From ba49df4767d4fa5bbd2af3a51709fb81f94264ec Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 7 Oct 2012 09:26:13 -0700 Subject: rcu: Update RCU_FAST_NO_HZ help text The RCU_FAST_NO_HZ help text included a warning about overhead on large systems, but that issue has since been resolved. The main remaining issue with RCU_FAST_NO_HZ is increased real-time latency. This commit therefore updates the help text accordingly. Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- init/Kconfig | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index 6fdd6e339326..b63e7982c1c6 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -582,14 +582,13 @@ config RCU_FAST_NO_HZ depends on NO_HZ && SMP default n help - This option causes RCU to attempt to accelerate grace periods - in order to allow CPUs to enter dynticks-idle state more - quickly. On the other hand, this option increases the overhead - of the dynticks-idle checking, particularly on systems with - large numbers of CPUs. - - Say Y if energy efficiency is critically important, particularly - if you have relatively few CPUs. + This option causes RCU to attempt to accelerate grace periods in + order to allow CPUs to enter dynticks-idle state more quickly. + On the other hand, this option increases the overhead of the + dynticks-idle checking, thus degrading scheduling latency. + + Say Y if energy efficiency is critically important, and you don't + care about real-time response. Say N if you are unsure. -- cgit v1.2.3 From af71befa282ddf51c09509978abe1e83de8fe7eb Mon Sep 17 00:00:00 2001 From: Paul Gortmaker Date: Wed, 24 Oct 2012 11:07:09 -0700 Subject: rcu: Wordsmith help text for RCU_USER_QS kernel parameter This commit adds a "try" missing from the end of the first paragraph of the RCU_USER_QS help text. [ paulmck: Also fix up the last paragraph a bit. ] Signed-off-by: Paul Gortmaker Signed-off-by: Paul E. McKenney --- init/Kconfig | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'init') diff --git a/init/Kconfig b/init/Kconfig index b63e7982c1c6..ec62139207d3 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -494,11 +494,11 @@ config RCU_USER_QS puts RCU in extended quiescent state when the CPU runs in userspace. It means that when a CPU runs in userspace, it is excluded from the global RCU state machine and thus doesn't - to keep the timer tick on for RCU. + try to keep the timer tick on for RCU. Unless you want to hack and help the development of the full - tickless feature, you shouldn't enable this option. It adds - unnecessary overhead. + tickless feature, you shouldn't enable this option. It also + adds unnecessary overhead. If unsure say N -- cgit v1.2.3 From 3fbfbf7a3b66ec424042d909f14ba2ddf4372ea8 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sun, 19 Aug 2012 21:35:53 -0700 Subject: rcu: Add callback-free CPUs RCU callback execution can add significant OS jitter and also can degrade both scheduling latency and, in asymmetric multiprocessors, energy efficiency. This commit therefore adds the ability for selected CPUs ("rcu_nocbs=" boot parameter) to have their callbacks offloaded to kthreads. If the "rcu_nocb_poll" boot parameter is also specified, these kthreads will do polling, removing the need for the offloaded CPUs to do wakeups. At least one CPU must be doing normal callback processing: currently CPU 0 cannot be selected as a no-CBs CPU. In addition, attempts to offline the last normal-CBs CPU will fail. This feature was inspired by Jim Houston's and Joe Korty's JRCU, and this commit includes fixes to problems located by Fengguang Wu's kbuild test robot. [ paulmck: Added gfp.h include file as suggested by Fengguang Wu. ] Signed-off-by: Paul E. McKenney Signed-off-by: Paul E. McKenney --- Documentation/kernel-parameters.txt | 21 ++ include/trace/events/rcu.h | 1 + init/Kconfig | 22 ++ kernel/rcutree.c | 63 ++++-- kernel/rcutree.h | 47 +++++ kernel/rcutree_plugin.h | 397 +++++++++++++++++++++++++++++++++++- kernel/rcutree_trace.c | 7 +- 7 files changed, 542 insertions(+), 16 deletions(-) (limited to 'init') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9776f068306b..9d2e5cb3a95f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2394,6 +2394,27 @@ bytes respectively. Such letter suffixes can also be entirely omitted. ramdisk_size= [RAM] Sizes of RAM disks in kilobytes See Documentation/blockdev/ramdisk.txt. + rcu_nocbs= [KNL,BOOT] + In kernels built with CONFIG_RCU_NOCB_CPU=y, set + the specified list of CPUs to be no-callback CPUs. + Invocation of these CPUs' RCU callbacks will + be offloaded to "rcuoN" kthreads created for + that purpose. This reduces OS jitter on the + offloaded CPUs, which can be useful for HPC and + real-time workloads. It can also improve energy + efficiency for asymmetric multiprocessors. + + rcu_nocbs_poll [KNL,BOOT] + Rather than requiring that offloaded CPUs + (specified by rcu_nocbs= above) explicitly + awaken the corresponding "rcuoN" kthreads, + make these kthreads poll for callbacks. + This improves the real-time response for the + offloaded CPUs by relieving them of the need to + wake up the corresponding kthread, but degrades + energy efficiency by requiring that the kthreads + periodically wake up to do the polling. + rcutree.blimit= [KNL,BOOT] Set maximum number of finished RCU callbacks to process in one batch. diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index 5bde94d8585b..d4f559b1ec34 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -549,6 +549,7 @@ TRACE_EVENT(rcu_torture_read, * "EarlyExit": rcu_barrier_callback() piggybacked, thus early exit. * "Inc1": rcu_barrier_callback() piggyback check counter incremented. * "Offline": rcu_barrier_callback() found offline CPU + * "OnlineNoCB": rcu_barrier_callback() found online