From 521d24ee598bd8a8b71d7ac76ce2c0da0e548406 Mon Sep 17 00:00:00 2001 From: Pranith Kumar Date: Tue, 8 Jul 2014 18:26:18 -0400 Subject: rcu: Return bool type in rcu_lockdep_current_cpu_online() Return true instead of 1 in rcu_lockdep_current_cpu_online() as this has bool as return type. Signed-off-by: Pranith Kumar Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index d231aa17b1d7..7e47e44bce03 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -349,7 +349,7 @@ bool rcu_lockdep_current_cpu_online(void); #else /* #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PROVE_RCU) */ static inline bool rcu_lockdep_current_cpu_online(void) { - return 1; + return true; } #endif /* #else #if defined(CONFIG_HOTPLUG_CPU) && defined(CONFIG_PROVE_RCU) */ -- cgit v1.2.3 From 85b39d305bfe809a11ff2770d380be3e2465beec Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 8 Jul 2014 15:17:59 -0700 Subject: rcu: Uninline rcu_read_lock_held() This commit uninlines rcu_read_lock_held(). According to "size vmlinux" this saves 28549 in .text: - 5541731 3014560 14757888 23314179 + 5513182 3026848 14757888 23297918 Note: it looks as if the data grows by 12288 bytes but this is not true, it does not actually grow. But .data starts with ALIGN(THREAD_SIZE) and since .text shrinks the padding grows, and thus .data grows too as it seen by /bin/size. diff System.map: - ffffffff81510000 D _sdata - ffffffff81510000 D init_thread_union + ffffffff81509000 D _sdata + ffffffff8150c000 D init_thread_union Perhaps we can change vmlinux.lds.S to .data itself, so that /bin/size can't "wrongly" report that .data grows if .text shinks. Signed-off-by: Oleg Nesterov Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 36 +----------------------------------- kernel/rcu/update.c | 32 ++++++++++++++++++++++++++++++++ 2 files changed, 33 insertions(+), 35 deletions(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 7e47e44bce03..321ed0d4e675 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -371,41 +371,7 @@ extern struct lockdep_map rcu_sched_lock_map; extern struct lockdep_map rcu_callback_map; int debug_lockdep_rcu_enabled(void); -/** - * rcu_read_lock_held() - might we be in RCU read-side critical section? - * - * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an RCU - * read-side critical section. In absence of CONFIG_DEBUG_LOCK_ALLOC, - * this assumes we are in an RCU read-side critical section unless it can - * prove otherwise. This is useful for debug checks in functions that - * require that they be called within an RCU read-side critical section. - * - * Checks debug_lockdep_rcu_enabled() to prevent false positives during boot - * and while lockdep is disabled. - * - * Note that rcu_read_lock() and the matching rcu_read_unlock() must - * occur in the same context, for example, it is illegal to invoke - * rcu_read_unlock() in process context if the matching rcu_read_lock() - * was invoked from within an irq handler. - * - * Note that rcu_read_lock() is disallowed if the CPU is either idle or - * offline from an RCU perspective, so check for those as well. - */ -static inline int rcu_read_lock_held(void) -{ - if (!debug_lockdep_rcu_enabled()) - return 1; - if (!rcu_is_watching()) - return 0; - if (!rcu_lockdep_current_cpu_online()) - return 0; - return lock_is_held(&rcu_lock_map); -} - -/* - * rcu_read_lock_bh_held() is defined out of line to avoid #include-file - * hell. - */ +int rcu_read_lock_held(void); int rcu_read_lock_bh_held(void); /** diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 4056d7992a6c..ea8ea7b16e11 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -136,6 +136,38 @@ int notrace debug_lockdep_rcu_enabled(void) } EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled); +/** + * rcu_read_lock_held() - might we be in RCU read-side critical section? + * + * If CONFIG_DEBUG_LOCK_ALLOC is selected, returns nonzero iff in an RCU + * read-side critical section. In absence of CONFIG_DEBUG_LOCK_ALLOC, + * this assumes we are in an RCU read-side critical section unless it can + * prove otherwise. This is useful for debug checks in functions that + * require that they be called within an RCU read-side critical section. + * + * Checks debug_lockdep_rcu_enabled() to prevent false positives during boot + * and while lockdep is disabled. + * + * Note that rcu_read_lock() and the matching rcu_read_unlock() must + * occur in the same context, for example, it is illegal to invoke + * rcu_read_unlock() in process context if the matching rcu_read_lock() + * was invoked from within an irq handler. + * + * Note that rcu_read_lock() is disallowed if the CPU is either idle or + * offline from an RCU perspective, so check for those as well. + */ +int rcu_read_lock_held(void) +{ + if (!debug_lockdep_rcu_enabled()) + return 1; + if (!rcu_is_watching()) + return 0; + if (!rcu_lockdep_current_cpu_online()) + return 0; + return lock_is_held(&rcu_lock_map); +} +EXPORT_SYMBOL_GPL(rcu_read_lock_held); + /** * rcu_read_lock_bh_held() - might we be in RCU-bh read-side critical section? * -- cgit v1.2.3 From 9fdd3bc9005824704f9802bec7b3e06f5edae434 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 29 Jul 2014 14:50:47 -0700 Subject: rcu: Break more call_rcu() deadlock involving scheduler and perf Commit 96d3fd0d315a9 (rcu: Break call_rcu() deadlock involving scheduler and perf) covered the case where __call_rcu_nocb_enqueue() needs to wake the rcuo kthread due to the queue being initially empty, but did not do anything for the case where the queue was overflowing. This commit therefore also defers wakeup for the overflow case. Signed-off-by: Paul E. McKenney --- include/trace/events/rcu.h | 3 +++ kernel/rcu/tree.h | 9 +++++++-- kernel/rcu/tree_plugin.h | 26 ++++++++++++++++++-------- 3 files changed, 28 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index aca382266411..9b56f37148cf 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -180,9 +180,12 @@ TRACE_EVENT(rcu_grace_period_init, * argument is a string as follows: * * "WakeEmpty": Wake rcuo kthread, first CB to empty list. + * "WakeEmptyIsDeferred": Wake rcuo kthread later, first CB to empty list. * "WakeOvf": Wake rcuo kthread, CB list is huge. + * "WakeOvfIsDeferred": Wake rcuo kthread later, CB list is huge. * "WakeNot": Don't wake rcuo kthread. * "WakeNotPoll": Don't wake rcuo kthread because it is polling. + * "DeferredWake": Carried out the "IsDeferred" wakeup. * "Poll": Start of new polling cycle for rcu_nocb_poll. * "Sleep": Sleep waiting for CBs for !rcu_nocb_poll. * "WokeEmpty": rcuo kthread woke to find empty list. diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index 6a86eb7bac45..e33562f2a655 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -350,7 +350,7 @@ struct rcu_data { int nocb_p_count_lazy; /* (approximate). */ wait_queue_head_t nocb_wq; /* For nocb kthreads to sleep on. */ struct task_struct *nocb_kthread; - bool nocb_defer_wakeup; /* Defer wakeup of nocb_kthread. */ + int nocb_defer_wakeup; /* Defer wakeup of nocb_kthread. */ /* The following fields are used by the leader, hence own cacheline. */ struct rcu_head *nocb_gp_head ____cacheline_internodealigned_in_smp; @@ -383,6 +383,11 @@ struct rcu_data { #define RCU_FORCE_QS 3 /* Need to force quiescent state. */ #define RCU_SIGNAL_INIT RCU_SAVE_DYNTICK +/* Values for nocb_defer_wakeup field in struct rcu_data. */ +#define RCU_NOGP_WAKE_NOT 0 +#define RCU_NOGP_WAKE 1 +#define RCU_NOGP_WAKE_FORCE 2 + #define RCU_JIFFIES_TILL_FORCE_QS (1 + (HZ > 250) + (HZ > 500)) /* For jiffies_till_first_fqs and */ /* and jiffies_till_next_fqs. */ @@ -589,7 +594,7 @@ static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp, static bool rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp, struct rcu_data *rdp, unsigned long flags); -static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp); +static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp); static void do_nocb_deferred_wakeup(struct rcu_data *rdp); static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp); static void rcu_spawn_nocb_kthreads(struct rcu_state *rsp); diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index bb564560aeb8..d67cc5c375c5 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -2121,16 +2121,23 @@ static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmpty")); } else { - rdp->nocb_defer_wakeup = true; + rdp->nocb_defer_wakeup = RCU_NOGP_WAKE; trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeEmptyIsDeferred")); } rdp->qlen_last_fqs_check = 0; } else if (len > rdp->qlen_last_fqs_check + qhimark) { /* ... or if many callbacks queued. */ - wake_nocb_leader(rdp, true); + if (!irqs_disabled_flags(flags)) { + wake_nocb_leader(rdp, true); + trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, + TPS("WakeOvf")); + } else { + rdp->nocb_defer_wakeup = RCU_NOGP_WAKE_FORCE; + trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, + TPS("WakeOvfIsDeferred")); + } rdp->qlen_last_fqs_check = LONG_MAX / 2; - trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeOvf")); } else { trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("WakeNot")); } @@ -2438,7 +2445,7 @@ static int rcu_nocb_kthread(void *arg) } /* Is a deferred wakeup of rcu_nocb_kthread() required? */ -static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) +static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) { return ACCESS_ONCE(rdp->nocb_defer_wakeup); } @@ -2446,11 +2453,14 @@ static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) /* Do a deferred wakeup of rcu_nocb_kthread(). */ static void do_nocb_deferred_wakeup(struct rcu_data *rdp) { + int ndw; + if (!rcu_nocb_need_deferred_wakeup(rdp)) return; - ACCESS_ONCE(rdp->nocb_defer_wakeup) = false; - wake_nocb_leader(rdp, false); - trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWakeEmpty")); + ndw = ACCESS_ONCE(rdp->nocb_defer_wakeup); + ACCESS_ONCE(rdp->nocb_defer_wakeup) = RCU_NOGP_WAKE_NOT; + wake_nocb_leader(rdp, ndw == RCU_NOGP_WAKE_FORCE); + trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWake")); } /* Initialize per-rcu_data variables for no-CBs CPUs. */ @@ -2557,7 +2567,7 @@ static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp) { } -static bool rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) +static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp) { return false; } -- cgit v1.2.3 From eea203fea3484598280a07fe503e025e886297fb Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Mon, 14 Jul 2014 09:16:15 -0400 Subject: rcu: Use pr_alert/pr_cont for printing logs User pr_alert/pr_cont for printing the logs from rcutorture module directly instead of writing it to a buffer and then printing it. This allows us from not having to allocate such buffers. Also remove a resulting empty function. I tested this using the parse-torture.sh script as follows: $ dmesg | grep torture > log.txt $ bash parse-torture.sh log.txt test $ There were no warnings which means that parsing went fine. Signed-off-by: Joe Perches Signed-off-by: Pranith Kumar Signed-off-by: Paul E. McKenney --- include/linux/torture.h | 2 +- kernel/rcu/rcutorture.c | 127 +++++++++++++++++++++--------------------------- kernel/torture.c | 16 +++--- 3 files changed, 64 insertions(+), 81 deletions(-) (limited to 'include') diff --git a/include/linux/torture.h