diff options
Diffstat (limited to 'kernel/rcu')
| -rw-r--r-- | kernel/rcu/Kconfig | 80 | ||||
| -rw-r--r-- | kernel/rcu/Kconfig.debug | 44 | ||||
| -rw-r--r-- | kernel/rcu/rcu.h | 51 | ||||
| -rw-r--r-- | kernel/rcu/rcuscale.c | 109 | ||||
| -rw-r--r-- | kernel/rcu/rcutorture.c | 854 | ||||
| -rw-r--r-- | kernel/rcu/refscale.c | 471 | ||||
| -rw-r--r-- | kernel/rcu/srcutiny.c | 54 | ||||
| -rw-r--r-- | kernel/rcu/srcutree.c | 564 | ||||
| -rw-r--r-- | kernel/rcu/tasks.h | 731 | ||||
| -rw-r--r-- | kernel/rcu/tiny.c | 37 | ||||
| -rw-r--r-- | kernel/rcu/tree.c | 288 | ||||
| -rw-r--r-- | kernel/rcu/tree.h | 20 | ||||
| -rw-r--r-- | kernel/rcu/tree_exp.h | 77 | ||||
| -rw-r--r-- | kernel/rcu/tree_nocb.h | 243 | ||||
| -rw-r--r-- | kernel/rcu/tree_plugin.h | 161 | ||||
| -rw-r--r-- | kernel/rcu/tree_stall.h | 76 | ||||
| -rw-r--r-- | kernel/rcu/update.c | 32 |
17 files changed, 2299 insertions, 1593 deletions
diff --git a/kernel/rcu/Kconfig b/kernel/rcu/Kconfig index b9b6bc55185d..762299291e09 100644 --- a/kernel/rcu/Kconfig +++ b/kernel/rcu/Kconfig @@ -18,7 +18,7 @@ config TREE_RCU config PREEMPT_RCU bool - default y if PREEMPTION + default y if (PREEMPT || PREEMPT_RT || PREEMPT_DYNAMIC) select TREE_RCU help This option selects the RCU implementation that is @@ -65,11 +65,24 @@ config TREE_SRCU help This option selects the full-fledged version of SRCU. +config FORCE_NEED_SRCU_NMI_SAFE + bool "Force selection of NEED_SRCU_NMI_SAFE" + depends on !TINY_SRCU + depends on RCU_EXPERT + depends on ARCH_HAS_NMI_SAFE_THIS_CPU_OPS + select NEED_SRCU_NMI_SAFE + default n + help + This option forces selection of the NEED_SRCU_NMI_SAFE + Kconfig option, allowing testing of srcu_read_lock_nmisafe() + and srcu_read_unlock_nmisafe() on architectures (like x86) + that select the ARCH_HAS_NMI_SAFE_THIS_CPU_OPS Kconfig option. + config NEED_SRCU_NMI_SAFE def_bool HAVE_NMI && !ARCH_HAS_NMI_SAFE_THIS_CPU_OPS && !TINY_SRCU config TASKS_RCU_GENERIC - def_bool TASKS_RCU || TASKS_RUDE_RCU || TASKS_TRACE_RCU + def_bool TASKS_RCU || TASKS_RUDE_RCU help This option enables generic infrastructure code supporting task-based RCU implementations. Not for manual selection. @@ -91,7 +104,7 @@ config NEED_TASKS_RCU config TASKS_RCU bool - default NEED_TASKS_RCU && (PREEMPTION || PREEMPT_AUTO) + default NEED_TASKS_RCU && PREEMPTION select IRQ_WORK config FORCE_TASKS_RUDE_RCU @@ -129,6 +142,29 @@ config TASKS_TRACE_RCU default n select IRQ_WORK +config TASKS_TRACE_RCU_NO_MB + bool "Override RCU Tasks Trace inclusion of read-side memory barriers" + depends on RCU_EXPERT && TASKS_TRACE_RCU + default ARCH_WANTS_NO_INSTR + help + This option prevents the use of read-side memory barriers in + rcu_read_lock_tasks_trace() and rcu_read_unlock_tasks_trace() + even in kernels built with CONFIG_ARCH_WANTS_NO_INSTR=n, that is, + in kernels that do not have noinstr set up in entry/exit code. + By setting this option, you are promising to carefully review + use of ftrace, BPF, and friends to ensure that no tracing + operation is attached to a function that runs in that portion + of the entry/exit code that RCU does not watch, that is, + where rcu_is_watching() returns false. Alternatively, you + might choose to never remove traces except by rebooting. + + Those wishing to disable read-side memory barriers for an entire + architecture can select this Kconfig option, hence the polarity. + + Say Y here if you need speed and will review use of tracing. + Say N here for certain esoteric testing of RCU itself. + Take the default if you are unsure. + config RCU_STALL_COMMON def_bool TREE_RCU help @@ -300,44 +336,32 @@ config RCU_NOCB_CPU_CB_BOOST Say Y here if you want to set RT priority for offloading kthreads. Say N here if you are building a !PREEMPT_RT kernel and are unsure. -config TASKS_TRACE_RCU_READ_MB - bool "Tasks Trace RCU readers use memory barriers in user and idle" - depends on RCU_EXPERT && TASKS_TRACE_RCU - default PREEMPT_RT || NR_CPUS < 8 - help - Use this option to further reduce the number of IPIs sent - to CPUs executing in userspace or idle during tasks trace - RCU grace periods. Given that a reasonable setting of - the rcupdate.rcu_task_ipi_delay kernel boot parameter - eliminates such IPIs for many workloads, proper setting - of this Kconfig option is important mostly for aggressive - real-time installations and for battery-powered devices, - hence the default chosen above. - - Say Y here if you hate IPIs. - Say N here if you hate read-side memory barriers. - Take the default if you are unsure. - config RCU_LAZY bool "RCU callback lazy invocation functionality" depends on RCU_NOCB_CPU default n help - To save power, batch RCU callbacks and flush after delay, memory - pressure, or callback list growing too big. + To save power, batch RCU callbacks and delay starting the + corresponding grace period for multiple seconds. The grace + period will be started after this delay, in case of memory + pressure, or if the corresponding CPU's callback list grows + too large. - Requires rcu_nocbs=all to be set. + These delays happen only on rcu_nocbs CPUs, that is, CPUs + whose callbacks have been offloaded. - Use rcutree.enable_rcu_lazy=0 to turn it off at boot time. + Use the rcutree.enable_rcu_lazy=0 kernel-boot parameter to + globally disable these delays. config RCU_LAZY_DEFAULT_OFF bool "Turn RCU lazy invocation off by default" depends on RCU_LAZY default n help - Allows building the kernel with CONFIG_RCU_LAZY=y yet keep it default - off. Boot time param rcutree.enable_rcu_lazy=1 can be used to switch - it back on. + Build the kernel with CONFIG_RCU_LAZY=y, but cause the kernel + to boot with these energy-efficiency delays disabled. Use the + rcutree.enable_rcu_lazy=0 kernel-boot parameter to override + the this option at boot time, thus re-enabling these delays. config RCU_DOUBLE_CHECK_CB_TIME bool "RCU callback-batch backup time check" diff --git a/kernel/rcu/Kconfig.debug b/kernel/rcu/Kconfig.debug index 6af90510a1ca..e078e988773d 100644 --- a/kernel/rcu/Kconfig.debug +++ b/kernel/rcu/Kconfig.debug @@ -54,7 +54,7 @@ config RCU_TORTURE_TEST Say N if you are unsure. config RCU_TORTURE_TEST_CHK_RDR_STATE - tristate "Check rcutorture reader state" + bool "Check rcutorture reader state" depends on RCU_TORTURE_TEST default n help @@ -70,7 +70,7 @@ config RCU_TORTURE_TEST_CHK_RDR_STATE Say N if you are unsure. config RCU_TORTURE_TEST_LOG_CPU - tristate "Log CPU for rcutorture failures" + bool "Log CPU for rcutorture failures" depends on RCU_TORTURE_TEST default n help @@ -84,6 +84,20 @@ config RCU_TORTURE_TEST_LOG_CPU Say Y here if you want CPU IDs logged. Say N if you are unsure. +config RCU_TORTURE_TEST_LOG_GP + bool "Log grace-period numbers for rcutorture failures" + depends on RCU_TORTURE_TEST + default n + help + This option causes rcutorture to decorate each entry of its + log of failure/close-call rcutorture reader segments with the + corresponding grace-period sequence numbers. This information + can be useful, but it does incur additional overhead, overhead + that can make both failures and close calls less probable. + + Say Y here if you want grace-period sequence numbers logged. + Say N if you are unsure. + config RCU_REF_SCALE_TEST tristate "Scalability tests for read-side synchronization (RCU and others)" depends on DEBUG_KERNEL @@ -199,4 +213,30 @@ config RCU_STRICT_GRACE_PERIOD when looking for certain types of RCU usage bugs, for example, too-short RCU read-side critical sections. + +config RCU_DYNTICKS_TORTURE + bool "Minimize RCU dynticks counter size" + depends on RCU_EXPERT && !COMPILE_TEST + default n + help + This option sets the width of the dynticks counter to its + minimum usable value. This minimum width greatly increases + the probability of flushing out bugs involving counter wrap, + but it also increases the probability of extending grace period + durations. This Kconfig option should therefore be avoided in + production due to the consequent increased probability of OOMs. + + This has no value for production and is only for testing. + +config TRIVIAL_PREEMPT_RCU + bool "Textbook trivial preemptible RCU in rcutorture" + depends on RCU_EXPERT && RCU_TORTURE_TEST + default n + help + This option enables a textbook preemptible RCU that is + implemented in rcutorture. Its sole purpose is to validate + code used in books, papers, and presentations. + + This has no value for production and is only for testing. + endmenu # "RCU Debugging" diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index feb3ac1dc5d5..fa6d30ce73d1 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -57,6 +57,9 @@ /* Low-order bit definition for polled grace-period APIs. */ #define RCU_GET_STATE_COMPLETED 0x1 +/* A complete grace period count */ +#define RCU_SEQ_GP (RCU_SEQ_STATE_MASK + 1) + extern int sysctl_sched_rt_runtime; /* @@ -157,12 +160,21 @@ static inline bool rcu_seq_done(unsigned long *sp, unsigned long s) * Given a snapshot from rcu_seq_snap(), determine whether or not a * full update-side operation has occurred, but do not allow the * (ULONG_MAX / 2) safety-factor/guard-band. + * + * The token returned by get_state_synchronize_rcu_full() is based on + * rcu_state.gp_seq but it is tested in poll_state_synchronize_rcu_full() + * against the root rnp->gp_seq. Since rcu_seq_start() is first called + * on rcu_state.gp_seq and only later reflected on the root rnp->gp_seq, + * it is possible that rcu_seq_snap(rcu_state.gp_seq) returns 2 full grace + * periods ahead of the root rnp->gp_seq. To prevent false-positives with the + * full polling API that a wrap around instantly completed the GP, when nothing + * like that happened, adjust for the 2 GPs in the ULONG_CMP_LT(). */ static inline bool rcu_seq_done_exact(unsigned long *sp, unsigned long s) { unsigned long cur_s = READ_ONCE(*sp); - return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (2 * RCU_SEQ_STATE_MASK + 1)); + return ULONG_CMP_GE(cur_s, s) || ULONG_CMP_LT(cur_s, s - (2 * RCU_SEQ_GP)); } /* @@ -490,6 +502,15 @@ do { \ ___locked; \ }) +#define raw_spin_trylock_irqsave_rcu_node(p, flags) \ +({ \ + bool ___locked = raw_spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ + \ + if (___locked) \ + smp_mb__after_unlock_lock(); \ + ___locked; \ +}) + #define raw_lockdep_assert_held_rcu_node(p) \ lockdep_assert_held(&ACCESS_PRIVATE(p, lock)) @@ -532,10 +553,6 @@ struct task_struct *get_rcu_tasks_rude_gp_kthread(void); void rcu_tasks_rude_get_gp_data(int *flags, unsigned long *gp_seq); #endif // # ifdef CONFIG_TASKS_RUDE_RCU -#ifdef CONFIG_TASKS_TRACE_RCU -void rcu_tasks_trace_get_gp_data(int *flags, unsigned long *gp_seq); -#endif - #ifdef CONFIG_TASKS_RCU_GENERIC void tasks_cblist_init_generic(void); #else /* #ifdef CONFIG_TASKS_RCU_GENERIC */ @@ -572,6 +589,8 @@ void do_trace_rcu_torture_read(const char *rcutorturename, unsigned long c_old, unsigned long c); void rcu_gp_set_torture_wait(int duration); +void rcu_set_gpwrap_lag(unsigned long lag); +int rcu_get_gpwrap_count(int cpu); #else static inline void rcutorture_get_gp_data(int *flags, unsigned long *gp_seq) { @@ -589,7 +608,11 @@ void do_trace_rcu_torture_read(const char *rcutorturename, do { } while (0) #endif static inline void rcu_gp_set_torture_wait(int duration) { } +static inline void rcu_set_gpwrap_lag(unsigned long lag) { } +static inline int rcu_get_gpwrap_count(int cpu) { return 0; } #endif +unsigned long long rcutorture_gather_gp_seqs(void); +void rcutorture_format_gp_seqs(unsigned long long seqs, char *cp, size_t len); #ifdef CONFIG_TINY_SRCU @@ -611,8 +634,6 @@ void srcutorture_get_gp_data(struct srcu_struct *sp, int *flags, static inline bool rcu_watching_zero_in_eqs(int cpu, int *vp) { return false; } static inline unsigned long rcu_get_gp_seq(void) { return 0; } static inline unsigned long rcu_exp_batches_completed(void) { return 0; } -static inline unsigned long -srcu_batches_completed(struct srcu_struct *sp) { return 0; } static inline void rcu_force_quiescent_state(void) { } static inline bool rcu_check_boost_fail(unsigned long gp_state, int *cpup) { return true; } static inline void show_rcu_gp_kthreads(void) { } @@ -624,7 +645,6 @@ static inline void rcu_gp_slow_unregister(atomic_t *rgssp) { } bool rcu_watching_zero_in_eqs(int cpu, int *vp); unsigned long rcu_get_gp_seq(void); unsigned long rcu_exp_batches_completed(void); -unsigned long srcu_batches_completed(struct srcu_struct *sp); bool rcu_check_boost_fail(unsigned long gp_state, int *cpup); void show_rcu_gp_kthreads(void); int rcu_get_gp_kthreads_prio(void); @@ -636,6 +656,12 @@ void rcu_gp_slow_register(atomic_t *rgssp); void rcu_gp_slow_unregister(atomic_t *rgssp); #endif /* #else #ifdef CONFIG_TINY_RCU */ +#ifdef CONFIG_TINY_SRCU +static inline unsigned long srcu_batches_completed(struct srcu_struct *sp) { return 0; } +#else // #ifdef CONFIG_TINY_SRCU +unsigned long srcu_batches_completed(struct srcu_struct *sp); +#endif // #else // #ifdef CONFIG_TINY_SRCU + #ifdef CONFIG_RCU_NOCB_CPU void rcu_bind_current_to_nocb(void); #else @@ -652,11 +678,6 @@ void show_rcu_tasks_rude_gp_kthread(void); #else static inline void show_rcu_tasks_rude_gp_kthread(void) {} #endif -#if !defined(CONFIG_TINY_RCU) && defined(CONFIG_TASKS_TRACE_RCU) -void show_rcu_tasks_trace_gp_kthread(void); -#else -static inline void show_rcu_tasks_trace_gp_kthread(void) {} -#endif #ifdef CONFIG_TINY_RCU static inline bool rcu_cpu_beenfullyonline(int cpu) { return true; } @@ -670,4 +691,8 @@ int rcu_stall_notifier_call_chain(unsigned long val, void *v); static inline int rcu_stall_notifier_call_chain(unsigned long val, void *v) { return NOTIFY_DONE; } #endif // #else // #if defined(CONFIG_RCU_STALL_COMMON) && defined(CONFIG_RCU_CPU_STALL_NOTIFIER) +#ifdef CONFIG_TRIVIAL_PREEMPT_RCU +void synchronize_rcu_trivial_preempt(void); +#endif // #ifdef CONFIG_TRIVIAL_PREEMPT_RCU + #endif /* __LINUX_RCU_H */ diff --git a/kernel/rcu/rcuscale.c b/kernel/rcu/rcuscale.c index 0f3059b1b80d..ac0b1c6b7dae 100644 --- a/kernel/rcu/rcuscale.c +++ b/kernel/rcu/rcuscale.c @@ -79,12 +79,6 @@ MODULE_AUTHOR("Paul E. McKenney <paulmck@linux.ibm.com>"); * test-end checks, and the pair of calls through pointers. */ -#ifdef MODULE -# define RCUSCALE_SHUTDOWN 0 -#else -# define RCUSCALE_SHUTDOWN 1 -#endif - torture_param(bool, gp_async, false, "Use asynchronous GP wait primitives"); torture_param(int, gp_async_max, 1000, "Max # outstanding waits per writer"); torture_param(bool, gp_exp, false, "Use expedited GP wait primitives"); @@ -92,8 +86,8 @@ torture_param(int, holdoff, 10, "Holdoff time before test start (s)"); torture_param(int, minruntime, 0, "Minimum run time (s)"); torture_param(int, nreaders, -1, "Number of RCU reader threads"); torture_param(int, nwriters, -1, "Number of RCU updater threads"); -torture_param(bool, shutdown, RCUSCALE_SHUTDOWN, - "Shutdown at end of scalability tests."); +torture_param(int, shutdown_secs, !IS_MODULE(CONFIG_RCU_SCALE_TEST) * 300, + "Shutdown at end of scalability tests or at specified timeout (s)."); torture_param(int, verbose, 1, "Enable verbose debugging printk()s"); torture_param(int, writer_holdoff, 0, "Holdoff (us) between GPs, zero to disable"); torture_param(int, writer_holdoff_jiffies, 0, "Holdoff (jiffies) between GPs, zero to disable"); @@ -123,7 +117,6 @@ static int nrealreaders; static int nrealwriters; static struct task_struct **writer_tasks; static struct task_struct **reader_tasks; -static struct task_struct *shutdown_task; static u64 **writer_durations; static bool *writer_done; @@ -132,7 +125,6 @@ static int *writer_n_durations; static atomic_t n_rcu_scale_reader_started; static atomic_t n_rcu_scale_writer_started; static atomic_t n_rcu_scale_writer_finished; -static wait_queue_head_t shutdown_wq; static u64 t_rcu_scale_writer_started; static u64 t_rcu_scale_writer_finished; static unsigned long b_rcu_gp_test_started; @@ -400,11 +392,6 @@ static void tasks_trace_scale_read_unlock(int idx) rcu_read_unlock_trace(); } -static void rcu_tasks_trace_scale_stats(void) -{ - rcu_tasks_trace_torture_stats_print(scale_type, SCALE_FLAG); -} - static struct rcu_scale_ops tasks_tracing_ops = { .ptype = RCU_TASKS_FLAVOR, .init = rcu_sync_scale_init, @@ -416,8 +403,6 @@ static struct rcu_scale_ops tasks_tracing_ops = { .gp_barrier = rcu_barrier_tasks_trace, .sync = synchronize_rcu_tasks_trace, .exp_sync = synchronize_rcu_tasks_trace, - .rso_gp_kthread = get_rcu_tasks_trace_gp_kthread, - .stats = IS_ENABLED(CONFIG_TINY_RCU) ? NULL : rcu_tasks_trace_scale_stats, .name = "tasks-tracing" }; @@ -526,6 +511,8 @@ static void rcu_scale_async_cb(struct rcu_head *rhp) rcu_scale_free(wmbp); } +static void rcu_scale_cleanup(void); + /* * RCU scale writer kthread. Repeatedly does a grace period. */ @@ -629,9 +616,11 @@ rcu_scale_writer(void *arg) b_rcu_gp_test_finished = cur_ops->get_gp_seq(); } - if (shutdown) { + if (shutdown_secs) { + writer_tasks[me] = NULL; smp_mb(); /* Assign before wake. */ - wake_up(&shutdown_wq); + rcu_scale_cleanup(); + kernel_power_off(); } } } @@ -675,8 +664,8 @@ static void rcu_scale_print_module_parms(struct rcu_scale_ops *cur_ops, const char *tag) { pr_alert("%s" SCALE_FLAG - "--- %s: gp_async=%d gp_async_max=%d gp_exp=%d holdoff=%d minruntime=%d nreaders=%d nwriters=%d writer_holdoff=%d writer_holdoff_jiffies=%d verbose=%d shutdown=%d\n", - scale_type, tag, gp_async, gp_async_max, gp_exp, holdoff, minruntime, nrealreaders, nrealwriters, writer_holdoff, writer_holdoff_jiffies, verbose, shutdown); + "--- %s: gp_async=%d gp_async_max=%d gp_exp=%d holdoff=%d minruntime=%d nreaders=%d nwriters=%d writer_holdoff=%d writer_holdoff_jiffies=%d verbose=%d shutdown_secs=%d\n", + scale_type, tag, gp_async, gp_async_max, gp_exp, holdoff, minruntime, nrealreaders, nrealwriters, writer_holdoff, writer_holdoff_jiffies, verbose, shutdown_secs); } /* @@ -729,6 +718,8 @@ static void kfree_call_rcu(struct rcu_head *rh) kfree(obj); } +static void kfree_scale_cleanup(void); + static int kfree_scale_thread(void *arg) { @@ -762,7 +753,7 @@ kfree_scale_thread(void *arg) } for (i = 0; i < kfree_alloc_num; i++) { - alloc_ptr = kmalloc(kfree_mult * sizeof(struct kfree_obj), GFP_KERNEL); + alloc_ptr = kzalloc_objs(struct kfree_obj, kfree_mult); if (!alloc_ptr) return -ENOMEM; @@ -796,11 +787,13 @@ kfree_scale_thread(void *arg) pr_alert("Total time taken by all kfree'ers: %llu ns, loops: %d, batches: %ld, memory footprint: %lldMB\n", (unsigned long long)(end_time - start_time), kfree_loops, rcuscale_seq_diff(b_rcu_gp_test_finished, b_rcu_gp_test_started), - (mem_begin - mem_during) >> (20 - PAGE_SHIFT)); + PAGES_TO_MB(mem_begin - mem_during)); - if (shutdown) { + if (shutdown_secs) { + kfree_reader_tasks[me] = NULL; smp_mb(); /* Assign before wake. */ - wake_up(&shutdown_wq); + kfree_scale_cleanup(); + kernel_power_off(); } } @@ -827,22 +820,6 @@ kfree_scale_cleanup(void) torture_cleanup_end(); } -/* - * shutdown kthread. Just waits to be awakened, then shuts down system. - */ -static int -kfree_scale_shutdown(void *arg) -{ - wait_event_idle(shutdown_wq, - atomic_read(&n_kfree_scale_thread_ended) >= kfree_nrealthreads); - - smp_mb(); /* Wake before output. */ - - kfree_scale_cleanup(); - kernel_power_off(); - return -EINVAL; -} - // Used if doing RCU-kfree'ing via call_rcu(). static unsigned long jiffies_at_lazy_cb; static struct rcu_head lazy_test1_rh; @@ -902,21 +879,18 @@ kfree_scale_init(void) kfree_nrealthreads = compute_real(kfree_nthreads); /* Start up the kthreads. */ - if (shutdown) { - init_waitqueue_head(&shutdown_wq); - firsterr = torture_create_kthread(kfree_scale_shutdown, NULL, - shutdown_task); + if (shutdown_secs) { + firsterr = torture_shutdown_init(shutdown_secs, kfree_scale_cleanup); if (torture_init_error(firsterr)) goto unwind; - schedule_timeout_uninterruptible(1); } pr_alert("kfree object size=%zu, kfree_by_call_rcu=%d\n", kfree_mult * sizeof(struct kfree_obj), kfree_by_call_rcu); - kfree_reader_tasks = kcalloc(kfree_nrealthreads, sizeof(kfree_reader_tasks[0]), - GFP_KERNEL); + kfree_reader_tasks = kzalloc_objs(kfree_reader_tasks[0], + kfree_nrealthreads); if (kfree_reader_tasks == NULL) { firsterr = -ENOMEM; goto unwind; @@ -1065,20 +1039,6 @@ rcu_scale_cleanup(void) torture_cleanup_end(); } -/* - * RCU scalability shutdown kthread. Just waits to be awakened, then shuts - * down system. - */ -static int -rcu_scale_shutdown(void *arg) -{ - wait_event_idle(shutdown_wq, atomic_read(&n_rcu_scale_writer_finished) >= nrealwriters); - smp_mb(); /* Wake before output. */ - rcu_scale_cleanup(); - kernel_power_off(); - return -EINVAL; -} - static int __init rcu_scale_init(void) { @@ -1128,16 +1088,12 @@ rcu_scale_init(void) /* Start up the kthreads. */ - if (shutdown) { - init_waitqueue_head(&shutdown_wq); - firsterr = torture_create_kthread(rcu_scale_shutdown, NULL, - shutdown_task); + if (shutdown_secs) { + firsterr = torture_shutdown_init(shutdown_secs, rcu_scale_cleanup); if (torture_init_error(firsterr)) goto unwind; - schedule_timeout_uninterruptible(1); } - reader_tasks = kcalloc(nrealreaders, sizeof(reader_tasks[0]), - GFP_KERNEL); + reader_tasks = kzalloc_objs(reader_tasks[0], nrealreaders); if (reader_tasks == NULL) { SCALEOUT_ERRSTRING("out of memory"); firsterr = -ENOMEM; @@ -1151,10 +1107,10 @@ rcu_scale_init(void) } while (atomic_read(&n_rcu_scale_reader_started) < nrealreaders) schedule_timeout_uninterruptible(1); - writer_tasks = kcalloc(nrealwriters, sizeof(writer_tasks[0]), GFP_KERNEL); + writer_tasks = kzalloc_objs(writer_tasks[0], nrealwriters); writer_durations = kcalloc(nrealwriters, sizeof(*writer_durations), GFP_KERNEL); - writer_n_durations = kcalloc(nrealwriters, sizeof(*writer_n_durations), GFP_KERNEL); - writer_done = kcalloc(nrealwriters, sizeof(writer_done[0]), GFP_KERNEL); + writer_n_durations = kzalloc_objs(*writer_n_durations, nrealwriters); + writer_done = kzalloc_objs(writer_done[0], nrealwriters); if (gp_async) { if (gp_async_max <= 0) { pr_warn("%s: gp_async_max = %d must be greater than zero.