// SPDX-License-Identifier: GPL-2.0-only/* * Generic helpers for smp ipi calls * * (C) Jens Axboe <jens.axboe@oracle.com> 2008 */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/irq_work.h>#include <linux/rcupdate.h>#include <linux/rculist.h>#include <linux/kernel.h>#include <linux/export.h>#include <linux/percpu.h>#include <linux/init.h>#include <linux/interrupt.h>#include <linux/gfp.h>#include <linux/slab.h>#include <linux/smp.h>#include <linux/cpu.h>#include <linux/sched.h>#include <linux/sched/idle.h>#include <linux/hypervisor.h>#include <linux/sched/clock.h>#include <linux/nmi.h>#include <linux/sched/debug.h>#include <linux/jump_label.h>#include <linux/string_choices.h> #include <trace/events/ipi.h>#define CREATE_TRACE_POINTS#include <trace/events/csd.h>#undef CREATE_TRACE_POINTS #include "smpboot.h"#include "sched/smp.h" #define CSD_TYPE(_csd) ((_csd)->node.u_flags & CSD_FLAG_TYPE_MASK) struct call_function_data { call_single_data_t __percpu *csd; cpumask_var_t cpumask; cpumask_var_t cpumask_ipi;}; static DEFINE_PER_CPU_ALIGNED(struct call_function_data, cfd_data); static DEFINE_PER_CPU_SHARED_ALIGNED(struct llist_head, call_single_queue); static DEFINE_PER_CPU(atomic_t, trigger_backtrace) = ATOMIC_INIT(1); static void __flush_smp_call_function_queue(bool warn_cpu_offline); int smpcfd_prepare_cpu(unsigned int cpu){ struct call_function_data *cfd = &per_cpu(cfd_data, cpu); if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL, cpu_to_node(cpu))) return -ENOMEM; if (!zalloc_cpumask_var_node(&cfd->cpumask_ipi, GFP_KERNEL, cpu_to_node(cpu))) { free_cpumask_var(cfd->cpumask); return -ENOMEM; } /* * Allocate the per-CPU CSD the first time a CPU comes up. It is * not freed when the CPU is offlined, so csd_lock_wait() can access * it even when the CPU was offlined after preemption was re-enabled. */ if (!cfd->csd) cfd->csd = alloc_percpu(call_single_data_t); if (!cfd->csd) { free_cpumask_var(cfd->cpumask); free_cpumask_var(cfd->cpumask_ipi); return -ENOMEM; } return 0;} int smpcfd_dead_cpu(unsigned int cpu){ struct call_function_data *cfd = &per_cpu(cfd_data, cpu); free_cpumask_var(cfd->cpumask); free_cpumask_var(cfd->cpumask_ipi); return 0;} int smpcfd_dying_cpu(unsigned int cpu){ /* * The IPIs for the smp-call-function callbacks queued by other CPUs * might arrive late, either due to hardware latencies or because this * CPU disabled interrupts (inside stop-machine) before the IPIs were * sent. So flush out any pending callbacks explicitly (without waiting * for the IPIs to arrive), to ensure that the outgoing CPU doesn't go * offline with work still pending. * * This runs with interrupts disabled inside the stopper task invoked by * stop_machine(), ensuring mutually exclusive CPU offlining and IPI flush. */ __flush_smp_call_function_queue(false); irq_work_run(); return 0;} void __init call_function_init(void){ int i; for_each_possible_cpu(i) init_llist_head(&per_cpu(call_single_queue, i)); smpcfd_prepare_cpu(smp_processor_id());} static __always_inline voidsend_call_function_single_ipi(int cpu){ if (call_function_single_prep_ipi(cpu)) { trace_ipi_send_cpu(cpu, _RET_IP_, generic_smp_call_function_single_interrupt); arch_send_call_function_single_ipi(cpu); }} static __always_inline voidsend_call_function_ipi_mask(struct cpumask *mask){ trace_ipi_send_cpumask(mask, _RET_IP_, generic_smp_call_function_single_interrupt); arch_send_call_function_ipi_mask(mask);} static __always_inline voidcsd_do_func(smp_call_func_t func, void *info, call_single_data_t *csd){ trace_csd_function_entry(func, csd); func(info); trace_csd_function_exit(func, csd);} static DEFINE_STATIC_KEY_MAYBE(CONFIG_CSD_LOCK_WAIT_DEBUG_DEFAULT, csdlock_debug_enabled); #ifdef CONFIG_CSD_LOCK_WAIT_DEBUG /* * Parse the csdlock_debug= kernel boot parameter. * * If you need to restore the old "ext" value that once provided * additional debugging information, reapply the following commits: * * de7b09ef658d ("locking/csd_lock: Prepare more CSD lock debugging") * a5aabace5fb8 ("locking/csd_lock: Add more data to CSD lock debugging") */static int __init csdlock_debug(char *str){ int ret; unsigned int val = 0; ret = get_option(&str, &val); if (ret) { if (val) static_branch_enable(&csdlock_debug_enabled); else static_branch_disable(&csdlock_debug_enabled); } return 1;}__setup("csdlock_debug=", csdlock_debug); static DEFINE_PER_CPU(call_single_data_t *, cur_csd);static DEFINE_PER_CPU(smp_call_func_t, cur_csd_func);static DEFINE_PER_CPU(void *, cur_csd_info); static ulong csd_lock_timeout = 5000; /* CSD lock timeout in milliseconds. */module_param(csd_lock_timeout, ulong, 0644);static int panic_on_ipistall; /* CSD panic timeout in milliseconds, 300000 for five minutes. */module_param(panic_on_ipistall, int, 0644); static atomic_t csd_bug_count = ATOMIC_INIT(0); /* Record current CSD work for current CPU, NULL to erase. */static void __csd_lock_record(call_single_data_t *csd){ if (!csd) { /* * Pairs with smp_load_acquire() of cur_csd in * csd_lock_wait_toolong(): orders any preceding CSD * callback/unlock before a remote reader observes NULL. */ smp_store_release(this_cpu_ptr(&cur_csd), NULL); return; } __this_cpu_write(cur_csd_func, csd->func); __this_cpu_write(cur_csd_info, csd->info); /* * Pairs with smp_load_acquire() of cur_csd in * csd_lock_wait_toolong(): publishes cur_csd_func and * cur_csd_info before the non-NULL pointer becomes visible. */ smp_store_release(this_cpu_ptr(&cur_csd), csd);} static __always_inline void csd_lock_record(call_single_data_t *csd){ if (static_branch_unlikely(&csdlock_debug_enabled)) __csd_lock_record(csd);} static int csd_lock_wait_getcpu(call_single_data_t *csd){ unsigned int csd_type; csd_type = CSD_TYPE(csd); if (csd_type == CSD_TYPE_ASYNC || csd_type == CSD_TYPE_SYNC) return csd->node.dst; /* Other CSD_TYPE_ values might not have ->dst. */ return -1;} static atomic_t n_csd_lock_stuck; /** * csd_lock_is_stuck - Has a CSD-lock acquisition been stuck too long? * * Returns: @true if a CSD-lock acquisition is stuck and has been stuck * long enough for a "non-responsive CSD lock" message to be printed. */bool csd_lock_is_stuck(void){ return !!atomic_read(&n_csd_lock_stuck);} /* * Complain if too much time spent waiting. Note that only * the CSD_TYPE_SYNC/ASYNC types provide the destination CPU, * so waiting on other types gets much less information. */static bool csd_lock_wait_toolong(call_single_data_t *csd, u64 ts0, u64 *ts1, int *bug_id, unsigned long *nmessages){ int cpu = -1; int cpux; bool firsttime; u64 ts2, ts_delta; call_single_data_t *cpu_cur_csd; unsigned int flags = READ_ONCE(csd->node.u_flags); unsigned long long csd_lock_timeout_ns = csd_lock_timeout * NSEC_PER_MSEC; if (!(flags & CSD_FLAG_LOCK)) { if (!unlikely(*bug_id)) return true; cpu = csd_lock_wait_getcpu(csd); pr_alert("csd: CSD lock (#%d) got unstuck on CPU#%02d, CPU#%02d released the lock.