b/include/linux/torture.h index 5ca58fcbaf1b..fec46f8c08eb 100644 --- a/include/linux/torture.h +++ b/include/linux/torture.h @@ -51,7 +51,7 @@ /* Definitions for online/offline exerciser. */ int torture_onoff_init(long ooholdoff, long oointerval); -char *torture_onoff_stats(char *page); +void torture_onoff_stats(void); bool torture_onoff_failures(void); /* Low-rider random number generator. */ diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 7e67711cbae8..ff4f0c756dee 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -242,7 +242,7 @@ struct rcu_torture_ops { void (*call)(struct rcu_head *head, void (*func)(struct rcu_head *rcu)); void (*cb_barrier)(void); void (*fqs)(void); - void (*stats)(char *page); + void (*stats)(void); int irq_capable; int can_boost; const char *name; @@ -525,21 +525,21 @@ static void srcu_torture_barrier(void) srcu_barrier(&srcu_ctl); } -static void srcu_torture_stats(char *page) +static void srcu_torture_stats(void) { int cpu; int idx = srcu_ctl.completed & 0x1; - page += sprintf(page, "%s%s per-CPU(idx=%d):", - torture_type, TORTURE_FLAG, idx); + pr_alert("%s%s per-CPU(idx=%d):", + torture_type, TORTURE_FLAG, idx); for_each_possible_cpu(cpu) { long c0, c1; c0 = (long)per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[!idx]; c1 = (long)per_cpu_ptr(srcu_ctl.per_cpu_ref, cpu)->c[idx]; - page += sprintf(page, " %d(%ld,%ld)", cpu, c0, c1); + pr_cont(" %d(%ld,%ld)", cpu, c0, c1); } - sprintf(page, "\n"); + pr_cont("\n"); } static void srcu_torture_synchronize_expedited(void) @@ -1031,10 +1031,15 @@ rcu_torture_reader(void *arg) } /* - * Create an RCU-torture statistics message in the specified buffer. + * Print torture statistics. Caller must ensure that there is only + * one call to this function at a given time!!! This is normally + * accomplished by relying on the module system to only have one copy + * of the module loaded, and then by giving the rcu_torture_stats + * kthread full control (or the init/cleanup functions when rcu_torture_stats + * thread is not running). */ static void -rcu_torture_printk(char *page) +rcu_torture_stats_print(void) { int cpu; int i; @@ -1052,55 +1057,60 @@ rcu_torture_printk(char *page) if (pipesummary[i] != 0) break; } - page += sprintf(page, "%s%s ", torture_type, TORTURE_FLAG); - page += sprintf(page, - "rtc: %p ver: %lu tfle: %d rta: %d rtaf: %d rtf: %d ", - rcu_torture_current, - rcu_torture_current_version, - list_empty(&rcu_torture_freelist), - atomic_read(&n_rcu_torture_alloc), - atomic_read(&n_rcu_torture_alloc_fail), - atomic_read(&n_rcu_torture_free)); - page += sprintf(page, "rtmbe: %d rtbke: %ld rtbre: %ld ", - atomic_read(&n_rcu_torture_mberror), - n_rcu_torture_boost_ktrerror, - n_rcu_torture_boost_rterror); - page += sprintf(page, "rtbf: %ld rtb: %ld nt: %ld ", - n_rcu_torture_boost_failure, - n_rcu_torture_boosts, - n_rcu_torture_timers); - page = torture_onoff_stats(page); - page += sprintf(page, "barrier: %ld/%ld:%ld", - n_barrier_successes, - n_barrier_attempts, - n_rcu_torture_barrier_error); - page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG); + + pr_alert("%s%s ", torture_type, TORTURE_FLAG); + pr_cont("rtc: %p ver: %lu tfle: %d rta: %d rtaf: %d rtf: %d ", + rcu_torture_current, + rcu_torture_current_version, + list_empty(&rcu_torture_freelist), + atomic_read(&n_rcu_torture_alloc), + atomic_read(&n_rcu_torture_alloc_fail), + atomic_read(&n_rcu_torture_free)); + pr_cont("rtmbe: %d rtbke: %ld rtbre: %ld ", + atomic_read(&n_rcu_torture_mberror), + n_rcu_torture_boost_ktrerror, + n_rcu_torture_boost_rterror); + pr_cont("rtbf: %ld rtb: %ld nt: %ld ", + n_rcu_torture_boost_failure, + n_rcu_torture_boosts, + n_rcu_torture_timers); + torture_onoff_stats(); + pr_cont("barrier: %ld/%ld:%ld\n", + n_barrier_successes, + n_barrier_attempts, + n_rcu_torture_barrier_error); + + pr_alert("%s%s ", torture_type, TORTURE_FLAG); if (atomic_read(&n_rcu_torture_mberror) != 0 || n_rcu_torture_barrier_error != 0 || n_rcu_torture_boost_ktrerror != 0 || n_rcu_torture_boost_rterror != 0 || n_rcu_torture_boost_failure != 0 || i > 1) { - page += sprintf(page, "!!! "); + pr_cont("%s", "!!! "); atomic_inc(&n_rcu_torture_error); WARN_ON_ONCE(1); } - page += sprintf(page, "Reader Pipe: "); + pr_cont("Reader Pipe: "); for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) - page += sprintf(page, " %ld", pipesummary[i]); - page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG); - page += sprintf(page, "Reader Batch: "); + pr_cont(" %ld", pipesummary[i]); + pr_cont("\n"); + + pr_alert("%s%s ", torture_type, TORTURE_FLAG); + pr_cont("Reader Batch: "); for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) - page += sprintf(page, " %ld", batchsummary[i]); - page += sprintf(page, "\n%s%s ", torture_type, TORTURE_FLAG); - page += sprintf(page, "Free-Block Circulation: "); + pr_cont(" %ld", batchsummary[i]); + pr_cont("\n"); + + pr_alert("%s%s ", torture_type, TORTURE_FLAG); + pr_cont("Free-Block Circulation: "); for (i = 0; i < RCU_TORTURE_PIPE_LEN + 1; i++) { - page += sprintf(page, " %d", - atomic_read(&rcu_torture_wcount[i])); + pr_cont(" %d", atomic_read(&rcu_torture_wcount[i])); } - page += sprintf(page, "\n"); + pr_cont("\n"); + if (cur_ops->stats) - cur_ops->stats(page); + cur_ops->stats(); if (rtcv_snap == rcu_torture_current_version && rcu_torture_current != NULL) { int __maybe_unused flags; @@ -1109,40 +1119,15 @@ rcu_torture_printk(char *page) rcutorture_get_gp_data(cur_ops->ttype, &flags, &gpnum, &completed); - page += sprintf(page, - "??? Writer stall state %d g%lu c%lu f%#x\n", - rcu_torture_writer_state, - gpnum, completed, flags); + pr_alert("??? Writer stall state %d g%lu c%lu f%#x\n", + rcu_torture_writer_state, + gpnum, completed, flags); show_rcu_gp_kthreads(); rcutorture_trace_dump(); } rtcv_snap = rcu_torture_current_version; } -/* - * Print torture statistics. Caller must ensure that there is only - * one call to this function at a given time!!! This is normally - * accomplished by relying on the module system to only have one copy - * of the module loaded, and then by giving the rcu_torture_stats - * kthread full control (or the init/cleanup functions when rcu_torture_stats - * thread is not running). - */ -static void -rcu_torture_stats_print(void) -{ - int size = nr_cpu_ids * 200 + 8192; - char *buf; - - buf = kmalloc(size, GFP_KERNEL); - if (!buf) { - pr_err("rcu-torture: Out of memory, need: %d", size); - return; - } - rcu_torture_printk(buf); - pr_alert("%s", buf); - kfree(buf); -} - /* * Periodically prints torture statistics, if periodic statistics printing * was specified via the stat_interval module parameter. diff --git a/kernel/torture.c b/kernel/torture.c index d600af21f022..ede8b25ec1ae 100644 --- a/kernel/torture.c +++ b/kernel/torture.c @@ -211,18 +211,16 @@ EXPORT_SYMBOL_GPL(torture_onoff_cleanup); /* * Print online/offline testing statistics. */ -char *torture_onoff_stats(char *page) +void torture_onoff_stats(void) { #ifdef CONFIG_HOTPLUG_CPU - page += sprintf(page, - "onoff: %ld/%ld:%ld/%ld %d,%d:%d,%d %lu:%lu (HZ=%d) ", - n_online_successes, n_online_attempts, - n_offline_successes, n_offline_attempts, - min_online, max_online, - min_offline, max_offline, - sum_online, sum_offline, HZ); + pr_cont("onoff: %ld/%ld:%ld/%ld %d,%d:%d,%d %lu:%lu (HZ=%d) ", + n_online_successes, n_online_attempts, + n_offline_successes, n_offline_attempts, + min_online, max_online, + min_offline, max_offline, + sum_online, sum_offline, HZ); #endif /* #ifdef CONFIG_HOTPLUG_CPU */ - return page; } EXPORT_SYMBOL_GPL(torture_onoff_stats); -- cgit v1.2.3 From 8315f42295d2667a7f942f154b73a86fd7cb2227 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 27 Jun 2014 13:42:20 -0700 Subject: rcu: Add call_rcu_tasks() This commit adds a new RCU-tasks flavor of RCU, which provides call_rcu_tasks(). This RCU flavor's quiescent states are voluntary context switch (not preemption!) and userspace execution (not the idle loop -- use some sort of schedule_on_each_cpu() if you need to handle the idle tasks. Note that unlike other RCU flavors, these quiescent states occur in tasks, not necessarily CPUs. Includes fixes from Steven Rostedt. This RCU flavor is assumed to have very infrequent latency-tolerant updaters. This assumption permits significant simplifications, including a single global callback list protected by a single global lock, along with a single task-private linked list containing all tasks that have not yet passed through a quiescent state. If experience shows this assumption to be incorrect, the required additional complexity will be added. Suggested-by: Steven Rostedt Signed-off-by: Paul E. McKenney --- include/linux/init_task.h | 9 +++ include/linux/rcupdate.h | 36 ++++++++++ include/linux/sched.h | 23 ++++--- init/Kconfig | 10 +++ kernel/rcu/tiny.c | 2 + kernel/rcu/tree.c | 2 + kernel/rcu/update.c | 171 ++++++++++++++++++++++++++++++++++++++++++++++ 7 files changed, 242 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 2bb4c4f3531a..dffd9258ee60 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -117,6 +117,14 @@ extern struct group_info init_groups; #else #define INIT_TASK_RCU_PREEMPT(tsk) #endif +#ifdef CONFIG_TASKS_RCU +#define INIT_TASK_RCU_TASKS(tsk) \ + .rcu_tasks_holdout = false, \ + .rcu_tasks_holdout_list = \ + LIST_HEAD_INIT(tsk.rcu_tasks_holdout_list), +#else +#define INIT_TASK_RCU_TASKS(tsk) +#endif extern struct cred init_cred; @@ -224,6 +232,7 @@ extern struct task_group root_task_group; INIT_FTRACE_GRAPH \ INIT_TRACE_RECURSION \ INIT_TASK_RCU_PREEMPT(tsk) \ + INIT_TASK_RCU_TASKS(tsk) \ INIT_CPUSET_SEQ(tsk) \ INIT_RT_MUTEXES(tsk) \ INIT_VTIME(tsk) \ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index d231aa17b1d7..3432063f4c87 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -197,6 +197,26 @@ void call_rcu_sched(struct rcu_head *head, void synchronize_sched(void); +/** + * call_rcu_tasks() - Queue an RCU for invocation task-based grace period + * @head: structure to be used for queueing the RCU updates. + * @func: actual callback function to be invoked after the grace period + * + * The callback function will be invoked some time after a full grace + * period elapses, in other words after all currently executing RCU + * read-side critical sections have completed. call_rcu_tasks() assumes + * that the read-side critical sections end at a voluntary context + * switch (not a preemption!), entry into idle, or transition to usermode + * execution. As such, there are no read-side primitives analogous to + * rcu_read_lock() and rcu_read_unlock() because this primitive is intended + * to determine that all tasks have passed through a safe state, not so + * much for data-strcuture synchronization. + * + * See the description of call_rcu() for more detailed information on + * memory ordering guarantees. + */ +void call_rcu_tasks(struct rcu_head *head, void (*func)(struct rcu_head *head)); + #ifdef CONFIG_PREEMPT_RCU void __rcu_read_lock(void); @@ -294,6 +314,22 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, rcu_irq_exit(); \ } while (0) +/* + * Note a voluntary context switch for RCU-tasks benefit. This is a + * macro rather than an inline function to avoid #include hell. + */ +#ifdef CONFIG_TASKS_RCU +#define rcu_note_voluntary_context_switch(t) \ + do { \ + preempt_disable(); /* Exclude synchronize_sched(); */ \ + if (ACCESS_ONCE((t)->rcu_tasks_holdout)) \ + ACCESS_ONCE((t)->rcu_tasks_holdout) = false; \ + preempt_enable(); \ + } while (0) +#else /* #ifdef CONFIG_TASKS_RCU */ +#define rcu_note_voluntary_context_switch(t) do { } while (0) +#endif /* #else #ifdef CONFIG_TASKS_RCU */ + #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP) bool __rcu_is_watching(void); #endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP) */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 5c2c885ee52b..eaacac4ae77d 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1270,6 +1270,11 @@ struct task_struct { #ifdef CONFIG_TREE_PREEMPT_RCU struct rcu_node *rcu_blocked_node; #endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ +#ifdef CONFIG_TASKS_RCU + unsigned long rcu_tasks_nvcsw; + bool rcu_tasks_holdout; + struct list_head rcu_tasks_holdout_list; +#endif /* #ifdef CONFIG_TASKS_RCU */ #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) struct sched_info sched_info; @@ -2000,28 +2005,24 @@ extern void task_clear_jobctl_pending(struct task_struct *task, unsigned int mask); #ifdef CONFIG_PREEMPT_RCU - #define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */ #define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */ +#endif /* #ifdef CONFIG_PREEMPT_RCU */ static inline void rcu_copy_process(struct task_struct *p) { +#ifdef CONFIG_PREEMPT_RCU p->rcu_read_lock_nesting = 0; p->rcu_read_unlock_special = 0; -#ifdef CONFIG_TREE_PREEMPT_RCU p->rcu_blocked_node = NULL; -#endif /* #ifdef CONFIG_TREE_PREEMPT_RCU */ INIT_LIST_HEAD(&p->rcu_node_entry); +#endif /* #ifdef CONFIG_PREEMPT_RCU */ +#ifdef CONFIG_TASKS_RCU + p->rcu_tasks_holdout = false; + INIT_LIST_HEAD(&p->rcu_tasks_holdout_list); +#endif /* #ifdef CONFIG_TASKS_RCU */ } -#else - -static inline void rcu_copy_process(struct task_struct *p) -{ -} - -#endif - static inline void tsk_restore_flags(struct task_struct *task, unsigned long orig_flags, unsigned long flags) { diff --git a/init/Kconfig b/init/Kconfig index e84c6423a2e5..c4539c4e177f 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -507,6 +507,16 @@ config PREEMPT_RCU This option enables preemptible-RCU code that is common between TREE_PREEMPT_RCU and, in the old days, TINY_PREEMPT_RCU. +config TASKS_RCU + bool "Task_based RCU implementation using voluntary context switch" + default n + help + This option enables a task-based RCU implementation that uses + only voluntary context switch (not preemption!), idle, and + user-mode execution as quiescent states. + + If unsure, say N. + config RCU_STALL_COMMON def_bool ( TREE_RCU || TREE_PREEMPT_RCU || RCU_TRACE ) help diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c index d9efcc13008c..717f00854fc0 100644 --- a/kernel/rcu/tiny.c +++ b/kernel/rcu/tiny.c @@ -254,6 +254,8 @@ void rcu_check_callbacks(int cpu, int user) rcu_sched_qs(cpu); else if (!in_softirq()) rcu_bh_qs(cpu); + if (user) + rcu_note_voluntary_context_switch(current); } /* diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 1b70cb6fbe3c..8ad91d1e317d 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2410,6 +2410,8 @@ void rcu_check_callbacks(int cpu, int user) rcu_preempt_check_callbacks(cpu); if (rcu_pending(cpu)) invoke_rcu_core(); + if (user) + rcu_note_voluntary_context_switch(current); trace_rcu_utilization(TPS("End scheduler-tick")); } diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 4056d7992a6c..19b3dacb0753 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -47,6 +47,7 @@ #include #include #include +#include #define CREATE_TRACE_POINTS @@ -347,3 +348,173 @@ static int __init check_cpu_stall_init(void) early_initcall(check_cpu_stall_init); #endif /* #ifdef CONFIG_RCU_STALL_COMMON */ + +#ifdef CONFIG_TASKS_RCU + +/* + * Simple variant of RCU whose quiescent states are voluntary context switch, + * user-space execution, and idle. As such, grace periods can take one good + * long time. There are no read-side primitives similar to rcu_read_lock() + * and rcu_read_unlock() because this implementation is intended to get + * the system into a safe state for some of the manipulations involved in + * tracing and the like. Finally, this implementation does not support + * high call_rcu_tasks() rates from multiple CPUs. If this is required, + * per-CPU callback lists will be needed. + */ + +/* Global list of callbacks and associated lock. */ +static struct rcu_head *rcu_tasks_cbs_head; +static struct rcu_head **rcu_tasks_cbs_tail = &rcu_tasks_cbs_head; +static DEFINE_RAW_SPINLOCK(rcu_tasks_cbs_lock); + +/* Post an RCU-tasks callback. */ +void call_rcu_tasks(struct rcu_head *rhp, void (*func)(struct rcu_head *rhp)) +{ + unsigned long flags; + + rhp->next = NULL; + rhp->func = func; + raw_spin_lock_irqsave(&rcu_tasks_cbs_lock, flags); + *rcu_tasks_cbs_tail = rhp; + rcu_tasks_cbs_tail = &rhp->next; + raw_spin_unlock_irqrestore(&rcu_tasks_cbs_lock, flags); +} +EXPORT_SYMBOL_GPL(call_rcu_tasks); + +/* See if the current task has stopped holding out, remove from list if so. */ +static void check_holdout_task(struct task_struct *t) +{ + if (!ACCESS_ONCE(t->rcu_tasks_holdout) || + t->rcu_tasks_nvcsw != ACCESS_ONCE(t->nvcsw) || + !ACCESS_ONCE(t->on_rq)) { + ACCESS_ONCE(t->rcu_tasks_holdout) = false; + list_del_rcu(&t->rcu_tasks_holdout_list); + put_task_struct(t); + } +} + +/* RCU-tasks kthread that detects grace periods and invokes callbacks. */ +static int __noreturn rcu_tasks_kthread(void *arg) +{ + unsigned long flags; + struct task_struct *g, *t; + struct rcu_head *list; + struct rcu_head *next; + LIST_HEAD(rcu_tasks_holdouts); + + /* FIXME: Add housekeeping affinity. */ + + /* + * Each pass through the following loop makes one check for + * newly arrived callbacks, and, if there are some, waits for + * one RCU-tasks grace period and then invokes the callbacks. + * This loop is terminated by the system going down. ;-) + */ + for (;;) { + + /* Pick up any new callbacks. */ + raw_spin_lock_irqsave(&rcu_tasks_cbs_lock, flags); + list = rcu_tasks_cbs_head; + rcu_tasks_cbs_head = NULL; + rcu_tasks_cbs_tail = &rcu_tasks_cbs_head; + raw_spin_unlock_irqrestore(&rcu_tasks_cbs_lock, flags); + + /* If there were none, wait a bit and start over. */ + if (!list) { + schedule_timeout_interruptible(HZ); + WARN_ON(signal_pending(current)); + continue; + } + + /* + * Wait for all pre-existing t->on_rq and t->nvcsw + * transitions to complete. Invoking synchronize_sched() + * suffices because all these transitions occur with + * interrupts disabled. Without this synchronize_sched(), + * a read-side critical section that started before the + * grace period might be incorrectly seen as having started + * after the grace period. + * + * This synchronize_sched() also dispenses with the + * need for a memory barrier on the first store to + * ->rcu_tasks_holdout, as it forces the store to happen + * after the beginning of the grace period. + */ + synchronize_sched(); + + /* + * There were callbacks, so we need to wait for an + * RCU-tasks grace period. Start off by scanning + * the task list for tasks that are not already + * voluntarily blocked. Mark these tasks and make + * a list of them in rcu_tasks_holdouts. + */ + rcu_read_lock(); + for_each_process_thread(g, t) { + if (t != current && ACCESS_ONCE(t->on_rq) && + !is_idle_task(t)) { + get_task_struct(t); + t->rcu_tasks_nvcsw = ACCESS_ONCE(t->nvcsw); + ACCESS_ONCE(t->rcu_tasks_holdout) = true; + list_add(&t->rcu_tasks_holdout_list, + &rcu_tasks_holdouts); + } + } + rcu_read_unlock(); + + /* + * Each pass through the following loop scans the list + * of holdout tasks, removing any that are no longer + * holdouts. When the list is empty, we are done. + */ + while (!list_empty(&rcu_tasks_holdouts)) { + schedule_timeout_interruptible(HZ); + WARN_ON(signal_pending(current)); + rcu_read_lock(); + list_for_each_entry_rcu(t, &rcu_tasks_holdouts, + rcu_tasks_holdout_list) + check_holdout_task(t); + rcu_read_unlock(); + } + + /* + * Because ->on_rq and ->nvcsw are not guaranteed + * to have a full memory barriers prior to them in the + * schedule() path, memory reordering on other CPUs could + * cause their RCU-tasks read-side critical sections to + * extend past the end of the grace period. However, + * because these ->nvcsw updates are carried out with + * interrupts disabled, we can use synchronize_sched() + * to force the needed ordering on all such CPUs. + * + * This synchronize_sched() also confines all + * ->rcu_tasks_holdout accesses to be within the grace + * period, avoiding the need for memory barriers for + * ->rcu_tasks_holdout accesses. + */ + synchronize_sched(); + + /* Invoke the callbacks. */ + while (list) { + next = list->next; + local_bh_disable(); + list->func(list); + local_bh_enable(); + list = next; + cond_resched(); + } + } +} + +/* Spawn rcu_tasks_kthread() at boot time. */ +static int __init rcu_spawn_tasks_kthread(void) +{ + struct task_struct __maybe_unused *t; + + t = kthread_run(rcu_tasks_kthread, NULL, "rcu_tasks_kthread"); + BUG_ON(IS_ERR(t)); + return 0; +} +early_initcall(rcu_spawn_tasks_kthread); + +#endif /* #ifdef CONFIG_TASKS_RCU */ -- cgit v1.2.3 From bde6c3aa993066acb0d6ce32ecabe03b9d5df92d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 1 Jul 2014 11:26:57 -0700 Subject: rcu: Provide cond_resched_rcu_qs() to force quiescent states in long loops RCU-tasks requires the occasional voluntary context switch from CPU-bound in-kernel tasks. In some cases, this requires instrumenting cond_resched(). However, there is some reluctance to countenance unconditionally instrumenting cond_resched() (see http://lwn.net/Articles/603252/), so this commit creates a separate cond_resched_rcu_qs() that may be used in place of cond_resched() in locations prone to long-duration in-kernel looping. This commit currently instruments only RCU-tasks. Future possibilities include also instrumenting RCU, RCU-bh, and RCU-sched in order to reduce IPI usage. Signed-off-by: Paul E. McKenney --- fs/file.c | 2 +- include/linux/rcupdate.h | 13 +++++++++++++ kernel/rcu/rcutorture.c | 4 ++-- kernel/rcu/tree.c | 12 ++++++------ kernel/rcu/tree_plugin.h | 2 +- mm/mlock.c | 2 +- 6 files changed, 24 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/fs/file.c b/fs/file.c index 66923fe3176e..1cafc4c9275b 100644 --- a/fs/file.c +++ b/fs/file.c @@ -367,7 +367,7 @@ static struct fdtable *close_files(struct files_struct * files) struct file * file = xchg(&fdt->fd[i], NULL); if (file) { filp_close(file, files); - cond_resched(); + cond_resched_rcu_qs(); } } i++; diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 3432063f4c87..473350462d04 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -330,6 +330,19 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, #define rcu_note_voluntary_context_switch(t) do { } while (0) #endif /* #else #ifdef CONFIG_TASKS_RCU */ +/** + * cond_resched_rcu_qs - Report potential quiescent states to RCU + * + * This macro resembles cond_resched(), except that it is defined to + * report potential quiescent states to RCU-tasks even if the cond_resched() + * machinery were to be shut off, as some advocate for PREEMPT kernels. + */ +#define cond_resched_rcu_qs() \ +do { \ + rcu_note_voluntary_context_switch(current); \ + cond_resched(); \ +} while (0) + #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP) bool __rcu_is_watching(void); #endif /* #if defined(CONFIG_DEBUG_LOCK_ALLOC) || defined(CONFIG_RCU_TRACE) || defined(CONFIG_SMP) */ diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 948a7693748e..178716713e11 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -667,7 +667,7 @@ static int rcu_torture_boost(void *arg) } call_rcu_time = jiffies; } - cond_resched(); + cond_resched_rcu_qs(); stutter_wait("rcu_torture_boost"); if (torture_must_stop()) goto checkwait; @@ -1019,7 +1019,7 @@ rcu_torture_reader(void *arg) __this_cpu_inc(rcu_torture_batch[completed]); preempt_enable(); cur_ops->readunlock(idx); - cond_resched(); + cond_resched_rcu_qs(); stutter_wait("rcu_torture_reader"); } while (!torture_must_stop()); if (irqreader && cur_ops->irq_capable) { diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 8ad91d1e317d..e23dad0661e2 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1647,7 +1647,7 @@ static int rcu_gp_init(struct rcu_state *rsp) rnp->level, rnp->grplo, rnp->grphi, rnp->qsmask); raw_spin_unlock_irq(&rnp->lock); - cond_resched(); + cond_resched_rcu_qs(); } mutex_unlock(&rsp->onoff_mutex); @@ -1736,7 +1736,7 @@ static void rcu_gp_cleanup(struct rcu_state *rsp) /* smp_mb() provided by prior unlock-lock pair. */ nocb += rcu_future_gp_cleanup(rsp, rnp); raw_spin_unlock_irq(&rnp->lock); - cond_resched(); + cond_resched_rcu_qs(); } rnp = rcu_get_root(rsp); raw_spin_lock_irq(&rnp->lock); @@ -1785,7 +1785,7 @@ static int __noreturn rcu_gp_kthread(void *arg) /* Locking provides needed memory barrier. */ if (rcu_gp_init(rsp)) break; - cond_resched(); + cond_resched_rcu_qs(); flush_signals(current); trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), @@ -1828,10 +1828,10 @@ static int __noreturn rcu_gp_kthread(void *arg) trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), TPS("fqsend")); - cond_resched(); + cond_resched_rcu_qs(); } else { /* Deal with stray signal. */ - cond_resched(); + cond_resched_rcu_qs(); flush_signals(current); trace_rcu_grace_period(rsp->name, ACCESS_ONCE(rsp->gpnum), @@ -2434,7 +2434,7 @@ static void force_qs_rnp(struct rcu_state *rsp, struct rcu_node *rnp; rcu_for_each_leaf_node(rsp, rnp) { - cond_resched(); + cond_resched_rcu_qs(); mask = 0; raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index a7997e272564..7672586d3920 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -1848,7 +1848,7 @@ static int rcu_oom_notify(struct notifier_block *self, get_online_cpus(); for_each_online_cpu(cpu) { smp_call_function_single(cpu, rcu_oom_notify_cpu, NULL, 1); - cond_resched(); + cond_resched_rcu_qs(); } put_online_cpus(); diff --git a/mm/mlock.c b/mm/mlock.c index ce84cb0b83ef..ab3150c26711 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -789,7 +789,7 @@ static int do_mlockall(int flags) /* Ignore errors */ mlock_fixup(vma, &prev, vma->vm_start, vma->vm_end, newflags); - cond_resched(); + cond_resched_rcu_qs(); } out: return 0; -- cgit v1.2.3 From 53c6d4edf874d3cbc031a53738c6cba9277faea5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 1 Jul 2014 12:22:23 -0700 Subject: rcu: Add synchronous grace-period waiting for RCU-tasks It turns out to be easier to add the synchronous grace-period waiting functions to RCU-tasks than to work around their absense in rcutorture, so this commit adds them. The key point is that the existence of call_rcu_tasks() means that rcutorture needs an rcu_barrier_tasks(). Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 2 ++ kernel/rcu/update.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 57 insertions(+) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 473350462d04..640152fedcde 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -216,6 +216,8 @@ void synchronize_sched(void); * memory ordering guarantees. */ void call_rcu_tasks(struct rcu_head *head, void (*func)(struct rcu_head *head)); +void synchronize_rcu_tasks(void); +void rcu_barrier_tasks(void); #ifdef CONFIG_PREEMPT_RCU diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 19b3dacb0753..5fd1ddbfcc55 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -381,6 +381,61 @@ void call_rcu_tasks(struct rcu_head *rhp, void (*func)(struct rcu_head *rhp)) } EXPORT_SYMBOL_GPL(call_rcu_tasks); +/** + * synchronize_rcu_tasks - wait until an rcu-tasks grace period has elapsed. + * + * Control will return to the caller some time after a full rcu-tasks + * grace period has elapsed, in other words after all currently + * executing rcu-tasks read-side critical sections have elapsed. These + * read-side critical sections are delimited by calls to schedule(), + * cond_resched_rcu_qs(), idle execution, userspace execution, calls + * to synchronize_rcu_tasks(), and (in theory, anyway) cond_resched(). + * + * This is a very specialized primitive, intended only for a few uses in + * tracing and other situations requiring manipulation of function + * preambles and profiling hooks. The synchronize_rcu_tasks() function + * is not (yet) intended for heavy use from multiple CPUs. + * + * Note that this guarantee implies further memory-ordering guarantees. + * On systems with more than one CPU, when synchronize_rcu_tasks() returns, + * each CPU is guaranteed to have executed a full memory barrier since the + * end of its last RCU-tasks read-side critical section whose beginning + * preceded the call to synchronize_rcu_tasks(). In addition, each CPU + * having an RCU-tasks read-side critical section that extends beyond + * the return from synchronize_rcu_tasks() is guaranteed to have executed + * a full memory barrier after the beginning of synchronize_rcu_tasks() + * and before the beginning of that RCU-tasks read-side critical section. + * Note that these guarantees include CPUs that are offline, idle, or + * executing in user mode, as well as CPUs that are executing in the kernel. + * + * Furthermore, if CPU A invoked synchronize_rcu_tasks(), which returned + * to its caller on CPU B, then both CPU A and CPU B are guaranteed + * to have executed a full memory barrier during the execution of + * synchronize_rcu_tasks() -- even if CPU A and CPU B are the same CPU + * (but again only if the system has more than one CPU). + */ +void synchronize_rcu_tasks(void) +{ + /* Complain if the scheduler has not started. */ + rcu_lockdep_assert(!rcu_scheduler_active, + "synchronize_rcu_tasks called too soon"); + + /* Wait for the grace period. */ + wait_rcu_gp(call_rcu_tasks); +} + +/** + * rcu_barrier_tasks - Wait for in-flight call_rcu_tasks() callbacks. + * + * Although the current implementation is guaranteed to wait, it is not + * obligated to, for example, if there are no pending callbacks. + */ +void rcu_barrier_tasks(void) +{ + /* There is only one callback queue, so this is easy. ;-) */ + synchronize_rcu_tasks(); +} + /* See if the current task has stopped holding out, remove from list if so. */ static void check_holdout_task(struct task_struct *t) { -- cgit v1.2.3 From 3f95aa81d265223fdb13ea2b59883766a05adbdf Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 4 Aug 2014 06:10:23 -0700 Subject: rcu: Make TASKS_RCU handle tasks that are almost done exiting Once a task has passed exit_notify() in the do_exit() code path, it is no longer on the task lists, and is therefore no longer visible to rcu_tasks_kthread(). This means that an almost-exited task might be preempted while within a trampoline, and this task won't be waited on by rcu_tasks_kthread(). This commit fixes this bug by adding an srcu_struct. An exiting task does srcu_read_lock() just before calling exit_notify(), and does the corresponding srcu_read_unlock() after doing the final preempt_disable(). This means that rcu_tasks_kthread() can do synchronize_srcu() to wait for all mostly-exited tasks to reach their final preempt_disable() region, and then use synchronize_sched() to wait for those tasks to finish exiting. Reported-by: Oleg Nesterov Suggested-by: Lai Jiangshan Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 3 +++ kernel/exit.c | 3 +++ kernel/rcu/update.c | 21 +++++++++++++++++++++ 3 files changed, 27 insertions(+) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 640152fedcde..54b2ebb20313 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -321,6 +321,8 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, * macro rather than an inline function to avoid #include hell. */ #ifdef CONFIG_TASKS_RCU +#define TASKS_RCU(x) x +extern struct srcu_struct tasks_rcu_exit_srcu; #define rcu_note_voluntary_context_switch(t) \ do { \ preempt_disable(); /* Exclude synchronize_sched(); */ \ @@ -329,6 +331,7 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, preempt_enable(); \ } while (0) #else /* #ifdef CONFIG_TASKS_RCU */ +#define TASKS_RCU(x) do { } while (0) #define rcu_note_voluntary_context_switch(t) do { } while (0) #endif /* #else #ifdef CONFIG_TASKS_RCU */ diff --git a/kernel/exit.c b/kernel/exit.c index 32c58f7433a3..d13f2eec4bb8 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -667,6 +667,7 @@ void do_exit(long code) { struct task_struct *tsk = current; int group_dead; + TASKS_RCU(int tasks_rcu_i); profile_task_exit(tsk); @@ -775,6 +776,7 @@ void do_exit(long code) */ flush_ptrace_hw_breakpoint(tsk); + TASKS_RCU(tasks_rcu_i = __srcu_read_lock(&tasks_rcu_exit_srcu)); exit_notify(tsk, group_dead); proc_exit_connector(tsk); #ifdef CONFIG_NUMA @@ -814,6 +816,7 @@ void do_exit(long code) if (tsk->nr_dirtied) __this_cpu_add(dirty_throttle_leaks, tsk->nr_dirtied); exit_rcu(); + TASKS_RCU(__srcu_read_unlock(&tasks_rcu_exit_srcu, tasks_rcu_i)); /* * The setting of TASK_RUNNING by try_to_wake_up() may be delayed diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 5fd1ddbfcc55..403fc4ae539e 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -367,6 +367,13 @@ static struct rcu_head *rcu_tasks_cbs_head; static struct rcu_head **rcu_tasks_cbs_tail = &rcu_tasks_cbs_head; static DEFINE_RAW_SPINLOCK(rcu_tasks_cbs_lock); +/* Track exiting tasks in order to allow them to be waited for. */ +DEFINE_SRCU(tasks_rcu_exit_srcu); + +/* Control stall timeouts. Disable with <= 0, otherwise jiffies till stall. */ +static int rcu_task_stall_timeout __read_mostly = HZ * 60 * 3; +module_param(rcu_task_stall_timeout, int, 0644); + /* Post an RCU-tasks callback. */ void call_rcu_tasks(struct rcu_head *rhp, void (*func)(struct rcu_head *rhp)) { @@ -517,6 +524,15 @@ static int __noreturn rcu_tasks_kthread(void *arg) } rcu_read_unlock(); + /* + * Wait for tasks that are in the process of exiting. + * This does only part of the job, ensuring that all + * tasks that were previously exiting reach the point + * where they have disabled preemption, allowing the + * later synchronize_sched() to finish the job. + */ + synchronize_srcu(&tasks_rcu_exit_srcu); + /* * Each pass through the following loop scans the list * of holdout tasks, removing any that are no longer @@ -546,6 +562,11 @@ static int __noreturn rcu_tasks_kthread(void *arg) * ->rcu_tasks_holdout accesses to be within the grace * period, avoiding the need for memory barriers for * ->rcu_tasks_holdout accesses. + * + * In addition, this synchronize_sched() waits for exiting + * tasks to complete their final preempt_disable() region + * of execution, cleaning up after the synchronize_srcu() + * above. */ synchronize_sched(); -- cgit v1.2.3 From 69c604557ce34015629b325b85ff1a4996038a3b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 1 Jul 2014 11:59:36 -0700 Subject: rcutorture: Add torture tests for RCU-tasks This commit adds torture tests for RCU-tasks. It also fixes a bug that would segfault for an RCU flavor lacking a callback-barrier function. Signed-off-by: Paul E. McKenney Reviewed-by: Josh Triplett --- include/linux/rcupdate.h | 1 + kernel/rcu/rcutorture.c | 50 +++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 50 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 54b2ebb20313..a3123f53a4ce 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -55,6 +55,7 @@ enum rcutorture_type { RCU_FLAVOR, RCU_BH_FLAVOR, RCU_SCHED_FLAVOR, + RCU_TASKS_FLAVOR, SRCU_FLAVOR, INVALID_RCU_FLAVOR }; diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 178716713e11..75b1abf78c48 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -601,6 +601,52 @@ static struct rcu_torture_ops sched_ops = { .name = "sched" }; +#ifdef CONFIG_TASKS_RCU + +/* + * Definitions for RCU-tasks torture testing. + */ + +static int tasks_torture_read_lock(void) +{ + return 0; +} + +static void tasks_torture_read_unlock(int idx) +{ +} + +static void rcu_tasks_torture_deferred_free(struct rcu_torture *p) +{ + call_rcu_tasks(&p->rtort_rcu, rcu_torture_cb); +} + +static struct rcu_torture_ops tasks_ops = { + .ttype = RCU_TASKS_FLAVOR, + .init = rcu_sync_torture_init, + .readlock = tasks_torture_read_lock, + .read_delay = rcu_read_delay, /* just reuse rcu's version. */ + .readunlock = tasks_torture_read_unlock, + .completed = rcu_no_completed, + .deferred_free = rcu_tasks_torture_deferred_free, + .sync = synchronize_rcu_tasks, + .exp_sync = synchronize_rcu_tasks, + .call = call_rcu_tasks, + .cb_barrier = rcu_barrier_tasks, + .fqs = NULL, + .stats = NULL, + .irq_capable = 1, + .name = "tasks" +}; + +#define RCUTORTURE_TASKS_OPS &tasks_ops, + +#else /* #ifdef CONFIG_TASKS_RCU */ + +#define RCUTORTURE_TASKS_OPS + +#endif /* #else #ifdef CONFIG_TASKS_RCU */ + /* * RCU torture priority-boost testing. Runs one real-time thread per * CPU for moderate bursts, repeatedly registering RCU callbacks and @@ -1295,7 +1341,8 @@ static int rcu_torture_barrier_cbs(void *arg) if (atomic_dec_and_test(&barrier_cbs_count)) wake_up(&barrier_wq); } while (!torture_must_stop()); - cur_ops->cb_barrier(); + if (cur_ops->cb_barrier != NULL) + cur_ops->cb_barrier(); destroy_rcu_head_on_stack(&rcu); torture_kthread_stopping("rcu_torture_barrier_cbs"); return 0; @@ -1534,6 +1581,7 @@ rcu_torture_init(void) int firsterr = 0; static struct rcu_torture_ops *torture_ops[] = { &rcu_ops, &rcu_bh_ops, &rcu_busted_ops, &srcu_ops, &sched_ops, + RCUTORTURE_TASKS_OPS }; if (!torture_init_begin(torture_type, verbose, &rcutorture_runnable)) -- cgit v1.2.3 From 176f8f7a52cc6d09d686f0d900abda6942a52fbb Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 4 Aug 2014 17:43:50 -0700 Subject: rcu: Make TASKS_RCU handle nohz_full= CPUs Currently TASKS_RCU would ignore a CPU running a task in nohz_full= usermode execution. There would be neither a context switch nor a scheduling-clock interrupt to tell TASKS_RCU that the task in question had passed through a quiescent state. The grace period would therefore extend indefinitely. This commit therefore makes RCU's dyntick-idle subsystem record the task_struct structure of the task that is running in dyntick-idle mode on each CPU. The TASKS_RCU grace period can then access this information and record a quiescent state on behalf of any CPU running in dyntick-idle usermode. Signed-off-by: Paul E. McKenney --- include/linux/init_task.h | 3 ++- include/linux/sched.h | 2 ++ kernel/rcu/tree.c | 2 ++ kernel/rcu/tree.h | 2 ++ kernel/rcu/tree_plugin.h | 16 ++++++++++++++++ kernel/rcu/update.c | 4 +++- 6 files changed, 27 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index dffd9258ee60..03b274873b06 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -121,7 +121,8 @@ extern struct group_info init_groups; #define INIT_TASK_RCU_TASKS(tsk) \ .rcu_tasks_holdout = false, \ .rcu_tasks_holdout_list = \ - LIST_HEAD_INIT(tsk.rcu_tasks_holdout_list), + LIST_HEAD_INIT(tsk.rcu_tasks_holdout_list), \ + .rcu_tasks_idle_cpu = -1, #else #define INIT_TASK_RCU_TASKS(tsk) #endif diff --git a/include/linux/sched.h b/include/linux/sched.h index eaacac4ae77d..ec8b34722bcc 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1274,6 +1274,7 @@ struct task_struct { unsigned long rcu_tasks_nvcsw; bool rcu_tasks_holdout; struct list_head rcu_tasks_holdout_list; + int rcu_tasks_idle_cpu; #endif /* #ifdef CONFIG_TASKS_RCU */ #if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) @@ -2020,6 +2021,7 @@ static inline void rcu_copy_process(struct task_struct *p) #ifdef CONFIG_TASKS_RCU p->rcu_tasks_holdout = false; INIT_LIST_HEAD(&p->rcu_tasks_holdout_list); + p->rcu_tasks_idle_cpu = -1; #endif /* #ifdef CONFIG_TASKS_RCU */ } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index e23dad0661e2..c880f5387b1f 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -526,6 +526,7 @@ static void rcu_eqs_enter_common(struct rcu_dynticks *rdtp, long long oldval, atomic_inc(&rdtp->dynticks); smp_mb__after_atomic(); /* Force ordering with next sojourn. */ WARN_ON_ONCE(atomic_read(&rdtp->dynticks) & 0x1); + rcu_dynticks_task_enter(); /* * It is illegal to enter an extended quiescent state while @@ -642,6 +643,7 @@ void rcu_irq_exit(void) static void rcu_eqs_exit_common(struct rcu_dynticks *rdtp, long long oldval, int user) { + rcu_dynticks_task_exit(); smp_mb__before_atomic(); /* Force ordering w/previous sojourn. */ atomic_inc(&rdtp->dynticks); /* CPUs seeing atomic_inc() must see later RCU read-side crit sects */ diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index 6a86eb7bac45..3a92000c354f 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -605,6 +605,8 @@ static void rcu_sysidle_report_gp(struct rcu_state *rsp, int isidle, static void rcu_bind_gp_kthread(void); static void rcu_sysidle_init_percpu_data(struct rcu_dynticks *rdtp); static bool rcu_nohz_full_cpu(struct rcu_state *rsp); +static void rcu_dynticks_task_enter(void); +static void rcu_dynticks_task_exit(void); #endif /* #ifndef RCU_TREE_NONCORE */ diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 7672586d3920..e466b40052a7 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -3036,3 +3036,19 @@ static void rcu_bind_gp_kthread(void) housekeeping_affine(current); #endif /* #else #ifdef CONFIG_NO_HZ_FULL_SYSIDLE */ } + +/* Record the current task on dyntick-idle entry. */ +static void rcu_dynticks_task_enter(void) +{ +#if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) + ACCESS_ONCE(current->rcu_tasks_idle_cpu) = smp_processor_id(); +#endif /* #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) */ +} + +/* Record no current task on dyntick-idle exit. */ +static void rcu_dynticks_task_exit(void) +{ +#if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) + ACCESS_ONCE(current->rcu_tasks_idle_cpu) = -1; +#endif /* #if defined(CONFIG_TASKS_RCU) && defined(CONFIG_NO_HZ_FULL) */ +} diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index e1d71741958f..2658de4a5975 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -463,7 +463,9 @@ static void check_holdout_task(struct task_struct *t, { if (!ACCESS_ONCE(t->rcu_tasks_holdout) || t->rcu_tasks_nvcsw != ACCESS_ONCE(t->nvcsw) || - !ACCESS_ONCE(t->on_rq)) { + !ACCESS_ONCE(t->on_rq) || + (IS_ENABLED(CONFIG_NO_HZ_FULL) && + !is_idle_task(t) && t->rcu_tasks_idle_cpu >= 0)) { ACCESS_ONCE(t->rcu_tasks_holdout) = false; list_del_rcu(&t->rcu_tasks_holdout_list); put_task_struct(t); -- cgit v1.2.3 From 01a81330344b09028881c953a51d1106a9e63518 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 5 Aug 2014 05:23:35 -0700 Subject: rcu: Remove redundant preempt_disable() from rcu_note_voluntary_context_switch() In theory, synchronize_sched() requires a read-side critical section to order against. In practice, preemption can be thought of as being disabled across every machine instruction, at least for those machine instructions that are not in the idle loop and not on offline CPUs. So this commit removes the redundant preempt_disable() from rcu_note_voluntary_context_switch(). Please note that the single instruction in question is the store of zero to ->rcu_tasks_holdout. The "if" is simply a performance optimization that avoids unnecessary stores. To see this, keep in mind that both the "if" condition and the store are in a quiescent state. Therefore, even if the task is preempted for a full grace period (presumably due to its having done a context switch beforehand), the store will be recording a legitimate quiescent state. Reported-by: Lai Jiangshan Signed-off-by: Paul E. McKenney Conflicts: include/linux/rcupdate.h --- include/linux/rcupdate.h | 2 -- 1 file changed, 2 deletions(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index a3123f53a4ce..132e1e34cdca 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -326,10 +326,8 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, extern struct srcu_struct tasks_rcu_exit_srcu; #define rcu_note_voluntary_context_switch(t) \ do { \ - preempt_disable(); /* Exclude synchronize_sched(); */ \ if (ACCESS_ONCE((t)->rcu_tasks_holdout)) \ ACCESS_ONCE((t)->rcu_tasks_holdout) = false; \ - preempt_enable(); \ } while (0) #else /* #ifdef CONFIG_TASKS_RCU */ #define TASKS_RCU(x) do { } while (0) -- cgit v1.2.3 From 1d082fd061884a587c490c4fc8a2056ce1e47624 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 14 Aug 2014 16:01:53 -0700 Subject: rcu: Remove local_irq_disable() in rcu_preempt_note_context_switch() The rcu_preempt_note_context_switch() function is on a scheduling fast path, so it would be good to avoid disabling irqs. The reason that irqs are disabled is to synchronize process-level and irq-handler access to the task_struct ->rcu_read_unlock_special bitmask. This commit therefore makes ->rcu_read_unlock_special instead be a union of bools with a short allowing single-access checks in RCU's __rcu_read_unlock(). This results in the process-level and irq-handler accesses being simple loads and stores, so that irqs need no longer be disabled. This commit therefore removes the irq disabling from rcu_preempt_note_context_switch(). Reported-by: Peter Zijlstra Signed-off-by: Paul E. McKenney --- include/linux/init_task.h | 2 +- include/linux/sched.h | 16 +++++++++------- kernel/rcu/tree_plugin.h | 32 +++++++++++++++----------------- kernel/rcu/update.c | 2 +- 4 files changed, 26 insertions(+), 26 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 03b274873b06..77fc43f8fb72 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -111,7 +111,7 @@ extern struct group_info init_groups; #ifdef CONFIG_PREEMPT_RCU #define INIT_TASK_RCU_PREEMPT(tsk) \ .rcu_read_lock_nesting = 0, \ - .rcu_read_unlock_special = 0, \ + .rcu_read_unlock_special.s = 0, \ .rcu_node_entry = LIST_HEAD_INIT(tsk.rcu_node_entry), \ INIT_TASK_RCU_TREE_PREEMPT() #else diff --git a/include/linux/sched.h b/include/linux/sched.h index ec8b34722bcc..42888d715fb1 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1212,6 +1212,13 @@ struct sched_dl_entity { struct hrtimer dl_timer; }; +union rcu_special { + struct { + bool