\n", @@ -1163,7 +1119,8 @@ rcu_scale_init(void) firsterr = -EINVAL; goto unwind; } - writer_freelists = kcalloc(nrealwriters, sizeof(writer_freelists[0]), GFP_KERNEL); + writer_freelists = kzalloc_objs(writer_freelists[0], + nrealwriters); } if (!writer_tasks || !writer_durations || !writer_n_durations || !writer_done || (gp_async && !writer_freelists)) { @@ -1184,8 +1141,8 @@ rcu_scale_init(void) init_llist_head(&wflp->ws_lhg); init_llist_head(&wflp->ws_lhp); - wflp->ws_mblocks = kcalloc(gp_async_max, sizeof(wflp->ws_mblocks[0]), - GFP_KERNEL); + wflp->ws_mblocks = kzalloc_objs(wflp->ws_mblocks[0], + gp_async_max); if (!wflp->ws_mblocks) { firsterr = -ENOMEM; goto unwind; @@ -1208,7 +1165,7 @@ rcu_scale_init(void) unwind: torture_init_end(); rcu_scale_cleanup(); - if (shutdown) { + if (shutdown_secs) { WARN_ON(!IS_MODULE(CONFIG_RCU_SCALE_TEST)); kernel_power_off(); } diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index d26fb1d33ed9..5f2848b828dc 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -55,22 +55,24 @@ MODULE_DESCRIPTION("Read-Copy Update module-based torture test facility"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Paul E. McKenney <paulmck@linux.ibm.com> and Josh Triplett <josh@joshtriplett.org>"); -/* Bits for ->extendables field, extendables param, and related definitions. */ -#define RCUTORTURE_RDR_SHIFT_1 8 /* Put SRCU index in upper bits. */ -#define RCUTORTURE_RDR_MASK_1 (0xff << RCUTORTURE_RDR_SHIFT_1) -#define RCUTORTURE_RDR_SHIFT_2 16 /* Put SRCU index in upper bits. */ -#define RCUTORTURE_RDR_MASK_2 (0xff << RCUTORTURE_RDR_SHIFT_2) -#define RCUTORTURE_RDR_BH 0x01 /* Extend readers by disabling bh. */ -#define RCUTORTURE_RDR_IRQ 0x02 /* ... disabling interrupts. */ -#define RCUTORTURE_RDR_PREEMPT 0x04 /* ... disabling preemption. */ -#define RCUTORTURE_RDR_RBH 0x08 /* ... rcu_read_lock_bh(). */ -#define RCUTORTURE_RDR_SCHED 0x10 /* ... rcu_read_lock_sched(). */ -#define RCUTORTURE_RDR_RCU_1 0x20 /* ... entering another RCU reader. */ -#define RCUTORTURE_RDR_RCU_2 0x40 /* ... entering another RCU reader. */ -#define RCUTORTURE_RDR_NBITS 7 /* Number of bits defined above. */ -#define RCUTORTURE_MAX_EXTEND \ +// Bits for ->extendables field, extendables param, and related definitions. +#define RCUTORTURE_RDR_SHIFT_1 8 // Put SRCU index in upper bits. +#define RCUTORTURE_RDR_MASK_1 (0xff << RCUTORTURE_RDR_SHIFT_1) +#define RCUTORTURE_RDR_SHIFT_2 16 // Put SRCU index in upper bits. +#define RCUTORTURE_RDR_MASK_2 (0xff << RCUTORTURE_RDR_SHIFT_2) +#define RCUTORTURE_RDR_BH 0x01 // Extend readers by disabling bh. +#define RCUTORTURE_RDR_IRQ 0x02 // ... disabling interrupts. +#define RCUTORTURE_RDR_PREEMPT 0x04 // ... disabling preemption. +#define RCUTORTURE_RDR_RBH 0x08 // ... rcu_read_lock_bh(). +#define RCUTORTURE_RDR_SCHED 0x10 // ... rcu_read_lock_sched(). +#define RCUTORTURE_RDR_RCU_1 0x20 // ... entering another RCU reader. +#define RCUTORTURE_RDR_RCU_2 0x40 // ... entering another RCU reader. +#define RCUTORTURE_RDR_UPDOWN 0x80 // ... up-read from task, down-read from timer. + // Note: Manual start, automatic end. +#define RCUTORTURE_RDR_NBITS 8 // Number of bits defined above. +#define RCUTORTURE_MAX_EXTEND \ (RCUTORTURE_RDR_BH | RCUTORTURE_RDR_IRQ | RCUTORTURE_RDR_PREEMPT | \ - RCUTORTURE_RDR_RBH | RCUTORTURE_RDR_SCHED) + RCUTORTURE_RDR_RBH | RCUTORTURE_RDR_SCHED) // Intentionally omit RCUTORTURE_RDR_UPDOWN. #define RCUTORTURE_RDR_ALLBITS \ (RCUTORTURE_MAX_EXTEND | RCUTORTURE_RDR_RCU_1 | RCUTORTURE_RDR_RCU_2 | \ RCUTORTURE_RDR_MASK_1 | RCUTORTURE_RDR_MASK_2) @@ -110,11 +112,16 @@ torture_param(bool, gp_sync, false, "Use synchronous GP wait primitives"); torture_param(int, irqreader, 1, "Allow RCU readers from irq handlers"); torture_param(int, leakpointer, 0, "Leak pointer dereferences from readers"); torture_param(int, n_barrier_cbs, 0, "# of callbacks/kthreads for barrier testing"); +torture_param(int, n_up_down, 32, "# of concurrent up/down hrtimer-based RCU readers"); torture_param(int, nfakewriters, 4, "Number of RCU fake writer threads"); torture_param(int, nreaders, -1, "Number of RCU reader threads"); torture_param(int, object_debug, 0, "Enable debug-object double call_rcu() testing"); torture_param(int, onoff_holdoff, 0, "Time after boot before CPU hotplugs (s)"); torture_param(int, onoff_interval, 0, "Time between CPU hotplugs (jiffies), 0=disable"); +torture_param(bool, gpwrap_lag, true, "Enable grace-period wrap lag testing"); +torture_param(int, gpwrap_lag_gps, 8, "Value to set for set_gpwrap_lag during an active testing period."); +torture_param(int, gpwrap_lag_cycle_mins, 30, "Total cycle duration for gpwrap lag testing (in minutes)"); +torture_param(int, gpwrap_lag_active_mins, 5, "Duration for which gpwrap lag is active within each cycle (in minutes)"); torture_param(int, nocbs_nthreads, 0, "Number of NOCB toggle threads, 0 to disable"); torture_param(int, nocbs_toggle, 1000, "Time between toggling nocb state (ms)"); torture_param(int, preempt_duration, 0, "Preemption duration (ms), zero to disable"); @@ -135,6 +142,7 @@ torture_param(int, stat_interval, 60, "Number of seconds between stats printk()s torture_param(int, stutter, 5, "Number of seconds to run/halt test"); torture_param(int, test_boost, 1, "Test RCU prio boost: 0=no, 1=maybe, 2=yes."); torture_param(int, test_boost_duration, 4, "Duration of each boost test, seconds."); +torture_param(int, test_boost_holdoff, 0, "Holdoff time from rcutorture start, seconds."); torture_param(int, test_boost_interval, 7, "Interval between boost tests, seconds."); torture_param(int, test_nmis, 0, "End-test NMI tests, 0 to disable."); torture_param(bool, test_no_idle_hz, true, "Test support for tickless idle CPUs"); @@ -147,9 +155,11 @@ MODULE_PARM_DESC(torture_type, "Type of RCU to torture (rcu, srcu, ...)"); static int nrealnocbers; static int nrealreaders; +static int nrealfakewriters; static struct task_struct *writer_task; static struct task_struct **fakewriter_tasks; static struct task_struct **reader_tasks; +static struct task_struct *updown_task; static struct task_struct **nocb_tasks; static struct task_struct *stats_task; static struct task_struct *fqs_task; @@ -272,6 +282,9 @@ struct rt_read_seg { bool rt_preempted; int rt_cpu; int rt_end_cpu; + unsigned long long rt_gp_seq; + unsigned long long rt_gp_seq_end; + u64 rt_ts; }; static int err_segs_recorded; static struct rt_read_seg err_segs[RCUTORTURE_RDR_MAX_SEGS]; @@ -369,11 +382,14 @@ struct rcu_torture_ops { void (*readunlock)(int idx); int (*readlock_held)(void); // lockdep. int (*readlock_nesting)(void); // actual nesting, if available, -1 if not. + int (*down_read)(void); + void (*up_read)(int idx); unsigned long (*get_gp_seq)(void); unsigned long (*gp_diff)(unsigned long new, unsigned long old); void (*deferred_free)(struct rcu_torture *p); void (*sync)(void); void (*exp_sync)(void); + void (*exp_current)(void); unsigned long (*get_gp_state_exp)(void); unsigned long (*start_gp_poll_exp)(void); void (*start_gp_poll_exp_full)(struct rcu_gp_oldstate *rgosp); @@ -406,6 +422,10 @@ struct rcu_torture_ops { void (*gp_slow_register)(atomic_t *rgssp); void (*gp_slow_unregister)(atomic_t *rgssp); bool (*reader_blocked)(void); + unsigned long long (*gather_gp_seqs)(void); + void (*format_gp_seqs)(unsigned long long seqs, char *cp, size_t len); + void (*set_gpwrap_lag)(unsigned long lag); + int (*get_gpwrap_count)(int cpu); long cbflood_max; int irq_capable; int can_boost; @@ -414,6 +434,7 @@ struct rcu_torture_ops { int no_pi_lock; int debug_objects; int start_poll_irqsoff; + int have_up_down; const char *name; }; @@ -451,7 +472,7 @@ rcu_read_delay(struct torture_random_state *rrsp, struct rt_read_seg *rtrsp) !(torture_random(rrsp) % (nrealreaders * 2000 * longdelay_ms))) { started = cur_ops->get_gp_seq(); ts = rcu_trace_clock_local(); - if (preempt_count() & (SOFTIRQ_MASK | HARDIRQ_MASK)) + if ((preempt_count() & HARDIRQ_MASK) || softirq_count()) longdelay_ms = 5; /* Avoid triggering BH limits. */ mdelay(longdelay_ms); rtrsp->rt_delay_ms = longdelay_ms; @@ -610,6 +631,10 @@ static struct rcu_torture_ops rcu_ops = { .reader_blocked = IS_ENABLED(CONFIG_RCU_TORTURE_TEST_LOG_CPU) ? has_rcu_reader_blocked : NULL, + .gather_gp_seqs = rcutorture_gather_gp_seqs, + .format_gp_seqs = rcutorture_format_gp_seqs, + .set_gpwrap_lag = rcu_set_gpwrap_lag, + .get_gpwrap_count = rcu_get_gpwrap_count, .irq_capable = 1, .can_boost = IS_ENABLED(CONFIG_RCU_BOOST), .extendables = RCUTORTURE_MAX_EXTEND, @@ -655,6 +680,8 @@ static struct rcu_torture_ops rcu_busted_ops = { .sync = synchronize_rcu_busted, .exp_sync = synchronize_rcu_busted, .call = call_rcu_busted, + .gather_gp_seqs = rcutorture_gather_gp_seqs, + .format_gp_seqs = rcutorture_format_gp_seqs, .irq_capable = 1, .extendables = RCUTORTURE_MAX_EXTEND, .name = "busted" @@ -665,10 +692,29 @@ static struct rcu_torture_ops rcu_busted_ops = { */ DEFINE_STATIC_SRCU(srcu_ctl); +DEFINE_STATIC_SRCU_FAST(srcu_ctlf); +DEFINE_STATIC_SRCU_FAST_UPDOWN(srcu_ctlfud); static struct srcu_struct srcu_ctld; static struct srcu_struct *srcu_ctlp = &srcu_ctl; static struct rcu_torture_ops srcud_ops; +static void srcu_torture_init(void) +{ + rcu_sync_torture_init(); + if (!reader_flavor || (reader_flavor & SRCU_READ_FLAVOR_NORMAL)) + VERBOSE_TOROUT_STRING("srcu_torture_init normal SRCU"); + if (reader_flavor & SRCU_READ_FLAVOR_NMI) + VERBOSE_TOROUT_STRING("srcu_torture_init NMI-safe SRCU"); + if (reader_flavor & SRCU_READ_FLAVOR_FAST) { + srcu_ctlp = &srcu_ctlf; + VERBOSE_TOROUT_STRING("srcu_torture_init fast SRCU"); + } + if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) { + srcu_ctlp = &srcu_ctlfud; + VERBOSE_TOROUT_STRING("srcu_torture_init fast-up/down SRCU"); + } +} + static void srcu_get_gp_data(int *flags, unsigned long *gp_seq) { srcutorture_get_gp_data(srcu_ctlp, flags, gp_seq); @@ -677,8 +723,11 @@ static void srcu_get_gp_data(int *flags, unsigned long *gp_seq) static int srcu_torture_read_lock(void) { int idx; + struct srcu_ctr __percpu *scp; int ret = 0; + WARN_ON_ONCE(reader_flavor & ~SRCU_READ_FLAVOR_ALL); + if ((reader_flavor & SRCU_READ_FLAVOR_NORMAL) || !(reader_flavor & SRCU_READ_FLAVOR_ALL)) { idx = srcu_read_lock(srcu_ctlp); WARN_ON_ONCE(idx & ~0x1); @@ -689,11 +738,18 @@ static int srcu_torture_read_lock(void) WARN_ON_ONCE(idx & ~0x1); ret += idx << 1; } - if (reader_flavor & SRCU_READ_FLAVOR_LITE) { - idx = srcu_read_lock_lite(srcu_ctlp); + if (reader_flavor & SRCU_READ_FLAVOR_FAST) { + scp = srcu_read_lock_fast(srcu_ctlp); + idx = __srcu_ptr_to_ctr(srcu_ctlp, scp); WARN_ON_ONCE(idx & ~0x1); ret += idx << 2; } + if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) { + scp = srcu_read_lock_fast_updown(srcu_ctlp); + idx = __srcu_ptr_to_ctr(srcu_ctlp, scp); + WARN_ON_ONCE(idx & ~0x1); + ret += idx << 3; + } return ret; } @@ -719,8 +775,11 @@ srcu_read_delay(struct torture_random_state *rrsp, struct rt_read_seg *rtrsp) static void srcu_torture_read_unlock(int idx) { WARN_ON_ONCE((reader_flavor && (idx & ~reader_flavor)) || (!reader_flavor && (idx & ~0x1))); - if (reader_flavor & SRCU_READ_FLAVOR_LITE) - srcu_read_unlock_lite(srcu_ctlp, (idx & 0x4) >> 2); + if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) + srcu_read_unlock_fast_updown(srcu_ctlp, + __srcu_ctr_to_ptr(srcu_ctlp, (idx & 0x8) >> 3)); + if (reader_flavor & SRCU_READ_FLAVOR_FAST) + srcu_read_unlock_fast(srcu_ctlp, __srcu_ctr_to_ptr(srcu_ctlp, (idx & 0x4) >> 2)); if (reader_flavor & SRCU_READ_FLAVOR_NMI) srcu_read_unlock_nmisafe(srcu_ctlp, (idx & 0x2) >> 1); if ((reader_flavor & SRCU_READ_FLAVOR_NORMAL) || !(reader_flavor & SRCU_READ_FLAVOR_ALL)) @@ -732,6 +791,50 @@ static int torture_srcu_read_lock_held(void) return srcu_read_lock_held(srcu_ctlp); } +static bool srcu_torture_have_up_down(void) +{ + int rf = reader_flavor; + + if (!rf) + rf = SRCU_READ_FLAVOR_NORMAL; + return !!(cur_ops->have_up_down & rf); +} + +static int srcu_torture_down_read(void) +{ + int idx; + struct srcu_ctr __percpu *scp; + + WARN_ON_ONCE(reader_flavor & ~SRCU_READ_FLAVOR_ALL); + WARN_ON_ONCE(reader_flavor & (reader_flavor - 1)); + + if ((reader_flavor & SRCU_READ_FLAVOR_NORMAL) || !(reader_flavor & SRCU_READ_FLAVOR_ALL)) { + idx = srcu_down_read(srcu_ctlp); + WARN_ON_ONCE(idx & ~0x1); + return idx; + } + if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) { + scp = srcu_down_read_fast(srcu_ctlp); + idx = __srcu_ptr_to_ctr(srcu_ctlp, scp); + WARN_ON_ONCE(idx & ~0x1); + return idx << 3; + } + WARN_ON_ONCE(1); + return 0; +} + +static void srcu_torture_up_read(int idx) +{ + WARN_ON_ONCE((reader_flavor && (idx & ~reader_flavor)) || (!reader_flavor && (idx & ~0x1))); + if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) + srcu_up_read_fast(srcu_ctlp, __srcu_ctr_to_ptr(srcu_ctlp, (idx & 0x8) >> 3)); + else if ((reader_flavor & SRCU_READ_FLAVOR_NORMAL) || + !(reader_flavor & SRCU_READ_FLAVOR_ALL)) + srcu_up_read(srcu_ctlp, idx & 0x1); + else + WARN_ON_ONCE(1); +} + static unsigned long srcu_torture_completed(void) { return srcu_batches_completed(srcu_ctlp); @@ -739,7 +842,14 @@ static unsigned long srcu_torture_completed(void) static void srcu_torture_deferred_free(struct rcu_torture *rp) { + unsigned long flags; + bool lockit = jiffies & 0x1; + + if (lockit) + raw_spin_lock_irqsave(¤t->pi_lock, flags); call_srcu(srcu_ctlp, &rp->rtort_rcu, rcu_torture_cb); + if (lockit) + raw_spin_unlock_irqrestore(¤t->pi_lock, flags); } static void srcu_torture_synchronize(void) @@ -783,17 +893,26 @@ static void srcu_torture_synchronize_expedited(void) synchronize_srcu_expedited(srcu_ctlp); } +static void srcu_torture_expedite_current(void) +{ + srcu_expedite_current(srcu_ctlp); +} + static struct rcu_torture_ops srcu_ops = { .ttype = SRCU_FLAVOR, - .init = rcu_sync_torture_init, + .init = srcu_torture_init, .readlock = srcu_torture_read_lock, .read_delay = srcu_read_delay, .readunlock = srcu_torture_read_unlock, + .down_read = srcu_torture_down_read, + .up_read = srcu_torture_up_read, .readlock_held = torture_srcu_read_lock_held, .get_gp_seq = srcu_torture_completed, + .gp_diff = rcu_seq_diff, .deferred_free = srcu_torture_deferred_free, .sync = srcu_torture_synchronize, .exp_sync = srcu_torture_synchronize_expedited, + .exp_current = srcu_torture_expedite_current, .same_gp_state = same_state_synchronize_srcu, .get_comp_state = get_completed_synchronize_srcu, .get_gp_state = srcu_torture_get_gp_state, @@ -808,13 +927,29 @@ static struct rcu_torture_ops srcu_ops = { .irq_capable = 1, .no_pi_lock = IS_ENABLED(CONFIG_TINY_SRCU), .debug_objects = 1, + .have_up_down = IS_ENABLED(CONFIG_TINY_SRCU) + ? 0 : SRCU_READ_FLAVOR_NORMAL | SRCU_READ_FLAVOR_FAST_UPDOWN, .name = "srcu" }; -static void srcu_torture_init(void) +static void srcud_torture_init(void) { rcu_sync_torture_init(); - WARN_ON(init_srcu_struct(&srcu_ctld)); + if (!reader_flavor || (reader_flavor & SRCU_READ_FLAVOR_NORMAL)) { + WARN_ON(init_srcu_struct(&srcu_ctld)); + VERBOSE_TOROUT_STRING("srcud_torture_init normal SRCU"); + } else if (reader_flavor & SRCU_READ_FLAVOR_NMI) { + WARN_ON(init_srcu_struct(&srcu_ctld)); + VERBOSE_TOROUT_STRING("srcud_torture_init NMI-safe SRCU"); + } else if (reader_flavor & SRCU_READ_FLAVOR_FAST) { + WARN_ON(init_srcu_struct_fast(&srcu_ctld)); + VERBOSE_TOROUT_STRING("srcud_torture_init fast SRCU"); + } else if (reader_flavor & SRCU_READ_FLAVOR_FAST_UPDOWN) { + WARN_ON(init_srcu_struct_fast_updown(&srcu_ctld)); + VERBOSE_TOROUT_STRING("srcud_torture_init fast-up/down SRCU"); + } else { + WARN_ON(init_srcu_struct(&srcu_ctld)); + } srcu_ctlp = &srcu_ctld; } @@ -827,16 +962,20 @@ static void srcu_torture_cleanup(void) /* As above, but dynamically allocated. */ static struct rcu_torture_ops srcud_ops = { .ttype = SRCU_FLAVOR, - .init = srcu_torture_init, + .init = srcud_torture_init, .cleanup = srcu_torture_cleanup, .readlock = srcu_torture_read_lock, .read_delay = srcu_read_delay, .readunlock = srcu_torture_read_unlock, .readlock_held = torture_srcu_read_lock_held, + .down_read = srcu_torture_down_read, + .up_read = srcu_torture_up_read, .get_gp_seq = srcu_torture_completed, + .gp_diff = rcu_seq_diff, .deferred_free = srcu_torture_deferred_free, .sync = srcu_torture_synchronize, .exp_sync = srcu_torture_synchronize_expedited, + .exp_current = srcu_torture_expedite_current, .same_gp_state = same_state_synchronize_srcu, .get_comp_state = get_completed_synchronize_srcu, .get_gp_state = srcu_torture_get_gp_state, @@ -851,6 +990,8 @@ static struct rcu_torture_ops srcud_ops = { .irq_capable = 1, .no_pi_lock = IS_ENABLED(CONFIG_TINY_SRCU), .debug_objects = 1, + .have_up_down = IS_ENABLED(CONFIG_TINY_SRCU) + ? 0 : SRCU_READ_FLAVOR_NORMAL | SRCU_READ_FLAVOR_FAST_UPDOWN, .name = "srcud" }; @@ -878,7 +1019,8 @@ static struct rcu_torture_ops busted_srcud_ops = { /* * Definitions for trivial CONFIG_PREEMPT=n-only torture testing. - * This implementation does not necessarily work well with CPU hotplug. + * This implementation does not work well with CPU hotplug nor + * with rcutorture's shuffling. */ static void synchronize_rcu_trivial(void) @@ -891,6 +1033,16 @@ static void synchronize_rcu_trivial(void) } } +static void rcu_sync_torture_init_trivial(void) +{ + rcu_sync_torture_init(); + // if (onoff_interval || shuffle_interval) { + if (WARN_ONCE(onoff_interval || shuffle_interval, "%s: Non-zero onoff_interval (%d) or shuffle_interval (%d) breaks trivial RCU, resetting to zero", __func__, onoff_interval, shuffle_interval)) { + onoff_interval = 0; + shuffle_interval = 0; + } +} + static int rcu_torture_read_lock_trivial(void) { preempt_disable(); @@ -904,7 +1056,7 @@ static void rcu_torture_read_unlock_trivial(int idx) static struct rcu_torture_ops trivial_ops = { .ttype = RCU_TRIVIAL_FLAVOR, - .init = rcu_sync_torture_init, + .init = rcu_sync_torture_init_trivial, .readlock = rcu_torture_read_lock_trivial, .read_delay = rcu_read_delay, /* just reuse rcu's version. */ .readunlock = rcu_torture_read_unlock_trivial, @@ -916,6 +1068,61 @@ static struct rcu_torture_ops trivial_ops = { .name = "trivial" }; +#ifdef CONFIG_TRIVIAL_PREEMPT_RCU + +/* + * Definitions for trivial CONFIG_PREEMPT=y torture testing. This + * implementation does not work well with large numbers of tasks or with + * long-term preemption. Either or both get you RCU CPU stall warnings. + */ + +static void rcu_sync_torture_init_trivial_preempt(void) +{ + rcu_sync_torture_init(); + if (WARN_ONCE(onoff_interval || shuffle_interval, "%s: Non-zero onoff_interval (%d) or shuffle_interval (%d) breaks trivial RCU, resetting to zero", __func__, onoff_interval, shuffle_interval)) { + onoff_interval = 0; + shuffle_interval = 0; + } +} + +static int rcu_torture_read_lock_trivial_preempt(void) +{ + struct task_struct *t = current; + + WRITE_ONCE(t->rcu_trivial_preempt_nesting, t->rcu_trivial_preempt_nesting + 1); + smp_mb(); + return 0; +} + +static void rcu_torture_read_unlock_trivial_preempt(int idx) +{ + struct task_struct *t = current; + + smp_store_release(&t->rcu_trivial_preempt_nesting, t->rcu_trivial_preempt_nesting - 1); +} + +static struct rcu_torture_ops trivial_preempt_ops = { + .ttype = RCU_TRIVIAL_FLAVOR, + .init = rcu_sync_torture_init_trivial_preempt, + .readlock = rcu_torture_read_lock_trivial_preempt, + .read_delay = rcu_read_delay, // just reuse rcu's version. + .readunlock = rcu_torture_read_unlock_trivial_preempt, + .readlock_held = torture_readlock_not_held, + .get_gp_seq = rcu_no_completed, + .sync = synchronize_rcu_trivial_preempt, + .exp_sync = synchronize_rcu_trivial_preempt, + .irq_capable = 0, // In theory it should be, but let's keep it trivial. + .name = "trivial-preempt" +}; + +#define TRIVIAL_PREEMPT_OPS &trivial_preempt_ops, + +#else // #ifdef CONFIG_TRIVIAL_PREEMPT_RCU + +#define TRIVIAL_PREEMPT_OPS + +#endif // #else // #ifdef CONFIG_TRIVIAL_PREEMPT_RCU + #ifdef CONFIG_TASKS_RCU /* @@ -1033,10 +1240,9 @@ static struct rcu_torture_ops tasks_tracing_ops = { .deferred_free = rcu_tasks_tracing_torture_deferred_free, .sync = synchronize_rcu_tasks_trace, .exp_sync = synchronize_rcu_tasks_trace, + .exp_current = rcu_tasks_trace_expedite_current, .call = call_rcu_tasks_trace, .cb_barrier = rcu_barrier_tasks_trace, - .gp_kthread_dbg = show_rcu_tasks_trace_gp_kthread, - .get_gp_data = rcu_tasks_trace_get_gp_data, .cbflood_max = 50000, .irq_capable = 1, .slow_gps = 1, @@ -1148,8 +1354,19 @@ static int rcu_torture_boost(void *arg) unsigned long gp_state; unsigned long gp_state_time; unsigned long oldstarttime; + unsigned long booststarttime = get_torture_init_jiffies() + test_boost_holdoff * HZ; - VERBOSE_TOROUT_STRING("rcu_torture_boost started"); + if (test_boost_holdoff <= 0 || time_after(jiffies, booststarttime)) { + VERBOSE_TOROUT_STRING("rcu_torture_boost started"); + } else { + VERBOSE_TOROUT_STRING("rcu_torture_boost started holdoff period"); + while (time_before(jiffies, booststarttime)) { + schedule_timeout_idle(HZ); + if (kthread_should_stop()) + goto cleanup; + } + VERBOSE_TOROUT_STRING("rcu_torture_boost finished holdoff period"); + } /* Set real-time priority. */ sched_set_fifo_low(current); @@ -1225,6 +1442,7 @@ checkwait: if (stutter_wait("rcu_torture_boost")) sched_set_fifo_low(current); } while (!torture_must_stop()); +cleanup: /* Clean up and exit. */ while (!kthread_should_stop()) { torture_shutdown_absorb("rcu_torture_boost"); @@ -1421,7 +1639,7 @@ static void do_rtws_sync(struct torture_random_state *trsp, void (*sync)(void)) static int rcu_torture_writer(void *arg) { - bool boot_ended; + bool booting_still = false; bool can_expedite = !rcu_gp_is_expedited() && !rcu_gp_is_normal(); unsigned long cookie; struct rcu_gp_oldstate cookie_full; @@ -1432,6 +1650,7 @@ rcu_torture_writer(void *arg) struct rcu_gp_oldstate gp_snap1_full; int i; int idx; + unsigned long j; int oldnice = task_nice(current); struct rcu_gp_oldstate *rgo = NULL; int rgo_size = 0; @@ -1464,16 +1683,26 @@ rcu_torture_writer(void *arg) return 0; } if (cur_ops->poll_active > 0) { - ulo = kzalloc(cur_ops->poll_active * sizeof(ulo[0]), GFP_KERNEL); + ulo = kcalloc(cur_ops->poll_active, sizeof(*ulo), GFP_KERNEL); if (!WARN_ON(!ulo)) ulo_size = cur_ops->poll_active; } if (cur_ops->poll_active_full > 0) { - rgo = kzalloc(cur_ops->poll_active_full * sizeof(rgo[0]), GFP_KERNEL); + rgo = kzalloc_objs(*rgo, cur_ops->poll_active_full); if (!WARN_ON(!rgo)) rgo_size = cur_ops->poll_active_full; } + // If the system is still booting, let it finish. + j = jiffies; + while (!torture_must_stop() && !rcu_inkernel_boot_has_ended()) { + booting_still = true; + schedule_timeout_interruptible(HZ); + } + if (booting_still) + pr_alert("%s" TORTURE_FLAG " Waited %lu jiffies for boot to complete.