\n", *bug_id, raw_smp_processor_id(), cpu); atomic_dec(&n_csd_lock_stuck); return true; } ts2 = ktime_get_mono_fast_ns(); /* How long since we last checked for a stuck CSD lock.*/ ts_delta = ts2 - *ts1; if (likely(ts_delta <= csd_lock_timeout_ns * (*nmessages + 1) * (!*nmessages ? 1 : (ilog2(num_online_cpus()) / 2 + 1)) || csd_lock_timeout_ns == 0)) return false; if (ts0 > ts2) { /* Our own sched_clock went backward; don't blame another CPU. */ ts_delta = ts0 - ts2; pr_alert("sched_clock on CPU %d went backward by %llu ns\n", raw_smp_processor_id(), ts_delta); *ts1 = ts2; return false; } firsttime = !*bug_id; if (firsttime) *bug_id = atomic_inc_return(&csd_bug_count); cpu = csd_lock_wait_getcpu(csd); if (WARN_ONCE(cpu < 0 || cpu >= nr_cpu_ids, "%s: cpu = %d\n", __func__, cpu)) cpux = 0; else cpux = cpu; /* * Pairs with smp_store_release() of cur_csd in __csd_lock_record(): * a non-NULL cur_csd here implies cur_csd_func and cur_csd_info * are the matching publication; a NULL value is ordered after any * preceding CSD callback/unlock on the remote CPU. */ cpu_cur_csd = smp_load_acquire(&per_cpu(cur_csd, cpux)); /* How long since this CSD lock was stuck. */ ts_delta = ts2 - ts0; pr_alert("csd: %s non-responsive CSD lock (#%d) on CPU#%d, waiting %lld ns for CPU#%02d %pS(%ps).\n", firsttime ? "Detected" : "Continued", *bug_id, raw_smp_processor_id(), (s64)ts_delta, cpu, csd->func, csd->info); (*nmessages)++; if (firsttime) atomic_inc(&n_csd_lock_stuck); /* * If the CSD lock is still stuck after 5 minutes, it is unlikely * to become unstuck. Use a signed comparison to avoid triggering * on underflows when the TSC is out of sync between sockets. */ BUG_ON(panic_on_ipistall > 0 && (s64)ts_delta > ((s64)panic_on_ipistall * NSEC_PER_MSEC)); if (cpu_cur_csd && csd != cpu_cur_csd) { pr_alert("\tcsd: CSD lock (#%d) handling prior %pS(%ps) request.\n", *bug_id, READ_ONCE(per_cpu(cur_csd_func, cpux)), READ_ONCE(per_cpu(cur_csd_info, cpux))); } else { pr_alert("\tcsd: CSD lock (#%d) %s.\n", *bug_id, !cpu_cur_csd ? "unresponsive" : "handling this request"); } if (cpu >= 0) { if (atomic_cmpxchg_acquire(&per_cpu(trigger_backtrace, cpu), 1, 0)) dump_cpu_task(cpu); if (!cpu_cur_csd) { pr_alert("csd: Re-sending CSD lock (#%d) IPI from CPU#%02d to CPU#%02d\n", *bug_id, raw_smp_processor_id(), cpu); arch_send_call_function_single_ipi(cpu); } } if (firsttime) dump_stack(); *ts1 = ts2; return false;} /* * csd_lock/csd_unlock used to serialize access to per-cpu csd resources * * For non-synchronous ipi calls the csd can still be in use by the * previous function call. For multi-cpu calls its even more interesting * as we'll have to ensure no other cpu is observing our csd. */static void __csd_lock_wait(call_single_data_t *csd){ unsigned long nmessages = 0; int bug_id = 0; u64 ts0, ts1; guard(preempt)(); ts1 = ts0 = ktime_get_mono_fast_ns(); for (;;) { if (csd_lock_wait_toolong(csd, ts0, &ts1, &bug_id, &nmessages)) break; cpu_relax(); } smp_acquire__after_ctrl_dep();} static __always_inline void csd_lock_wait(call_single_data_t *csd){ if (static_branch_unlikely(&csdlock_debug_enabled)) { __csd_lock_wait(csd); return; } smp_cond_load_acquire(&csd->node.u_flags, !(VAL & CSD_FLAG_LOCK));}#elsestatic __always_inline void __csd_lock_wait(call_single_data_t *csd){} static void csd_lock_record(call_single_data_t *csd){} static __always_inline void csd_lock_wait(call_single_data_t *csd){ smp_cond_load_acquire(&csd->node.u_flags, !