blocked; + bool need_qs; + } b; + short s; +}; struct rcu_node; enum perf_event_task_context { @@ -1264,7 +1271,7 @@ struct task_struct { #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; - char rcu_read_unlock_special; + union rcu_special rcu_read_unlock_special; struct list_head rcu_node_entry; #endif /* #ifdef CONFIG_PREEMPT_RCU */ #ifdef CONFIG_TREE_PREEMPT_RCU @@ -2005,16 +2012,11 @@ extern void task_clear_jobctl_trapping(struct task_struct *task); extern void task_clear_jobctl_pending(struct task_struct *task, unsigned int mask); -#ifdef CONFIG_PREEMPT_RCU -#define RCU_READ_UNLOCK_BLOCKED (1 << 0) /* blocked while in RCU read-side. */ -#define RCU_READ_UNLOCK_NEED_QS (1 << 1) /* RCU core needs CPU response. */ -#endif /* #ifdef CONFIG_PREEMPT_RCU */ - static inline void rcu_copy_process(struct task_struct *p) { #ifdef CONFIG_PREEMPT_RCU p->rcu_read_lock_nesting = 0; - p->rcu_read_unlock_special = 0; + p->rcu_read_unlock_special.s = 0; p->rcu_blocked_node = NULL; INIT_LIST_HEAD(&p->rcu_node_entry); #endif /* #ifdef CONFIG_PREEMPT_RCU */ diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index e466b40052a7..0981c0cd70fe 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -155,9 +155,8 @@ EXPORT_SYMBOL_GPL(rcu_batches_completed); * not in a quiescent state. There might be any number of tasks blocked * while in an RCU read-side critical section. * - * Unlike the other rcu_*_qs() functions, callers to this function - * must disable irqs in order to protect the assignment to - * ->rcu_read_unlock_special. + * As with the other rcu_*_qs() functions, callers to this function + * must disable preemption. */ static void rcu_preempt_qs(int cpu) { @@ -166,7 +165,7 @@ static void rcu_preempt_qs(int cpu) if (rdp->passed_quiesce == 0) trace_rcu_grace_period(TPS("rcu_preempt"), rdp->gpnum, TPS("cpuqs")); rdp->passed_quiesce = 1; - current->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_NEED_QS; + current->rcu_read_unlock_special.b.need_qs = false; } /* @@ -190,14 +189,14 @@ static void rcu_preempt_note_context_switch(int cpu) struct rcu_node *rnp; if (t->rcu_read_lock_nesting > 0 && - (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) { + !t->rcu_read_unlock_special.b.blocked) { /* Possibly blocking in an RCU read-side critical section. */ rdp = per_cpu_ptr(rcu_preempt_state.rda, cpu); rnp = rdp->mynode; raw_spin_lock_irqsave(&rnp->lock, flags); smp_mb__after_unlock_lock(); - t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BLOCKED; + t->rcu_read_unlock_special.b.blocked = true; t->rcu_blocked_node = rnp; /* @@ -239,7 +238,7 @@ static void rcu_preempt_note_context_switch(int cpu) : rnp->gpnum + 1); raw_spin_unlock_irqrestore(&rnp->lock, flags); } else if (t->rcu_read_lock_nesting < 0 && - t->rcu_read_unlock_special) { + t->rcu_read_unlock_special.s) { /* * Complete exit from RCU read-side critical section on @@ -257,9 +256,7 @@ static void rcu_preempt_note_context_switch(int cpu) * grace period, then the fact that the task has been enqueued * means that we continue to block the current grace period. */ - local_irq_save(flags); rcu_preempt_qs(cpu); - local_irq_restore(flags); } /* @@ -340,7 +337,7 @@ void rcu_read_unlock_special(struct task_struct *t) bool drop_boost_mutex = false; #endif /* #ifdef CONFIG_RCU_BOOST */ struct rcu_node *rnp; - int special; + union rcu_special special; /* NMI handlers cannot block and cannot safely manipulate state. */ if (in_nmi()) @@ -350,12 +347,13 @@ void rcu_read_unlock_special(struct task_struct *t) /* * If RCU core is waiting for this CPU to exit critical section, - * let it know that we have done so. + * let it know that we have done so. Because irqs are disabled, + * t->rcu_read_unlock_special cannot change. */ special = t->rcu_read_unlock_special; - if (special & RCU_READ_UNLOCK_NEED_QS) { + if (special.b.need_qs) { rcu_preempt_qs(smp_processor_id()); - if (!t->rcu_read_unlock_special) { + if (!t->rcu_read_unlock_special.s) { local_irq_restore(flags); return; } @@ -368,8 +366,8 @@ void rcu_read_unlock_special(struct task_struct *t) } /* Clean up if blocked during RCU read-side critical section. */ - if (special & RCU_READ_UNLOCK_BLOCKED) { - t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BLOCKED; + if (special.b.blocked) { + t->rcu_read_unlock_special.b.blocked = false; /* * Remove this task from the list it blocked on. The @@ -658,7 +656,7 @@ static void rcu_preempt_check_callbacks(int cpu) } if (t->rcu_read_lock_nesting > 0 && per_cpu(rcu_preempt_data, cpu).qs_pending) - t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; + t->rcu_read_unlock_special.b.need_qs = true; } #ifdef CONFIG_RCU_BOOST @@ -941,7 +939,7 @@ void exit_rcu(void) return; t->rcu_read_lock_nesting = 1; barrier(); - t->rcu_read_unlock_special = RCU_READ_UNLOCK_BLOCKED; + t->rcu_read_unlock_special.b.blocked = true; __rcu_read_unlock(); } diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index 9487b4898e51..6fb911558562 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -93,7 +93,7 @@ void __rcu_read_unlock(void) barrier(); /* critical section before exit code. */ t->rcu_read_lock_nesting = INT_MIN; barrier(); /* assign before ->rcu_read_unlock_special load */ - if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) + if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special.s))) rcu_read_unlock_special(t); barrier(); /* ->rcu_read_unlock_special load before assign */ t->rcu_read_lock_nesting = 0; -- cgit v1.2.3 From 284a8c93af47306beed967a303d84730b32bab39 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 14 Aug 2014 16:38:46 -0700 Subject: rcu: Per-CPU operation cleanups to rcu_*_qs() functions The rcu_bh_qs(), rcu_preempt_qs(), and rcu_sched_qs() functions use old-style per-CPU variable access and write to ->passed_quiesce even if it is already set. This commit therefore updates to use the new-style per-CPU variable access functions and avoids the spurious writes. This commit also eliminates the "cpu" argument to these functions because they are always invoked on the indicated CPU. Reported-by: Peter Zijlstra Signed-off-by: Paul E. McKenney --- include/linux/rcupdate.h | 4 ++-- include/linux/rcutiny.h | 2 +- kernel/rcu/tiny.c | 10 +++++----- kernel/rcu/tree.c | 34 ++++++++++++++++++---------------- kernel/rcu/tree_plugin.h | 27 +++++++++++++++------------ kernel/softirq.c | 2 +- 6 files changed, 42 insertions(+), 37 deletions(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 132e1e34cdca..2fab0e37afe0 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -261,8 +261,8 @@ static inline int rcu_preempt_depth(void) /* Internal to kernel */ void rcu_init(void); -void rcu_sched_qs(int cpu); -void rcu_bh_qs(int cpu); +void rcu_sched_qs(void); +void rcu_bh_qs(void); void rcu_check_callbacks(int cpu, int user); struct notifier_block; void rcu_idle_enter(void); diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index d40a6a451330..38cc5b1e252d 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -80,7 +80,7 @@ static inline void kfree_call_rcu(struct rcu_head *head, static inline void rcu_note_context_switch(int cpu) { - rcu_sched_qs(cpu); + rcu_sched_qs(); } /* diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c index 717f00854fc0..61b8d2ccc2cb 100644 --- a/kernel/rcu/tiny.c +++ b/kernel/rcu/tiny.c @@ -72,7 +72,7 @@ static void rcu_idle_enter_common(long long newval) current->pid, current->comm, idle->pid, idle->comm); /* must be idle task! */ } - rcu_sched_qs(0); /* implies rcu_bh_qsctr_inc(0) */ + rcu_sched_qs(); /* implies rcu_bh_inc() */ barrier(); rcu_dynticks_nesting = newval; } @@ -217,7 +217,7 @@ static int rcu_qsctr_help(struct rcu_ctrlblk *rcp) * are at it, given that any rcu quiescent state is also an rcu_bh * quiescent state. Use "+" instead of "||" to defeat short circuiting. */ -void rcu_sched_qs(int cpu) +void rcu_sched_qs(void) { unsigned long flags; @@ -231,7 +231,7 @@ void rcu_sched_qs(int cpu) /* * Record an rcu_bh quiescent state. */ -void rcu_bh_qs(int cpu) +void rcu_bh_qs(void) { unsigned long flags; @@ -251,9 +251,9 @@ void rcu_check_callbacks(int cpu, int user) { RCU_TRACE(check_cpu_stalls()); if (user || rcu_is_cpu_rrupt_from_idle()) - rcu_sched_qs(cpu); + rcu_sched_qs(); else if (!in_softirq()) - rcu_bh_qs(cpu); + rcu_bh_qs(); if (user) rcu_note_voluntary_context_switch(current); } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index c880f5387b1f..4c340625ffd4 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -188,22 +188,24 @@ static int rcu_gp_in_progress(struct rcu_state *rsp) * one since the start of the grace period, this just sets a flag. * The caller must have disabled preemption. */ -void rcu_sched_qs(int cpu) +void rcu_sched_qs(void) { - struct rcu_data *rdp = &per_cpu(rcu_sched_data, cpu); - - if (rdp->passed_quiesce == 0) - trace_rcu_grace_period(TPS("rcu_sched"), rdp->gpnum, TPS("cpuqs")); - rdp->passed_quiesce = 1; + if (!__this_cpu_read(rcu_sched_data.passed_quiesce)) { + trace_rcu_grace_period(TPS("rcu_sched"), + __this_cpu_read(rcu_sched_data.gpnum), + TPS("cpuqs")); + __this_cpu_write(rcu_sched_data.passed_quiesce, 1); + } } -void rcu_bh_qs(int cpu) +void rcu_bh_qs(void) { - struct rcu_data *rdp = &per_cpu(rcu_bh_data, cpu); - - if (rdp->passed_quiesce == 0) - trace_rcu_grace_period(TPS("rcu_bh"), rdp->gpnum, TPS("cpuqs")); - rdp->passed_quiesce = 1; + if (!__this_cpu_read(rcu_bh_data.passed_quiesce)) { + trace_rcu_grace_period(TPS("rcu_bh"), + __this_cpu_read(rcu_bh_data.gpnum), + TPS("cpuqs")); + __this_cpu_write(rcu_bh_data.passed_quiesce, 1); + } } static DEFINE_PER_CPU(int, rcu_sched_qs_mask); @@ -278,7 +280,7 @@ static void rcu_momentary_dyntick_idle(void) void rcu_note_context_switch(int cpu) { trace_rcu_utilization(TPS("Start context switch")); - rcu_sched_qs(cpu); + rcu_sched_qs(); rcu_preempt_note_context_switch(cpu); if (unlikely(raw_cpu_read(rcu_sched_qs_mask))) rcu_momentary_dyntick_idle(); @@ -2395,8 +2397,8 @@ void rcu_check_callbacks(int cpu, int user) * at least not while the corresponding CPU is online. */ - rcu_sched_qs(cpu); - rcu_bh_qs(cpu); + rcu_sched_qs(); + rcu_bh_qs(); } else if (!in_softirq()) { @@ -2407,7 +2409,7 @@ void rcu_check_callbacks(int cpu, int user) * critical section, so note it. */ - rcu_bh_qs(cpu); + rcu_bh_qs(); } rcu_preempt_check_callbacks(cpu); if (rcu_pending(cpu)) diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 0981c0cd70fe..25e692a36280 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -158,14 +158,16 @@ EXPORT_SYMBOL_GPL(rcu_batches_completed); * As with the other rcu_*_qs() functions, callers to this function * must disable preemption. */ -static void rcu_preempt_qs(int cpu) -{ - struct rcu_data *rdp = &per_cpu(rcu_preempt_data, cpu); - - if (rdp->passed_quiesce == 0) - trace_rcu_grace_period(TPS("rcu_preempt"), rdp->gpnum, TPS("cpuqs")); - rdp->passed_quiesce = 1; - current->rcu_read_unlock_special.b.need_qs = false; +static void rcu_preempt_qs(void) +{ + if (!