\n", + torture_type, jiffies - j); + do { rcu_torture_writer_state = RTWS_FIXED_DELAY; torture_hrtimeout_us(500, 1000, &rand); @@ -1582,6 +1811,8 @@ rcu_torture_writer(void *arg) ulo[i] = cur_ops->get_comp_state(); gp_snap = cur_ops->start_gp_poll(); rcu_torture_writer_state = RTWS_POLL_WAIT; + if (cur_ops->exp_current && !(torture_random(&rand) & 0xff)) + cur_ops->exp_current(); while (!cur_ops->poll_gp_state(gp_snap)) { gp_snap1 = cur_ops->get_gp_state(); for (i = 0; i < ulo_size; i++) @@ -1602,6 +1833,8 @@ rcu_torture_writer(void *arg) cur_ops->get_comp_state_full(&rgo[i]); cur_ops->start_gp_poll_full(&gp_snap_full); rcu_torture_writer_state = RTWS_POLL_WAIT_FULL; + if (cur_ops->exp_current && !(torture_random(&rand) & 0xff)) + cur_ops->exp_current(); while (!cur_ops->poll_gp_state_full(&gp_snap_full)) { cur_ops->get_gp_state_full(&gp_snap1_full); for (i = 0; i < rgo_size; i++) @@ -1662,13 +1895,11 @@ rcu_torture_writer(void *arg) !rcu_gp_is_normal(); } rcu_torture_writer_state = RTWS_STUTTER; - boot_ended = rcu_inkernel_boot_has_ended(); stutter_waited = stutter_wait("rcu_torture_writer"); if (stutter_waited && !atomic_read(&rcu_fwd_cb_nodelay) && !cur_ops->slow_gps && !torture_must_stop() && - boot_ended && time_after(jiffies, stallsdone)) for (i = 0; i < ARRAY_SIZE(rcu_tortures); i++) if (list_empty(&rcu_tortures[i].rtort_free) && @@ -1678,6 +1909,7 @@ rcu_torture_writer(void *arg) cur_ops->gp_kthread_dbg(); WARN(1, "%s: rtort_pipe_count: %d\n", __func__, rcu_tortures[i].rtort_pipe_count); rcu_ftrace_dump(DUMP_ALL); + break; } if (stutter_waited) sched_set_normal(current, oldnice); @@ -1728,7 +1960,7 @@ rcu_torture_fakewriter(void *arg) do { torture_hrtimeout_jiffies(torture_random(&rand) % 10, &rand); if (cur_ops->cb_barrier != NULL && - torture_random(&rand) % (nfakewriters * 8) == 0) { + torture_random(&rand) % (nrealfakewriters * 8) == 0) { cur_ops->cb_barrier(); } else { switch (synctype[torture_random(&rand) % nsynctypes]) { @@ -1871,12 +2103,14 @@ static void rcu_torture_reader_do_mbchk(long myid, struct rcu_torture *rtp, // Verify the specified RCUTORTURE_RDR* state. #define ROEC_ARGS "%s %s: Current %#x To add %#x To remove %#x preempt_count() %#x\n", __func__, s, curstate, new, old, preempt_count() -static void rcutorture_one_extend_check(char *s, int curstate, int new, int old, bool insoftirq) +static void rcutorture_one_extend_check(char *s, int curstate, int new, int old) { - if (!IS_ENABLED(CONFIG_RCU_TORTURE_TEST_CHK_RDR_STATE)) + int mask; + + if (!IS_ENABLED(CONFIG_RCU_TORTURE_TEST_CHK_RDR_STATE) || in_nmi()) return; - WARN_ONCE(!(curstate & RCUTORTURE_RDR_IRQ) && irqs_disabled(), ROEC_ARGS); + WARN_ONCE(!(curstate & RCUTORTURE_RDR_IRQ) && irqs_disabled() && !in_hardirq(), ROEC_ARGS); WARN_ONCE((curstate & RCUTORTURE_RDR_IRQ) && !irqs_disabled(), ROEC_ARGS); // If CONFIG_PREEMPT_COUNT=n, further checks are unreliable. @@ -1884,26 +2118,45 @@ static void rcutorture_one_extend_check(char *s, int curstate, int new, int old, return; WARN_ONCE((curstate & (RCUTORTURE_RDR_BH | RCUTORTURE_RDR_RBH)) && - !(preempt_count() & SOFTIRQ_MASK), ROEC_ARGS); + !softirq_count(), ROEC_ARGS); WARN_ONCE((curstate & (RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED)) && !(preempt_count() & PREEMPT_MASK), ROEC_ARGS); WARN_ONCE(cur_ops->readlock_nesting && (curstate & (RCUTORTURE_RDR_RCU_1 | RCUTORTURE_RDR_RCU_2)) && cur_ops->readlock_nesting() == 0, ROEC_ARGS); - // Timer handlers have all sorts of stuff disabled, so ignore + // Interrupt handlers have all sorts of stuff disabled, so ignore // unintended disabling. - if (insoftirq) + if (in_serving_softirq() || in_hardirq()) return; WARN_ONCE(cur_ops->extendables && !(curstate & (RCUTORTURE_RDR_BH | RCUTORTURE_RDR_RBH)) && - (preempt_count() & SOFTIRQ_MASK), ROEC_ARGS); - WARN_ONCE(cur_ops->extendables && - !(curstate & (RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED)) && + softirq_count(), ROEC_ARGS); + + /* + * non-preemptible RCU in a preemptible kernel uses preempt_disable() + * as rcu_read_lock(). + */ + mask = RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED; + if (!IS_ENABLED(CONFIG_PREEMPT_RCU)) + mask |= RCUTORTURE_RDR_RCU_1 | RCUTORTURE_RDR_RCU_2; + + WARN_ONCE(cur_ops->extendables && !(curstate & mask) && (preempt_count() & PREEMPT_MASK), ROEC_ARGS); - WARN_ONCE(cur_ops->readlock_nesting && - !(curstate & (RCUTORTURE_RDR_RCU_1 | RCUTORTURE_RDR_RCU_2)) && + + /* + * non-preemptible RCU in a preemptible kernel uses "preempt_count() & + * PREEMPT_MASK" as ->readlock_nesting(). + */ + mask = RCUTORTURE_RDR_RCU_1 | RCUTORTURE_RDR_RCU_2; + if (!IS_ENABLED(CONFIG_PREEMPT_RCU)) + mask |= RCUTORTURE_RDR_PREEMPT | RCUTORTURE_RDR_SCHED; + + if (IS_ENABLED(CONFIG_PREEMPT_RT) && softirq_count()) + mask |= RCUTORTURE_RDR_BH | RCUTORTURE_RDR_RBH; + + WARN_ONCE(cur_ops->readlock_nesting && !(curstate & mask) && cur_ops->readlock_nesting() > 0, ROEC_ARGS); } @@ -1916,8 +2169,7 @@ static void rcutorture_one_extend_check(char *s, int curstate, int new, int old, * beginning or end of the critical section and if there was actually a * change, do a ->read_delay(). */ -static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, - struct torture_random_state *trsp, +static void rcutorture_one_extend(int *readstate, int newstate, struct torture_random_state *trsp, struct rt_read_seg *rtrsp) { bool first; @@ -1931,8 +2183,8 @@ static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, first = idxold1 == 0; WARN_ON_ONCE(idxold2 < 0); - WARN_ON_ONCE(idxold2 & ~RCUTORTURE_RDR_ALLBITS); - rcutorture_one_extend_check("before change", idxold1, statesnew, statesold, insoftirq); + WARN_ON_ONCE(idxold2 & ~(RCUTORTURE_RDR_ALLBITS | RCUTORTURE_RDR_UPDOWN)); + rcutorture_one_extend_check("before change", idxold1, statesnew, statesold); rtrsp->rt_readstate = newstate; /* First, put new protection in place to avoid critical-section gap. */ @@ -1952,8 +2204,7 @@ static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, idxnew2 = (cur_ops->readlock() << RCUTORTURE_RDR_SHIFT_2) & RCUTORTURE_RDR_MASK_2; // Complain unless both the old and the new protection is in place. - rcutorture_one_extend_check("during change", - idxold1 | statesnew, statesnew, statesold, insoftirq); + rcutorture_one_extend_check("during change", idxold1 | statesnew, statesnew, statesold); // Sample CPU under both sets of protections to reduce confusion. if (IS_ENABLED(CONFIG_RCU_TORTURE_TEST_LOG_CPU)) { @@ -1965,6 +2216,13 @@ static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, rtrsp[-1].rt_preempted = cur_ops->reader_blocked(); } } + // Sample grace-period sequence number, as good a place as any. + if (IS_ENABLED(CONFIG_RCU_TORTURE_TEST_LOG_GP) && cur_ops->gather_gp_seqs) { + rtrsp->rt_gp_seq = cur_ops->gather_gp_seqs(); + rtrsp->rt_ts = ktime_get_mono_fast_ns(); + if (!first) + rtrsp[-1].rt_gp_seq_end = rtrsp->rt_gp_seq; + } /* * Next, remove old protection, in decreasing order of strength @@ -2000,6 +2258,11 @@ static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, if (lockit) raw_spin_unlock_irqrestore(¤t->pi_lock, flags); } + if (statesold & RCUTORTURE_RDR_UPDOWN) { + cur_ops->up_read((idxold1 & RCUTORTURE_RDR_MASK_1) >> RCUTORTURE_RDR_SHIFT_1); + WARN_ON_ONCE(idxnew1 != -1); + idxold1 = 0; + } /* Delay if neither beginning nor end and there was a change. */ if ((statesnew || statesold) && *readstate && newstate) @@ -2016,7 +2279,7 @@ static void rcutorture_one_extend(int *readstate, int newstate, bool insoftirq, WARN_ON_ONCE(*readstate < 0); if (WARN_ON_ONCE(*readstate & ~RCUTORTURE_RDR_ALLBITS)) pr_info("Unexpected readstate value of %#x\n", *readstate); - rcutorture_one_extend_check("after change", *readstate, statesnew, statesold, insoftirq); + rcutorture_one_extend_check("after change", *readstate, statesnew, statesold); } /* Return the biggest extendables mask given current RCU and boot parameters. */ @@ -2083,8 +2346,7 @@ rcutorture_extend_mask(int oldmask, struct torture_random_state *trsp) * critical section. */ static struct rt_read_seg * -rcutorture_loop_extend(int *readstate, bool insoftirq, struct torture_random_state *trsp, - struct rt_read_seg *rtrsp) +rcutorture_loop_extend(int *readstate, struct torture_random_state *trsp, struct rt_read_seg *rtrsp) { int i; int j; @@ -2098,58 +2360,77 @@ rcutorture_loop_extend(int *readstate, bool insoftirq, struct torture_random_sta i = ((i | (i >> 3)) & RCUTORTURE_RDR_MAX_LOOPS) + 1; for (j = 0; j < i; j++) { mask = rcutorture_extend_mask(*readstate, trsp); - rcutorture_one_extend(readstate, mask, insoftirq, trsp, &rtrsp[j]); + WARN_ON_ONCE(mask & RCUTORTURE_RDR_UPDOWN); + rcutorture_one_extend(readstate, mask, trsp, &rtrsp[j]); } return &rtrsp[j]; } -/* - * Do one read-side critical section, returning false if there was - * no data to read. Can be invoked both from process context and - * from a timer handler. - */ -static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid) -{ - bool checkpolling = !(torture_random(trsp) & 0xfff); +struct rcu_torture_one_read_state { + bool checkpolling; unsigned long cookie; struct rcu_gp_oldstate cookie_full; - int i; unsigned long started; - unsigned long completed; - int newstate; struct rcu_torture *p; - int pipe_count; - bool preempted = false; - int readstate = 0; - struct rt_read_seg rtseg[RCUTORTURE_RDR_MAX_SEGS] = { { 0 } }; - struct rt_read_seg *rtrsp = &rtseg[0]; - struct rt_read_seg *rtrsp1; + int readstate; + struct rt_read_seg rtseg[RCUTORTURE_RDR_MAX_SEGS]; + struct rt_read_seg *rtrsp; unsigned long long ts; +}; - WARN_ON_ONCE(!rcu_is_watching()); - newstate = rcutorture_extend_mask(readstate, trsp); - rcutorture_one_extend(&readstate, newstate, myid < 0, trsp, rtrsp++); - if (checkpolling) { +static void init_rcu_torture_one_read_state(struct rcu_torture_one_read_state *rtorsp, + struct torture_random_state *trsp) +{ + memset(rtorsp, 0, sizeof(*rtorsp)); + rtorsp->checkpolling = !(torture_random(trsp) & 0xfff); + rtorsp->rtrsp = &rtorsp->rtseg[0]; +} + +/* + * Set up the first segment of a series of overlapping read-side + * critical sections. The caller must have actually initiated the + * outermost read-side critical section. + */ +static bool rcu_torture_one_read_start(struct rcu_torture_one_read_state *rtorsp, + struct torture_random_state *trsp, long myid) +{ + if (rtorsp->checkpolling) { if (cur_ops->get_gp_state && cur_ops->poll_gp_state) - cookie = cur_ops->get_gp_state(); + rtorsp->cookie = cur_ops->get_gp_state(); if (cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full) - cur_ops->get_gp_state_full(&cookie_full); + cur_ops->get_gp_state_full(&rtorsp->cookie_full); } - started = cur_ops->get_gp_seq(); - ts = rcu_trace_clock_local(); - p = rcu_dereference_check(rcu_torture_current, - !cur_ops->readlock_held || cur_ops->readlock_held()); - if (p == NULL) { + rtorsp->started = cur_ops->get_gp_seq(); + rtorsp->ts = rcu_trace_clock_local(); + rtorsp->p = rcu_dereference_check(rcu_torture_current, + !cur_ops->readlock_held || cur_ops->readlock_held() || + (rtorsp->readstate & RCUTORTURE_RDR_UPDOWN)); + if (rtorsp->p == NULL) { /* Wait for rcu_torture_writer to get underway */ - rcutorture_one_extend(&readstate, 0, myid < 0, trsp, rtrsp); + rcutorture_one_extend(&rtorsp->readstate, 0, trsp, rtorsp->rtrsp); return false; } - if (p->rtort_mbtest == 0) + if (rtorsp->p->rtort_mbtest == 0) atomic_inc(&n_rcu_torture_mberror); - rcu_torture_reader_do_mbchk(myid, p, trsp); - rtrsp = rcutorture_loop_extend(&readstate, myid < 0, trsp, rtrsp); + rcu_torture_reader_do_mbchk(myid, rtorsp->p, trsp); + return true; +} + +/* + * Complete the last segment of a series of overlapping read-side + * critical sections and check for errors. + */ +static void rcu_torture_one_read_end(struct rcu_torture_one_read_state *rtorsp, + struct torture_random_state *trsp) +{ + int i; + unsigned long completed; + int pipe_count; + bool preempted = false; + struct rt_read_seg *rtrsp1; + preempt_disable(); - pipe_count = READ_ONCE(p->rtort_pipe_count); + pipe_count = READ_ONCE(rtorsp->p->rtort_pipe_count); if (pipe_count > RCU_TORTURE_PIPE_LEN) { // Should not happen in a correct RCU implementation, // happens quite often for torture_type=busted. @@ -2157,28 +2438,28 @@ static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid) } completed = cur_ops->get_gp_seq(); if (pipe_count > 1) { - do_trace_rcu_torture_read(cur_ops->name, &p->rtort_rcu, - ts, started, completed); + do_trace_rcu_torture_read(cur_ops->name, &rtorsp->p->rtort_rcu, + rtorsp->ts, rtorsp->started, completed); rcu_ftrace_dump(DUMP_ALL); } __this_cpu_inc(rcu_torture_count[pipe_count]); - completed = rcutorture_seq_diff(completed, started); + completed = rcutorture_seq_diff(completed, rtorsp->started); if (completed > RCU_TORTURE_PIPE_LEN) { /* Should not happen, but... */ completed = RCU_TORTURE_PIPE_LEN; } __this_cpu_inc(rcu_torture_batch[completed]); preempt_enable(); - if (checkpolling) { + if (rtorsp->checkpolling) { if (cur_ops->get_gp_state && cur_ops->poll_gp_state) - WARN_ONCE(cur_ops->poll_gp_state(cookie), + WARN_ONCE(cur_ops->poll_gp_state(rtorsp->cookie), "%s: Cookie check 2 failed %s(%d) %lu->%lu\n", __func__, rcu_torture_writer_state_getname(), rcu_torture_writer_state, - cookie, cur_ops->get_gp_state()); + rtorsp->cookie, cur_ops->get_gp_state()); if (cur_ops->get_gp_state_full && cur_ops->poll_gp_state_full) - WARN_ONCE(cur_ops->poll_gp_state_full(&cookie_full), + WARN_ONCE(cur_ops->poll_gp_state_full(&rtorsp->cookie_full), "%s: Cookie check 6 failed %s(%d) online %*pbl\n", __func__, rcu_torture_writer_state_getname(), @@ -2187,21 +2468,41 @@ static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid) } if (cur_ops->reader_blocked) preempted = cur_ops->reader_blocked(); - rcutorture_one_extend(&readstate, 0, myid < 0, trsp, rtrsp); - WARN_ON_ONCE(readstate); + rcutorture_one_extend(&rtorsp->readstate, 0, trsp, rtorsp->rtrsp); + WARN_ON_ONCE(rtorsp->readstate); // This next splat is expected behavior if leakpointer, especially // for CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels. - WARN_ON_ONCE(leakpointer && READ_ONCE(p->rtort_pipe_count) > 1); + WARN_ON_ONCE(leakpointer && READ_ONCE(rtorsp->p->rtort_pipe_count) > 1); /* If error or close call, record the sequence of reader protections. */ if ((pipe_count > 1 || completed > 1) && !xchg(&err_segs_recorded, 1)) { i = 0; - for (rtrsp1 = &rtseg[0]; rtrsp1 < rtrsp; rtrsp1++) + for (rtrsp1 = &rtorsp->rtseg[0]; rtrsp1 < rtorsp->rtrsp; rtrsp1++) err_segs[i++] = *rtrsp1; rt_read_nsegs = i; rt_read_preempted = preempted; } +} +/* + * Do one read-side critical section, returning false if there was + * no data to read. Can be invoked both from process context and + * from a timer handler. + */ +static bool rcu_torture_one_read(struct torture_random_state *trsp, long myid) +{ + int newstate; + struct rcu_torture_one_read_state rtors; + + WARN_ON_ONCE(!rcu_is_watching()); + init_rcu_torture_one_read_state(&rtors, trsp); + newstate = rcutorture_extend_mask(rtors.readstate, trsp); + WARN_ON_ONCE(newstate & RCUTORTURE_RDR_UPDOWN); + rcutorture_one_extend(&rtors.readstate, newstate, trsp, rtors.rtrsp++); + if (!rcu_torture_one_read_start(&rtors, trsp, myid)) + return false; + rtors.rtrsp = rcutorture_loop_extend(&rtors.readstate, trsp, rtors.rtrsp); + rcu_torture_one_read_end(&rtors, trsp); return true; } @@ -2215,12 +2516,15 @@ static DEFINE_TORTURE_RANDOM_PERCPU(rcu_torture_timer_rand); */ static void rcu_torture_timer(struct timer_list *unused) { + WARN_ON_ONCE(!in_serving_softirq()); + WARN_ON_ONCE(in_hardirq()); + WARN_ON_ONCE(in_nmi()); atomic_long_inc(&n_rcu_torture_timers); (void)rcu_torture_one_read(this_cpu_ptr(&rcu_torture_timer_rand), -1); /* Test call_rcu() invocation from interrupt handler. */ if (cur_ops->call) { - struct rcu_head *rhp = kmalloc(sizeof(*rhp), GFP_NOWAIT); + struct rcu_head *rhp = kmalloc_obj(*rhp, GFP_NOWAIT); if (rhp) cur_ops->call(rhp, rcu_torture_timer_cb); @@ -2246,7 +2550,7 @@ rcu_torture_reader(void *arg) set_user_nice(current, MAX_NICE); if (irqreader && cur_ops->irq_capable) timer_setup_on_stack(&t, rcu_torture_timer, 0); - tick_dep_set_task(current, TICK_DEP_BIT_RCU); + tick_dep_set_task(current, TICK_DEP_BIT_RCU); // CPU bound, so need tick. do { if (irqreader && cur_ops->irq_capable) { if (!timer_pending(&t)) @@ -2258,19 +2562,166 @@ rcu_torture_reader(void *arg) torture_hrtimeout_us(500, 1000, &rand); lastsleep = jiffies + 10; } - while (torture_num_online_cpus() < mynumonline && !torture_must_stop()) + while (!torture_must_stop() && + (torture_num_online_cpus() < mynumonline || !rcu_inkernel_boot_has_ended())) schedule_timeout_interruptible(HZ / 5); stutter_wait("rcu_torture_reader"); } while (!torture_must_stop()); if (irqreader && cur_ops->irq_capable) { - del_timer_sync(&t); - destroy_timer_on_stack(&t); + timer_delete_sync(&t); + timer_destroy_on_stack(&t); } tick_dep_clear_task(current, TICK_DEP_BIT_RCU); torture_kthread_stopping("rcu_torture_reader"); return 0; } +struct rcu_torture_one_read_state_updown { + struct hrtimer rtorsu_hrt; + bool rtorsu_inuse; + ktime_t rtorsu_kt; + int rtorsu_cpu; + unsigned long rtorsu_j; + unsigned long rtorsu_ndowns; + unsigned long rtorsu_nups; + unsigned long rtorsu_nmigrates; + struct torture_random_state rtorsu_trs; + struct rcu_torture_one_read_state rtorsu_rtors; +}; + +static struct rcu_torture_one_read_state_updown *updownreaders; +static DEFINE_TORTURE_RANDOM(rcu_torture_updown_rand); +static int rcu_torture_updown(void *arg); + +static enum hrtimer_restart rcu_torture_updown_hrt(struct hrtimer *hrtp) +{ + int cpu = raw_smp_processor_id(); + struct rcu_torture_one_read_state_updown *rtorsup; + + rtorsup = container_of(hrtp, struct rcu_torture_one_read_state_updown, rtorsu_hrt); + rcu_torture_one_read_end(&rtorsup->rtorsu_rtors, &rtorsup->rtorsu_trs); + WARN_ONCE(rtorsup->rtorsu_nups >= rtorsup->rtorsu_ndowns, "%s: Up without matching down #%zu.\n", __func__, rtorsup - updownreaders); + WRITE_ONCE(rtorsup->rtorsu_nups, rtorsup->rtorsu_nups + 1); + WRITE_ONCE(rtorsup->rtorsu_nmigrates, + rtorsup->rtorsu_nmigrates + (cpu != rtorsup->rtorsu_cpu)); + smp_store_release(&rtorsup->rtorsu_inuse, false); + return HRTIMER_NORESTART; +} + +static int rcu_torture_updown_init(void) +{ + int i; + struct torture_random_state *rand = &rcu_torture_updown_rand; + int ret; + + if (n_up_down < 0) + return 0; + if (!srcu_torture_have_up_down()) { + VERBOSE_TOROUT_STRING("rcu_torture_updown_init: Disabling up/down reader tests due to lack of primitives"); + return 0; + } + updownreaders = kzalloc_objs(*updownreaders, n_up_down); + if (!updownreaders) { + VERBOSE_TOROUT_STRING("rcu_torture_updown_init: Out of memory, disabling up/down reader tests"); + return -ENOMEM; + } + for (i = 0; i < n_up_down; i++) { + init_rcu_torture_one_read_state(&updownreaders[i].rtorsu_rtors, rand); + hrtimer_setup(&updownreaders[i].rtorsu_hrt, rcu_torture_updown_hrt, CLOCK_MONOTONIC, + HRTIMER_MODE_REL | HRTIMER_MODE_HARD); + torture_random_init(&updownreaders[i].rtorsu_trs); + init_rcu_torture_one_read_state(&updownreaders[i].rtorsu_rtors, + &updownreaders[i].rtorsu_trs); + } + ret = torture_create_kthread(rcu_torture_updown, rand, updown_task); + if (ret) { + kfree(updownreaders); + updownreaders = NULL; + } + return ret; +} + +static void rcu_torture_updown_cleanup(void) +{ + struct rcu_torture_one_read_state_updown *rtorsup; + + for (rtorsup = updownreaders; rtorsup < &updownreaders[n_up_down]; rtorsup++) { + if (!smp_load_acquire(&rtorsup->rtorsu_inuse)) + continue; + if (hrtimer_cancel(&rtorsup->rtorsu_hrt) || WARN_ON_ONCE(rtorsup->rtorsu_inuse)) { + rcu_torture_one_read_end(&rtorsup->rtorsu_rtors, &rtorsup->rtorsu_trs); + WARN_ONCE(rtorsup->rtorsu_nups >= rtorsup->rtorsu_ndowns, "%s: Up without matching down #%zu.