(VAL & CSD_FLAG_LOCK));}#endif static __always_inline void csd_lock(call_single_data_t *csd){ if (IS_ENABLED(CONFIG_CSD_LOCK_WAIT_DEBUG) && static_branch_unlikely(&csdlock_debug_enabled)) { for (;;) { unsigned int flags; __csd_lock_wait(csd); flags = READ_ONCE(csd->node.u_flags); if (!(flags & CSD_FLAG_LOCK) && try_cmpxchg_acquire(&csd->node.u_flags, &flags, flags | CSD_FLAG_LOCK)) break; } } else { csd_lock_wait(csd); csd->node.u_flags |= CSD_FLAG_LOCK; } /* * prevent CPU from reordering the above assignment * to ->flags with any subsequent assignments to other * fields of the specified call_single_data_t structure: */ smp_wmb();} static __always_inline void csd_unlock(call_single_data_t *csd){ WARN_ON(!(csd->node.u_flags & CSD_FLAG_LOCK)); /* * ensure we're all done before releasing data: */ smp_store_release(&csd->node.u_flags, 0);} static DEFINE_PER_CPU_SHARED_ALIGNED(call_single_data_t, csd_data); #ifdef CONFIG_CSD_LOCK_WAIT_DEBUGstatic call_single_data_t *get_single_csd_data(int cpu){ if (static_branch_unlikely(&csdlock_debug_enabled) && (unsigned int)cpu < nr_cpu_ids) return per_cpu_ptr(&csd_data, cpu); return this_cpu_ptr(&csd_data);}#elsestatic call_single_data_t *get_single_csd_data(int cpu){ return this_cpu_ptr(&csd_data);}#endif void __smp_call_single_queue(int cpu, struct llist_node *node){ /* * We have to check the type of the CSD before queueing it, because * once queued it can have its flags cleared by * flush_smp_call_function_queue() * even if we haven't sent the smp_call IPI yet (e.g. the stopper * executes migration_cpu_stop() on the remote CPU). */ if (trace_csd_queue_cpu_enabled()) { call_single_data_t *csd; smp_call_func_t func; csd = container_of(node, call_single_data_t, node.llist); func = CSD_TYPE(csd) == CSD_TYPE_TTWU ? sched_ttwu_pending : csd->func; trace_call__csd_queue_cpu(cpu, _RET_IP_, func, csd); } /* * The list addition should be visible to the target CPU when it pops * the head of the list to pull the entry off it in the IPI handler * because of normal cache coherency rules implied by the underlying * llist ops. * * If IPIs can go out of order to the cache coherency protocol * in an architecture, sufficient synchronisation should be added * to arch code to make it appear to obey cache coherency WRT * locking and barrier primitives. Generic code isn't really * equipped to do the right thing... */ if (llist_add(node, &per_cpu(call_single_queue, cpu))) send_call_function_single_ipi(cpu);} /* * Insert a previously allocated call_single_data_t element * for execution on the given CPU. data must already have * ->func, ->info, and ->flags set. */static int generic_exec_single(int cpu, call_single_data_t *csd){ /* * Preemption already disabled here so stopper cannot run on this CPU, * ensuring mutually exclusive CPU offlining and last IPI flush. */ if (cpu == smp_processor_id()) { smp_call_func_t func = csd->func; void *info = csd->info; unsigned long flags; /* * We can unlock early even for the synchronous on-stack case, * since we're doing this from the same CPU.. */ csd_lock_record(csd); csd_unlock(csd); local_irq_save(flags); csd_do_func(func, info, NULL); csd_lock_record(NULL); local_irq_restore(flags); return 0; } if ((unsigned)cpu >= nr_cpu_ids || !cpu_online(cpu)) {Lines 1–500 of 1305. Load more