__this_cpu_read(rcu_preempt_data.passed_quiesce)) { + trace_rcu_grace_period(TPS("rcu_preempt"), + __this_cpu_read(rcu_preempt_data.gpnum), + TPS("cpuqs")); + __this_cpu_write(rcu_preempt_data.passed_quiesce, 1); + barrier(); /* Coordinate with rcu_preempt_check_callbacks(). */ + current->rcu_read_unlock_special.b.need_qs = false; + } } /* @@ -256,7 +258,7 @@ static void rcu_preempt_note_context_switch(int cpu) * grace period, then the fact that the task has been enqueued * means that we continue to block the current grace period. */ - rcu_preempt_qs(cpu); + rcu_preempt_qs(); } /* @@ -352,7 +354,7 @@ void rcu_read_unlock_special(struct task_struct *t) */ special = t->rcu_read_unlock_special; if (special.b.need_qs) { - rcu_preempt_qs(smp_processor_id()); + rcu_preempt_qs(); if (!t->rcu_read_unlock_special.s) { local_irq_restore(flags); return; @@ -651,11 +653,12 @@ static void rcu_preempt_check_callbacks(int cpu) struct task_struct *t = current; if (t->rcu_read_lock_nesting == 0) { - rcu_preempt_qs(cpu); + rcu_preempt_qs(); return; } if (t->rcu_read_lock_nesting > 0 && - per_cpu(rcu_preempt_data, cpu).qs_pending) + per_cpu(rcu_preempt_data, cpu).qs_pending && + !per_cpu(rcu_preempt_data, cpu).passed_quiesce) t->rcu_read_unlock_special.b.need_qs = true; } diff --git a/kernel/softirq.c b/kernel/softirq.c index 5918d227730f..348ec763b104 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -278,7 +278,7 @@ restart: pending >>= softirq_bit; } - rcu_bh_qs(smp_processor_id()); + rcu_bh_qs(); local_irq_disable(); pending = local_softirq_pending(); -- cgit v1.2.3 From f4579fc57cf4244057b713b1f73f4dc9f0b11e97 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 25 Jul 2014 11:21:47 -0700 Subject: rcu: Fix attempt to avoid unsolicited offloading of callbacks Commit b58cc46c5f6b (rcu: Don't offload callbacks unless specifically requested) failed to adjust the callback lists of the CPUs that are known to be no-CBs CPUs only because they are also nohz_full= CPUs. This failure can result in callbacks that are posted during early boot getting stranded on nxtlist for CPUs whose no-CBs property becomes apparent late, and there can also be spurious warnings about offline CPUs posting callbacks. This commit fixes these problems by adding an early-boot rcu_init_nohz() that properly initializes the no-CBs CPUs. Note that kernels built with CONFIG_RCU_NOCB_CPU_ALL=y or with CONFIG_RCU_NOCB_CPU=n do not exhibit this bug. Neither do kernels booted without the nohz_full= boot parameter. Signed-off-by: Paul E. McKenney Reviewed-by: Pranith Kumar Tested-by: Paul Gortmaker --- include/linux/rcupdate.h | 8 +++++ init/Kconfig | 4 +-- init/main.c | 1 + kernel/rcu/tree_plugin.h | 92 ++++++++++++++++++++++++++++++++---------------- 4 files changed, 72 insertions(+), 33 deletions(-) (limited to 'include') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index d231aa17b1d7..cc7bed1c90dc 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -269,6 +269,14 @@ static inline void rcu_user_hooks_switch(struct task_struct *prev, struct task_struct *next) { } #endif /* CONFIG_RCU_USER_QS */ +#ifdef CONFIG_RCU_NOCB_CPU +void rcu_init_nohz(void); +#else /* #ifdef CONFIG_RCU_NOCB_CPU */ +static inline void rcu_init_nohz(void) +{ +} +#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */ + /** * RCU_NONIDLE - Indicate idle-loop code that needs RCU readers * @a: Code that RCU needs to pay attention to. diff --git a/init/Kconfig b/init/Kconfig index e84c6423a2e5..64ee4d967786 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -737,7 +737,7 @@ choice config RCU_NOCB_CPU_NONE bool "No build_forced no-CBs CPUs" - depends on RCU_NOCB_CPU && !NO_HZ_FULL_ALL + depends on RCU_NOCB_CPU help This option does not force any of the CPUs to be no-CBs CPUs. Only CPUs designated by the rcu_nocbs= boot parameter will be @@ -751,7 +751,7 @@ config RCU_NOCB_CPU_NONE config RCU_NOCB_CPU_ZERO bool "CPU 0 is a build_forced no-CBs CPU" - depends on RCU_NOCB_CPU && !NO_HZ_FULL_ALL + depends on RCU_NOCB_CPU help This option forces CPU 0 to be a no-CBs CPU, so that its RCU callbacks are invoked by a per-CPU kthread whose name begins diff --git a/init/main.c b/init/main.c index bb1aed928f21..e3c4cdd94d5b 100644 --- a/init/main.c +++ b/init/main.c @@ -578,6 +578,7 @@ asmlinkage __visible void __init start_kernel(void) idr_init_cache(); rcu_init(); tick_nohz_init(); + rcu_init_nohz(); context_tracking_init(); radix_tree_init(); /* init some links before init_ISA_irqs() */ diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index a7997e272564..06d077ccf8d5 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -85,33 +85,6 @@ static void __init rcu_bootup_announce_oddness(void) pr_info("\tBoot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); if (nr_cpu_ids != NR_CPUS) pr_info("\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids); -#ifdef CONFIG_RCU_NOCB_CPU -#ifndef CONFIG_RCU_NOCB_CPU_NONE - if (!have_rcu_nocb_mask) { - zalloc_cpumask_var(&rcu_nocb_mask, GFP_KERNEL); - have_rcu_nocb_mask = true; - } -#ifdef CONFIG_RCU_NOCB_CPU_ZERO - pr_info("\tOffload RCU callbacks from CPU 0\n"); - cpumask_set_cpu(0, rcu_nocb_mask); -#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ -#ifdef CONFIG_RCU_NOCB_CPU_ALL - pr_info("\tOffload RCU callbacks from all CPUs\n"); - cpumask_copy(rcu_nocb_mask, cpu_possible_mask); -#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ -#endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ - if (have_rcu_nocb_mask) { - if (!cpumask_subset(rcu_nocb_mask, cpu_possible_mask)) { - pr_info("\tNote: kernel parameter 'rcu_nocbs=' contains nonexistent CPUs.\n"); - cpumask_and(rcu_nocb_mask, cpu_possible_mask, - rcu_nocb_mask); - } - cpulist_scnprintf(nocb_buf, sizeof(nocb_buf), rcu_nocb_mask); - pr_info("\tOffload RCU callbacks from CPUs: %s.\n", nocb_buf); - if (rcu_nocb_poll) - pr_info("\tPoll for callbacks from no-CBs CPUs.\n"); - } -#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ } #ifdef CONFIG_TREE_PREEMPT_RCU @@ -2451,6 +2424,67 @@ static void do_nocb_deferred_wakeup(struct rcu_data *rdp) trace_rcu_nocb_wake(rdp->rsp->name, rdp->cpu, TPS("DeferredWakeEmpty")); } +void __init rcu_init_nohz(void) +{ + int cpu; + bool need_rcu_nocb_mask = true; + struct rcu_state *rsp; + +#ifdef CONFIG_RCU_NOCB_CPU_NONE + need_rcu_nocb_mask = false; +#endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ + +#if defined(CONFIG_NO_HZ_FULL) + if (tick_nohz_full_running && cpumask_weight(tick_nohz_full_mask)) + need_rcu_nocb_mask = true; +#endif /* #if defined(CONFIG_NO_HZ_FULL) */ + + if (!have_rcu_nocb_mask && need_rcu_nocb_mask) { + zalloc_cpumask_var(&rcu_nocb_mask, GFP_KERNEL); + have_rcu_nocb_mask = true; + } + if (!have_rcu_nocb_mask) + return; + +#ifdef CONFIG_RCU_NOCB_CPU_ZERO + pr_info("\tOffload RCU callbacks from CPU 0\n"); + cpumask_set_cpu(0, rcu_nocb_mask); +#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ +#ifdef CONFIG_RCU_NOCB_CPU_ALL + pr_info("\tOffload RCU callbacks from all CPUs\n"); + cpumask_copy(rcu_nocb_mask, cpu_possible_mask); +#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ +#if defined(CONFIG_NO_HZ_FULL) + if (tick_nohz_full_running) + cpumask_or(rcu_nocb_mask, rcu_nocb_mask, tick_nohz_full_mask); +#endif /* #if defined(CONFIG_NO_HZ_FULL) */ + + if (!cpumask_subset(rcu_nocb_mask, cpu_possible_mask)) { + pr_info("\tNote: kernel parameter 'rcu_nocbs=' contains nonexistent CPUs.\n"); + cpumask_and(rcu_nocb_mask, cpu_possible_mask, + rcu_nocb_mask); + } + cpulist_scnprintf(nocb_buf, sizeof(nocb_buf), rcu_nocb_mask); + pr_info("\tOffload RCU callbacks from CPUs: %s.\n", nocb_buf); + if (rcu_nocb_poll) + pr_info("\tPoll for callbacks from no-CBs CPUs.\n"); + + for_each_rcu_flavor(rsp) { + for_each_cpu(cpu, rcu_nocb_mask) { + struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); + + /* + * If there are early callbacks, they will need + * to be moved to the nocb lists. + */ + WARN_ON_ONCE(rdp->nxttail[RCU_NEXT_TAIL] != + &rdp->nxtlist && + rdp->nxttail[RCU_NEXT_TAIL] != NULL); + init_nocb_callback_list(rdp); + } + } +} + /* Initialize per-rcu_data variables for no-CBs CPUs. */ static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp) { @@ -2479,10 +2513,6 @@ static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp) if (rcu_nocb_mask == NULL) return; -#if defined(CONFIG_NO_HZ_FULL) && !defined(CONFIG_NO_HZ_FULL_ALL) - if (tick_nohz_full_running) - cpumask_or(rcu_nocb_mask, rcu_nocb_mask, tick_nohz_full_mask); -#endif /* #if defined(CONFIG_NO_HZ_FULL) && !defined(CONFIG_NO_HZ_FULL_ALL) */ if (ls == -1) { ls = int_sqrt(nr_cpu_ids); rcu_nocb_leader_stride = ls; -- cgit v1.2.3 From d36a7a0d5e8b5bff1671723d733eb61621b0cee4 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Thu, 11 Sep 2014 20:40:21 -0700 Subject: torture: Address race in module cleanup When performing module cleanups by calling torture_cleanup() the 'torture_type' string in nullified However, callers are not necessarily done, and might still need to reference the variable. This impacts both rcutorture and locktorture, causing printing things like: [ 94.226618] (null)-torture: Stopping lock_torture_writer task [ 94.226624] (null)-torture: Stopping lock_torture_stats task Thus delay this operation until the very end of the cleanup process. The consequence (which shouldn't matter for this kid of program) is, of course, that we delay the window between rmmod and modprobing, for instance in module_torture_begin(). Signed-off-by: Davidlohr Bueso Signed-off-by: Paul E. McKenney --- include/linux/torture.h | 3 ++- kernel/locking/locktorture.c | 3 ++- kernel/rcu/rcutorture.c | 3 ++- kernel/torture.c | 16 +++++++++++++--- 4 files changed, 19 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/torture.h b/include/linux/torture.h index fec46f8c08eb..7759fc3c622d 100644 --- a/include/linux/torture.h +++ b/include/linux/torture.h @@ -77,7 +77,8 @@ int torture_stutter_init(int s); /* Initialization and cleanup. */ bool torture_init_begin(char *ttype, bool v, int *runnable); void torture_init_end(void); -bool torture_cleanup(void); +bool torture_cleanup_begin(void); +void torture_cleanup_end(void); bool torture_must_stop(void); bool torture_must_stop_irq(void); void torture_kthread_stopping(char *title); diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c index de703a769c1d..988267cc92c1 100644 --- a/kernel/locking/locktorture.c +++ b/kernel/locking/locktorture.c @@ -361,7 +361,7 @@ static void lock_torture_cleanup(void) { int i; - if (torture_cleanup()) + if (torture_cleanup_begin()) return; if (writer_tasks) { @@ -384,6 +384,7 @@ static void lock_torture_cleanup(void) else lock_torture_print_module_parms(cur_ops, "End of test: SUCCESS"); + torture_cleanup_end(); } static int __init lock_torture_init(void) diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 6d1509500d2b..04c4b5afb759 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -1514,7 +1514,7 @@ rcu_torture_cleanup(void) int i; rcutorture_record_test_transition(); - if (torture_cleanup()) { + if (torture_cleanup_begin()) { if (cur_ops->cb_barrier != NULL) cur_ops->cb_barrier(); return; @@ -1566,6 +1566,7 @@ rcu_torture_cleanup(void) "End of test: RCU_HOTPLUG"); else rcu_torture_print_module_parms(cur_ops, "End of test: SUCCESS"); + torture_cleanup_end(); } #ifdef CONFIG_DEBUG_OBJECTS_RCU_HEAD diff --git a/kernel/torture.c b/kernel/torture.c index ede8b25ec1ae..dd70993c266c 100644 --- a/kernel/torture.c +++ b/kernel/torture.c @@ -633,8 +633,13 @@ EXPORT_SYMBOL_GPL(torture_init_end); * * This must be called before the caller starts shutting down its own * kthreads. + * + * Both torture_cleanup_begin() and