no-CBs CPU. * "OnlineQ": rcu_barrier_callback() found online CPU with callbacks. * "OnlineNQ": rcu_barrier_callback() found online CPU, no callbacks. * "IRQ": An rcu_barrier_callback() callback posted on remote CPU. diff --git a/init/Kconfig b/init/Kconfig index ec62139207d3..5ac6ee094225 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -654,6 +654,28 @@ config RCU_BOOST_DELAY Accept the default if unsure. +config RCU_NOCB_CPU + bool "Offload RCU callback processing from boot-selected CPUs" + depends on TREE_RCU || TREE_PREEMPT_RCU + default n + help + Use this option to reduce OS jitter for aggressive HPC or + real-time workloads. It can also be used to offload RCU + callback invocation to energy-efficient CPUs in battery-powered + asymmetric multiprocessors. + + This option offloads callback invocation from the set of + CPUs specified at boot time by the rcu_nocbs parameter. + For each such CPU, a kthread ("rcuoN") will be created to + invoke callbacks, where the "N" is the CPU being offloaded. + Nothing prevents this kthread from running on the specified + CPUs, but (1) the kthreads may be preempted between each + callback, and (2) affinity or cgroups can be used to force + the kthreads to run on whatever set of CPUs is desired. + + Say Y here if you want reduced OS jitter on selected CPUs. + Say N here if you are unsure. + endmenu # "RCU Subsystem" config IKCONFIG diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 5ffadcc3bb26..7733eb56e156 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -303,7 +303,8 @@ EXPORT_SYMBOL_GPL(rcu_sched_force_quiescent_state); static int cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp) { - return &rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL]; + return &rdp->nxtlist != rdp->nxttail[RCU_DONE_TAIL] && + rdp->nxttail[RCU_DONE_TAIL] != NULL; } /* @@ -312,8 +313,11 @@ cpu_has_callbacks_ready_to_invoke(struct rcu_data *rdp) static int cpu_needs_another_gp(struct rcu_state *rsp, struct rcu_data *rdp) { - return *rdp->nxttail[RCU_DONE_TAIL + - (ACCESS_ONCE(rsp->completed) != rdp->completed)] && + struct rcu_head **ntp; + + ntp = rdp->nxttail[RCU_DONE_TAIL + + (ACCESS_ONCE(rsp->completed) != rdp->completed)]; + return rdp->nxttail[RCU_DONE_TAIL] && ntp && *ntp && !rcu_gp_in_progress(rsp); } @@ -1123,6 +1127,7 @@ static void init_callback_list(struct rcu_data *rdp) rdp->nxtlist = NULL; for (i = 0; i < RCU_NEXT_SIZE; i++) rdp->nxttail[i] = &rdp->nxtlist; + init_nocb_callback_list(rdp); } /* @@ -1633,6 +1638,10 @@ static void rcu_send_cbs_to_orphanage(int cpu, struct rcu_state *rsp, struct rcu_node *rnp, struct rcu_data *rdp) { + /* No-CBs CPUs do not have orphanable callbacks. */ + if (is_nocb_cpu(rdp->cpu)) + return; + /* * Orphan the callbacks. First adjust the counts. This is safe * because _rcu_barrier() excludes CPU-hotplug operations, so it @@ -1684,6 +1693,10 @@ static void rcu_adopt_orphan_cbs(struct rcu_state *rsp) int i; struct rcu_data *rdp = __this_cpu_ptr(rsp->rda); + /* No-CBs CPUs are handled specially. */ + if (rcu_nocb_adopt_orphan_cbs(rsp, rdp)) + return; + /* Do the accounting first. */ rdp->qlen_lazy += rsp->qlen_lazy; rdp->qlen += rsp->qlen; @@ -2162,9 +2175,15 @@ static void __call_rcu_core(struct rcu_state *rsp, struct rcu_data *rdp, } } +/* + * Helper function for call_rcu() and friends. The cpu argument will + * normally be -1, indicating "currently running CPU". It may specify + * a CPU only if that CPU is a no-CBs CPU. Currently, only _rcu_barrier() + * is expected to specify a CPU. + */ static void __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), - struct rcu_state *rsp, bool lazy) + struct rcu_state *rsp, int cpu, bool lazy) { unsigned long flags; struct rcu_data *rdp; @@ -2184,9 +2203,14 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), rdp = this_cpu_ptr(rsp->rda); /* Add the callback to our list. */ - if (unlikely(rdp->nxttail[RCU_NEXT_TAIL] == NULL)) { + if (unlikely(rdp->nxttail[RCU_NEXT_TAIL] == NULL) || cpu != -1) { + int offline; + + if (cpu != -1) + rdp = per_cpu_ptr(rsp->rda, cpu); + offline = !__call_rcu_nocb(rdp, head, lazy); + WARN_ON_ONCE(offline); /* _call_rcu() is illegal on offline CPU; leak the callback. */ - WARN_ON_ONCE(1); local_irq_restore(flags); return; } @@ -2215,7 +2239,7 @@ __call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu), */ void call_rcu_sched(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_sched_state, 0); + __call_rcu(head, func, &rcu_sched_state, -1, 0); } EXPORT_SYMBOL_GPL(call_rcu_sched); @@ -2224,7 +2248,7 @@ EXPORT_SYMBOL_GPL(call_rcu_sched); */ void call_rcu_bh(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_bh_state, 0); + __call_rcu(head, func, &rcu_bh_state, -1, 0); } EXPORT_SYMBOL_GPL(call_rcu_bh); @@ -2676,9 +2700,17 @@ static void _rcu_barrier(struct rcu_state *rsp) * When that callback is invoked, we will know that all of the * corresponding CPU's preceding callbacks have been invoked. */ - for_each_online_cpu(cpu) { + for_each_possible_cpu(cpu) { + if (!cpu_online(cpu) && !is_nocb_cpu(cpu)) + continue; rdp = per_cpu_ptr(rsp->rda, cpu); - if (ACCESS_ONCE(rdp->qlen)) { + if (is_nocb_cpu(cpu)) { + _rcu_barrier_trace(rsp, "OnlineNoCB", cpu, + rsp->n_barrier_done); + atomic_inc(&rsp->barrier_cpu_count); + __call_rcu(&rdp->barrier_head, rcu_barrier_callback, + rsp, cpu, 0); + } else if (ACCESS_ONCE(rdp->qlen)) { _rcu_barrier_trace(rsp, "OnlineQ", cpu, rsp->n_barrier_done); smp_call_function_single(cpu, rcu_barrier_func, rsp, 1); @@ -2752,6 +2784,7 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) #endif rdp->cpu = cpu; rdp->rsp = rsp; + rcu_boot_init_nocb_percpu_data(rdp); raw_spin_unlock_irqrestore(&rnp->lock, flags); } @@ -2833,6 +2866,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, struct