We can only do that when @basej is past base->clk
+	 * otherwise we might rewind base->clk.
 	 */
-	if (time_after(nextevt, jiffies))
-		base->clk = jiffies;
-	else if (time_after(nextevt, base->clk))
-		base->clk = nextevt;
+	if (time_after(basej, base->clk)) {
+		if (time_after(nextevt, basej))
+			base->clk = basej;
+		else if (time_after(nextevt, base->clk))
+			base->clk = nextevt;
+	}
 
 	if (time_before_eq(nextevt, basej)) {
 		expires = basem;
@@ -1633,7 +1647,7 @@ static inline void __run_timers(struct timer_base *base)
 /*
  * This function runs timers and the timer-tq in bottom half context.
  */
-static void run_timer_softirq(struct softirq_action *h)
+static __latent_entropy void run_timer_softirq(struct softirq_action *h)
 {
 	struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);