\n", __func__, rtorsup - updownreaders); + WRITE_ONCE(rtorsup->rtorsu_nups, rtorsup->rtorsu_nups + 1); + smp_store_release(&rtorsup->rtorsu_inuse, false); + } + + } + kfree(updownreaders); + updownreaders = NULL; +} + +// Do one reader for rcu_torture_updown(). +static void rcu_torture_updown_one(struct rcu_torture_one_read_state_updown *rtorsup) +{ + int idx; + int rawidx; + ktime_t t; + + init_rcu_torture_one_read_state(&rtorsup->rtorsu_rtors, &rtorsup->rtorsu_trs); + rawidx = cur_ops->down_read(); + WRITE_ONCE(rtorsup->rtorsu_ndowns, rtorsup->rtorsu_ndowns + 1); + idx = (rawidx << RCUTORTURE_RDR_SHIFT_1) & RCUTORTURE_RDR_MASK_1; + rtorsup->rtorsu_rtors.readstate = idx | RCUTORTURE_RDR_UPDOWN; + rtorsup->rtorsu_rtors.rtrsp++; + rtorsup->rtorsu_cpu = raw_smp_processor_id(); + if (!rcu_torture_one_read_start(&rtorsup->rtorsu_rtors, &rtorsup->rtorsu_trs, -1)) { + WARN_ONCE(rtorsup->rtorsu_nups >= rtorsup->rtorsu_ndowns, "%s: Up without matching down #%zu.\n", __func__, rtorsup - updownreaders); + WRITE_ONCE(rtorsup->rtorsu_nups, rtorsup->rtorsu_nups + 1); + schedule_timeout_idle(HZ); + return; + } + smp_store_release(&rtorsup->rtorsu_inuse, true); + t = torture_random(&rtorsup->rtorsu_trs) & 0xfffff; // One per million. + if (t < 10 * 1000) + t = 200 * 1000 * 1000; + hrtimer_start(&rtorsup->rtorsu_hrt, t, HRTIMER_MODE_REL | HRTIMER_MODE_HARD); + smp_mb(); // Sample jiffies after posting hrtimer. + rtorsup->rtorsu_j = jiffies; // Not used by hrtimer handler. + rtorsup->rtorsu_kt = t; +} + +/* + * RCU torture up/down reader kthread, starting RCU readers in kthread + * context and ending them in hrtimer handlers. Otherwise similar to + * rcu_torture_reader(). + */ +static int +rcu_torture_updown(void *arg) +{ + unsigned long j; + struct rcu_torture_one_read_state_updown *rtorsup; + + VERBOSE_TOROUT_STRING("rcu_torture_updown task started"); + do { + for (rtorsup = updownreaders; rtorsup < &updownreaders[n_up_down]; rtorsup++) { + if (torture_must_stop()) + break; + j = smp_load_acquire(&jiffies); // Time before ->rtorsu_inuse. + if (smp_load_acquire(&rtorsup->rtorsu_inuse)) { + WARN_ONCE(time_after(j, rtorsup->rtorsu_j + 1 + HZ * 10), + "hrtimer queued at jiffies %lu for %lld ns took %lu jiffies\n", rtorsup->rtorsu_j, rtorsup->rtorsu_kt, j - rtorsup->rtorsu_j); + continue; + } + rcu_torture_updown_one(rtorsup); + } + torture_hrtimeout_ms(1, 1000, &rcu_torture_updown_rand); + stutter_wait("rcu_torture_updown"); + } while (!torture_must_stop()); + rcu_torture_updown_cleanup(); + torture_kthread_stopping("rcu_torture_updown"); + return 0; +} + /* * Randomly Toggle CPUs' callback-offload state. This uses hrtimers to * increase race probabilities and fuzzes the interval between toggling. @@ -2333,6 +2784,11 @@ rcu_torture_stats_print(void) int i; long pipesummary[RCU_TORTURE_PIPE_LEN + 1] = { 0 }; long batchsummary[RCU_TORTURE_PIPE_LEN + 1] = { 0 }; + long n_gpwraps = 0; + unsigned long ndowns = 0; + unsigned long nunexpired = 0; + unsigned long nmigrates = 0; + unsigned long nups = 0; struct rcu_torture *rtcp; static unsigned long rtcv_snap = ULONG_MAX; static bool splatted; @@ -2343,11 +2799,21 @@ rcu_torture_stats_print(void) pipesummary[i] += READ_ONCE(per_cpu(rcu_torture_count, cpu)[i]); batchsummary[i] += READ_ONCE(per_cpu(rcu_torture_batch, cpu)[i]); } + if (cur_ops->get_gpwrap_count) + n_gpwraps += cur_ops->get_gpwrap_count(cpu); + } + if (updownreaders) { + for (i = 0; i < n_up_down; i++) { + ndowns += READ_ONCE(updownreaders[i].rtorsu_ndowns); + nups += READ_ONCE(updownreaders[i].rtorsu_nups); + nunexpired += READ_ONCE(updownreaders[i].rtorsu_inuse); + nmigrates += READ_ONCE(updownreaders[i].rtorsu_nmigrates); + } } for (i = RCU_TORTURE_PIPE_LEN; i >= 0; i--) { if (pipesummary[i] != 0) break; - } + } // The value of variable "i" is used later, so don't clobber it! pr_alert("%s%s ", torture_type, TORTURE_FLAG); rtcp = rcu_access_pointer(rcu_torture_current); @@ -2368,14 +2834,17 @@ rcu_torture_stats_print(void) n_rcu_torture_boost_failure, n_rcu_torture_boosts, atomic_long_read(&n_rcu_torture_timers)); + if (updownreaders) + pr_cont("ndowns: %lu nups: %lu nhrt: %lu nmigrates: %lu ", ndowns, nups, nunexpired, nmigrates); torture_onoff_stats(); pr_cont("barrier: %ld/%ld:%ld ", data_race(n_barrier_successes), data_race(n_barrier_attempts), data_race(n_rcu_torture_barrier_error)); pr_cont("read-exits: %ld ", data_race(n_read_exits)); // Statistic. - pr_cont("nocb-toggles: %ld:%ld\n", + pr_cont("nocb-toggles: %ld:%ld ", atomic_long_read(&n_nocb_offload), atomic_long_read(&n_nocb_deoffload)); + pr_cont("gpwraps: %ld\n", n_gpwraps); pr_alert("%s%s ", torture_type, TORTURE_FLAG); if (atomic_read(&n_rcu_torture_mberror) || @@ -2413,7 +2882,8 @@ rcu_torture_stats_print(void) cur_ops->stats(); if (rtcv_snap == rcu_torture_current_version && rcu_access_pointer(rcu_torture_current) && - !rcu_stall_is_suppressed()) { + !rcu_stall_is_suppressed() && + rcu_inkernel_boot_has_ended()) { int __maybe_unused flags = 0; unsigned long __maybe_unused gp_seq = 0; @@ -2483,7 +2953,7 @@ static void rcu_torture_mem_dump_obj(void) mem_dump_obj(&z); kmem_cache_free(kcp, rhp); kmem_cache_destroy(kcp); - rhp = kmalloc(sizeof(*rhp), GFP_KERNEL); + rhp = kmalloc_obj(*rhp); if (WARN_ON_ONCE(!rhp)) return; pr_alert("mem_dump_obj() kmalloc test: rcu_torture_stats = %px, &rhp = %px, rhp = %px\n", stats_task, &rhp, rhp); @@ -2512,7 +2982,7 @@ rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, const char *tag) "shuffle_interval=%d stutter=%d irqreader=%d " "fqs_duration=%d fqs_holdoff=%d fqs_stutter=%d " "test_boost=%d/%d test_boost_interval=%d " - "test_boost_duration=%d shutdown_secs=%d " + "test_boost_duration=%d test_boost_holdoff=%d shutdown_secs=%d " "stall_cpu=%d stall_cpu_holdoff=%d stall_cpu_irqsoff=%d " "stall_cpu_block=%d stall_cpu_repeat=%d " "n_barrier_cbs=%d " @@ -2521,12 +2991,12 @@ rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, const char *tag) "reader_flavor=%x " "nocbs_nthreads=%d nocbs_toggle=%d " "test_nmis=%d " - "preempt_duration=%d preempt_interval=%d\n", - torture_type, tag, nrealreaders, nfakewriters, + "preempt_duration=%d preempt_interval=%d n_up_down=%d\n", + torture_type, tag, nrealreaders, nrealfakewriters, stat_interval, verbose, test_no_idle_hz, shuffle_interval, stutter, irqreader, fqs_duration, fqs_holdoff, fqs_stutter, test_boost, cur_ops->can_boost, - test_boost_interval, test_boost_duration, shutdown_secs, + test_boost_interval, test_boost_duration, test_boost_holdoff, shutdown_secs, stall_cpu, stall_cpu_holdoff, stall_cpu_irqsoff, stall_cpu_block, stall_cpu_repeat, n_barrier_cbs, @@ -2535,7 +3005,7 @@ rcu_torture_print_module_parms(struct rcu_torture_ops *cur_ops, const char *tag) reader_flavor, nocbs_nthreads, nocbs_toggle, test_nmis, - preempt_duration, preempt_interval); + preempt_duration, preempt_interval, n_up_down); } static int rcutorture_booster_cleanup(unsigned int cpu) @@ -2975,7 +3445,7 @@ static void rcu_torture_fwd_prog_cr(struct rcu_fwd *rfp) cver = READ_ONCE(rcu_torture_current_version); gps = cur_ops->get_gp_seq(); rfp->rcu_launder_gp_seq_start = gps; - tick_dep_set_task(current, TICK_DEP_BIT_RCU); + tick_dep_set_task(current, TICK_DEP_BIT_RCU); // CPU bound, so need tick. while (time_before(jiffies, stopat) && !shutdown_time_arrived() && !READ_ONCE(rcu_fwd_emergency_stop) && !torture_must_stop()) { @@ -2991,7 +3461,7 @@ static void rcu_torture_fwd_prog_cr(struct rcu_fwd *rfp) n_launders++; n_launders_sa++; } else if (!cur_ops->cbflood_max || cur_ops->cbflood_max > n_max_cbs) { - rfcp = kmalloc(sizeof(*rfcp), GFP_KERNEL); + rfcp = kmalloc_obj(*rfcp); if (WARN_ON_ONCE(!rfcp)) { schedule_timeout_interruptible(1); continue; @@ -3103,6 +3573,8 @@ static int rcu_torture_fwd_prog(void *args) int tested_tries = 0; VERBOSE_TOROUT_STRING("rcu_torture_fwd_progress task started"); + while (!rcu_inkernel_boot_has_ended()) + schedule_timeout_interruptible(HZ / 10); rcu_bind_current_to_nocb(); if (!IS_ENABLED(CONFIG_SMP) || !IS_ENABLED(CONFIG_RCU_BOOST)) set_user_nice(current, MAX_NICE); @@ -3177,8 +3649,8 @@ static int __init rcu_torture_fwd_prog_init(void) fwd_progress_holdoff = 1; if (fwd_progress_div <= 0) fwd_progress_div = 4; - rfp = kcalloc(fwd_progress, sizeof(*rfp), GFP_KERNEL); - fwd_prog_tasks = kcalloc(fwd_progress, sizeof(*fwd_prog_tasks), GFP_KERNEL); + rfp = kzalloc_objs(*rfp, fwd_progress); + fwd_prog_tasks = kzalloc_objs(*fwd_prog_tasks, fwd_progress); if (!rfp || !fwd_prog_tasks) { kfree(rfp); kfree(fwd_prog_tasks); @@ -3344,10 +3816,9 @@ static int rcu_torture_barrier_init(void) atomic_set(&barrier_cbs_count, 0); atomic_set(&barrier_cbs_invoked, 0); barrier_cbs_tasks = - kcalloc(n_barrier_cbs, sizeof(barrier_cbs_tasks[0]), - GFP_KERNEL); + kzalloc_objs(barrier_cbs_tasks[0], n_barrier_cbs); barrier_cbs_wq = - kcalloc(n_barrier_cbs, sizeof(barrier_cbs_wq[0]), GFP_KERNEL); + kzalloc_objs(barrier_cbs_wq[0], n_barrier_cbs); if (barrier_cbs_tasks == NULL || !barrier_cbs_wq) return -ENOMEM; for (i = 0; i < n_barrier_cbs; i++) { @@ -3546,6 +4017,57 @@ static int rcu_torture_preempt(void *unused) static enum cpuhp_state rcutor_hp; +static struct hrtimer gpwrap_lag_timer; +static bool gpwrap_lag_active; + +/* Timer handler for toggling RCU grace-period sequence overflow test lag value */ +static enum hrtimer_restart rcu_gpwrap_lag_timer(struct hrtimer *timer) +{ + ktime_t next_delay; + + if (gpwrap_lag_active) { + pr_alert("rcu-torture: Disabling gpwrap lag (value=0)\n"); + cur_ops->set_gpwrap_lag(0); + gpwrap_lag_active = false; + next_delay = ktime_set((gpwrap_lag_cycle_mins - gpwrap_lag_active_mins) * 60, 0); + } else { + pr_alert("rcu-torture: Enabling gpwrap lag (value=%d)\n", gpwrap_lag_gps); + cur_ops->set_gpwrap_lag(gpwrap_lag_gps); + gpwrap_lag_active = true; + next_delay = ktime_set(gpwrap_lag_active_mins * 60, 0); + } + + if (torture_must_stop_irq()) + return HRTIMER_NORESTART; + + hrtimer_forward_now(timer, next_delay); + return HRTIMER_RESTART; +} + +static int rcu_gpwrap_lag_init(void) +{ + if (!gpwrap_lag) + return 0; + + if (gpwrap_lag_cycle_mins <= 0 || gpwrap_lag_active_mins <= 0) { + pr_alert("rcu-torture: lag timing parameters must be positive\n"); + return -EINVAL; + } + + hrtimer_setup(&gpwrap_lag_timer, rcu_gpwrap_lag_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + gpwrap_lag_active = false; + hrtimer_start(&gpwrap_lag_timer, + ktime_set((gpwrap_lag_cycle_mins - gpwrap_lag_active_mins) * 60, 0), HRTIMER_MODE_REL); + + return 0; +} + +static void rcu_gpwrap_lag_cleanup(void) +{ + hrtimer_cancel(&gpwrap_lag_timer); + cur_ops->set_gpwrap_lag(0); + gpwrap_lag_active = false; +} static void rcu_torture_cleanup(void) { @@ -3553,6 +4075,7 @@ rcu_torture_cleanup(void) int flags = 0; unsigned long gp_seq = 0; int i; + int j; if (torture_cleanup_begin()) { if (cur_ops->cb_barrier != NULL) { @@ -3586,6 +4109,10 @@ rcu_torture_cleanup(void) nocb_tasks = NULL; } + if (updown_task) { + torture_stop_kthread(rcu_torture_updown, updown_task); + updown_task = NULL; + } if (reader_tasks) { for (i = 0; i < nrealreaders; i++) torture_stop_kthread(rcu_torture_reader, @@ -3597,7 +4124,7 @@ rcu_torture_cleanup(void) rcu_torture_reader_mbchk = NULL; if (fakewriter_tasks) { - for (i = 0; i < nfakewriters; i++) + for (i = 0; i < nrealfakewriters; i++) torture_stop_kthread(rcu_torture_fakewriter, fakewriter_tasks[i]); kfree(fakewriter_tasks); @@ -3635,7 +4162,11 @@ rcu_torture_cleanup(void) pr_alert("\t: No segments recorded!!!\n"); firsttime = 1; for (i = 0; i < rt_read_nsegs; i++) { - pr_alert("\t%d: %#4x", i, err_segs[i].rt_readstate); + if (IS_ENABLED(CONFIG_RCU_TORTURE_TEST_LOG_GP)) + pr_alert("\t%lluus ", div64_u64(err_segs[i].rt_ts, 1000ULL)); + else + pr_alert("\t"); + pr_cont("%d: %#4x", i, err_segs[i].rt_readstate); if (err_segs[i].rt_delay_jiffies != 0) { pr_cont("%s%ldjiffies", firsttime ? "" : "+", err_segs[i].rt_delay_jiffies); @@ -3648,6 +4179,27 @@ rcu_torture_cleanup(void) else pr_cont(" ..."); } + if (IS_ENABLED(CONFIG_RCU_TORTURE_TEST_LOG_GP) && + cur_ops->gather_gp_seqs && cur_ops->format_gp_seqs) { + char buf1[20+1]; + char buf2[20+1]; + char sepchar = '-'; + + cur_ops->format_gp_seqs(err_segs[i].rt_gp_seq, + buf1, ARRAY_SIZE(buf1)); + cur_ops->format_gp_seqs(err_segs[i].rt_gp_seq_end, + buf2, ARRAY_SIZE(buf2)); + if (err_segs[i].rt_gp_seq == err_segs[i].rt_gp_seq_end) { + if (buf2[0]) { + for (j = 0; buf2[j]; j++) + buf2[j] = '.'; + if (j) + buf2[j - 1] = ' '; + } + sepchar = ' '; + } + pr_cont(" %s%c%s", buf1, sepchar, buf2); + } if (err_segs[i].rt_delay_ms != 0) { pr_cont(" %s%ldms", firsttime ? "" : "+", err_segs[i].rt_delay_ms); @@ -3689,6 +4241,9 @@ rcu_torture_cleanup(void) torture_cleanup_end(); if (cur_ops->gp_slow_unregister) cur_ops->gp_slow_unregister(NULL); + + if (gpwrap_lag && cur_ops->set_gpwrap_lag) + rcu_gpwrap_lag_cleanup(); } static void rcu_torture_leak_cb(struct rcu_head *rhp) @@ -3730,7 +4285,7 @@ static void rcu_test_debug_objects(void) (!cur_ops->call || !cur_ops->cb_barrier))) return; - struct rcu_head *rhp = kmalloc(sizeof(*rhp), GFP_KERNEL); + struct rcu_head *rhp = kmalloc_obj(*rhp); init_rcu_head_on_stack(&rh1); init_rcu_head_on_stack(&rh2); @@ -3956,7 +4511,7 @@ rcu_torture_init(void) static struct rcu_torture_ops *torture_ops[] = { &rcu_ops, &rcu_busted_ops, &srcu_ops, &srcud_ops, &busted_srcud_ops, TASKS_OPS TASKS_RUDE_OPS TASKS_TRACING_OPS - &trivial_ops, + &trivial_ops, TRIVIAL_PREEMPT_OPS }; if (!torture_init_begin(torture_type, verbose)) @@ -3994,6 +4549,14 @@ rcu_torture_init(void) rcu_torture_init_srcu_lockdep(); + if (nfakewriters >= 0) { + nrealfakewriters = nfakewriters; + } else { + nrealfakewriters = num_online_cpus() - 2 - nfakewriters; + if (nrealfakewriters <= 0) + nrealfakewriters = 1; + } + if (nreaders >= 0) { nrealreaders = nreaders; } else { @@ -4046,30 +4609,24 @@ rcu_torture_init(void) /* Start up the kthreads. */ rcu_torture_write_types(); - firsterr = torture_create_kthread(rcu_torture_writer, NULL, - writer_task); - if (torture_init_error(firsterr)) - goto unwind; - if (nfakewriters > 0) { - fakewriter_tasks = kcalloc(nfakewriters, - sizeof(fakewriter_tasks[0]), - GFP_KERNEL); + if (nrealfakewriters > 0) { + fakewriter_tasks = kzalloc_objs(fakewriter_tasks[0], + nrealfakewriters); if (fakewriter_tasks == NULL) { TOROUT_ERRSTRING("out of memory"); firsterr = -ENOMEM; goto unwind; } } - for (i = 0; i < nfakewriters; i++) { + for (i = 0; i < nrealfakewriters; i++) { firsterr = torture_create_kthread(rcu_torture_fakewriter, NULL, fakewriter_tasks[i]); if (torture_init_error(firsterr)) goto unwind; } - reader_tasks = kcalloc(nrealreaders, sizeof(reader_tasks[0]), - GFP_KERNEL); - rcu_torture_reader_mbchk = kcalloc(nrealreaders, sizeof(*rcu_torture_reader_mbchk), - GFP_KERNEL); + reader_tasks = kzalloc_objs(reader_tasks[0], nrealreaders); + rcu_torture_reader_mbchk = kzalloc_objs(*rcu_torture_reader_mbchk, + nrealreaders); if (!reader_tasks || !rcu_torture_reader_mbchk) { TOROUT_ERRSTRING("out of memory"); firsterr = -ENOMEM; @@ -4082,13 +4639,22 @@ rcu_torture_init(void) if (torture_init_error(firsterr)) goto unwind; } + + firsterr = torture_create_kthread(rcu_torture_writer, NULL, + writer_task); + if (torture_init_error(firsterr)) + goto unwind; + + firsterr = rcu_torture_updown_init(); + if (torture_init_error(firsterr)) + goto unwind; nrealnocbers = nocbs_nthreads; if (WARN_ON(nrealnocbers < 0)) nrealnocbers = 1; if (WARN_ON(nocbs_toggle < 0)) nocbs_toggle = HZ; if (nrealnocbers > 0) { - nocb_tasks = kcalloc(nrealnocbers, sizeof(nocb_tasks[0]), GFP_KERNEL); + nocb_tasks = kzalloc_objs(nocb_tasks[0], nrealnocbers); if (nocb_tasks == NULL) { TOROUT_ERRSTRING("out of memory"); firsterr = -ENOMEM; @@ -4176,9 +4742,17 @@ rcu_torture_init(void) } if (object_debug) rcu_test_debug_objects(); - torture_init_end(); + if (cur_ops->gp_slow_register && !WARN_ON_ONCE(!cur_ops->gp_slow_unregister)) cur_ops->gp_slow_register(&rcu_fwd_cb_nodelay); + + if (gpwrap_lag && cur_ops->set_gpwrap_lag) { + firsterr = rcu_gpwrap_lag_init(); + if (torture_init_error(firsterr)) + goto unwind; + } + + torture_init_end(); return 0; unwind: diff --git a/kernel/rcu/refscale.c b/kernel/rcu/refscale.c index 1b47376acdc4..a2d9d75d88a1 100644 --- a/kernel/rcu/refscale.c +++ b/kernel/rcu/refscale.c @@ -85,22 +85,16 @@ torture_param(int, holdoff, IS_BUILTIN(CONFIG_RCU_REF_SCALE_TEST) ? 10 : 0, // Number of typesafe_lookup structures, that is, the degree of concurrency. torture_param(long, lookup_instances, 0, "Number of typesafe_lookup structures."); // Number of loops per experiment, all readers execute operations concurrently. -torture_param(long, loops, 10000, "Number of loops per experiment."); +torture_param(int, loops, 10000, "Number of loops per experiment."); // Number of readers, with -1 defaulting to about 75% of the CPUs. torture_param(int, nreaders, -1, "Number of readers, -1 for 75% of CPUs."); // Number of runs. torture_param(int, nruns, 30, "Number of experiments to run."); // Reader delay in nanoseconds, 0 for no delay. torture_param(int, readdelay, 0, "Read-side delay in nanoseconds."); - -#ifdef MODULE -# define REFSCALE_SHUTDOWN 0 -#else -# define REFSCALE_SHUTDOWN 1 -#endif - -torture_param(bool, shutdown, REFSCALE_SHUTDOWN, - "Shutdown at end of scalability tests."); +// Maximum shutdown delay in seconds, or zero for no shutdown. +torture_param(int, shutdown_secs, !IS_MODULE(CONFIG_REPRO_TEST) * 300, + "Shutdown at end of scalability tests or at specified timeout (s)."); struct reader_task { struct task_struct *task; @@ -109,12 +103,8 @@ struct reader_task { u64 last_duration_ns; }; -static struct task_struct *shutdown_task; -static wait_queue_head_t shutdown_wq; - static struct task_struct *main_task; static wait_queue_head_t main_wq; -static int shutdown_start; static struct reader_task *reader_tasks; @@ -136,6 +126,7 @@ struct ref_scale_ops { void (*cleanup)(void); void (*readsection)(const int nloops); void (*delaysection)(const int nloops, const int udl, const int ndl); + bool enable_irqs; const char *name; }; @@ -184,6 +175,8 @@ static const struct ref_scale_ops rcu_ops = { // Definitions for SRCU ref scale testing. DEFINE_STATIC_SRCU(srcu_refctl_scale); +DEFINE_STATIC_SRCU_FAST(srcu_fast_refctl_scale); +DEFINE_STATIC_SRCU_FAST_UPDOWN(srcu_fast_updown_refctl_scale); static struct srcu_struct *srcu_ctlp = &srcu_refctl_scale; static void srcu_ref_scale_read_section(const int nloops) @@ -216,34 +209,76 @@ static const struct ref_scale_ops srcu_ops = { .name = "srcu" }; -static void srcu_lite_ref_scale_read_section(const int nloops) +static bool srcu_fast_sync_scale_init(void) +{ + srcu_ctlp = &srcu_fast_refctl_scale; + return true; +} + +static void srcu_fast_ref_scale_read_section(const int nloops) { int i; - int idx; + struct srcu_ctr __percpu *scp; for (i = nloops; i >= 0; i--) { - idx = srcu_read_lock_lite(srcu_ctlp); - srcu_read_unlock_lite(srcu_ctlp, idx); + scp = srcu_read_lock_fast(srcu_ctlp); + srcu_read_unlock_fast(srcu_ctlp, scp); } } -static void srcu_lite_ref_scale_delay_section(const int nloops, const int udl, const int ndl) +static void srcu_fast_ref_scale_delay_section(const int nloops, const int udl, const int ndl) { int i; - int idx; + struct srcu_ctr __percpu *scp; for (i = nloops; i >= 0; i--) { - idx = srcu_read_lock_lite(srcu_ctlp); + scp = srcu_read_lock_fast(srcu_ctlp); un_delay(udl, ndl); - srcu_read_unlock_lite(srcu_ctlp, idx); + srcu_read_unlock_fast(srcu_ctlp, scp); } } -static const struct ref_scale_ops srcu_lite_ops = { - .init = rcu_sync_scale_init, - .readsection = srcu_lite_ref_scale_read_section, - .delaysection = srcu_lite_ref_scale_delay_section, - .name = "srcu-lite" +static const struct ref_scale_ops srcu_fast_ops = { + .init = srcu_fast_sync_scale_init, + .readsection = srcu_fast_ref_scale_read_section, + .delaysection = srcu_fast_ref_scale_delay_section, + .name = "srcu-fast" +}; + +static bool srcu_fast_updown_sync_scale_init(void) +{ + srcu_ctlp = &srcu_fast_updown_refctl_scale; + return true; +} + +static void srcu_fast_updown_ref_scale_read_section(const int nloops) +{ + int i; + struct srcu_ctr __percpu *scp; + + for (i = nloops; i >= 0; i--) { + scp = srcu_read_lock_fast_updown(srcu_ctlp); + srcu_read_unlock_fast_updown(srcu_ctlp, scp); + } +} + +static void srcu_fast_updown_ref_scale_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + struct srcu_ctr __percpu *scp; + + for (i = nloops; i >= 0; i--) { + scp = srcu_read_lock_fast_updown(srcu_ctlp); + un_delay(udl, ndl); + srcu_read_unlock_fast_updown(srcu_ctlp, scp); + } +} + +static const struct ref_scale_ops srcu_fast_updown_ops = { + .init = srcu_fast_updown_sync_scale_init, + .readsection = srcu_fast_updown_ref_scale_read_section, + .delaysection = srcu_fast_updown_ref_scale_delay_section, + .name = "srcu-fast-updown" }; #ifdef CONFIG_TASKS_RCU @@ -323,6 +358,9 @@ static const struct ref_scale_ops rcu_trace_ops = { // Definitions for reference count static atomic_t refcnt; +// Definitions acquire-release. +static DEFINE_PER_CPU(unsigned long, test_acqrel); + static void ref_refcnt_section(const int nloops) { int i; @@ -351,6 +389,184 @@ static const struct ref_scale_ops refcnt_ops = { .name = "refcnt" }; +static void ref_percpuinc_section(const int nloops) +{ + int i; + + for (i = nloops; i >= 0; i--) { + this_cpu_inc(test_acqrel); + this_cpu_dec(test_acqrel); + } +} + +static void ref_percpuinc_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + for (i = nloops; i >= 0; i--) { + this_cpu_inc(test_acqrel); + un_delay(udl, ndl); + this_cpu_dec(test_acqrel); + } +} + +static const struct ref_scale_ops percpuinc_ops = { + .init = rcu_sync_scale_init, + .readsection = ref_percpuinc_section, + .delaysection = ref_percpuinc_delay_section, + .name = "percpuinc" +}; + +// Note that this can lose counts in preemptible kernels. +static void ref_incpercpu_section(const int nloops) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap = this_cpu_ptr(&test_acqrel); + + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + } +} + +static void ref_incpercpu_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap = this_cpu_ptr(&test_acqrel); + + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + un_delay(udl, ndl); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + } +} + +static const struct ref_scale_ops incpercpu_ops = { + .init = rcu_sync_scale_init, + .readsection = ref_incpercpu_section, + .delaysection = ref_incpercpu_delay_section, + .name = "incpercpu" +}; + +static void ref_incpercpupreempt_section(const int nloops) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + preempt_disable(); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + preempt_enable(); + } +} + +static void ref_incpercpupreempt_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + preempt_disable(); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + un_delay(udl, ndl); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + preempt_enable(); + } +} + +static const struct ref_scale_ops incpercpupreempt_ops = { + .init = rcu_sync_scale_init, + .readsection = ref_incpercpupreempt_section, + .delaysection = ref_incpercpupreempt_delay_section, + .name = "incpercpupreempt" +}; + +static void ref_incpercpubh_section(const int nloops) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + local_bh_disable(); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + local_bh_enable(); + } +} + +static void ref_incpercpubh_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + local_bh_disable(); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + un_delay(udl, ndl); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + local_bh_enable(); + } +} + +static const struct ref_scale_ops incpercpubh_ops = { + .init = rcu_sync_scale_init, + .readsection = ref_incpercpubh_section, + .delaysection = ref_incpercpubh_delay_section, + .enable_irqs = true, + .name = "incpercpubh" +}; + +static void ref_incpercpuirqsave_section(const int nloops) +{ + int i; + unsigned long flags; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + local_irq_save(flags); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + local_irq_restore(flags); + } +} + +static void ref_incpercpuirqsave_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + unsigned long flags; + + for (i = nloops; i >= 0; i--) { + unsigned long *tap; + + local_irq_save(flags); + tap = this_cpu_ptr(&test_acqrel); + WRITE_ONCE(*tap, READ_ONCE(*tap) + 1); + un_delay(udl, ndl); + WRITE_ONCE(*tap, READ_ONCE(*tap) - 1); + local_irq_restore(flags); + } +} + +static const struct ref_scale_ops incpercpuirqsave_ops = { + .init = rcu_sync_scale_init, + .readsection = ref_incpercpuirqsave_section, + .delaysection = ref_incpercpuirqsave_delay_section, + .name = "incpercpuirqsave" +}; + // Definitions for rwlock static rwlock_t test_rwlock; @@ -494,9 +710,6 @@ static const struct ref_scale_ops lock_irq_ops = { .name = "lock-irq" }; -// Definitions acquire-release. -static DEFINE_PER_CPU(unsigned long, test_acqrel); - static void ref_acqrel_section(const int nloops) { unsigned long x; @@ -629,6 +842,133 @@ static const struct ref_scale_ops jiffies_ops = { .name = "jiffies" }; +static void ref_preempt_section(const int nloops) +{ + int i; + + migrate_disable(); + for (i = nloops; i >= 0; i--) { + preempt_disable(); + preempt_enable(); + } + migrate_enable(); +} + +static void ref_preempt_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + migrate_disable(); + for (i = nloops; i >= 0; i--) { + preempt_disable(); + un_delay(udl, ndl); + preempt_enable(); + } + migrate_enable(); +} + +static const struct ref_scale_ops preempt_ops = { + .readsection = ref_preempt_section, + .delaysection = ref_preempt_delay_section, + .name = "preempt" +}; + +static void ref_bh_section(const int nloops) +{ + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_bh_disable(); + local_bh_enable(); + } + preempt_enable(); +} + +static void ref_bh_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_bh_disable(); + un_delay(udl, ndl); + local_bh_enable(); + } + preempt_enable(); +} + +static const struct ref_scale_ops bh_ops = { + .readsection = ref_bh_section, + .delaysection = ref_bh_delay_section, + .enable_irqs = true, + .name = "bh" +}; + +static void ref_irq_section(const int nloops) +{ + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_irq_disable(); + local_irq_enable(); + } + preempt_enable(); +} + +static void ref_irq_delay_section(const int nloops, const int udl, const int ndl) +{ + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_irq_disable(); + un_delay(udl, ndl); + local_irq_enable(); + } + preempt_enable(); +} + +static const struct ref_scale_ops irq_ops = { + .readsection = ref_irq_section, + .delaysection = ref_irq_delay_section, + .name = "irq" +}; + +static void ref_irqsave_section(const int nloops) +{ + unsigned long flags; + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_irq_save(flags); + local_irq_restore(flags); + } + preempt_enable(); +} + +static void ref_irqsave_delay_section(const int nloops, const int udl, const int ndl) +{ + unsigned long flags; + int i; + + preempt_disable(); + for (i = nloops; i >= 0; i--) { + local_irq_save(flags); + un_delay(udl, ndl); + local_irq_restore(flags); + } + preempt_enable(); +} + +static const struct ref_scale_ops irqsave_ops = { + .readsection = ref_irqsave_section, + .delaysection = ref_irqsave_delay_section, + .name = "irqsave" +}; + //////////////////////////////////////////////////////////////////////// // // Methods leveraging SLAB_TYPESAFE_BY_RCU. @@ -793,7 +1133,7 @@ static bool typesafe_init(void) else if (si == 0) si = nr_cpu_ids; rtsarray_size = si; - rtsarray = kcalloc(si, sizeof(*rtsarray), GFP_KERNEL); + rtsarray = kzalloc_objs(*rtsarray, si); if (!rtsarray) return false; for (idx = 0; idx < rtsarray_size; idx++) { @@ -924,15 +1264,18 @@ repeat: if (!atomic_dec_return(&n_warmedup)) while (atomic_read_acquire(&n_warmedup)) rcu_scale_one_reader(); - // Also keep interrupts disabled. This also has the effect - // of preventing entries into slow path for rcu_read_unlock(). - local_irq_save(flags); + // Also keep interrupts disabled when it is safe to do so, which + // it is not for local_bh_enable(). This also has the effect of + // preventing entries into slow path for rcu_read_unlock(). + if (!cur_ops->enable_irqs) + local_irq_save(flags); start = ktime_get_mono_fast_ns(); rcu_scale_one_reader(); duration = ktime_get_mono_fast_ns() - start; - local_irq_restore(flags); + if (!cur_ops->enable_irqs) + local_irq_restore(flags); rt->last_duration_ns = WARN_ON_ONCE(duration < 0) ? 0 : duration; // To reduce runtime-skew noise, do maintain-load invocations until @@ -1004,6 +1347,8 @@ static u64 process_durations(int n) return sum; } +static void ref_scale_cleanup(void); + // The main_func is the main orchestrator, it performs a bunch of // experiments. For every experiment, it orders all the readers // involved to start and waits for them to finish the experiment. It @@ -1021,7 +1366,7 @@ static int main_func(void *arg) set_user_nice(current, MAX_NICE); VERBOSE_SCALEOUT("main_func task started"); - result_avg = kzalloc(nruns * sizeof(*result_avg), GFP_KERNEL); + result_avg = kcalloc(nruns, sizeof(*result_avg), GFP_KERNEL); buf = kzalloc(800 + 64, GFP_KERNEL); if (!result_avg || !buf) { SCALEOUT_ERRSTRING("out of memory"); @@ -1090,9 +1435,10 @@ static int main_func(void *arg) oom_exit: // This will shutdown everything including us. - if (shutdown) { - shutdown_start = 1; - wake_up(&shutdown_wq); + if (shutdown_secs) { + main_task = NULL; // Avoid self-kill deadlock. + ref_scale_cleanup(); + kernel_power_off(); } // Wait for torture to stop us @@ -1110,8 +1456,8 @@ static void ref_scale_print_module_parms(const struct ref_scale_ops *cur_ops, const char *tag) { pr_alert("%s" SCALE_FLAG - "--- %s: verbose=%d verbose_batched=%d shutdown=%d holdoff=%d lookup_instances=%ld loops=%ld nreaders=%d nruns=%d readdelay=%d\n", scale_type, tag, - verbose, verbose_batched, shutdown, holdoff, lookup_instances, loops, nreaders, nruns, readdelay); + "--- %s: verbose=%d verbose_batched=%d shutdown_secs=%d holdoff=%d lookup_instances=%ld loops=%d nreaders=%d nruns=%d readdelay=%d\n", scale_type, tag, + verbose, verbose_batched, shutdown_secs, holdoff, lookup_instances, loops, nreaders, nruns, readdelay); } static void @@ -1133,9 +1479,9 @@ ref_scale_cleanup(void) reader_tasks[i].task); } kfree(reader_tasks); + reader_tasks = NULL; torture_stop_kthread("main_task", main_task); - kfree(main_task); // Do scale-type-specific cleanup operations. if (cur_ops->cleanup != NULL) @@ -1144,28 +1490,19 @@ ref_scale_cleanup(void) torture_cleanup_end(); } -// Shutdown kthread. Just waits to be awakened, then shuts down system. -static int -ref_scale_shutdown(void *arg) -{ - wait_event_idle(shutdown_wq, shutdown_start); - - smp_mb(); // Wake before output. - ref_scale_cleanup(); - kernel_power_off(); - - return -EINVAL; -} - static int __init ref_scale_init(void) { long i; int firsterr = 0; static const struct ref_scale_ops *scale_ops[] = { - &rcu_ops, &srcu_ops, &srcu_lite_ops, RCU_TRACE_OPS RCU_TASKS_OPS - &refcnt_ops, &rwlock_ops, &rwsem_ops, &lock_ops, &lock_irq_ops, - &acqrel_ops, &sched_clock_ops, &clock_ops, &jiffies_ops, + &rcu_ops, &srcu_ops, &srcu_fast_ops, &srcu_fast_updown_ops, + RCU_TRACE_OPS RCU_TASKS_OPS + &refcnt_ops, &percpuinc_ops, &incpercpu_ops, &incpercpupreempt_ops, + &incpercpubh_ops, &incpercpuirqsave_ops, + &rwlock_ops, &rwsem_ops, &lock_ops, &lock_irq_ops, &acqrel_ops, + &sched_clock_ops, &clock_ops, &jiffies_ops, + &preempt_ops, &bh_ops, &irq_ops, &irqsave_ops, &typesafe_ref_ops, &typesafe_lock_ops, &typesafe_seqlock_ops, }; @@ -1196,26 +1533,26 @@ ref_scale_init(void) ref_scale_print_module_parms(cur_ops, "Start of test"); // Shutdown task - if (shutdown) { - init_waitqueue_head(&shutdown_wq); - firsterr = torture_create_kthread(ref_scale_shutdown, NULL, - shutdown_task); + if (shutdown_secs) { + firsterr = torture_shutdown_init(shutdown_secs, ref_scale_cleanup); if (torture_init_error(firsterr)) goto unwind; - schedule_timeout_uninterruptible(1); } // Reader tasks (default to ~75% of online CPUs). if (nreaders < 0) nreaders = (num_online_cpus() >> 1) + (num_online_cpus() >> 2); - if (WARN_ONCE(loops <= 0, "%s: loops = %ld, adjusted to 1\n", __func__, loops)) + if (WARN_ONCE(loops <= 0, "%s: loops = %d, adjusted to 1\n", __func__, loops)) loops = 1; if (WARN_ONCE(nreaders <= 0, "%s: nreaders = %d, adjusted to 1\n", __func__, nreaders)) nreaders = 1; if (WARN_ONCE(nruns <= 0, "%s: nruns = %d, adjusted to 1\n", __func__, nruns)) nruns = 1; - reader_tasks = kcalloc(nreaders, sizeof(reader_tasks[0]), - GFP_KERNEL); + if (WARN_ONCE(loops > INT_MAX / nreaders, + "%s: nreaders * loops will overflow, adjusted loops to %d", + __func__, INT_MAX / nreaders)) + loops = INT_MAX / nreaders; + reader_tasks = kzalloc_objs(reader_tasks[0], nreaders); if (!reader_tasks) { SCALEOUT_ERRSTRING("out of memory"); firsterr = -ENOMEM; @@ -1244,7 +1581,7 @@ ref_scale_init(void) unwind: torture_init_end(); ref_scale_cleanup(); - if (shutdown) { + if (shutdown_secs) { WARN_ON(!IS_MODULE(CONFIG_RCU_REF_SCALE_TEST)); kernel_power_off(); } diff --git a/kernel/rcu/srcutiny.c b/kernel/rcu/srcutiny.c index 4dcbf8aa80ff..a2e2d516e51b 100644 --- a/kernel/rcu/srcutiny.c +++ b/kernel/rcu/srcutiny.c @@ -9,6 +9,7 @@ */ #include <linux/export.h> +#include <linux/irq_work.h> #include <linux/mutex.h> #include <linux/preempt.h> #include <linux/rcupdate_wait.h> @@ -20,7 +21,11 @@ #include "rcu_segcblist.h" #include "rcu.h" +#ifndef CONFIG_TREE_RCU int rcu_scheduler_active __read_mostly; +#else // #ifndef CONFIG_TREE_RCU +extern int rcu_scheduler_active; +#endif // #else // #ifndef CONFIG_TREE_RCU static LIST_HEAD(srcu_boot_list); static bool srcu_init_done; @@ -37,6 +42,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp) ssp->srcu_idx_max = 0; INIT_WORK(&ssp->srcu_work, srcu_drive_gp); INIT_LIST_HEAD(&ssp->srcu_work.entry); + init_irq_work(&ssp->srcu_irq_work, srcu_tiny_irq_work); return 0; } @@ -80,6 +86,7 @@ EXPORT_SYMBOL_GPL(init_srcu_struct); void cleanup_srcu_struct(struct srcu_struct *ssp) { WARN_ON(ssp->srcu_lock_nesting[0] || ssp->srcu_lock_nesting[1]); + irq_work_sync(&ssp->srcu_irq_work); flush_work(&ssp->srcu_work); WARN_ON(ssp->srcu_gp_running); WARN_ON(ssp->srcu_gp_waiting); @@ -98,19 +105,19 @@ void __srcu_read_unlock(struct srcu_struct *ssp, int idx) { int newval; - preempt_disable(); // Needed for PREEMPT_AUTO + preempt_disable(); // Needed for PREEMPT_LAZY newval = READ_ONCE(ssp->srcu_lock_nesting[idx]) - 1; WRITE_ONCE(ssp->srcu_lock_nesting[idx], newval); preempt_enable(); - if (!newval && READ_ONCE(ssp->srcu_gp_waiting) && in_task()) + if (!newval && READ_ONCE(ssp->srcu_gp_waiting) && in_task() && !irqs_disabled()) swake_up_one(&ssp->srcu_wq); } EXPORT_SYMBOL_GPL(__srcu_read_unlock); /* * Workqueue handler to drive one grace period and invoke any callbacks - * that become ready as a result. Single-CPU and !PREEMPTION operation - * means that we get away with murder on synchronization. ;-) + * that become ready as a result. Single-CPU operation and preemption + * disabling mean that we get away with murder on synchronization. ;-) */ void srcu_drive_gp(struct work_struct *wp) { @@ -120,7 +127,7 @@ void srcu_drive_gp(struct work_struct *wp) struct srcu_struct *ssp; ssp = container_of(wp, struct srcu_struct, srcu_work); - preempt_disable(); // Needed for PREEMPT_AUTO + preempt_disable(); // Needed for PREEMPT_LAZY if (ssp->srcu_gp_running || ULONG_CMP_GE(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max))) { preempt_enable(); return; /* Already running or nothing to do. */ @@ -137,8 +144,13 @@ void srcu_drive_gp(struct work_struct *wp) WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); WRITE_ONCE(ssp->srcu_gp_waiting, true); /* srcu_read_unlock() wakes! */ preempt_enable(); - swait_event_exclusive(ssp->srcu_wq, !READ_ONCE(ssp->srcu_lock_nesting[idx])); - preempt_disable(); // Needed for PREEMPT_AUTO + do { + // Deadlock issues prevent __srcu_read_unlock() from + // doing an unconditional wakeup, so polling is required. + swait_event_timeout_exclusive(ssp->srcu_wq, + !READ_ONCE(ssp->srcu_lock_nesting[idx]), HZ / 10); + } while (READ_ONCE(ssp->srcu_lock_nesting[idx])); + preempt_disable(); // Needed for PREEMPT_LAZY WRITE_ONCE(ssp->srcu_gp_waiting, false); /* srcu_read_unlock() cheap. */ WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); preempt_enable(); @@ -159,7 +171,7 @@ void srcu_drive_gp(struct work_struct *wp) * at interrupt level, but the ->srcu_gp_running checks will * straighten that out. */ - preempt_disable(); // Needed for PREEMPT_AUTO + preempt_disable(); // Needed for PREEMPT_LAZY WRITE_ONCE(ssp->srcu_gp_running, false); idx = ULONG_CMP_LT(ssp->srcu_idx, READ_ONCE(ssp->srcu_idx_max)); preempt_enable(); @@ -168,24 +180,36 @@ void srcu_drive_gp(struct work_struct *wp) } EXPORT_SYMBOL_GPL(srcu_drive_gp); +/* + * Use an irq_work to defer schedule_work() to avoid acquiring the workqueue + * pool->lock while the caller might hold scheduler locks, causing lockdep + * splats due to workqueue_init() doing a wakeup. + */ +void srcu_tiny_irq_work(struct irq_work *irq_work) +{ + struct srcu_struct *ssp; + + ssp = container_of(irq_work, struct srcu_struct, srcu_irq_work); + schedule_work(&ssp->srcu_work); +} +EXPORT_SYMBOL_GPL(srcu_tiny_irq_work); + static void srcu_gp_start_if_needed(struct srcu_struct *ssp) { unsigned long cookie; - preempt_disable(); // Needed for PREEMPT_AUTO + lockdep_assert_preemption_disabled(); // Needed for PREEMPT_LAZY cookie = get_state_synchronize_srcu(ssp); if (ULONG_CMP_GE(READ_ONCE(ssp->srcu_idx_max), cookie)) { - preempt_enable(); return; } WRITE_ONCE(ssp->srcu_idx_max, cookie); if (!READ_ONCE(ssp->srcu_gp_running)) { if (likely(srcu_init_done)) - schedule_work(&ssp->srcu_work); + irq_work_queue(&ssp->srcu_irq_work); else if (list_empty(&ssp->srcu_work.entry)) list_add(&ssp->srcu_work.entry, &srcu_boot_list); } - preempt_enable(); } /* @@ -199,7 +223,7 @@ void call_srcu(struct srcu_struct *ssp, struct rcu_head *rhp, rhp->func = func; rhp->next = NULL; - preempt_disable(); // Needed for PREEMPT_AUTO + preempt_disable(); // Needed for PREEMPT_LAZY local_irq_save(flags); *ssp->srcu_cb_tail = rhp; ssp->srcu_cb_tail = &rhp->next; @@ -261,7 +285,7 @@ unsigned long start_poll_synchronize_srcu(struct srcu_struct *ssp) { unsigned long ret; - preempt_disable(); // Needed for PREEMPT_AUTO + preempt_disable(); // Needed for PREEMPT_LAZY ret = get_state_synchronize_srcu(ssp); srcu_gp_start_if_needed(ssp); preempt_enable(); @@ -282,11 +306,13 @@ bool poll_state_synchronize_srcu(struct srcu_struct *ssp, unsigned long cookie) } EXPORT_SYMBOL_GPL(poll_state_synchronize_srcu); +#ifndef CONFIG_TREE_RCU /* Lockdep diagnostics. */ void __init rcu_scheduler_starting(void) { rcu_scheduler_active = RCU_SCHEDULER_RUNNING; } +#endif // #ifndef CONFIG_TREE_RCU /* * Queue work for srcu_struct structures with early boot callbacks. diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index b83c74c4dcc0..7c2f7cc131f7 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -19,6 +19,7 @@ #include <linux/mutex.h> #include <linux/percpu.h> #include <linux/preempt.h> +#include <linux/irq_work.h> #include <linux/rcupdate_wait.h> #include <linux/sched.h> #include <linux/smp.h> @@ -75,49 +76,15 @@ static bool __read_mostly srcu_init_done; static void srcu_invoke_callbacks(struct work_struct *work); static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay); static void process_srcu(struct work_struct *work); +static void srcu_irq_work(struct irq_work *work); static void srcu_delay_timer(struct timer_list *t); -/* Wrappers for lock acquisition and release, see raw_spin_lock_rcu_node(). */ -#define spin_lock_rcu_node(p) \ -do { \ - spin_lock(&ACCESS_PRIVATE(p, lock)); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_unlock_rcu_node(p) spin_unlock(&ACCESS_PRIVATE(p, lock)) - -#define spin_lock_irq_rcu_node(p) \ -do { \ - spin_lock_irq(&ACCESS_PRIVATE(p, lock)); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_unlock_irq_rcu_node(p) \ - spin_unlock_irq(&ACCESS_PRIVATE(p, lock)) - -#define spin_lock_irqsave_rcu_node(p, flags) \ -do { \ - spin_lock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_trylock_irqsave_rcu_node(p, flags) \ -({ \ - bool ___locked = spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ - \ - if (___locked) \ - smp_mb__after_unlock_lock(); \ - ___locked; \ -}) - -#define spin_unlock_irqrestore_rcu_node(p, flags) \ - spin_unlock_irqrestore(&ACCESS_PRIVATE(p, lock), flags) \ - /* * Initialize SRCU per-CPU data. Note that statically allocated * srcu_struct structures might already have srcu_read_lock() and - * srcu_read_unlock() running against them. So if the is_static parameter - * is set, don't initialize ->srcu_lock_count[] and ->srcu_unlock_count[]. + * srcu_read_unlock() running against them. So if the is_static + * parameter is set, don't initialize ->srcu_ctrs[].srcu_locks and + * ->srcu_ctrs[].srcu_unlocks. */ static void init_srcu_struct_data(struct srcu_struct *ssp) { @@ -128,11 +95,9 @@ static void init_srcu_struct_data(struct srcu_struct *ssp) * Initialize the per-CPU srcu_data array, which feeds into the * leaves of the srcu_node tree. */ - BUILD_BUG_ON(ARRAY_SIZE(sdp->srcu_lock_count) != - ARRAY_SIZE(sdp->srcu_unlock_count)); for_each_possible_cpu(cpu) { sdp = per_cpu_ptr(ssp->sda, cpu); - spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); rcu_segcblist_init(&sdp->srcu_cblist); sdp->srcu_cblist_invoking = false; sdp->srcu_gp_seq_needed = ssp->srcu_sup->srcu_gp_seq; @@ -174,7 +139,8 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) /* Initialize geometry if it has not already been initialized. */ rcu_init_geometry(); - ssp->srcu_sup->node = kcalloc(rcu_num_nodes, sizeof(*ssp->srcu_sup->node), gfp_flags); + ssp->srcu_sup->node = kzalloc_objs(*ssp->srcu_sup->node, rcu_num_nodes, + gfp_flags); if (!ssp->srcu_sup->node) return false; @@ -186,7 +152,7 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) /* Each pass through this loop initializes one srcu_node structure. */ srcu_for_each_node_breadth_first(ssp, snp) { - spin_lock_init(&ACCESS_PRIVATE(snp, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(snp, lock)); BUILD_BUG_ON(ARRAY_SIZE(snp->srcu_have_cbs) != ARRAY_SIZE(snp->srcu_data_have_cbs)); for (i = 0; i < ARRAY_SIZE(snp->srcu_have_cbs); i++) { @@ -238,33 +204,38 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) { if (!is_static) - ssp->srcu_sup = kzalloc(sizeof(*ssp->srcu_sup), GFP_KERNEL); + ssp->srcu_sup = kzalloc_obj(*ssp->srcu_sup); if (!ssp->srcu_sup) return -ENOMEM; if (!is_static) - spin_lock_init(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_sup->srcu_cb_mutex); mutex_init(&ssp->srcu_sup->srcu_gp_mutex); - ssp->srcu_idx = 0; ssp->srcu_sup->srcu_gp_seq = SRCU_GP_SEQ_INITIAL_VAL; ssp->srcu_sup->srcu_barrier_seq = 0; mutex_init(&ssp->srcu_sup->srcu_barrier_mutex); atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 0); INIT_DELAYED_WORK(&ssp->srcu_sup->work, process_srcu); + init_irq_work(&ssp->srcu_sup->irq_work, srcu_irq_work); ssp->srcu_sup->sda_is_static = is_static; - if (!is_static) + if (!is_static) { ssp->sda = alloc_percpu(struct srcu_data); + ssp->srcu_ctrp = &ssp->sda->srcu_ctrs[0]; + } if (!ssp->sda) goto err_free_sup; init_srcu_struct_data(ssp); ssp->srcu_sup->srcu_gp_seq_needed_exp = SRCU_GP_SEQ_INITIAL_VAL; ssp->srcu_sup->srcu_last_gp_end = ktime_get_mono_fast_ns(); if (READ_ONCE(ssp->srcu_sup->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { - if (!init_srcu_struct_nodes(ssp, GFP_ATOMIC)) + if (!preemptible()) + WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_ALLOC); + else if (init_srcu_struct_nodes(ssp, GFP_KERNEL)) + WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); + else goto err_free_sda; - WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); } ssp->srcu_sup->srcu_ssp = ssp; smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, @@ -286,32 +257,92 @@ err_free_sup: #ifdef CONFIG_DEBUG_LOCK_ALLOC -int __init_srcu_struct(struct srcu_struct *ssp, const char *name, - struct lock_class_key *key) +static int +__init_srcu_struct_common(struct srcu_struct *ssp, const char *name, struct lock_class_key *key) { /* Don't re-initialize a lock while it is held. */ debug_check_no_locks_freed((void *)ssp, sizeof(*ssp)); lockdep_init_map(&ssp->dep_map, name, key, 0); return init_srcu_struct_fields(ssp, false); } + +int __init_srcu_struct(struct srcu_struct *ssp, const char *name, struct lock_class_key *key) +{ + ssp->srcu_reader_flavor = 0; + return __init_srcu_struct_common(ssp, name, key); +} EXPORT_SYMBOL_GPL(__init_srcu_struct); +int __init_srcu_struct_fast(struct srcu_struct *ssp, const char *name, struct lock_class_key *key) +{ + ssp->srcu_reader_flavor = SRCU_READ_FLAVOR_FAST; + return __init_srcu_struct_common(ssp, name, key); +} +EXPORT_SYMBOL_GPL(__init_srcu_struct_fast); + +int __init_srcu_struct_fast_updown(struct srcu_struct *ssp, const char *name, + struct lock_class_key *key) +{ + ssp->srcu_reader_flavor = SRCU_READ_FLAVOR_FAST_UPDOWN; + return __init_srcu_struct_common(ssp, name, key); +} +EXPORT_SYMBOL_GPL(__init_srcu_struct_fast_updown); + #else /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ /** * init_srcu_struct - initialize a sleep-RCU structure * @ssp: structure to initialize. * - * Must invoke this on a given srcu_struct before passing that srcu_struct + * Use this in place of DEFINE_SRCU() and DEFINE_STATIC_SRCU() + * for non-static srcu_struct structures that are to be passed to + * srcu_read_lock(), srcu_read_lock_nmisafe(), and friends. It is necessary + * to invoke this on a given srcu_struct before passing that srcu_struct * to any other function. Each srcu_struct represents a separate domain * of SRCU protection. */ int init_srcu_struct(struct srcu_struct *ssp) { + ssp->srcu_reader_flavor = 0; return init_srcu_struct_fields(ssp, false); } EXPORT_SYMBOL_GPL(init_srcu_struct); +/** + * init_srcu_struct_fast - initialize a fast-reader sleep-RCU structure + * @ssp: structure to initialize. + * + * Use this in place of DEFINE_SRCU_FAST() and DEFINE_STATIC_SRCU_FAST() + * for non-static srcu_struct structures that are to be passed to + * srcu_read_lock_fast() and friends. It is necessary to invoke this on a + * given srcu_struct before passing that srcu_struct to any other function. + * Each srcu_struct represents a separate domain of SRCU protection. + */ +int init_srcu_struct_fast(struct srcu_struct *ssp) +{ + ssp->srcu_reader_flavor = SRCU_READ_FLAVOR_FAST; + return init_srcu_struct_fields(ssp, false); +} +EXPORT_SYMBOL_GPL(init_srcu_struct_fast); + +/** + * init_srcu_struct_fast_updown - initialize a fast-reader up/down sleep-RCU structure + * @ssp: structure to initialize. + * + * Use this function in place of DEFINE_SRCU_FAST_UPDOWN() and + * DEFINE_STATIC_SRCU_FAST_UPDOWN() for non-static srcu_struct + * structures that are to be passed to srcu_read_lock_fast_updown(), + * srcu_down_read_fast(), and friends. It is necessary to invoke this on a + * given srcu_struct before passing that srcu_struct to any other function. + * Each srcu_struct represents a separate domain of SRCU protection. + */ +int init_srcu_struct_fast_updown(struct srcu_struct *ssp) +{ + ssp->srcu_reader_flavor = SRCU_READ_FLAVOR_FAST_UPDOWN; + return init_srcu_struct_fields(ssp, false); +} +EXPORT_SYMBOL_GPL(init_srcu_struct_fast_updown); + #endif /* #else #ifdef CONFIG_DEBUG_LOCK_ALLOC */ /* @@ -333,20 +364,20 @@ static void srcu_transition_to_big(struct srcu_struct *ssp) /* Double-checked locking on ->srcu_size-state. */ if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) { - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } __srcu_transition_to_big(ssp); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* * Check to see if the just-encountered contention event justifies * a transition to SRCU_SIZE_BIG. */ -static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) +static void raw_spin_lock_irqsave_check_contention(struct srcu_struct *ssp) { unsigned long j; @@ -368,16 +399,16 @@ static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) * to SRCU_SIZE_BIG. But only if the srcutree.convert_to_big module * parameter permits this. */ -static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned long *flags) +static void raw_spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned long *flags) { struct srcu_struct *ssp = sdp->ssp; - if (spin_trylock_irqsave_rcu_node(sdp, *flags)) + if (raw_spin_trylock_irqsave_rcu_node(sdp, *flags)) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_check_contention(ssp); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_rcu_node(sdp, *flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_check_contention(ssp); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_rcu_node(sdp, *flags); } /* @@ -386,12 +417,12 @@ static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned lon * to SRCU_SIZE_BIG. But only if the srcutree.convert_to_big module * parameter permits this. */ -static void spin_lock_irqsave_ssp_contention(struct srcu_struct *ssp, unsigned long *flags) +static void raw_spin_lock_irqsave_ssp_contention(struct srcu_struct *ssp, unsigned long *flags) { - if (spin_trylock_irqsave_rcu_node(ssp->srcu_sup, *flags)) + if (raw_spin_trylock_irqsave_rcu_node(ssp->srcu_sup, *flags)) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_check_contention(ssp); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_check_contention(ssp); } /* @@ -409,13 +440,13 @@ static void check_init_srcu_struct(struct srcu_struct *ssp) /* The smp_load_acquire() pairs with the smp_store_release(). */ if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed))) /*^^^*/ return; /* Already initialized. */ - spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (!rcu_seq_state(ssp->srcu_sup->srcu_gp_seq_needed)) { - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } init_srcu_struct_fields(ssp, true); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -429,10 +460,10 @@ static bool srcu_gp_is_expedited(struct srcu_struct *ssp) } /* - * Computes approximate total of the readers' ->srcu_lock_count[] values - * for the rank of per-CPU counters specified by idx, and returns true if - * the caller did the proper barrier (gp), and if the count of the locks - * matches that of the unlocks passed in. + * Computes approximate total of the readers' ->srcu_ctrs[].srcu_locks + * values for the rank of per-CPU counters specified by idx, and returns + * true if the caller did the proper barrier (gp), and if the count of + * the locks matches that of the unlocks passed in. */ static bool srcu_readers_lock_idx(struct srcu_struct *ssp, int idx, bool gp, unsigned long unlocks) { @@ -443,31 +474,31 @@ static bool srcu_readers_lock_idx(struct srcu_struct *ssp, int idx, bool gp, uns for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); - sum += atomic_long_read(&sdp->srcu_lock_count[idx]); + sum += atomic_long_read(&sdp->srcu_ctrs[idx].srcu_locks); if (IS_ENABLED(CONFIG_PROVE_RCU)) mask = mask | READ_ONCE(sdp->srcu_reader_flavor); } WARN_ONCE(IS_ENABLED(CONFIG_PROVE_RCU) && (mask & (mask - 1)), "Mixed reader flavors for srcu_struct at %ps.\n", ssp); - if (mask & SRCU_READ_FLAVOR_LITE && !gp) + if (mask & SRCU_READ_FLAVOR_SLOWGP && !gp) return false; return sum == unlocks; } /* - * Returns approximate total of the readers' ->srcu_unlock_count[] values - * for the rank of per-CPU counters specified by idx. + * Returns approximate total of the readers' ->srcu_ctrs[].srcu_unlocks + * values for the rank of per-CPU counters specified by idx. */ static unsigned long srcu_readers_unlock_idx(struct srcu_struct *ssp, int idx, unsigned long *rdm) { int cpu; - unsigned long mask = 0; + unsigned long mask = ssp->srcu_reader_flavor; unsigned long sum = 0; for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); - sum += atomic_long_read(&sdp->srcu_unlock_count[idx]); + sum += atomic_long_read(&sdp->srcu_ctrs[idx].srcu_unlocks); mask = mask | READ_ONCE(sdp->srcu_reader_flavor); } WARN_ONCE(IS_ENABLED(CONFIG_PROVE_RCU) && (mask & (mask - 1)), @@ -487,7 +518,7 @@ static bool srcu_readers_active_idx_check(struct srcu_struct *ssp, int idx) unsigned long unlocks; unlocks = srcu_readers_unlock_idx(ssp, idx, &rdm); - did_gp = !!(rdm & SRCU_READ_FLAVOR_LITE); + did_gp = !!(rdm & SRCU_READ_FLAVOR_SLOWGP); /* * Make sure that a lock is always counted if the corresponding @@ -502,6 +533,8 @@ static bool srcu_readers_active_idx_check(struct srcu_struct *ssp, int idx) */ if (!did_gp) smp_mb(); /* A */ + else if (srcu_gp_is_expedited(ssp)) + synchronize_rcu_expedited(); /* X */ else synchronize_rcu(); /* X */ @@ -509,48 +542,49 @@ static bool srcu_readers_active_idx_check(struct srcu_struct *ssp, int idx) * If the locks are the same as the unlocks, then there must have * been no readers on this index at some point in this function. * But there might be more readers, as a task might have read - * the current ->srcu_idx but not yet have incremented its CPU's - * ->srcu_lock_count[idx] counter. In fact, it is possible + * the current ->srcu_ctrp but not yet have incremented its CPU's + * ->srcu_ctrs[idx].srcu_locks counter. In fact, it is possible * that most of the tasks have been preempted between fetching - * ->srcu_idx and incrementing ->srcu_lock_count[idx]. And there - * could be almost (ULONG_MAX / sizeof(struct task_struct)) tasks - * in a system whose address space was fully populated with memory. - * Call this quantity Nt. + * ->srcu_ctrp and incrementing ->srcu_ctrs[idx].srcu_locks. And + * there could be almost (ULONG_MAX / sizeof(struct task_struct)) + * tasks in a system whose address space was fully populated + * with memory. Call this quantity Nt. * - * So suppose that the updater is preempted at this point in the - * code for a long time. That now-preempted updater has already - * flipped ->srcu_idx (possibly during the preceding grace period), - * done an smp_mb() (again, possibly during the preceding grace - * period), and summed up the ->srcu_unlock_count[idx] counters. - * How many times can a given one of the aforementioned Nt tasks - * increment the old ->srcu_idx value's ->srcu_lock_count[idx] - * counter, in the absence of nesting? + * So suppose that the updater is preempted at this + * point in the code for a long time. That now-preempted + * updater has already flipped ->srcu_ctrp (possibly during + * the preceding grace period), done an smp_mb() (again, + * possibly during the preceding grace period), and summed up + * the ->srcu_ctrs[idx].srcu_unlocks counters. How many times + * can a given one of the aforementioned Nt tasks increment the + * old ->srcu_ctrp value's ->srcu_ctrs[idx].srcu_locks counter, + * in the absence of nesting? * * It can clearly do so once, given that it has already fetched - * the old value of ->srcu_idx and is just about to use that value - * to index its increment of ->srcu_lock_count[idx]. But as soon as - * it leaves that SRCU read-side critical section, it will increment - * ->srcu_unlock_count[idx], which must follow the updater's above - * read from that same value. Thus, as soon the reading task does - * an smp_mb() and a later fetch from ->srcu_idx, that task will be - * guaranteed to get the new index. Except that the increment of - * ->srcu_unlock_count[idx] in __srcu_read_unlock() is after the - * smp_mb(), and the fetch from ->srcu_idx in __srcu_read_lock() - * is before the smp_mb(). Thus, that task might not see the new - * value of ->srcu_idx until the -second- __srcu_read_lock(), - * which in turn means that this task might well increment - * ->srcu_lock_count[idx] for the old value of ->srcu_idx twice, - * not just once. + * the old value of ->srcu_ctrp and is just about to use that + * value to index its increment of ->srcu_ctrs[idx].srcu_locks. + * But as soon as it leaves that SRCU read-side critical section, + * it will increment ->srcu_ctrs[idx].srcu_unlocks, which must + * follow the updater's above read from that same value. Thus, + as soon the reading task does an smp_mb() and a later fetch from + * ->srcu_ctrp, that task will be guaranteed to get the new index. + * Except that the increment of ->srcu_ctrs[idx].srcu_unlocks + * in __srcu_read_unlock() is after the smp_mb(), and the fetch + * from ->srcu_ctrp in __srcu_read_lock() is before the smp_mb(). + * Thus, that task might not see the new value of ->srcu_ctrp until + * the -second- __srcu_read_lock(), which in turn means that this + * task might well increment ->srcu_ctrs[idx].srcu_locks for the + * old value of ->srcu_ctrp twice, not just once. * * However, it is important to note that a given smp_mb() takes * effect not just for the task executing it, but also for any * later task running on that same CPU. * - * That is, there can be almost Nt + Nc further increments of - * ->srcu_lock_count[idx] for the old index, where Nc is the number - * of CPUs. But this is OK because the size of the task_struct - * structure limits the value of Nt and current systems limit Nc - * to a few thousand. + * That is, there can be almost Nt + Nc further increments + * of ->srcu_ctrs[idx].srcu_locks for the old index, where Nc + * is the number of CPUs. But this is OK because the size of + * the task_struct structure limits the value of Nt and current + * systems limit Nc to a few thousand. * * OK, but what about nesting? This does impose a limit on * nesting of half of the size of the task_struct structure @@ -581,10 +615,10 @@ static bool srcu_readers_active(struct srcu_struct *ssp) for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); - sum += atomic_long_read(&sdp->srcu_lock_count[0]); - sum += atomic_long_read(&sdp->srcu_lock_count[1]); - sum -= atomic_long_read(&sdp->srcu_unlock_count[0]); - sum -= atomic_long_read(&sdp->srcu_unlock_count[1]); + sum += atomic_long_read(&sdp->srcu_ctrs[0].srcu_locks); + sum += atomic_long_read(&sdp->srcu_ctrs[1].srcu_locks); + sum -= atomic_long_read(&sdp->srcu_ctrs[0].srcu_unlocks); + sum -= atomic_long_read(&sdp->srcu_ctrs[1].srcu_unlocks); } return sum; } @@ -647,6 +681,7 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) unsigned long jbase = SRCU_INTERVAL; struct srcu_usage *sup = ssp->srcu_sup; + lockdep_assert_held(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); if (srcu_gp_is_expedited(ssp)) jbase = 0; if (rcu_seq_state(READ_ONCE(sup->srcu_gp_seq))) { @@ -674,17 +709,23 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) void cleanup_srcu_struct(struct srcu_struct *ssp) { int cpu; + unsigned long delay; struct srcu_usage *sup = ssp->srcu_sup; - if (WARN_ON(!srcu_get_delay(ssp))) + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); + delay = srcu_get_delay(ssp); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); + if (WARN_ON(!delay)) return; /* Just leak it! */ if (WARN_ON(srcu_readers_active(ssp))) return; /* Just leak it! */ + /* Wait for irq_work to finish first as it may queue a new work. */ + irq_work_sync(&sup->irq_work); flush_delayed_work(&sup->work); for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); - del_timer_sync(&sdp->delay_work); + timer_delete_sync(&sdp->delay_work); flush_work(&sdp->work); if (WARN_ON(rcu_segcblist_n_cbs(&sdp->srcu_cblist))) return; /* Forgot srcu_barrier(), so just leak it! */ @@ -721,11 +762,16 @@ void __srcu_check_read_flavor(struct srcu_struct *ssp, int read_flavor) struct srcu_data *sdp; /* NMI-unsafe use in NMI is a bad sign, as is multi-bit read_flavor values. */ - WARN_ON_ONCE((read_flavor != SRCU_READ_FLAVOR_NMI) && in_nmi()); + WARN_ON_ONCE(read_flavor != SRCU_READ_FLAVOR_NMI && + read_flavor != SRCU_READ_FLAVOR_FAST && in_nmi()); WARN_ON_ONCE(read_flavor & (read_flavor - 1)); sdp = raw_cpu_ptr(ssp->sda); old_read_flavor = READ_ONCE(sdp->srcu_reader_flavor); + WARN_ON_ONCE(ssp->srcu_reader_flavor && read_flavor != ssp->srcu_reader_flavor); + WARN_ON_ONCE(old_read_flavor && ssp->srcu_reader_flavor && + old_read_flavor != ssp->srcu_reader_flavor); + WARN_ON_ONCE(read_flavor == SRCU_READ_FLAVOR_FAST && !ssp->srcu_reader_flavor); if (!old_read_flavor) { old_read_flavor = cmpxchg(&sdp->srcu_reader_flavor, 0, read_flavor); if (!old_read_flavor) @@ -743,12 +789,11 @@ EXPORT_SYMBOL_GPL(__srcu_check_read_flavor); */ int __srcu_read_lock(struct srcu_struct *ssp) { - int idx; + struct srcu_ctr __percpu *scp = READ_ONCE(ssp->srcu_ctrp); - idx = READ_ONCE(ssp->srcu_idx) & 0x1; - this_cpu_inc(ssp->sda->srcu_lock_count[idx].counter); + this_cpu_inc(scp->srcu_locks.counter); smp_mb(); /* B */ /* Avoid leaking the critical section. */ - return idx; + return __srcu_ptr_to_ctr(ssp, scp); } EXPORT_SYMBOL_GPL(__srcu_read_lock); @@ -760,7 +805,7 @@ EXPORT_SYMBOL_GPL(__srcu_read_lock); void __srcu_read_unlock(struct srcu_struct *ssp, int idx) { smp_mb(); /* C */ /* Avoid leaking the critical section. */ - this_cpu_inc(ssp->sda->srcu_unlock_count[idx].counter); + this_cpu_inc(__srcu_ctr_to_ptr(ssp, idx)->srcu_unlocks.counter); } EXPORT_SYMBOL_GPL(__srcu_read_unlock); @@ -773,13 +818,12 @@ EXPORT_SYMBOL_GPL(__srcu_read_unlock); */ int __srcu_read_lock_nmisafe(struct srcu_struct *ssp) { - int idx; - struct srcu_data *sdp = raw_cpu_ptr(ssp->sda); + struct srcu_ctr __percpu *scpp = READ_ONCE(ssp->srcu_ctrp); + struct srcu_ctr *scp = raw_cpu_ptr(scpp); - idx = READ_ONCE(ssp->srcu_idx) & 0x1; - atomic_long_inc(&sdp->srcu_lock_count[idx]); + atomic_long_inc(&scp->srcu_locks); smp_mb__after_atomic(); /* B */ /* Avoid leaking the critical section. */ - return idx; + return __srcu_ptr_to_ctr(ssp, scpp); } EXPORT_SYMBOL_GPL(__srcu_read_lock_nmisafe); @@ -790,10 +834,8 @@ EXPORT_SYMBOL_GPL(__srcu_read_lock_nmisafe); */ void __srcu_read_unlock_nmisafe(struct srcu_struct *ssp, int idx) { - struct srcu_data *sdp = raw_cpu_ptr(ssp->sda); - smp_mb__before_atomic(); /* C */ /* Avoid leaking the critical section. */ - atomic_long_inc(&sdp->srcu_unlock_count[idx]); + atomic_long_inc(&raw_cpu_ptr(__srcu_ctr_to_ptr(ssp, idx))->srcu_unlocks); } EXPORT_SYMBOL_GPL(__srcu_read_unlock_nmisafe); @@ -855,11 +897,9 @@ static void srcu_schedule_cbs_snp(struct srcu_struct *ssp, struct srcu_node *snp { int cpu; - for (cpu = snp->grplo; cpu <= snp->grphi; cpu++) { - if (!(mask & (1UL << (cpu - snp->grplo)))) - continue; - srcu_schedule_cbs_sdp(per_cpu_ptr(ssp->sda, cpu), delay); - } + for (cpu = snp->grplo; cpu <= snp->grphi; cpu++) + if ((mask & (1UL << (cpu - snp->grplo))) && rcu_cpu_beenfullyonline(cpu)) + srcu_schedule_cbs_sdp(per_cpu_ptr(ssp->sda, cpu), delay); } /* @@ -890,7 +930,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) mutex_lock(&sup->srcu_cb_mutex); /* End the current grace period. */ - spin_lock_irq_rcu_node(sup); + raw_spin_lock_irq_rcu_node(sup); idx = rcu_seq_state(sup->srcu_gp_seq); WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); if (srcu_gp_is_expedited(ssp)) @@ -901,7 +941,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) gpseq = rcu_seq_current(&sup->srcu_gp_seq); if (ULONG_CMP_LT(sup->srcu_gp_seq_needed_exp, gpseq)) WRITE_ONCE(sup->srcu_gp_seq_needed_exp, gpseq); - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); mutex_unlock(&sup->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ @@ -913,7 +953,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) } else { idx = rcu_seq_ctr(gpseq) % ARRAY_SIZE(snp->srcu_have_cbs); srcu_for_each_node_breadth_first(ssp, snp) { - spin_lock_irq_rcu_node(snp); + raw_spin_lock_irq_rcu_node(snp); cbs = false; last_lvl = snp >= sup->level[rcu_num_lvls - 1]; if (last_lvl) @@ -928,7 +968,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) else mask = snp->srcu_data_have_cbs[idx]; snp->srcu_data_have_cbs[idx] = 0; - spin_unlock_irq_rcu_node(snp); + raw_spin_unlock_irq_rcu_node(snp); if (cbs) srcu_schedule_cbs_snp(ssp, snp, mask, cbdelay); } @@ -938,27 +978,27 @@ static void srcu_gp_end(struct srcu_struct *ssp) if (!