rcu_data *rdp = per_cpu_ptr(rcu_state->rda, cpu); struct rcu_node *rnp = rdp->mynode; struct rcu_state *rsp; + int ret = NOTIFY_OK; trace_rcu_utilization("Start CPU hotplug"); switch (action) { @@ -2846,7 +2880,10 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, rcu_boost_kthread_setaffinity(rnp, -1); break; case CPU_DOWN_PREPARE: - rcu_boost_kthread_setaffinity(rnp, cpu); + if (nocb_cpu_expendable(cpu)) + rcu_boost_kthread_setaffinity(rnp, cpu); + else + ret = NOTIFY_BAD; break; case CPU_DYING: case CPU_DYING_FROZEN: @@ -2870,7 +2907,7 @@ static int __cpuinit rcu_cpu_notify(struct notifier_block *self, break; } trace_rcu_utilization("End CPU hotplug"); - return NOTIFY_OK; + return ret; } /* @@ -2890,6 +2927,7 @@ static int __init rcu_spawn_gp_kthread(void) raw_spin_lock_irqsave(&rnp->lock, flags); rsp->gp_kthread = t; raw_spin_unlock_irqrestore(&rnp->lock, flags); + rcu_spawn_nocb_kthreads(rsp); } return 0; } @@ -3085,6 +3123,7 @@ void __init rcu_init(void) rcu_init_one(&rcu_sched_state, &rcu_sched_data); rcu_init_one(&rcu_bh_state, &rcu_bh_data); __rcu_init_preempt(); + rcu_init_nocb(); open_softirq(RCU_SOFTIRQ, rcu_process_callbacks); /* diff --git a/kernel/rcutree.h b/kernel/rcutree.h index d274af357210..488f2ec6b663 100644 --- a/kernel/rcutree.h +++ b/kernel/rcutree.h @@ -317,6 +317,18 @@ struct rcu_data { struct rcu_head oom_head; #endif /* #ifdef CONFIG_RCU_FAST_NO_HZ */ + /* 7) Callback offloading. */ +#ifdef CONFIG_RCU_NOCB_CPU + struct rcu_head *nocb_head; /* CBs waiting for kthread. */ + struct rcu_head **nocb_tail; + atomic_long_t nocb_q_count; /* # CBs waiting for kthread */ + atomic_long_t nocb_q_count_lazy; /* (approximate). */ + int nocb_p_count; /* # CBs being invoked by kthread */ + int nocb_p_count_lazy; /* (approximate). */ + wait_queue_head_t nocb_wq; /* For nocb kthreads to sleep on. */ + struct task_struct *nocb_kthread; +#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ + int cpu; struct rcu_state *rsp; }; @@ -369,6 +381,12 @@ struct rcu_state { struct rcu_data __percpu *rda; /* pointer of percu rcu_data. */ void (*call)(struct rcu_head *head, /* call_rcu() flavor. */ void (*func)(struct rcu_head *head)); +#ifdef CONFIG_RCU_NOCB_CPU + void (*call_remote)(struct rcu_head *head, + void (*func)(struct rcu_head *head)); + /* call_rcu() flavor, but for */ + /* placing on remote CPU. */ +#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ /* The following fields are guarded by the root rcu_node's lock. */ @@ -439,6 +457,8 @@ struct rcu_state { #define RCU_GP_FLAG_FQS 0x2 /* Need grace-period quiescent-state forcing. */ extern struct list_head rcu_struct_flavors; + +/* Sequence through rcu_state structures for each RCU flavor. */ #define for_each_rcu_flavor(rsp) \ list_for_each_entry((rsp), &rcu_struct_flavors, flavors) @@ -515,5 +535,32 @@ static void print_cpu_stall_info(struct rcu_state *rsp, int cpu); static void print_cpu_stall_info_end(void); static void zero_cpu_stall_ticks(struct rcu_data *rdp); static void increment_cpu_stall_ticks(void); +static bool is_nocb_cpu(int cpu); +static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp, + bool lazy); +static bool rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp, + struct rcu_data *rdp); +static bool nocb_cpu_expendable(int cpu); +static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp); +static void rcu_spawn_nocb_kthreads(struct rcu_state *rsp); +static void init_nocb_callback_list(struct rcu_data *rdp); +static void __init rcu_init_nocb(void); #endif /* #ifndef RCU_TREE_NONCORE */ + +#ifdef CONFIG_RCU_TRACE +#ifdef CONFIG_RCU_NOCB_CPU +/* Sum up queue lengths for tracing. */ +static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll) +{ + *ql = atomic_long_read(&rdp->nocb_q_count) + rdp->nocb_p_count; + *qll = atomic_long_read(&rdp->nocb_q_count_lazy) + rdp->nocb_p_count_lazy; +} +#else /* #ifdef CONFIG_RCU_NOCB_CPU */ +static inline void rcu_nocb_q_lengths(struct rcu_data *rdp, long *ql, long *qll) +{ + *ql = 0; + *qll = 0; +} +#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */ +#endif /* #ifdef CONFIG_RCU_TRACE */ diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 5ce3352505e9..6cdc372de34c 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -25,6 +25,7 @@ */ #include +#include #include #include @@ -36,6 +37,14 @@ #define RCU_BOOST_PRIO RCU_KTHREAD_PRIO #endif +#ifdef CONFIG_RCU_NOCB_CPU +static cpumask_var_t rcu_nocb_mask; /* CPUs to have callbacks offloaded. */ +static bool have_rcu_nocb_mask; /* Was rcu_nocb_mask allocated? */ +static bool rcu_nocb_poll; /* Offload kthread are to poll. */ +module_param(rcu_nocb_poll, bool, 0444); +static char __initdata nocb_buf[NR_CPUS * 5]; +#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ + /* * Check the RCU kernel configuration parameters and print informative * messages about anything out of the ordinary. If you like #ifdef, you @@ -76,6 +85,18 @@ static void __init rcu_bootup_announce_oddness(void) printk(KERN_INFO "\tExperimental boot-time adjustment of leaf fanout to %d.\n", rcu_fanout_leaf); if (nr_cpu_ids != NR_CPUS) printk(KERN_INFO "\tRCU restricting CPUs from NR_CPUS=%d to nr_cpu_ids=%d.\n", NR_CPUS, nr_cpu_ids); +#ifdef CONFIG_RCU_NOCB_CPU + if (have_rcu_nocb_mask) { + if (cpumask_test_cpu(0, rcu_nocb_mask)) { + cpumask_clear_cpu(0, rcu_nocb_mask); + pr_info("\tCPU 0: illegal no-CBs CPU (cleared).\n"); + } + cpulist_scnprintf(nocb_buf, sizeof(nocb_buf), rcu_nocb_mask); + pr_info("\tExperimental no-CBs CPUs: %s.\n", nocb_buf); + if (rcu_nocb_poll) + pr_info("\tExperimental polled no-CBs CPUs.