// SPDX-License-Identifier: GPL-2.0-only /* * Generic helpers for smp ipi calls * * (C) Jens Axboe <jens.axboe@oracle.com> 2008 */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/irq_work.h> #include <linux/rcupdate.h> #include <linux/rculist.h> #include <linux/kernel.h> #include <linux/export.h> #include <linux/percpu.h> #include <linux/init.h> #include <linux/interrupt.h> #include <linux/gfp.h> #include <linux/slab.h> #include <linux/smp.h> #include <linux/cpu.h> #include <linux/sched.h> #include <linux/sched/idle.h> #include <linux/hypervisor.h> #include <linux/sched/clock.h> #include <linux/nmi.h> #include <linux/sched/debug.h> #include <linux/jump_label.h> #include <linux/string_choices.h> #include <trace/events/ipi.h> #define CREATE_TRACE_POINTS #include <trace/events/csd.h> #undef CREATE_TRACE_POINTS #include "smpboot.h" #include "sched/smp.h" #define CSD_TYPE(_csd) ((_csd)->node.u_flags & CSD_FLAG_TYPE_MASK) struct call_function_data { call_single_data_t __percpu *csd; cpumask_var_t cpumask; cpumask_var_t cpumask_ipi; }; static DEFINE_PER_CPU_ALIGNED(struct call_function_data, cfd_data); static DEFINE_PER_CPU_SHARED_ALIGNED(struct llist_head, call_single_queue); static DEFINE_PER_CPU(atomic_t, trigger_backtrace) = ATOMIC_INIT(1); static void __flush_smp_call_function_queue(bool warn_cpu_offline); int smpcfd_prepare_cpu(unsigned int cpu) { struct call_function_data *cfd = &per_cpu(cfd_data, cpu); if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL, cpu_to_node(cpu))) return -ENOMEM; if (!zalloc_cpumask_var_node(&cfd->cpumask_ipi, GFP_KERNEL, cpu_to_node(cpu))) { free_cpumask_var(cfd->cpumask); return -ENOMEM; } /* * Allocate the per-CPU CSD the first time a CPU comes up. It is * not freed when the CPU is offlined, so csd_lock_wait() can access * it even when the CPU was offlined after preemption was re-enabled. */ if (!cfd->csd) cfd->csd = alloc_percpu(call_single_data_t); if (!cfd->csd) { free_cpumask_var(cfd->cpumask); free_cpumask_var(cfd->cpumask_ipi); return -ENOMEM; } return 0; } int smpcfd_dead_cpu(unsigned int cpu) { struct call_function_data *cfd = &per_cpu(cfd_data, cpu); free_cpumask_var(cfd->cpumask); free_cpumask_var(cfd->cpumask_ipi); return 0; } int smpcfd_dying_cpu(unsigned int cpu) { /* * The IPIs for the smp-call-function callbacks queued by other CPUs * might arrive late, either due to hardware latencies or because this * CPU disabled interrupts (inside stop-machine) before the IPIs were * sent. So flush out any pending callbacks explicitly (without waiting * for the IPIs to arrive), to ensure that the outgoing CPU doesn't go * offline with work still pending. * * This runs with interrupts disabled inside the stopper task invoked by * stop_machine(), ensuring mutually exclusive CPU offlining and IPI flush. */ __flush_smp_call_function_queue(false); irq_work_run(); return 0; } void __init call_function_init(void) { int i; for_each_possible_cpu(i) init_llist_head(&per_cpu(call_single_queue, i)); smpcfd_prepare_cpu(smp_processor_id()); } static __always_inline void send_call_function_single_ipi(int cpu) { if (call_function_single_prep_ipi(cpu)) { trace_ipi_send_cpu(cpu, _RET_IP_, generic_smp_call_function_single_interrupt); arch_send_call_function_single_ipi(cpu); } } static __always_inline void send_call_function_ipi_mask(struct cpumask *mask) { trace_ipi_send_cpumask(mask, _RET_IP_, generic_smp_call_function_single_interrupt); arch_send_call_function_ipi_mask(mask); } static __always_inline void csd_do_func(smp_call_func_t func, void *info, call_single_data_t *csd) { trace_csd_function_entry(func, csd); func(info); trace_csd_function_exit(func, csd); } static DEFINE_STATIC_KEY_MAYBE(CONFIG_CSD_LOCK_WAIT_DEBUG_DEFAULT, csdlock_debug_enabled); #ifdef CONFIG_CSD_LOCK_WAIT_DEBUG /* * Parse the csdlock_debug= kernel boot parameter. * * If you need to restore the old "ext" value that once provided * additional debugging information, reapply the following commits: * * de7b09ef658d ("locking/csd_lock: Prepare more CSD lock debugging") * a5aabace5fb8 ("locking/csd_lock: Add more data to CSD lock debugging") */ static int __init csdlock_debug(char *str) { int ret; unsigned int val = 0; ret = get_option(&str, &val); if (ret) { if (val) static_branch_enable(&csdlock_debug_enabled); else static_branch_disable(&csdlock_debug_enabled); } return 1; } __setup("csdlock_debug=", csdlock_debug); static DEFINE_PER_CPU(call_single_data_t *, cur_csd); static DEFINE_PER_CPU(smp_call_func_t, cur_csd_func); static DEFINE_PER_CPU(void *, cur_csd_info); static ulong csd_lock_timeout = 5000; /* CSD lock timeout in milliseconds. */ module_param(csd_lock_timeout, ulong, 0644); static int panic_on_ipistall; /* CSD panic timeout in milliseconds, 300000 for five minutes. */ module_param(panic_on_ipistall, int, 0644); static atomic_t csd_bug_count = ATOMIC_INIT(0); /* Record current CSD work for current CPU, NULL to erase. */ static void __csd_lock_record(call_single_data_t *csd) { if (!csd) { /* * Pairs with smp_load_acquire() of cur_csd in * csd_lock_wait_toolong(): orders any preceding CSD * callback/unlock before a remote reader observes NULL. */ smp_store_release(this_cpu_ptr(&cur_csd), NULL); return; } __this_cpu_write(cur_csd_func, csd->func); __this_cpu_write(cur_csd_info, csd->info); /* * Pairs with smp_load_acquire() of cur_csd in * csd_lock_wait_toolong(): publishes cur_csd_func and * cur_csd_info before the non-NULL pointer becomes visible. */ smp_store_release(this_cpu_ptr(&cur_csd), csd); } static __always_inline void csd_lock_record(call_single_data_t *csd) { if (static_branch_unlikely(&csdlock_debug_enabled)) __csd_lock_record(csd); } static int csd_lock_wait_getcpu(call_single_data_t *csd) { unsigned int csd_type; csd_type = CSD_TYPE(csd); if (csd_type == CSD_TYPE_ASYNC || csd_type == CSD_TYPE_SYNC) return csd->node.dst; /* Other CSD_TYPE_ values might not have ->dst. */ return -1; } static atomic_t n_csd_lock_stuck; /** * csd_lock_is_stuck - Has a CSD-lock acquisition been stuck too long? * * Returns: @true if a CSD-lock acquisition is stuck and has been stuck * long enough for a "non-responsive CSD lock" message to be printed. */ bool csd_lock_is_stuck(void) { return !!atomic_read(&n_csd_lock_stuck); } /* * Complain if too much time spent waiting. Note that only * the CSD_TYPE_SYNC/ASYNC types provide the destination CPU, * so waiting on other types gets much less information. */ static bool csd_lock_wait_toolong(call_single_data_t *csd, u64 ts0, u64 *ts1, int *bug_id, unsigned long *nmessages) { int cpu = -1; int cpux; bool firsttime; u64 ts2, ts_delta; call_single_data_t *cpu_cur_csd; unsigned int flags = READ_ONCE(csd->node.u_flags); unsigned long long csd_lock_timeout_ns = csd_lock_timeout * NSEC_PER_MSEC; if (!(flags & CSD_FLAG_LOCK)) { if (!unlikely(*bug_id)) return true; cpu = csd_lock_wait_getcpu(csd); pr_alert("csd: CSD lock (#%d) got unstuck on CPU#%02d, CPU#%02d released the lock.