torture_cleanup_end() must be paired, + * in order to correctly perform the cleanup. They are separated because + * threads can still need to reference the torture_type type, thus nullify + * only after completing all other relevant calls. */ -bool torture_cleanup(void) +bool torture_cleanup_begin(void) { mutex_lock(&fullstop_mutex); if (ACCESS_ONCE(fullstop) == FULLSTOP_SHUTDOWN) { @@ -649,12 +654,17 @@ bool torture_cleanup(void) torture_shuffle_cleanup(); torture_stutter_cleanup(); torture_onoff_cleanup(); + return false; +} +EXPORT_SYMBOL_GPL(torture_cleanup_begin); + +void torture_cleanup_end(void) +{ mutex_lock(&fullstop_mutex); torture_type = NULL; mutex_unlock(&fullstop_mutex); - return false; } -EXPORT_SYMBOL_GPL(torture_cleanup); +EXPORT_SYMBOL_GPL(torture_cleanup_end); /* * Is it time for the current torture test to stop? -- cgit v1.2.3 From 59da22a02032cf1a069ec431f93d403b321ff6b4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 12 Sep 2014 10:36:15 -0700 Subject: rcutorture: Rename rcutorture_runnable parameter This commit changes rcutorture_runnable to torture_runnable, which is consistent with the names of the other parameters and is a bit shorter as well. Signed-off-by: Paul E. McKenney --- Documentation/kernel-parameters.txt | 2 +- include/linux/rcupdate.h | 3 --- kernel/rcu/rcutorture.c | 8 ++++---- kernel/sysctl.c | 9 --------- tools/testing/selftests/rcutorture/configs/rcu/ver_functions.sh | 2 +- 5 files changed, 6 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e1147bc62633..7aba744afcde 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2938,7 +2938,7 @@ bytes respectively. Such letter suffixes can also be entirely omitted. Set time (s) between CPU-hotplug operations, or zero to disable CPU-hotplug testing. - rcutorture.rcutorture_runnable= [BOOT] + rcutorture.torture_runnable= [BOOT] Start rcutorture running at boot time. rcutorture.shuffle_interval= [KNL] diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 5cafd60c1ee4..a4a819ffb2d1 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -47,9 +47,6 @@ #include extern int rcu_expedited; /* for sysctl */ -#ifdef CONFIG_RCU_TORTURE_TEST -extern int rcutorture_runnable; /* for sysctl */ -#endif /* #ifdef CONFIG_RCU_TORTURE_TEST */ enum rcutorture_type { RCU_FLAVOR, diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 04c4b5afb759..240fa9094f83 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -168,9 +168,9 @@ static int rcu_torture_writer_state; #else #define RCUTORTURE_RUNNABLE_INIT 0 #endif -int rcutorture_runnable = RCUTORTURE_RUNNABLE_INIT; -module_param(rcutorture_runnable, int, 0444); -MODULE_PARM_DESC(rcutorture_runnable, "Start rcutorture at boot"); +static int torture_runnable = RCUTORTURE_RUNNABLE_INIT; +module_param(torture_runnable, int, 0444); +MODULE_PARM_DESC(torture_runnable, "Start rcutorture at boot"); #if defined(CONFIG_RCU_BOOST) && !defined(CONFIG_HOTPLUG_CPU) #define rcu_can_boost() 1 @@ -1636,7 +1636,7 @@ rcu_torture_init(void) RCUTORTURE_TASKS_OPS }; - if (!torture_init_begin(torture_type, verbose, &rcutorture_runnable)) + if (!torture_init_begin(torture_type, verbose, &torture_runnable)) return -EBUSY; /* Process args and tell the world that the torturer is on the job. */ diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 75875a741b5e..ab456664609d 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1055,15 +1055,6 @@ static struct ctl_table kern_table[] = { .child = key_sysctls, }, #endif -#ifdef CONFIG_RCU_TORTURE_TEST - { - .procname = "rcutorture_runnable", - .data = &rcutorture_runnable, - .maxlen = sizeof(int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, -#endif #ifdef CONFIG_PERF_EVENTS /* * User-space scripts rely on the existence of this file diff --git a/tools/testing/selftests/rcutorture/configs/rcu/ver_functions.sh b/tools/testing/selftests/rcutorture/configs/rcu/ver_functions.sh index 8977d8d31b19..ffb85ed786fa 100644 --- a/tools/testing/selftests/rcutorture/configs/rcu/ver_functions.sh +++ b/tools/testing/selftests/rcutorture/configs/rcu/ver_functions.sh @@ -51,7 +51,7 @@ per_version_boot_params () { `rcutorture_param_n_barrier_cbs "$1"` \ rcutorture.stat_interval=15 \ rcutorture.shutdown_secs=$3 \ - rcutorture.rcutorture_runnable=1 \ + rcutorture.torture_runnable=1 \ rcutorture.test_no_idle_hz=1 \ rcutorture.verbose=1 } -- cgit v1.2.3 From dd56af42bd829c6e770ed69812bd65a04eaeb1e4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 25 Aug 2014 20:25:06 -0700 Subject: rcu: Eliminate deadlock between CPU hotplug and expedited grace periods Currently, the expedited grace-period primitives do get_online_cpus(). This greatly simplifies their implementation, but means that calls to them holding locks that are acquired by CPU-hotplug notifiers (to say nothing of calls to these primitives from CPU-hotplug notifiers) can deadlock. But this is starting to become inconvenient, as can be seen here: https://lkml.org/lkml/2014/8/5/754. The problem in this case is that some developers need to acquire a mutex from a CPU-hotplug notifier, but also need to hold it across a synchronize_rcu_expedited(). As noted above, this currently results in deadlock. This commit avoids the deadlock and retains the simplicity by creating a try_get_online_cpus(), which returns false if the get_online_cpus() reference count could not immediately be incremented. If a call to try_get_online_cpus() returns true, the expedited primitives operate as before. If a call returns false, the expedited primitives fall back to normal grace-period operations. This falling back of course results in increased grace-period latency, but only during times when CPU hotplug operations are actually in flight. The effect should therefore be negligible during normal operation. Signed-off-by: Paul E. McKenney Cc: Josh Triplett Cc: "Rafael J. Wysocki" Tested-by: Lan Tianyu --- include/linux/cpu.h | 2 ++ include/linux/lockdep.h | 1 + kernel/cpu.c | 16 +++++++++++++++- kernel/rcu/tree.c | 19 ++++++++++++------- kernel/rcu/tree_plugin.h | 11 +++++------ 5 files changed, 35 insertions(+), 14 deletions(-) (limited to 'include') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 95978ad7fcdd..b2d9a43012b2 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -213,6 +213,7 @@ extern struct bus_type cpu_subsys; extern void cpu_hotplug_begin(void); extern void cpu_hotplug_done(void); extern void get_online_cpus(void); +extern bool try_get_online_cpus(void); extern void put_online_cpus(void); extern void cpu_hotplug_disable(void); extern void cpu_hotplug_enable(void); @@ -230,6 +231,7 @@ int cpu_down(unsigned int cpu); static inline void cpu_hotplug_begin(void) {} static inline void cpu_hotplug_done(void) {} #define get_online_cpus() do { } while (0) +#define try_get_online_cpus() true #define put_online_cpus() do { } while (0) #define cpu_hotplug_disable() do { } while (0) #define cpu_hotplug_enable() do { } while (0) diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 008388f920d7..4f86465cc317 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -505,6 +505,7 @@ static inline void print_irqtrace_events(struct task_struct *curr) #define lock_map_acquire(l) lock_acquire_exclusive(l, 0, 0, NULL, _THIS_IP_) #define lock_map_acquire_read(l) lock_acquire_shared_recursive(l, 0, 0, NULL, _THIS_IP_) +#define lock_map_acquire_tryread(l) lock_acquire_shared_recursive(l, 0, 1, NULL, _THIS_IP_) #define lock_map_release(l) lock_release(l, 1, _THIS_IP_) #ifdef CONFIG_PROVE_LOCKING diff --git a/kernel/cpu.c b/kernel/cpu.c index 81e2a388a0f6..356450f09c1f 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -79,6 +79,8 @@ static struct { /* Lockdep annotations for get/put_online_cpus() and cpu_hotplug_begin/end() */ #define cpuhp_lock_acquire_read() lock_map_acquire_read(&cpu_hotplug.dep_map) +#define cpuhp_lock_acquire_tryread() \ + lock_map_acquire_tryread(&cpu_hotplug.dep_map) #define cpuhp_lock_acquire() lock_map_acquire(&cpu_hotplug.dep_map) #define cpuhp_lock_release() lock_map_release(&cpu_hotplug.dep_map) @@ -91,10 +93,22 @@ void get_online_cpus(void) mutex_lock(&cpu_hotplug.lock); cpu_hotplug.refcount++; mutex_unlock(&cpu_hotplug.lock); - } EXPORT_SYMBOL_GPL(get_online_cpus); +bool try_get_online_cpus(void) +{ + if (cpu_hotplug.active_writer == current) + return true; + if (!mutex_trylock(&cpu_hotplug.lock)) + return false; + cpuhp_lock_acquire_tryread(); + cpu_hotplug.refcount++; + mutex_unlock(&cpu_hotplug.lock); + return true; +} +EXPORT_SYMBOL_GPL(try_get_online_cpus); + void put_online_cpus(void) { if (cpu_hotplug.active_writer == current) diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index d7a3b13bc94c..133e47223095 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -2940,11 +2940,6 @@ static int synchronize_sched_expedited_cpu_stop(void *data) * restructure your code to batch your updates, and then use a single * synchronize_sched() instead. * - * Note that it is illegal to call this function while holding any lock - * that is acquired by a CPU-hotplug notifier. And yes, it is also illegal - * to call this function from a CPU-hotplug notifier. Failing to observe - * these restriction will result in deadlock. - * * This implementation can be thought of as an application of ticket * locking to RCU, with sync_sched_expedited_started and * sync_sched_expedited_done taking on the roles of the halves @@ -2994,7 +2989,12 @@ void synchronize_sched_expedited(void) */ snap = atomic_long_inc_return(&rsp->expedited_start); firstsnap = snap; - get_online_cpus(); + if (!try_get_online_cpus()) { + /* CPU hotplug operation in flight, fall back to normal GP. */ + wait_rcu_gp(call_rcu_sched); + atomic_long_inc(&rsp->expedited_normal); + return; + } WARN_ON_ONCE(cpu_is_offline(raw_smp_processor_id())); /* @@ -3041,7 +3041,12 @@ void synchronize_sched_expedited(void) * and they started after our first try, so their grace * period works for us. */ - get_online_cpus(); + if (!try_get_online_cpus()) { + /* CPU hotplug operation in flight, use normal GP. */ + wait_rcu_gp(call_rcu_sched); + atomic_long_inc(&rsp->expedited_normal); + return; + } snap = atomic_long_read(&rsp->expedited_start); smp_mb(); /* ensure read is before try_stop_cpus(). */ } diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index e2c5910546f6..387dd4599344 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -793,11 +793,6 @@ sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp) * In fact, if you are using synchronize_rcu_expedited() in a loop, * please restructure your code to batch your updates, and then Use a * single synchronize_rcu() instead. - * - * Note that it is illegal to call this function while holding any lock - * that is acquired by a CPU-hotplug notifier. And yes, it is also illegal - * to call this function from a CPU-hotplug notifier. Failing to observe - * these restriction will result in deadlock. */ void synchronize_rcu_expedited(void) { @@ -819,7 +814,11 @@ void synchronize_rcu_expedited(void) * being boosted. This simplifies the process of moving tasks * from leaf to root rcu_node structures. */ - get_online_cpus(); + if (!try_get_online_cpus()) { + /* CPU-hotplug operation in flight, fall back to normal GP. */ + wait_rcu_gp(call_rcu); + return; + } /* * Acquire lock, falling back to synchronize_rcu() if too many -- cgit v1.2.3