\n"); + } +#endif /* #ifdef CONFIG_RCU_NOCB_CPU */ } #ifdef CONFIG_TREE_PREEMPT_RCU @@ -642,7 +663,7 @@ static void rcu_preempt_do_callbacks(void) */ void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_preempt_state, 0); + __call_rcu(head, func, &rcu_preempt_state, -1, 0); } EXPORT_SYMBOL_GPL(call_rcu); @@ -656,7 +677,7 @@ EXPORT_SYMBOL_GPL(call_rcu); void kfree_call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_preempt_state, 1); + __call_rcu(head, func, &rcu_preempt_state, -1, 1); } EXPORT_SYMBOL_GPL(kfree_call_rcu); @@ -1025,7 +1046,7 @@ static void rcu_preempt_check_callbacks(int cpu) void kfree_call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu)) { - __call_rcu(head, func, &rcu_sched_state, 1); + __call_rcu(head, func, &rcu_sched_state, -1, 1); } EXPORT_SYMBOL_GPL(kfree_call_rcu); @@ -2104,3 +2125,373 @@ static void increment_cpu_stall_ticks(void) } #endif /* #else #ifdef CONFIG_RCU_CPU_STALL_INFO */ + +#ifdef CONFIG_RCU_NOCB_CPU + +/* + * Offload callback processing from the boot-time-specified set of CPUs + * specified by rcu_nocb_mask. For each CPU in the set, there is a + * kthread created that pulls the callbacks from the corresponding CPU, + * waits for a grace period to elapse, and invokes the callbacks. + * The no-CBs CPUs do a wake_up() on their kthread when they insert + * a callback into any empty list, unless the rcu_nocb_poll boot parameter + * has been specified, in which case each kthread actively polls its + * CPU. (Which isn't so great for energy efficiency, but which does + * reduce RCU's overhead on that CPU.) + * + * This is intended to be used in conjunction with Frederic Weisbecker's + * adaptive-idle work, which would seriously reduce OS jitter on CPUs + * running CPU-bound user-mode computations. + * + * Offloading of callback processing could also in theory be used as + * an energy-efficiency measure because CPUs with no RCU callbacks + * queued are more aggressive about entering dyntick-idle mode. + */ + + +/* Parse the boot-time rcu_nocb_mask CPU list from the kernel parameters. */ +static int __init rcu_nocb_setup(char *str) +{ + alloc_bootmem_cpumask_var(&rcu_nocb_mask); + have_rcu_nocb_mask = true; + cpulist_parse(str, rcu_nocb_mask); + return 1; +} +__setup("rcu_nocbs=", rcu_nocb_setup); + +/* Is the specified CPU a no-CPUs CPU? */ +static bool is_nocb_cpu(int cpu) +{ + if (have_rcu_nocb_mask) + return cpumask_test_cpu(cpu, rcu_nocb_mask); + return false; +} + +/* + * Enqueue the specified string of rcu_head structures onto the specified + * CPU's no-CBs lists. The CPU is specified by rdp, the head of the + * string by rhp, and the tail of the string by rhtp. The non-lazy/lazy + * counts are supplied by rhcount and rhcount_lazy. + * + * If warranted, also wake up the kthread servicing this CPUs queues. + */ +static void __call_rcu_nocb_enqueue(struct rcu_data *rdp, + struct rcu_head *rhp, + struct rcu_head **rhtp, + int rhcount, int rhcount_lazy) +{ + int len; + struct rcu_head **old_rhpp; + struct task_struct *t; + + /* Enqueue the callback on the nocb list and update counts. */ + old_rhpp = xchg(&rdp->nocb_tail, rhtp); + ACCESS_ONCE(*old_rhpp) = rhp; + atomic_long_add(rhcount, &rdp->nocb_q_count); + atomic_long_add(rhcount_lazy, &rdp->nocb_q_count_lazy); + + /* If we are not being polled and there is a kthread, awaken it ... */ + t = ACCESS_ONCE(rdp->nocb_kthread); + if (rcu_nocb_poll | !t) + return; + len = atomic_long_read(&rdp->nocb_q_count); + if (old_rhpp == &rdp->nocb_head) { + wake_up(&rdp->nocb_wq); /* ... only if queue was empty ... */ + rdp->qlen_last_fqs_check = 0; + } else if (len > rdp->qlen_last_fqs_check + qhimark) { + wake_up_process(t); /* ... or if many callbacks queued. */ + rdp->qlen_last_fqs_check = LONG_MAX / 2; + } + return; +} + +/* + * This is a helper for __call_rcu(), which invokes this when the normal + * callback queue is inoperable. If this is not a no-CBs CPU, this + * function returns failure back to __call_rcu(), which can complain + * appropriately. + * + * Otherwise, this function queues the callback where the corresponding + * "rcuo" kthread can find it. + */ +static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp, + bool lazy) +{ + + if (!is_nocb_cpu(rdp->cpu)) + return 0; + __call_rcu_nocb_enqueue(rdp, rhp, &rhp->next, 1, lazy); + return 1; +} + +/* + * Adopt orphaned callbacks on a no-CBs CPU, or return 0 if this is + * not a no-CBs CPU. + */ +static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp, + struct rcu_data *rdp) +{ + long ql = rsp->qlen; + long qll = rsp->qlen_lazy; + + /* If this is not a no-CBs CPU, tell the caller to do it the old way. */ + if (!is_nocb_cpu(smp_processor_id())) + return 0; + rsp->qlen = 0; + rsp->qlen_lazy = 0; + + /* First, enqueue the donelist, if any. This preserves CB ordering. */ + if (rsp->orphan_donelist != NULL) { + __call_rcu_nocb_enqueue(rdp, rsp->orphan_donelist, + rsp->orphan_donetail, ql, qll); + ql = qll = 0; + rsp->orphan_donelist = NULL; + rsp->orphan_donetail = &rsp->orphan_donelist; + } + if (rsp->orphan_nxtlist != NULL) { + __call_rcu_nocb_enqueue(rdp, rsp->orphan_nxtlist, + rsp->orphan_nxttail, ql, qll); + ql = qll = 0; + rsp->orphan_nxtlist = NULL; + rsp->orphan_nxttail = &rsp->orphan_nxtlist; + } + return 1; +} + +/* + * There must be at least one non-no-CBs CPU in operation at any given + * time, because no-CBs CPUs are not capable of initiating grace periods + * independently. This function therefore complains if the specified + * CPU is the last non-no-CBs CPU, allowing the CPU-hotplug system to + * avoid offlining the last such CPU. (Recursion is a wonderful thing, + * but you have to have a base case!) + */ +static bool nocb_cpu_expendable(int cpu) +{ + cpumask_var_t non_nocb_cpus; + int ret; + + /* + * If there are no no-CB CPUs or if this CPU is not a no-CB CPU, + * then offlining this CPU is harmless. Let it happen. + */ + if (!have_rcu_nocb_mask || is_nocb_cpu(cpu)) + return 1; + + /* If no