(gpseq & counter_wrap_check)) for_each_possible_cpu(cpu) { sdp = per_cpu_ptr(ssp->sda, cpu); - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); if (ULONG_CMP_GE(gpseq, sdp->srcu_gp_seq_needed + 100)) sdp->srcu_gp_seq_needed = gpseq; if (ULONG_CMP_GE(gpseq, sdp->srcu_gp_seq_needed_exp + 100)) sdp->srcu_gp_seq_needed_exp = gpseq; - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); } /* Callback initiation done, allow grace periods after next. */ mutex_unlock(&sup->srcu_cb_mutex); /* Start a new grace period if needed. */ - spin_lock_irq_rcu_node(sup); + raw_spin_lock_irq_rcu_node(sup); gpseq = rcu_seq_current(&sup->srcu_gp_seq); if (!rcu_seq_state(gpseq) && ULONG_CMP_LT(gpseq, sup->srcu_gp_seq_needed)) { srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); srcu_reschedule(ssp, 0); } else { - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); } /* Transition to big if needed. */ @@ -989,19 +1029,19 @@ static void srcu_funnel_exp_start(struct srcu_struct *ssp, struct srcu_node *snp if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) || (!srcu_invl_snp_seq(sgsne) && ULONG_CMP_GE(sgsne, s))) return; - spin_lock_irqsave_rcu_node(snp, flags); + raw_spin_lock_irqsave_rcu_node(snp, flags); sgsne = snp->srcu_gp_seq_needed_exp; if (!srcu_invl_snp_seq(sgsne) && ULONG_CMP_GE(sgsne, s)) { - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); return; } WRITE_ONCE(snp->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); } - spin_lock_irqsave_ssp_contention(ssp, &flags); + raw_spin_lock_irqsave_ssp_contention(ssp, &flags); if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, s)) WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -1039,12 +1079,12 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, for (snp = snp_leaf; snp != NULL; snp = snp->srcu_parent) { if (WARN_ON_ONCE(rcu_seq_done(&sup->srcu_gp_seq, s)) && snp != snp_leaf) return; /* GP already done and CBs recorded. */ - spin_lock_irqsave_rcu_node(snp, flags); + raw_spin_lock_irqsave_rcu_node(snp, flags); snp_seq = snp->srcu_have_cbs[idx]; if (!srcu_invl_snp_seq(snp_seq) && ULONG_CMP_GE(snp_seq, s)) { if (snp == snp_leaf && snp_seq == s) snp->srcu_data_have_cbs[idx] |= sdp->grpmask; - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); if (snp == snp_leaf && snp_seq != s) { srcu_schedule_cbs_sdp(sdp, do_norm ? SRCU_INTERVAL : 0); return; @@ -1059,11 +1099,11 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, sgsne = snp->srcu_gp_seq_needed_exp; if (!do_norm && (srcu_invl_snp_seq(sgsne) || ULONG_CMP_LT(sgsne, s))) WRITE_ONCE(snp->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); } /* Top of tree, must ensure the grace period will be started. */ - spin_lock_irqsave_ssp_contention(ssp, &flags); + raw_spin_lock_irqsave_ssp_contention(ssp, &flags); if (ULONG_CMP_LT(sup->srcu_gp_seq_needed, s)) { /* * Record need for grace period s. Pair with load @@ -1084,25 +1124,31 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, // it isn't. And it does not have to be. After all, it // can only be executed during early boot when there is only // the one boot CPU running with interrupts still disabled. + // + // Use an irq_work here to avoid acquiring runqueue lock with + // srcu rcu_node::lock held. BPF instrument could introduce the + // opposite dependency, hence we need to break the possible + // locking dependency here. if (likely(srcu_init_done)) - queue_delayed_work(rcu_gp_wq, &sup->work, - !!srcu_get_delay(ssp)); + irq_work_queue(&sup->irq_work); else if (list_empty(&sup->work.work.entry)) list_add(&sup->work.work.entry, &srcu_boot_list); } - spin_unlock_irqrestore_rcu_node(sup, flags); + raw_spin_unlock_irqrestore_rcu_node(sup, flags); } /* * Wait until all readers counted by array index idx complete, but * loop an additional time if there is an expedited grace period pending. - * The caller must ensure that ->srcu_idx is not changed while checking. + * The caller must ensure that ->srcu_ctrp is not changed while checking. */ static bool try_check_zero(struct srcu_struct *ssp, int idx, int trycount) { unsigned long curdelay; + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); curdelay = !srcu_get_delay(ssp); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); for (;;) { if (srcu_readers_active_idx_check(ssp, idx)) @@ -1114,30 +1160,30 @@ static bool try_check_zero(struct srcu_struct *ssp, int idx, int trycount) } /* - * Increment the ->srcu_idx counter so that future SRCU readers will + * Increment the ->srcu_ctrp counter so that future SRCU readers will * use the other rank of the ->srcu_(un)lock_count[] arrays. This allows * us to wait for pre-existing readers in a starvation-free manner. */ static void srcu_flip(struct srcu_struct *ssp) { /* - * Because the flip of ->srcu_idx is executed only if the + * Because the flip of ->srcu_ctrp is executed only if the * preceding call to srcu_readers_active_idx_check() found that - * the ->srcu_unlock_count[] and ->srcu_lock_count[] sums matched - * and because that summing uses atomic_long_read(), there is - * ordering due to a control dependency between that summing and - * the WRITE_ONCE() in this call to srcu_flip(). This ordering - * ensures that if this updater saw a given reader's increment from - * __srcu_read_lock(), that reader was using a value of ->srcu_idx - * from before the previous call to srcu_flip(), which should be - * quite rare. This ordering thus helps forward progress because - * the grace period could otherwise be delayed by additional - * calls to __srcu_read_lock() using that old (soon to be new) - * value of ->srcu_idx. + * the ->srcu_ctrs[].srcu_unlocks and ->srcu_ctrs[].srcu_locks sums + * matched and because that summing uses atomic_long_read(), + * there is ordering due to a control dependency between that + * summing and the WRITE_ONCE() in this call to srcu_flip(). + * This ordering ensures that if this updater saw a given reader's + * increment from __srcu_read_lock(), that reader was using a value + * of ->srcu_ctrp from before the previous call to srcu_flip(), + * which should be quite rare. This ordering thus helps forward + * progress because the grace period could otherwise be delayed + * by additional calls to __srcu_read_lock() using that old (soon + * to be new) value of ->srcu_ctrp. * * This sum-equality check and ordering also ensures that if * a given call to __srcu_read_lock() uses the new value of - * ->srcu_idx, this updater's earlier scans cannot have seen + * ->srcu_ctrp, this updater's earlier scans cannot have seen * that reader's increments, which is all to the good, because * this grace period need not wait on that reader. After all, * if those earlier scans had seen that reader, there would have @@ -1152,7 +1198,8 @@ static void srcu_flip(struct srcu_struct *ssp) */ smp_mb(); /* E */ /* Pairs with B and C. */ - WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); // Flip the counter. + WRITE_ONCE(ssp->srcu_ctrp, + &ssp->sda->srcu_ctrs[!(ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0])]); /* * Ensure that if the updater misses an __srcu_read_unlock() @@ -1161,6 +1208,16 @@ static void srcu_flip(struct srcu_struct *ssp) * counter update. Note that both this memory barrier and the * one in srcu_readers_active_idx_check() provide the guarantee * for __srcu_read_lock(). + * + * Note that this is a performance optimization, in which we spend + * an otherwise unnecessary smp_mb() in order to reduce the number + * of full per-CPU-variable scans in srcu_readers_lock_idx() and + * srcu_readers_unlock_idx(). But this performance optimization + * is not so optimal for SRCU-fast, where we would be spending + * not smp_mb(), but rather synchronize_rcu(). At the same time, + * the overhead of the smp_mb() is in the noise, so there is no + * point in omitting it in the SRCU-fast case. So the same code + * is executed either way. */ smp_mb(); /* D */ /* Pairs with C. */ } @@ -1198,16 +1255,16 @@ static bool srcu_should_expedite(struct srcu_struct *ssp) check_init_srcu_struct(ssp); /* If _lite() readers, don't do unsolicited expediting. */ - if (this_cpu_read(ssp->sda->srcu_reader_flavor) & SRCU_READ_FLAVOR_LITE) + if (this_cpu_read(ssp->sda->srcu_reader_flavor) & SRCU_READ_FLAVOR_SLOWGP) return false; /* If the local srcu_data structure has callbacks, not idle. */ sdp = raw_cpu_ptr(ssp->sda); - spin_lock_irqsave_rcu_node(sdp, flags); + raw_spin_lock_irqsave_rcu_node(sdp, flags); if (rcu_segcblist_pend_cbs(&sdp->srcu_cblist)) { - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); return false; /* Callbacks already present, so not idle. */ } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); /* * No local callbacks, so probabilistically probe global state. @@ -1263,11 +1320,13 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, */ idx = __srcu_read_lock_nmisafe(ssp); ss_state = smp_load_acquire(&ssp->srcu_sup->srcu_size_state); - if (ss_state < SRCU_SIZE_WAIT_CALL) + // If !rcu_cpu_beenfullyonline(), interrupts are still disabled, + // so no migration is possible in either direction from this CPU. + if (ss_state < SRCU_SIZE_WAIT_CALL || !rcu_cpu_beenfullyonline(raw_smp_processor_id())) sdp = per_cpu_ptr(ssp->sda, get_boot_cpu_id()); else sdp = raw_cpu_ptr(ssp->sda); - spin_lock_irqsave_sdp_contention(sdp, &flags); + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); if (rhp) rcu_segcblist_enqueue(&sdp->srcu_cblist, rhp); /* @@ -1327,7 +1386,7 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, sdp->srcu_gp_seq_needed_exp = s; needexp = true; } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); /* Ensure that snp node tree is fully initialized before traversing it */ if (ss_state < SRCU_SIZE_WAIT_BARRIER) @@ -1398,8 +1457,12 @@ static void __call_srcu(struct srcu_struct *ssp, struct rcu_head *rhp, * read-side critical sections are delimited by srcu_read_lock() and * srcu_read_unlock(), and may be nested. * - * The callback will be invoked from process context, but must nevertheless - * be fast and must not block. + * The callback will be invoked from process context, but with bh + * disabled. The callback function must therefore be fast and must + * not block. + * + * See the description of call_rcu() for more detailed information on + * memory ordering guarantees. */ void call_srcu(struct srcu_struct *ssp, struct rcu_head *rhp, rcu_callback_t func) @@ -1435,7 +1498,7 @@ static void __synchronize_srcu(struct srcu_struct *ssp, bool do_norm) /* * Make sure that later code is ordered after the SRCU grace - * period. This pairs with the spin_lock_irq_rcu_node() + * period. This pairs with the raw_spin_lock_irq_rcu_node() * in srcu_invoke_callbacks(). Unlike Tree RCU, this is needed * because the current CPU might have been totally uninvolved with * (and thus unordered against) that grace period. @@ -1465,8 +1528,9 @@ EXPORT_SYMBOL_GPL(synchronize_srcu_expedited); * * Wait for the count to drain to zero of both indexes. To avoid the * possible starvation of synchronize_srcu(), it waits for the count of - * the index=((->srcu_idx & 1) ^ 1) to drain to zero at first, - * and then flip the srcu_idx and wait for the count of the other index. + * the index=!(ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0]) to drain to zero + * at first, and then flip the ->srcu_ctrp and wait for the count of the + * other index. * * Can block; must be called from process context. * @@ -1579,7 +1643,7 @@ EXPORT_SYMBOL_GPL(start_poll_synchronize_srcu); bool poll_state_synchronize_srcu(struct srcu_struct *ssp, unsigned long cookie) { if (cookie != SRCU_GET_STATE_COMPLETED && - !rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, cookie)) + !rcu_seq_done_exact(&ssp->srcu_sup->srcu_gp_seq, cookie)) return false; // Ensure that the end of the SRCU grace period happens before // any subsequent code that the caller might execute. @@ -1613,7 +1677,7 @@ static void srcu_barrier_cb(struct rcu_head *rhp) */ static void srcu_barrier_one_cpu(struct srcu_struct *ssp, struct srcu_data *sdp) { - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); atomic_inc(&ssp->srcu_sup->srcu_barrier_cpu_cnt); sdp->srcu_barrier_head.func = srcu_barrier_cb; debug_rcu_head_queue(&sdp->srcu_barrier_head); @@ -1622,7 +1686,7 @@ static void srcu_barrier_one_cpu(struct srcu_struct *ssp, struct srcu_data *sdp) debug_rcu_head_unqueue(&sdp->srcu_barrier_head); atomic_dec(&ssp->srcu_sup->srcu_barrier_cpu_cnt); } - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); } /** @@ -1666,6 +1730,64 @@ void srcu_barrier(struct srcu_struct *ssp) } EXPORT_SYMBOL_GPL(srcu_barrier); +/* Callback for srcu_expedite_current() usage. */ +static void srcu_expedite_current_cb(struct rcu_head *rhp) +{ + unsigned long flags; + bool needcb = false; + struct srcu_data *sdp = container_of(rhp, struct srcu_data, srcu_ec_head); + + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); + if (sdp->srcu_ec_state == SRCU_EC_IDLE) { + WARN_ON_ONCE(1); + } else if (sdp->srcu_ec_state == SRCU_EC_PENDING) { + sdp->srcu_ec_state = SRCU_EC_IDLE; + } else { + WARN_ON_ONCE(sdp->srcu_ec_state != SRCU_EC_REPOST); + sdp->srcu_ec_state = SRCU_EC_PENDING; + needcb = true; + } + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); + // If needed, requeue ourselves as an expedited SRCU callback. + if (needcb) + __call_srcu(sdp->ssp, &sdp->srcu_ec_head, srcu_expedite_current_cb, false); +} + +/** + * srcu_expedite_current - Expedite the current SRCU grace period + * @ssp: srcu_struct to expedite. + * + * Cause the current SRCU grace period to become expedited. The grace + * period following the current one might also be expedited. If there is + * no current grace period, one might be created. If the current grace + * period is currently sleeping, that sleep will complete before expediting + * will take effect. + */ +void srcu_expedite_current(struct srcu_struct *ssp) +{ + unsigned long flags; + bool needcb = false; + struct srcu_data *sdp; + + migrate_disable(); + sdp = this_cpu_ptr(ssp->sda); + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); + if (sdp->srcu_ec_state == SRCU_EC_IDLE) { + sdp->srcu_ec_state = SRCU_EC_PENDING; + needcb = true; + } else if (sdp->srcu_ec_state == SRCU_EC_PENDING) { + sdp->srcu_ec_state = SRCU_EC_REPOST; + } else { + WARN_ON_ONCE(sdp->srcu_ec_state != SRCU_EC_REPOST); + } + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); + // If needed, queue an expedited SRCU callback. + if (needcb) + __call_srcu(ssp, &sdp->srcu_ec_head, srcu_expedite_current_cb, false); + migrate_enable(); +} +EXPORT_SYMBOL_GPL(srcu_expedite_current); + /** * srcu_batches_completed - return batches completed. * @ssp: srcu_struct on which to report batch completion. @@ -1675,7 +1797,7 @@ EXPORT_SYMBOL_GPL(srcu_barrier); */ unsigned long srcu_batches_completed(struct srcu_struct *ssp) { - return READ_ONCE(ssp->srcu_idx); + return READ_ONCE(ssp->srcu_sup->srcu_gp_seq); } EXPORT_SYMBOL_GPL(srcu_batches_completed); @@ -1692,7 +1814,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) /* * Because readers might be delayed for an extended period after - * fetching ->srcu_idx for their index, at any point in time there + * fetching ->srcu_ctrp for their index, at any point in time there * might well be readers using both idx=0 and idx=1. We therefore * need to wait for readers to clear from both index values before * invoking a callback. @@ -1702,17 +1824,17 @@ static void srcu_advance_state(struct srcu_struct *ssp) */ idx = rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq)); /* ^^^ */ if (idx == SRCU_STATE_IDLE) { - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq)); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; } idx = rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)); if (idx == SRCU_STATE_IDLE) srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (idx != SRCU_STATE_IDLE) { mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* Someone else started the grace period. */ @@ -1720,16 +1842,16 @@ static void srcu_advance_state(struct srcu_struct *ssp) } if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) == SRCU_STATE_SCAN1) { - idx = 1 ^ (ssp->srcu_idx & 1); + idx = !(ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0]); if (!try_check_zero(ssp, idx, 1)) { mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* readers present, retry later. */ } srcu_flip(ssp); - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); rcu_seq_set_state(&ssp->srcu_sup->srcu_gp_seq, SRCU_STATE_SCAN2); ssp->srcu_sup->srcu_n_exp_nodelay = 0; - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); } if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) == SRCU_STATE_SCAN2) { @@ -1738,7 +1860,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) * SRCU read-side critical sections are normally short, * so check at least twice in quick succession after a flip. */ - idx = 1 ^ (ssp->srcu_idx & 1); + idx = !(ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0]); if (!try_check_zero(ssp, idx, 2)) { mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* readers present, retry later. */ @@ -1767,7 +1889,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) ssp = sdp->ssp; rcu_cblist_init(&ready_cbs); - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); WARN_ON_ONCE(!rcu_segcblist_segempty(&sdp->srcu_cblist, RCU_NEXT_TAIL)); rcu_segcblist_advance(&sdp->srcu_cblist, rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)); @@ -1778,7 +1900,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) */ if (sdp->srcu_cblist_invoking || !rcu_segcblist_ready_cbs(&sdp->srcu_cblist)) { - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); return; /* Someone else on the job or nothing to do. */ } @@ -1786,7 +1908,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) sdp->srcu_cblist_invoking = true; rcu_segcblist_extract_done_cbs(&sdp->srcu_cblist, &ready_cbs); len = ready_cbs.len; - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); rhp = rcu_cblist_dequeue(&ready_cbs); for (; rhp != NULL; rhp = rcu_cblist_dequeue(&ready_cbs)) { debug_rcu_head_unqueue(rhp); @@ -1801,11 +1923,11 @@ static void srcu_invoke_callbacks(struct work_struct *work) * Update counts, accelerate new callbacks, and if needed, * schedule another round of callback invocation. */ - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); rcu_segcblist_add_len(&sdp->srcu_cblist, -len); sdp->srcu_cblist_invoking = false; more = rcu_segcblist_ready_cbs(&sdp->srcu_cblist); - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); /* An SRCU barrier or callbacks from previous nesting work pending */ if (more) srcu_schedule_cbs_sdp(sdp, 0); @@ -1819,7 +1941,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) { bool pushgp = true; - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { if (!WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq))) { /* All requests fulfilled, time to go idle. */ @@ -1829,7 +1951,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) /* Outstanding request and no GP. Start one. */ srcu_gp_start(ssp); } - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (pushgp) queue_delayed_work(rcu_gp_wq, &ssp->srcu_sup->work, delay); @@ -1849,7 +1971,9 @@ static void process_srcu(struct work_struct *work) ssp = sup->srcu_ssp; srcu_advance_state(ssp); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); curdelay = srcu_get_delay(ssp); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (curdelay) { WRITE_ONCE(sup->reschedule_count, 0); } else { @@ -1867,6 +1991,23 @@ static void process_srcu(struct work_struct *work) srcu_reschedule(ssp, curdelay); } +static void srcu_irq_work(struct irq_work *work) +{ + struct srcu_struct *ssp; + struct srcu_usage *sup; + unsigned long delay; + unsigned long flags; + + sup = container_of(work, struct srcu_usage, irq_work); + ssp = sup->srcu_ssp; + + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + delay = srcu_get_delay(ssp); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + + queue_delayed_work(rcu_gp_wq, &sup->work, !!delay); +} + void srcutorture_get_gp_data(struct srcu_struct *ssp, int *flags, unsigned long *gp_seq) { @@ -1896,7 +2037,7 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf) int ss_state = READ_ONCE(ssp->srcu_sup->srcu_size_state); int ss_state_idx = ss_state; - idx = ssp->srcu_idx & 0x1; + idx = ssp->srcu_ctrp - &ssp->sda->srcu_ctrs[0]; if (ss_state < 0 || ss_state >= ARRAY_SIZE(srcu_size_state_name)) ss_state_idx = ARRAY_SIZE(srcu_size_state_name) - 1; pr_alert("%s%s Tree SRCU g%ld state %d (%s)", @@ -1914,8 +2055,8 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf) struct srcu_data *sdp; sdp = per_cpu_ptr(ssp->sda, cpu); - u0 = data_race(atomic_long_read(&sdp->srcu_unlock_count[!idx])); - u1 = data_race(atomic_long_read(&sdp->srcu_unlock_count[idx])); + u0 = data_race(atomic_long_read(&sdp->srcu_ctrs[!idx].srcu_unlocks)); + u1 = data_race(atomic_long_read(&sdp->srcu_ctrs[idx].srcu_unlocks)); /* * Make sure that a lock is always counted if the corresponding @@ -1923,8 +2064,8 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf) */ smp_rmb(); - l0 = data_race(atomic_long_read(&sdp->srcu_lock_count[!idx])); - l1 = data_race(atomic_long_read(&sdp->srcu_lock_count[idx])); + l0 = data_race(atomic_long_read(&sdp->srcu_ctrs[!idx].srcu_locks)); + l1 = data_race(atomic_long_read(&sdp->srcu_ctrs[idx].srcu_locks)); c0 = l0 - u0; c1 = l1 - u1; @@ -2001,6 +2142,7 @@ static int srcu_module_coming(struct module *mod) ssp->sda = alloc_percpu(struct srcu_data); if (WARN_ON_ONCE(!ssp->sda)) return -ENOMEM; + ssp->srcu_ctrp = &ssp->sda->srcu_ctrs[0]; } return 0; } diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h index 59314da5eb60..48f0d803c8e2 100644 --- a/kernel/rcu/tasks.h +++ b/kernel/rcu/tasks.h @@ -161,11 +161,6 @@ static void tasks_rcu_exit_srcu_stall(struct timer_list *unused); static DEFINE_TIMER(tasks_rcu_exit_srcu_stall_timer, tasks_rcu_exit_srcu_stall); #endif -/* Avoid IPIing CPUs early in the grace period. */ -#define RCU_TASK_IPI_DELAY (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB) ? HZ / 2 : 0) -static int rcu_task_ipi_delay __read_mostly = RCU_TASK_IPI_DELAY; -module_param(rcu_task_ipi_delay, int, 0644); - /* Control stall timeouts. Disable with <= 0, otherwise jiffies till stall. */ #define RCU_TASK_BOOT_STALL_TIMEOUT (HZ * 30) #define RCU_TASK_STALL_TIMEOUT (HZ * 60 * 10) @@ -264,7 +259,8 @@ static void cblist_init_generic(struct rcu_tasks *rtp) } lim = rcu_task_enqueue_lim; - rtp->rtpcp_array = kcalloc(num_possible_cpus(), sizeof(struct rcu_tasks_percpu *), GFP_KERNEL); + rtp->rtpcp_array = kzalloc_objs(struct rcu_tasks_percpu *, + num_possible_cpus()); BUG_ON(!rtp->rtpcp_array); for_each_possible_cpu(cpu) { @@ -295,9 +291,9 @@ static void cblist_init_generic(struct rcu_tasks *rtp) shift = ilog2(rcu_task_cpu_ids / lim); if (((rcu_task_cpu_ids - 1) >> shift) >= lim) shift++; - WRITE_ONCE(rtp->percpu_enqueue_shift, shift); - WRITE_ONCE(rtp->percpu_dequeue_lim, lim); - smp_store_release(&rtp->percpu_enqueue_lim, lim); + rtp->percpu_enqueue_shift = shift; + rtp->percpu_dequeue_lim = lim; + rtp->percpu_enqueue_lim = lim; pr_info("%s: Setting shift to %d and lim to %d rcu_task_cb_adjust=%d rcu_task_cpu_ids=%d.