\n", *bug_id, raw_smp_processor_id(), cpu); atomic_dec(&n_csd_lock_stuck); return true; } ts2 = ktime_get_mono_fast_ns(); /* How long since we last checked for a stuck CSD lock.*/ ts_delta = ts2 - *ts1; if (likely(ts_delta <= csd_lock_timeout_ns * (*nmessages + 1) * (!*nmessages ? 1 : (ilog2(num_online_cpus()) / 2 + 1)) || csd_lock_timeout_ns == 0)) return false; if (ts0 > ts2) { /* Our own sched_clock went backward; don't blame another CPU. */ ts_delta = ts0 - ts2; pr_alert("sched_clock on CPU %d went backward by %llu ns\n", raw_smp_processor_id(), ts_delta); *ts1 = ts2; return false; } firsttime = !*bug_id; if (firsttime) *bug_id = atomic_inc_return(&csd_bug_count); cpu = csd_lock_wait_getcpu(csd); if (WARN_ONCE(cpu < 0 || cpu >= nr_cpu_ids, "%s: cpu = %d\n", __func__, cpu)) cpux = 0; else cpux = cpu; /* * Pairs with smp_store_release() of cur_csd in __csd_lock_record(): * a non-NULL cur_csd here implies cur_csd_func and cur_csd_info * are the matching publication; a NULL value is ordered after any * preceding CSD callback/unlock on the remote CPU. */ cpu_cur_csd = smp_load_acquire(&per_cpu(cur_csd, cpux)); /* How long since this CSD lock was stuck. */ ts_delta = ts2 - ts0; pr_alert("csd: %s non-responsive CSD lock (#%d) on CPU#%d, waiting %lld ns for CPU#%02d %pS(%ps).\n", firsttime ? "Detected" : "Continued", *bug_id, raw_smp_processor_id(), (s64)ts_delta, cpu, csd->func, csd->info); (*nmessages)++; if (firsttime) atomic_inc(&n_csd_lock_stuck); /* * If the CSD lock is still stuck after 5 minutes, it is unlikely * to become unstuck. Use a signed comparison to avoid triggering * on underflows when the TSC is out of sync between sockets. */ BUG_ON(panic_on_ipistall > 0 && (s64)ts_delta > ((s64)panic_on_ipistall * NSEC_PER_MSEC)); if (cpu_cur_csd && csd != cpu_cur_csd) { pr_alert("\tcsd: CSD lock (#%d) handling prior %pS(%ps) request.\n", *bug_id, READ_ONCE(per_cpu(cur_csd_func, cpux)), READ_ONCE(per_cpu(cur_csd_info, cpux))); } else { pr_alert("\tcsd: CSD lock (#%d) %s.\n", *bug_id, !cpu_cur_csd ? "unresponsive" : "handling this request"); } if (cpu >= 0) { if (atomic_cmpxchg_acquire(&per_cpu(trigger_backtrace, cpu), 1, 0)) dump_cpu_task(cpu); if (!cpu_cur_csd) { pr_alert("csd: Re-sending CSD lock (#%d) IPI from CPU#%02d to CPU#%02d\n", *bug_id, raw_smp_processor_id(), cpu); arch_send_call_function_single_ipi(cpu); } } if (firsttime) dump_stack(); *ts1 = ts2; return false; } /* * csd_lock/csd_unlock used to serialize access to per-cpu csd resources * * For non-synchronous ipi calls the csd can still be in use by the * previous function call. For multi-cpu calls its even more interesting * as we'll have to ensure no other cpu is observing our csd. */ static void __csd_lock_wait(call_single_data_t *csd) { unsigned long nmessages = 0; int bug_id = 0; u64 ts0, ts1; guard(preempt)(); ts1 = ts0 = ktime_get_mono_fast_ns(); for (;;) { if (csd_lock_wait_toolong(csd, ts0, &ts1, &bug_id, &nmessages)) break; cpu_relax(); } smp_acquire__after_ctrl_dep(); } static __always_inline void csd_lock_wait(call_single_data_t *csd) { if (static_branch_unlikely(&csdlock_debug_enabled)) { __csd_lock_wait(csd); return; } smp_cond_load_acquire(&csd->node.u_flags, !(VAL & CSD_FLAG_LOCK)); } #else static __always_inline void __csd_lock_wait(call_single_data_t *csd) { } static void csd_lock_record(call_single_data_t *csd) { } static __always_inline void csd_lock_wait(call_single_data_t *csd) { smp_cond_load_acquire(&csd->node.u_flags, !