memory, play it safe and keep the CPU around. */ + if (!alloc_cpumask_var(&non_nocb_cpus, GFP_NOIO)) + return 0; + cpumask_andnot(non_nocb_cpus, cpu_online_mask, rcu_nocb_mask); + cpumask_clear_cpu(cpu, non_nocb_cpus); + ret = !cpumask_empty(non_nocb_cpus); + free_cpumask_var(non_nocb_cpus); + return ret; +} + +/* + * Helper structure for remote registry of RCU callbacks. + * This is needed for when a no-CBs CPU needs to start a grace period. + * If it just invokes call_rcu(), the resulting callback will be queued, + * which can result in deadlock. + */ +struct rcu_head_remote { + struct rcu_head *rhp; + call_rcu_func_t *crf; + void (*func)(struct rcu_head *rhp); +}; + +/* + * Register a callback as specified by the rcu_head_remote struct. + * This function is intended to be invoked via smp_call_function_single(). + */ +static void call_rcu_local(void *arg) +{ + struct rcu_head_remote *rhrp = + container_of(arg, struct rcu_head_remote, rhp); + + rhrp->crf(rhrp->rhp, rhrp->func); +} + +/* + * Set up an rcu_head_remote structure and the invoke call_rcu_local() + * on CPU 0 (which is guaranteed to be a non-no-CBs CPU) via + * smp_call_function_single(). + */ +static void invoke_crf_remote(struct rcu_head *rhp, + void (*func)(struct rcu_head *rhp), + call_rcu_func_t crf) +{ + struct rcu_head_remote rhr; + + rhr.rhp = rhp; + rhr.crf = crf; + rhr.func = func; + smp_call_function_single(0, call_rcu_local, &rhr, 1); +} + +/* + * Helper functions to be passed to wait_rcu_gp(), each of which + * invokes invoke_crf_remote() to register a callback appropriately. + */ +static void __maybe_unused +call_rcu_preempt_remote(struct rcu_head *rhp, + void (*func)(struct rcu_head *rhp)) +{ + invoke_crf_remote(rhp, func, call_rcu); +} +static void call_rcu_bh_remote(struct rcu_head *rhp, + void (*func)(struct rcu_head *rhp)) +{ + invoke_crf_remote(rhp, func, call_rcu_bh); +} +static void call_rcu_sched_remote(struct rcu_head *rhp, + void (*func)(struct rcu_head *rhp)) +{ + invoke_crf_remote(rhp, func, call_rcu_sched); +} + +/* + * Per-rcu_data kthread, but only for no-CBs CPUs. Each kthread invokes + * callbacks queued by the corresponding no-CBs CPU. + */ +static int rcu_nocb_kthread(void *arg) +{ + int c, cl; + struct rcu_head *list; + struct rcu_head *next; + struct rcu_head **tail; + struct rcu_data *rdp = arg; + + /* Each pass through this loop invokes one batch of callbacks */ + for (;;) { + /* If not polling, wait for next batch of callbacks. */ + if (!rcu_nocb_poll) + wait_event(rdp->nocb_wq, rdp->nocb_head); + list = ACCESS_ONCE(rdp->nocb_head); + if (!list) { + schedule_timeout_interruptible(1); + continue; + } + + /* + * Extract queued callbacks, update counts, and wait + * for a grace period to elapse. + */ + ACCESS_ONCE(rdp->nocb_head) = NULL; + tail = xchg(&rdp->nocb_tail, &rdp->nocb_head); + c = atomic_long_xchg(&rdp->nocb_q_count, 0); + cl = atomic_long_xchg(&rdp->nocb_q_count_lazy, 0); + ACCESS_ONCE(rdp->nocb_p_count) += c; + ACCESS_ONCE(rdp->nocb_p_count_lazy) += cl; + wait_rcu_gp(rdp->rsp->call_remote); + + /* Each pass through the following loop invokes a callback. */ + trace_rcu_batch_start(rdp->rsp->name, cl, c, -1); + c = cl = 0; + while (list) { + next = list->next; + /* Wait for enqueuing to complete, if needed. */ + while (next == NULL && &list->next != tail) { + schedule_timeout_interruptible(1); + next = list->next; + } + debug_rcu_head_unqueue(list); + local_bh_disable(); + if (__rcu_reclaim(rdp->rsp->name, list)) + cl++; + c++; + local_bh_enable(); + list = next; + } + trace_rcu_batch_end(rdp->rsp->name, c, !!list, 0, 0, 1); + ACCESS_ONCE(rdp->nocb_p_count) -= c; + ACCESS_ONCE(rdp->nocb_p_count_lazy) -= cl; + rdp->n_cbs_invoked += c; + } + return 0; +} + +/* Initialize per-rcu_data variables for no-CBs CPUs. */ +static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp) +{ + rdp->nocb_tail = &rdp->nocb_head; + init_waitqueue_head(&rdp->nocb_wq); +} + +/* Create a kthread for each RCU flavor for each no-CBs CPU. */ +static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp) +{ + int cpu; + struct rcu_data *rdp; + struct task_struct *t; + + if (rcu_nocb_mask == NULL) + return; + for_each_cpu(cpu, rcu_nocb_mask) { + rdp = per_cpu_ptr(rsp->rda, cpu); + t = kthread_run(rcu_nocb_kthread, rdp, "rcuo%d", cpu); + BUG_ON(IS_ERR(t)); + ACCESS_ONCE(rdp->nocb_kthread) = t; + } +} + +/* Prevent __call_rcu() from enqueuing callbacks on no-CBs CPUs */ +static void init_nocb_callback_list(struct rcu_data *rdp) +{ + if (rcu_nocb_mask == NULL || + !cpumask_test_cpu(rdp->cpu, rcu_nocb_mask)) + return; + rdp->nxttail[RCU_NEXT_TAIL] = NULL; +} + +/* Initialize the ->call_remote fields in the rcu_state structures. */ +static void __init rcu_init_nocb(void) +{ +#ifdef CONFIG_PREEMPT_RCU + rcu_preempt_state.call_remote = call_rcu_preempt_remote; +#endif /* #ifdef CONFIG_PREEMPT_RCU */ + rcu_bh_state.call_remote = call_rcu_bh_remote; + rcu_sched_state.call_remote = call_rcu_sched_remote; +} + +#else /* #ifdef CONFIG_RCU_NOCB_CPU */ + +static bool is_nocb_cpu(int cpu) +{ + return false; +} + +static bool __call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *rhp, + bool lazy) +{ + return 0; +} + +static bool __maybe_unused rcu_nocb_adopt_orphan_cbs(struct rcu_state *rsp, + struct rcu_data *rdp) +{ + return 0; +} + +static bool nocb_cpu_expendable(int cpu) +{ + return 1; +} + +static void __init rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp) +{ +} + +static void __init rcu_spawn_nocb_kthreads(struct rcu_state *rsp) +{ +} + +static void init_nocb_callback_list(struct rcu_data *rdp) +{ +} + +static void __init rcu_init_nocb(void) +{ +} + +#endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */ diff --git a/kernel/rcutree_trace.c b/kernel/rcutree_trace.c index f9512687a6e5..3189f9aa3e84 100644 --- a/kernel/rcutree_trace.c +++ b/kernel/rcutree_trace.c @@ -113,6 +113,8 @@ static char convert_kthread_status(unsigned int kthread_status) static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) { + long ql, qll; + if (!rdp->beenonline) return; seq_printf(m, "%3d%cc=%ld g=%ld pq=%d qp=%d", @@ -126,8 +128,11 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp) rdp->dynticks->dynticks_nmi_nesting, rdp->dynticks_fqs); seq_printf(m, " of=%lu", rdp->offline_fqs); + rcu_nocb_q_lengths(rdp, &ql, &qll); + qll += rdp->qlen_lazy; + ql += rdp->qlen; seq_printf(m, " ql=%ld/%ld qs=%c%c%c%c", - rdp->qlen_lazy, rdp->qlen, + qll, ql, ".N"[rdp->nxttail[RCU_NEXT_READY_TAIL] != rdp->nxttail[RCU_NEXT_TAIL]], ".R"[rdp->nxttail[RCU_WAIT_TAIL] != -- cgit v1.2.3 From 91d1aa43d30505b0b825db8898ffc80a8eca96c7 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Tue, 27 Nov 2012 19:33:25 +0100 Subject: context_tracking: New context tracking susbsystem Create a new subsystem that probes on kernel boundaries to keep track of the transitions between level contexts with two basic initial contexts: user or kernel. This is an abstraction of some RCU code that use such tracking to implement its userspace extended quiescent state. We need to pull this up from RCU into this new level of indirection because this tracking is also going to be used to implement an "on demand" generic virtual cputime accounting. A necessary step to shutdown the tick while still accounting the cputime. Signed-off-by: Frederic Weisbecker Cc: Andrew Morton Cc: H. Peter Anvin Cc: Ingo Molnar Cc: Paul E. McKenney Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Thomas Gleixner Cc: Li Zhong Cc: Gilad Ben-Yossef Reviewed-by: Steven Rostedt [ paulmck: fix whitespace error and email address. ] Signed-off-by: Paul E. McKenney --- arch/Kconfig | 15 +++--- arch/x86/Kconfig | 2 +- arch/x86/include/asm/context_tracking.h | 31 ++++++++++++ arch/x86/include/asm/rcu.h | 32 ------------- arch/x86/kernel/entry_64.S | 2 +- arch/x86/kernel/ptrace.c | 8 ++-- arch/x86/kernel/signal.c | 5 +- arch/x86/kernel/traps.c | 2 +- arch/x86/mm/fault.c | 2 +- include/linux/context_tracking.h | 18 +++++++ include/linux/rcupdate.h | 2 - init/Kconfig | 28 +++++------ kernel/Makefile | 1 + kernel/context_tracking.c | 83 +++++++++++++++++++++++++++++++++ kernel/rcutree.c | 64 +------------------------ kernel/sched/core.c | 11 +++-- 16 files changed, 174 insertions(+), 132 deletions(-) create mode 100644 arch/x86/include/asm/context_tracking.h delete mode 100644 arch/x86/include/asm/rcu.h create mode 100644 include/linux/context_tracking.h create mode 100644 kernel/context_tracking.c (limited to 'init') diff --git a/arch/Kconfig b/arch/Kconfig index 366ec06a5185..cc74aaea116c 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -300,15 +300,16 @@ config SECCOMP_FILTER See Documentation/prctl/seccomp_filter.txt for details. -config HAVE_RCU_USER_QS +config HAVE_CONTEXT_TRACKING bool help - Provide kernel entry/exit hooks necessary for userspace - RCU extended quiescent state. Syscalls need to be wrapped inside - rcu_user_exit()-rcu_user_enter() through the slow path using - TIF_NOHZ flag. Exceptions handlers must be wrapped as well. Irqs - are already protected inside rcu_irq_enter/rcu_irq_exit() but - preemption or signal handling on irq exit still need to be protected. + Provide kernel/user boundaries probes necessary for subsystems + that need it, such as userspace RCU extended quiescent state. + Syscalls need to be wrapped inside user_exit()-user_enter() through + the slow path using TIF_NOHZ flag. Exceptions handlers must be + wrapped as well. Irqs are already protected inside + rcu_irq_enter/rcu_irq_exit() but preemption or signal handling on + irq exit still need to be protected. config HAVE_VIRT_CPU_ACCOUNTING bool diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 46c3bff3ced2..110cfad24f26 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -106,7 +106,7 @@ config X86 select KTIME_SCALAR if X86_32 select GENERIC_STRNCPY_FROM_USER select GENERIC_STRNLEN_USER - select HAVE_RCU_USER_QS if X86_64 + select HAVE_CONTEXT_TRACKING if X86_64 select HAVE_IRQ_TIME_ACCOUNTING select GENERIC_KERNEL_THREAD select GENERIC_KERNEL_EXECVE diff --git a/arch/x86/include/asm/context_tracking.h b/arch/x86/include/asm/context_tracking.h new file mode 100644 index 000000000000..1616562683e9 --- /dev/null +++ b/arch/x86/include/asm/context_tracking.h @@ -0,0 +1,31 @@ +#ifndef _ASM_X86_CONTEXT_TRACKING_H +#define _ASM_X86_CONTEXT_TRACKING_H + +#ifndef __ASSEMBLY__ +#include +#include + +static inline void exception_enter(struct pt_regs *regs) +{ + user_exit(); +} + +static inline void exception_exit(struct pt_regs *regs) +{ +#ifdef CONFIG_CONTEXT_TRACKING + if (user_mode(regs)) + user_enter(); +#endif +} + +#else /* __ASSEMBLY__ */ + +#ifdef CONFIG_CONTEXT_TRACKING +# define SCHEDULE_USER call schedule_user +#else +# define SCHEDULE_USER call schedule +#endif + +#endif /* !__ASSEMBLY__ */ + +#endif diff --git a/arch/x86/include/asm/rcu.h b/arch/x86/include/asm/rcu.h deleted file mode 100644 index d1ac07a23979..000000000000 --- a/arch/x86/include/asm/rcu.h +++ /dev/null @@ -1,32 +0,0 @@ -#ifndef _ASM_X86_RCU_H -#define _ASM_X86_RCU_H - -#ifndef __ASSEMBLY__ - -#include -#include - -static inline void exception_enter(struct pt_regs *regs) -{ - rcu_user_exit(); -} - -static inline void exception_exit(struct pt_regs *regs) -{ -#ifdef CONFIG_RCU_USER_QS - if (user_mode(regs)) - rcu_user_enter(); -#endif -} - -#else /* __ASSEMBLY__ */ - -#ifdef CONFIG_RCU_USER_QS -# define SCHEDULE_USER call schedule_user -#else -# define SCHEDULE_USER call schedule -#endif - -#endif /* !