(VAL & CSD_FLAG_LOCK)); } #endif static __always_inline void csd_lock(call_single_data_t *csd) { if (IS_ENABLED(CONFIG_CSD_LOCK_WAIT_DEBUG) && static_branch_unlikely(&csdlock_debug_enabled)) { for (;;) { unsigned int flags; __csd_lock_wait(csd); flags = READ_ONCE(csd->node.u_flags); if (!(flags & CSD_FLAG_LOCK) && try_cmpxchg_acquire(&csd->node.u_flags, &flags, flags | CSD_FLAG_LOCK)) break; } } else { csd_lock_wait(csd); csd->node.u_flags |= CSD_FLAG_LOCK; } /* * prevent CPU from reordering the above assignment * to ->flags with any subsequent assignments to other * fields of the specified call_single_data_t structure: */ smp_wmb(); } static __always_inline void csd_unlock(call_single_data_t *csd) { WARN_ON(!(csd->node.u_flags & CSD_FLAG_LOCK)); /* * ensure we're all done before releasing data: */ smp_store_release(&csd->node.u_flags, 0); } static DEFINE_PER_CPU_SHARED_ALIGNED(call_single_data_t, csd_data); #ifdef CONFIG_CSD_LOCK_WAIT_DEBUG static call_single_data_t *get_single_csd_data(int cpu) { if (static_branch_unlikely(&csdlock_debug_enabled) && (unsigned int)cpu < nr_cpu_ids) return per_cpu_ptr(&csd_data, cpu); return this_cpu_ptr(&csd_data); } #else static call_single_data_t *get_single_csd_data(int cpu) { return this_cpu_ptr(&csd_data); } #endif void __smp_call_single_queue(int cpu, struct llist_node *node) { /* * We have to check the type of the CSD before queueing it, because * once queued it can have its flags cleared by * flush_smp_call_function_queue() * even if we haven't sent the smp_call IPI yet (e.g. the stopper * executes migration_cpu_stop() on the remote CPU). */ if (trace_csd_queue_cpu_enabled()) { call_single_data_t *csd; smp_call_func_t func; csd = container_of(node, call_single_data_t, node.llist); func = CSD_TYPE(csd) == CSD_TYPE_TTWU ? sched_ttwu_pending : csd->func; trace_call__csd_queue_cpu(cpu, _RET_IP_, func, csd); } /* * The list addition should be visible to the target CPU when it pops * the head of the list to pull the entry off it in the IPI handler * because of normal cache coherency rules implied by the underlying * llist ops. * * If IPIs can go out of order to the cache coherency protocol * in an architecture, sufficient synchronisation should be added * to arch code to make it appear to obey cache coherency WRT * locking and barrier primitives. Generic code isn't really * equipped to do the right thing... */ if (llist_add(node, &per_cpu(call_single_queue, cpu))) send_call_function_single_ipi(cpu); } /* * Insert a previously allocated call_single_data_t element * for execution on the given CPU. data must already have * ->func, ->info, and ->flags set. */ static int generic_exec_single(int cpu, call_single_data_t *csd) { /* * Preemption already disabled here so stopper cannot run on this CPU, * ensuring mutually exclusive CPU offlining and last IPI flush. */ if (cpu == smp_processor_id()) { smp_call_func_t func = csd->func; void *info = csd->info; unsigned long flags; /* * We can unlock early even for the synchronous on-stack case, * since we're doing this from the same CPU.. */ csd_lock_record(csd); csd_unlock(csd); local_irq_save(flags); csd_do_func(func, info, NULL); csd_lock_record(NULL); local_irq_restore(flags); return 0; } if ((unsigned)cpu >= nr_cpu_ids || !cpu_online(cpu)) {