__ASSEMBLY__ */ - -#endif diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index 0c58952d64e8..98faeb30139d 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -56,7 +56,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index eff5b8c68652..65b88a5dc1a8 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include @@ -1461,7 +1461,7 @@ long syscall_trace_enter(struct pt_regs *regs) { long ret = 0; - rcu_user_exit(); + user_exit(); /* * If we stepped into a sysenter/syscall insn, it trapped in @@ -1516,7 +1516,7 @@ void syscall_trace_leave(struct pt_regs *regs) * or do_notify_resume(), in which case we can be in RCU * user mode. */ - rcu_user_exit(); + user_exit(); audit_syscall_exit(regs); @@ -1534,5 +1534,5 @@ void syscall_trace_leave(struct pt_regs *regs) if (step || test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall_exit(regs, step); - rcu_user_enter(); + user_enter(); } diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 29ad351804e9..20ecac112e74 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -816,7 +817,7 @@ static void do_signal(struct pt_regs *regs) void do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) { - rcu_user_exit(); + user_exit(); #ifdef CONFIG_X86_MCE /* notify userspace of pending MCEs */ @@ -840,7 +841,7 @@ do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) if (thread_info_flags & _TIF_USER_RETURN_NOTIFY) fire_user_return_notifiers(); - rcu_user_enter(); + user_enter(); } void signal_fault(struct pt_regs *regs, void __user *frame, char *where) diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 8276dc6794cc..eb8586693e0b 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -55,7 +55,7 @@ #include #include #include -#include +#include #include diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 8e13ecb41bee..7a529cbab7ad 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -18,7 +18,7 @@ #include /* pgd_*(), ... */ #include /* kmemcheck_*(), ... */ #include /* VSYSCALL_START */ -#include /* exception_enter(), ... */ +#include /* exception_enter(), ... */ /* * Page fault error code bits: diff --git a/include/linux/context_tracking.h b/include/linux/context_tracking.h new file mode 100644 index 000000000000..e24339ccb7f0 --- /dev/null +++ b/include/linux/context_tracking.h @@ -0,0 +1,18 @@ +#ifndef _LINUX_CONTEXT_TRACKING_H +#define _LINUX_CONTEXT_TRACKING_H + +#ifdef CONFIG_CONTEXT_TRACKING +#include + +extern void user_enter(void); +extern void user_exit(void); +extern void context_tracking_task_switch(struct task_struct *prev, + struct task_struct *next); +#else +static inline void user_enter(void) { } +static inline void user_exit(void) { } +static inline void context_tracking_task_switch(struct task_struct *prev, + struct task_struct *next) { } +#endif /* !CONFIG_CONTEXT_TRACKING */ + +#endif diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 8fe7c1840d30..275aa3f1062d 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -222,8 +222,6 @@ extern void rcu_user_enter(void); extern void rcu_user_exit(void); extern void rcu_user_enter_after_irq(void); extern void rcu_user_exit_after_irq(void); -extern void rcu_user_hooks_switch(struct task_struct *prev, - struct task_struct *next); #else static inline void rcu_user_enter(void) { } static inline void rcu_user_exit(void) { } diff --git a/init/Kconfig b/init/Kconfig index 5ac6ee094225..2054e048bb98 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -486,9 +486,13 @@ config PREEMPT_RCU This option enables preemptible-RCU code that is common between the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations. +config CONTEXT_TRACKING + bool + config RCU_USER_QS bool "Consider userspace as in RCU extended quiescent state" - depends on HAVE_RCU_USER_QS && SMP + depends on HAVE_CONTEXT_TRACKING && SMP + select CONTEXT_TRACKING help This option sets hooks on kernel / userspace boundaries and puts RCU in extended quiescent state when the CPU runs in @@ -497,24 +501,20 @@ config RCU_USER_QS try to keep the timer tick on for RCU. Unless you want to hack and help the development of the full - tickless feature, you shouldn't enable this option. It also + dynticks mode, you shouldn't enable this option. It also adds unnecessary overhead. If unsure say N -config RCU_USER_QS_FORCE - bool "Force userspace extended QS by default" - depends on RCU_USER_QS +config CONTEXT_TRACKING_FORCE + bool "Force context tracking" + depends on CONTEXT_TRACKING help - Set the hooks in user/kernel boundaries by default in order to - test this feature that treats userspace as an extended quiescent - state until we have a real user like a full adaptive nohz option. - - Unless you want to hack and help the development of the full - tickless feature, you shouldn't enable this option. It adds - unnecessary overhead. - - If unsure say N + Probe on user/kernel boundaries by default in order to + test the features that rely on it such as userspace RCU extended + quiescent states. + This test is there for debugging until we have a real user like the + full dynticks mode. config RCU_FANOUT int "Tree-based hierarchical RCU fanout value" diff --git a/kernel/Makefile b/kernel/Makefile index 0dfeca4324ee..f90bbfc9727f 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -110,6 +110,7 @@ obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o obj-$(CONFIG_PADATA) += padata.o obj-$(CONFIG_CRASH_DUMP) += crash_dump.o obj-$(CONFIG_JUMP_LABEL) += jump_label.o +obj-$(CONFIG_CONTEXT_TRACKING) += context_tracking.o $(obj)/configs.o: $(obj)/config_data.h diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c new file mode 100644 index 000000000000..e0e07fd55508 --- /dev/null +++ b/kernel/context_tracking.c @@ -0,0 +1,83 @@ +#include +#include +#include +#include +#include + +struct context_tracking { + /* + * When active is false, hooks are not set to + * minimize overhead: TIF flags are cleared + * and calls to user_enter/exit are ignored. This + * may be further optimized using static keys. + */ + bool active; + enum { + IN_KERNEL = 0, + IN_USER, + } state; +}; + +static DEFINE_PER_CPU(struct context_tracking, context_tracking) = { +#ifdef CONFIG_CONTEXT_TRACKING_FORCE + .active = true, +#endif +}; + +void user_enter(void) +{ + unsigned long flags; + + /* + * Some contexts may involve an exception occuring in an irq, + * leading to that nesting: + * rcu_irq_enter() rcu_user_exit() rcu_user_exit() rcu_irq_exit() + * This would mess up the dyntick_nesting count though. And rcu_irq_*() + * helpers are enough to protect RCU uses inside the exception. So + * just return immediately if we detect we are in an IRQ. + */ + if (in_interrupt()) + return; + + WARN_ON_ONCE(!current->mm); + + local_irq_save(flags); + if (__this_cpu_read(context_tracking.active) && + __this_cpu_read(context_tracking.state) != IN_USER) { + __this_cpu_write(context_tracking.state, IN_USER); + rcu_user_enter(); + } + local_irq_restore(flags); +} + +void user_exit(void) +{ + unsigned long flags; + + /* + * Some contexts may involve an exception occuring in an irq, + * leading to that nesting: + * rcu_irq_enter() rcu_user_exit() rcu_user_exit() rcu_irq_exit() + * This would mess up the dyntick_nesting count though. And rcu_irq_*() + * helpers are enough to protect RCU uses inside the exception. So + * just return immediately if we detect we are in an IRQ. + */ + if (in_interrupt()) + return; + + local_irq_save(flags); + if (__this_cpu_read(context_tracking.state) == IN_USER) { + __this_cpu_write(context_tracking.state, IN_KERNEL); + rcu_user_exit(); + } + local_irq_restore(flags); +} + +void context_tracking_task_switch(struct task_struct *prev, + struct task_struct *next) +{ + if (__this_cpu_read(context_tracking.active)) { + clear_tsk_thread_flag(prev, TIF_NOHZ); + set_tsk_thread_flag(next, TIF_NOHZ); + } +} diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 7733eb56e156..e441b77b614e 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -207,9 +207,6 @@ EXPORT_SYMBOL_GPL(rcu_note_context_switch); DEFINE_PER_CPU(struct rcu_dynticks, rcu_dynticks) = { .dynticks_nesting = DYNTICK_TASK_EXIT_IDLE, .dynticks = ATOMIC_INIT(1), -#if defined(CONFIG_RCU_USER_QS) && !defined(CONFIG_RCU_USER_QS_FORCE) - .ignore_user_qs = true, -#endif }; static long blimit = 10; /* Maximum callbacks per rcu_do_batch. */ @@ -420,29 +417,7 @@ EXPORT_SYMBOL_GPL(rcu_idle_enter); */ void rcu_user_enter(void) { - unsigned long flags; - struct rcu_dynticks *rdtp; - - /* - * Some contexts may involve an exception occuring in an irq, - * leading to that nesting: - * rcu_irq_enter() rcu_user_exit() rcu_user_exit() rcu_irq_exit() - * This would mess up the dyntick_nesting count though. And rcu_irq_*() - * helpers are enough to protect RCU uses inside the exception. So - * just return immediately if we detect we are in an IRQ. - */ - if (in_interrupt()) - return; - - WARN_ON_ONCE(!current->mm); - - local_irq_save(flags); - rdtp = &__get_cpu_var(rcu_dynticks); - if (!rdtp->ignore_user_qs && !rdtp->in_user) { - rdtp->in_user = true; - rcu_eqs_enter(true); - } - local_irq_restore(flags); + rcu_eqs_enter(1); } /** @@ -579,27 +554,7 @@ EXPORT_SYMBOL_GPL(rcu_idle_exit); */ void rcu_user_exit(void) { - unsigned long flags; - struct rcu_dynticks *rdtp; - - /* - * Some contexts may involve an exception occuring in an irq, - * leading to that nesting: - * rcu_irq_enter() rcu_user_exit() rcu_user_exit() rcu_irq_exit() - * This would mess up the dyntick_nesting count though. And rcu_irq_*() - * helpers are enough to protect RCU uses inside the exception. So - * just return immediately if we detect we are in an IRQ. - */ - if (in_interrupt()) - return; - - local_irq_save(flags); - rdtp = &__get_cpu_var(rcu_dynticks); - if (rdtp->in_user) { - rdtp->in_user = false; - rcu_eqs_exit(true); - } - local_irq_restore(flags); + rcu_eqs_exit(1); } /** @@ -722,21 +677,6 @@ int rcu_is_cpu_idle(void) } EXPORT_SYMBOL(rcu_is_cpu_idle); -#ifdef CONFIG_RCU_USER_QS -void rcu_user_hooks_switch(struct task_struct *prev, - struct task_struct *next) -{ - struct rcu_dynticks *rdtp; - - /* Interrupts are disabled in context switch */ - rdtp = &__get_cpu_var(rcu_dynticks); - if (!rdtp->ignore_user_qs) { - clear_tsk_thread_flag(prev, TIF_NOHZ); - set_tsk_thread_flag(next, TIF_NOHZ); - } -} -#endif /* #ifdef CONFIG_RCU_USER_QS */ - #if defined(CONFIG_PROVE_RCU) && defined(CONFIG_HOTPLUG_CPU) /* diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 36f260864f65..80f80dfca70e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -72,6 +72,7 @@ #include #include #include +#include #include #include @@ -1886,8 +1887,8 @@ context_switch(struct rq *rq, struct task_struct *prev, spin_release(&rq->lock.dep_map, 1, _THIS_IP_); #endif + context_tracking_task_switch(prev, next); /* Here we just switch the register state and the stack. */ - rcu_user_hooks_switch(prev, next); switch_to(prev, next, prev); barrier(); @@ -2911,7 +2912,7 @@ asmlinkage void __sched schedule(void) } EXPORT_SYMBOL(schedule); -#ifdef CONFIG_RCU_USER_QS +#ifdef CONFIG_CONTEXT_TRACKING asmlinkage void __sched schedule_user(void) { /* @@ -2920,9 +2921,9 @@ asmlinkage void __sched schedule_user(void) * we haven't yet exited the RCU idle mode. Do it here manually until * we find a better solution. */ - rcu_user_exit(); + user_exit(); schedule(); - rcu_user_enter(); + user_enter(); } #endif @@ -3027,7 +3028,7 @@ asmlinkage void __sched preempt_schedule_irq(void) /* Catch callers which need to be fixed */ BUG_ON(ti->preempt_count || !irqs_disabled()); - rcu_user_exit(); + user_exit(); do { add_preempt_count(PREEMPT_ACTIVE); local_irq_enable(); -- cgit v1.2.3