\n", rtp->name, data_race(rtp->percpu_enqueue_shift), data_race(rtp->percpu_enqueue_lim), @@ -316,7 +312,8 @@ static void call_rcu_tasks_generic_timer(struct timer_list *tlp) unsigned long flags; bool needwake = false; struct rcu_tasks *rtp; - struct rcu_tasks_percpu *rtpcp = from_timer(rtpcp, tlp, lazy_timer); + struct rcu_tasks_percpu *rtpcp = timer_container_of(rtpcp, tlp, + lazy_timer); rtp = rtpcp->rtpp; raw_spin_lock_irqsave_rcu_node(rtpcp, flags); @@ -552,13 +549,13 @@ static void rcu_tasks_invoke_cbs(struct rcu_tasks *rtp, struct rcu_tasks_percpu rtpcp_next = rtp->rtpcp_array[index]; if (rtpcp_next->cpu < smp_load_acquire(&rtp->percpu_dequeue_lim)) { cpuwq = rcu_cpu_beenfullyonline(rtpcp_next->cpu) ? rtpcp_next->cpu : WORK_CPU_UNBOUND; - queue_work_on(cpuwq, system_wq, &rtpcp_next->rtp_work); + queue_work_on(cpuwq, system_percpu_wq, &rtpcp_next->rtp_work); index++; if (index < num_possible_cpus()) { rtpcp_next = rtp->rtpcp_array[index]; if (rtpcp_next->cpu < smp_load_acquire(&rtp->percpu_dequeue_lim)) { cpuwq = rcu_cpu_beenfullyonline(rtpcp_next->cpu) ? rtpcp_next->cpu : WORK_CPU_UNBOUND; - queue_work_on(cpuwq, system_wq, &rtpcp_next->rtp_work); + queue_work_on(cpuwq, system_percpu_wq, &rtpcp_next->rtp_work); } } } @@ -717,7 +714,6 @@ static void __init rcu_tasks_bootup_oddness(void) #endif /* #ifdef CONFIG_TASKS_TRACE_RCU */ } - /* Dump out rcutorture-relevant state common to all RCU-tasks flavors. */ static void show_rcu_tasks_generic_gp_kthread(struct rcu_tasks *rtp, char *s) { @@ -800,9 +796,7 @@ static void rcu_tasks_torture_stats_print_generic(struct rcu_tasks *rtp, char *t #endif // #ifndef CONFIG_TINY_RCU -static void exit_tasks_rcu_finish_trace(struct task_struct *t); - -#if defined(CONFIG_TASKS_RCU) || defined(CONFIG_TASKS_TRACE_RCU) +#if defined(CONFIG_TASKS_RCU) //////////////////////////////////////////////////////////////////////// // @@ -897,7 +891,7 @@ static void rcu_tasks_wait_gp(struct rcu_tasks *rtp) rtp->postgp_func(rtp); } -#endif /* #if defined(CONFIG_TASKS_RCU) || defined(CONFIG_TASKS_TRACE_RCU) */ +#endif /* #if defined(CONFIG_TASKS_RCU) */ #ifdef CONFIG_TASKS_RCU @@ -1086,7 +1080,7 @@ static void rcu_tasks_postscan(struct list_head *hop) } if (!IS_ENABLED(CONFIG_TINY_RCU)) - del_timer_sync(&tasks_rcu_exit_srcu_stall_timer); + timer_delete_sync(&tasks_rcu_exit_srcu_stall_timer); } /* See if tasks are still holding out, complain if so. */ @@ -1321,13 +1315,11 @@ void exit_tasks_rcu_finish(void) raw_spin_lock_irqsave_rcu_node(rtpcp, flags); list_del_init(&t->rcu_tasks_exit_list); raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); - - exit_tasks_rcu_finish_trace(t); } #else /* #ifdef CONFIG_TASKS_RCU */ void exit_tasks_rcu_start(void) { } -void exit_tasks_rcu_finish(void) { exit_tasks_rcu_finish_trace(current); } +void exit_tasks_rcu_finish(void) { } #endif /* #else #ifdef CONFIG_TASKS_RCU */ #ifdef CONFIG_TASKS_RUDE_RCU @@ -1448,682 +1440,11 @@ EXPORT_SYMBOL_GPL(rcu_tasks_rude_get_gp_data); #endif /* #ifdef CONFIG_TASKS_RUDE_RCU */ -//////////////////////////////////////////////////////////////////////// -// -// Tracing variant of Tasks RCU. This variant is designed to be used -// to protect tracing hooks, including those of BPF. This variant -// therefore: -// -// 1. Has explicit read-side markers to allow finite grace periods -// in the face of in-kernel loops for PREEMPT=n builds. -// -// 2. Protects code in the idle loop, exception entry/exit, and -// CPU-hotplug code paths, similar to the capabilities of SRCU. -// -// 3. Avoids expensive read-side instructions, having overhead similar -// to that of Preemptible RCU. -// -// There are of course downsides. For example, the grace-period code -// can send IPIs to CPUs, even when those CPUs are in the idle loop or -// in nohz_full userspace. If needed, these downsides can be at least -// partially remedied. -// -// Perhaps most important, this variant of RCU does not affect the vanilla -// flavors, rcu_preempt and rcu_sched. The fact that RCU Tasks Trace -// readers can operate from idle, offline, and exception entry/exit in no -// way allows rcu_preempt and rcu_sched readers to also do so. -// -// The implementation uses rcu_tasks_wait_gp(), which relies on function -// pointers in the rcu_tasks structure. The rcu_spawn_tasks_trace_kthread() -// function sets these function pointers up so that rcu_tasks_wait_gp() -// invokes these functions in this order: -// -// rcu_tasks_trace_pregp_step(): -// Disables CPU hotplug, adds all currently executing tasks to the -// holdout list, then checks the state of all tasks that blocked -// or were preempted within their current RCU Tasks Trace read-side -// critical section, adding them to the holdout list if appropriate. -// Finally, this function re-enables CPU hotplug. -// The ->pertask_func() pointer is NULL, so there is no per-task processing. -// rcu_tasks_trace_postscan(): -// Invokes synchronize_rcu() to wait for late-stage exiting tasks -// to finish exiting. -// check_all_holdout_tasks_trace(), repeatedly until holdout list is empty: -// Scans the holdout list, attempting to identify a quiescent state -// for each task on the list. If there is a quiescent state, the -// corresponding task is removed from the holdout list. Once this -// list is empty, the grace period has completed. -// rcu_tasks_trace_postgp(): -// Provides the needed full memory barrier and does debug checks. -// -// The exit_tasks_rcu_finish_trace() synchronizes with exiting tasks. -// -// Pre-grace-period update-side code is ordered before the grace period -// via the ->cbs_lock and barriers in rcu_tasks_kthread(). Pre-grace-period -// read-side code is ordered before the grace period by atomic operations -// on .b.need_qs flag of each task involved in this process, or by scheduler -// context-switch ordering (for locked-down non-running readers). - -// The lockdep state must be outside of #ifdef to be useful. -#ifdef CONFIG_DEBUG_LOCK_ALLOC -static struct lock_class_key rcu_lock_trace_key; -struct lockdep_map rcu_trace_lock_map = - STATIC_LOCKDEP_MAP_INIT("rcu_read_lock_trace", &rcu_lock_trace_key); -EXPORT_SYMBOL_GPL(rcu_trace_lock_map); -#endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ - -#ifdef CONFIG_TASKS_TRACE_RCU - -// Record outstanding IPIs to each CPU. No point in sending two... -static DEFINE_PER_CPU(bool, trc_ipi_to_cpu); - -// The number of detections of task quiescent state relying on -// heavyweight readers executing explicit memory barriers. -static unsigned long n_heavy_reader_attempts; -static unsigned long n_heavy_reader_updates; -static unsigned long n_heavy_reader_ofl_updates; -static unsigned long n_trc_holdouts; - -void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func); -DEFINE_RCU_TASKS(rcu_tasks_trace, rcu_tasks_wait_gp, call_rcu_tasks_trace, - "RCU Tasks Trace"); - -/* Load from ->trc_reader_special.b.need_qs with proper ordering. */ -static u8 rcu_ld_need_qs(struct task_struct *t) -{ - smp_mb(); // Enforce full grace-period ordering. - return smp_load_acquire(&t->trc_reader_special.b.need_qs); -} - -/* Store to ->trc_reader_special.b.need_qs with proper ordering. */ -static void rcu_st_need_qs(struct task_struct *t, u8 v) -{ - smp_store_release(&t->trc_reader_special.b.need_qs, v); - smp_mb(); // Enforce full grace-period ordering. -} - -/* - * Do a cmpxchg() on ->trc_reader_special.b.need_qs, allowing for - * the four-byte operand-size restriction of some platforms. - * - * Returns the old value, which is often ignored. - */ -u8 rcu_trc_cmpxchg_need_qs(struct task_struct *t, u8 old, u8 new) -{ - return cmpxchg(&t->trc_reader_special.b.need_qs, old, new); -} -EXPORT_SYMBOL_GPL(rcu_trc_cmpxchg_need_qs); - -/* - * If we are the last reader, signal the grace-period kthread. - * Also remove from the per-CPU list of blocked tasks. - */ -void rcu_read_unlock_trace_special(struct task_struct *t) -{ - unsigned long flags; - struct rcu_tasks_percpu *rtpcp; - union rcu_special trs; - - // Open-coded full-word version of rcu_ld_need_qs(). - smp_mb(); // Enforce full grace-period ordering. - trs = smp_load_acquire(&t->trc_reader_special); - - if (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB) && t->trc_reader_special.b.need_mb) - smp_mb(); // Pairs with update-side barriers. - // Update .need_qs before ->trc_reader_nesting for irq/NMI handlers. - if (trs.b.need_qs == (TRC_NEED_QS_CHECKED | TRC_NEED_QS)) { - u8 result = rcu_trc_cmpxchg_need_qs(t, TRC_NEED_QS_CHECKED | TRC_NEED_QS, - TRC_NEED_QS_CHECKED); - - WARN_ONCE(result != trs.b.need_qs, "%s: result = %d", __func__, result); - } - if (trs.b.blocked) { - rtpcp = per_cpu_ptr(rcu_tasks_trace.rtpcpu, t->trc_blkd_cpu); - raw_spin_lock_irqsave_rcu_node(rtpcp, flags); - list_del_init(&t->trc_blkd_node); - WRITE_ONCE(t->trc_reader_special.b.blocked, false); - raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); - } - WRITE_ONCE(t->trc_reader_nesting, 0); -} -EXPORT_SYMBOL_GPL(rcu_read_unlock_trace_special); - -/* Add a newly blocked reader task to its CPU's list. */ -void rcu_tasks_trace_qs_blkd(struct task_struct *t) -{ - unsigned long flags; - struct rcu_tasks_percpu *rtpcp; - - local_irq_save(flags); - rtpcp = this_cpu_ptr(rcu_tasks_trace.rtpcpu); - raw_spin_lock_rcu_node(rtpcp); // irqs already disabled - t->trc_blkd_cpu = smp_processor_id(); - if (!rtpcp->rtp_blkd_tasks.next) - INIT_LIST_HEAD(&rtpcp->rtp_blkd_tasks); - list_add(&t->trc_blkd_node, &rtpcp->rtp_blkd_tasks); - WRITE_ONCE(t->trc_reader_special.b.blocked, true); - raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); -} -EXPORT_SYMBOL_GPL(rcu_tasks_trace_qs_blkd); - -/* Add a task to the holdout list, if it is not already on the list. */ -static void trc_add_holdout(struct task_struct *t, struct list_head *bhp) -{ - if (list_empty(&t->trc_holdout_list)) { - get_task_struct(t); - list_add(&t->trc_holdout_list, bhp); - n_trc_holdouts++; - } -} - -/* Remove a task from the holdout list, if it is in fact present. */ -static void trc_del_holdout(struct task_struct *t) -{ - if (!list_empty(&t->trc_holdout_list)) { - list_del_init(&t->trc_holdout_list); - put_task_struct(t); - n_trc_holdouts--; - } -} - -/* IPI handler to check task state. */ -static void trc_read_check_handler(void *t_in) -{ - int nesting; - struct task_struct *t = current; - struct task_struct *texp = t_in; - - // If the task is no longer running on this CPU, leave. - if (unlikely(texp != t)) - goto reset_ipi; // Already on holdout list, so will check later. - - // If the task is not in a read-side critical section, and - // if this is the last reader, awaken the grace-period kthread. - nesting = READ_ONCE(t->trc_reader_nesting); - if (likely(!nesting)) { - rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS_CHECKED); - goto reset_ipi; - } - // If we are racing with an rcu_read_unlock_trace(), try again later. - if (unlikely(nesting < 0)) - goto reset_ipi; - - // Get here if the task is in a read-side critical section. - // Set its state so that it will update state for the grace-period - // kthread upon exit from that critical section. - rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS | TRC_NEED_QS_CHECKED); - -reset_ipi: - // Allow future IPIs to be sent on CPU and for task. - // Also order this IPI handler against any later manipulations of - // the intended task. - smp_store_release(per_cpu_ptr(&trc_ipi_to_cpu, smp_processor_id()), false); // ^^^ - smp_store_release(&texp->trc_ipi_to_cpu, -1); // ^^^ -} - -/* Callback function for scheduler to check locked-down task. */ -static int trc_inspect_reader(struct task_struct *t, void *bhp_in) -{ - struct list_head *bhp = bhp_in; - int cpu = task_cpu(t); - int nesting; - bool ofl = cpu_is_offline(cpu); - - if (task_curr(t) && !ofl) { - // If no chance of heavyweight readers, do it the hard way. - if (!IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB)) - return -EINVAL; - - // If heavyweight readers are enabled on the remote task, - // we can inspect its state despite its currently running. - // However, we cannot safely change its state. - n_heavy_reader_attempts++; - // Check for "running" idle tasks on offline CPUs. - if (!rcu_watching_zero_in_eqs(cpu, &t->trc_reader_nesting)) - return -EINVAL; // No quiescent state, do it the hard way. - n_heavy_reader_updates++; - nesting = 0; - } else { - // The task is not running, so C-language access is safe. - nesting = t->trc_reader_nesting; - WARN_ON_ONCE(ofl && task_curr(t) && (t != idle_task(task_cpu(t)))); - if (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB) && ofl) - n_heavy_reader_ofl_updates++; - } - - // If not exiting a read-side critical section, mark as checked - // so that the grace-period kthread will remove it from the - // holdout list. - if (!nesting) { - rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS_CHECKED); - return 0; // In QS, so done. - } - if (nesting < 0) - return -EINVAL; // Reader transitioning, try again later. - - // The task is in a read-side critical section, so set up its - // state so that it will update state upon exit from that critical - // section. - if (!rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS | TRC_NEED_QS_CHECKED)) - trc_add_holdout(t, bhp); - return 0; -} - -/* Attempt to extract the state for the specified task. */ -static void trc_wait_for_one_reader(struct task_struct *t, - struct list_head *bhp) -{ - int cpu; - - // If a previous IPI is still in flight, let it complete. - if (smp_load_acquire(&t->trc_ipi_to_cpu) != -1) // Order IPI - return; - - // The current task had better be in a quiescent state. - if (t == current) { - rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS_CHECKED); - WARN_ON_ONCE(READ_ONCE(t->trc_reader_nesting)); - return; - } - - // Attempt to nail down the task for inspection. - get_task_struct(t); - if (!task_call_func(t, trc_inspect_reader, bhp)) { - put_task_struct(t); - return; - } - put_task_struct(t); - - // If this task is not yet on the holdout list, then we are in - // an RCU read-side critical section. Otherwise, the invocation of - // trc_add_holdout() that added it to the list did the necessary - // get_task_struct(). Either way, the task cannot be freed out - // from under this code. - - // If currently running, send an IPI, either way, add to list. - trc_add_holdout(t, bhp); - if (task_curr(t) && - time_after(jiffies + 1, rcu_tasks_trace.gp_start + rcu_task_ipi_delay)) { - // The task is currently running, so try IPIing it. - cpu = task_cpu(t); - - // If there is already an IPI outstanding, let it happen. - if (per_cpu(trc_ipi_to_cpu, cpu) || t->trc_ipi_to_cpu >= 0) - return; - - per_cpu(trc_ipi_to_cpu, cpu) = true; - t->trc_ipi_to_cpu = cpu; - rcu_tasks_trace.n_ipis++; - if (smp_call_function_single(cpu, trc_read_check_handler, t, 0)) { - // Just in case there is some other reason for - // failure than the target CPU being offline. - WARN_ONCE(1, "%s(): smp_call_function_single() failed for CPU: %d\n", - __func__, cpu); - rcu_tasks_trace.n_ipis_fails++; - per_cpu(trc_ipi_to_cpu, cpu) = false; - t->trc_ipi_to_cpu = -1; - } - } -} - -/* - * Initialize for first-round processing for the specified task. - * Return false if task is NULL or already taken care of, true otherwise. - */ -static bool rcu_tasks_trace_pertask_prep(struct task_struct *t, bool notself) -{ - // During early boot when there is only the one boot CPU, there - // is no idle task for the other CPUs. Also, the grace-period - // kthread is always in a quiescent state. In addition, just return - // if this task is already on the list. - if (unlikely(t == NULL) || (t == current && notself) || !list_empty(&t->trc_holdout_list)) - return false; - - rcu_st_need_qs(t, 0); - t->trc_ipi_to_cpu = -1; - return true; -} - -/* Do first-round processing for the specified task. */ -static void rcu_tasks_trace_pertask(struct task_struct *t, struct list_head *hop) -{ - if (rcu_tasks_trace_pertask_prep(t, true)) - trc_wait_for_one_reader(t, hop); -} - -/* Initialize for a new RCU-tasks-trace grace period. */ -static void rcu_tasks_trace_pregp_step(struct list_head *hop) -{ - LIST_HEAD(blkd_tasks); - int cpu; - unsigned long flags; - struct rcu_tasks_percpu *rtpcp; - struct task_struct *t; - - // There shouldn't be any old IPIs, but... - for_each_possible_cpu(cpu) - WARN_ON_ONCE(per_cpu(trc_ipi_to_cpu, cpu)); - - // Disable CPU hotplug across the CPU scan for the benefit of - // any IPIs that might be needed. This also waits for all readers - // in CPU-hotplug code paths. - cpus_read_lock(); - - // These rcu_tasks_trace_pertask_prep() calls are serialized to - // allow safe access to the hop list. - for_each_online_cpu(cpu) { - rcu_read_lock(); - // Note that cpu_curr_snapshot() picks up the target - // CPU's current task while its runqueue is locked with - // an smp_mb__after_spinlock(). This ensures that either - // the grace-period kthread will see that task's read-side - // critical section or the task will see the updater's pre-GP - // accesses. The trailing smp_mb() in cpu_curr_snapshot() - // does not currently play a role other than simplify - // that function's ordering semantics. If these simplified - // ordering semantics continue to be redundant, that smp_mb() - // might be removed. - t = cpu_curr_snapshot(cpu); - if (rcu_tasks_trace_pertask_prep(t, true)) - trc_add_holdout(t, hop); - rcu_read_unlock(); - cond_resched_tasks_rcu_qs(); - } - - // Only after all running tasks have been accounted for is it - // safe to take care of the tasks that have blocked within their - // current RCU tasks trace read-side critical section. - for_each_possible_cpu(cpu) { - rtpcp = per_cpu_ptr(rcu_tasks_trace.rtpcpu, cpu); - raw_spin_lock_irqsave_rcu_node(rtpcp, flags); - list_splice_init(&rtpcp->rtp_blkd_tasks, &blkd_tasks); - while (!list_empty(&blkd_tasks)) { - rcu_read_lock(); - t = list_first_entry(&blkd_tasks, struct task_struct, trc_blkd_node); - list_del_init(&t->trc_blkd_node); - list_add(&t->trc_blkd_node, &rtpcp->rtp_blkd_tasks); - raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); - rcu_tasks_trace_pertask(t, hop); - rcu_read_unlock(); - raw_spin_lock_irqsave_rcu_node(rtpcp, flags); - } - raw_spin_unlock_irqrestore_rcu_node(rtpcp, flags); - cond_resched_tasks_rcu_qs(); - } - - // Re-enable CPU hotplug now that the holdout list is populated. - cpus_read_unlock(); -} - -/* - * Do intermediate processing between task and holdout scans. - */ -static void rcu_tasks_trace_postscan(struct list_head *hop) -{ - // Wait for late-stage exiting tasks to finish exiting. - // These might have passed the call to exit_tasks_rcu_finish(). - - // If you remove the following line, update rcu_trace_implies_rcu_gp()!!! - synchronize_rcu(); - // Any tasks that exit after this point will set - // TRC_NEED_QS_CHECKED in ->trc_reader_special.b.need_qs. -} - -/* Communicate task state back to the RCU tasks trace stall warning request. */ -struct trc_stall_chk_rdr { - int nesting; - int ipi_to_cpu; - u8 needqs; -}; - -static int trc_check_slow_task(struct task_struct *t, void *arg) -{ - struct trc_stall_chk_rdr *trc_rdrp = arg; - - if (task_curr(t) && cpu_online(task_cpu(t))) - return false; // It is running, so decline to inspect it. - trc_rdrp->nesting = READ_ONCE(t->trc_reader_nesting); - trc_rdrp->ipi_to_cpu = READ_ONCE(t->trc_ipi_to_cpu); - trc_rdrp->needqs = rcu_ld_need_qs(t); - return true; -} - -/* Show the state of a task stalling the current RCU tasks trace GP. */ -static void show_stalled_task_trace(struct task_struct *t, bool *firstreport) -{ - int cpu; - struct trc_stall_chk_rdr trc_rdr; - bool is_idle_tsk = is_idle_task(t); - - if (*firstreport) { - pr_err("INFO: rcu_tasks_trace detected stalls on tasks:\n"); - *firstreport = false; - } - cpu = task_cpu(t); - if (!task_call_func(t, trc_check_slow_task, &trc_rdr)) - pr_alert("P%d: %c%c\n", - t->pid, - ".I"[t->trc_ipi_to_cpu >= 0], - ".i"[is_idle_tsk]); - else - pr_alert("P%d: %c%c%c%c nesting: %d%c%c cpu: %d%s\n", - t->pid, - ".I"[trc_rdr.ipi_to_cpu >= 0], - ".i"[is_idle_tsk], - ".N"[cpu >= 0 && tick_nohz_full_cpu(cpu)], - ".B"[!!data_race(t->trc_reader_special.b.blocked)], - trc_rdr.nesting, - " !CN"[trc_rdr.needqs & 0x3], - " ?"[trc_rdr.needqs > 0x3], - cpu, cpu_online(cpu) ? "" : "(offline)"); - sched_show_task(t); -} - -/* List stalled IPIs for RCU tasks trace. */ -static void show_stalled_ipi_trace(void) -{ - int cpu; - - for_each_possible_cpu(cpu) - if (per_cpu(trc_ipi_to_cpu, cpu)) - pr_alert("\tIPI outstanding to CPU %d\n", cpu); -} - -/* Do one scan of the holdout list. */ -static void check_all_holdout_tasks_trace(struct list_head *hop, - bool needreport, bool *firstreport) -{ - struct task_struct *g, *t; - - // Disable CPU hotplug across the holdout list scan for IPIs. - cpus_read_lock(); - - list_for_each_entry_safe(t, g, hop, trc_holdout_list) { - // If safe and needed, try to check the current task. - if (READ_ONCE(t->trc_ipi_to_cpu) == -1 && - !(rcu_ld_need_qs(t) & TRC_NEED_QS_CHECKED)) - trc_wait_for_one_reader(t, hop); - - // If check succeeded, remove this task from the list. - if (smp_load_acquire(&t->trc_ipi_to_cpu) == -1 && - rcu_ld_need_qs(t) == TRC_NEED_QS_CHECKED) - trc_del_holdout(t); - else if (needreport) - show_stalled_task_trace(t, firstreport); - cond_resched_tasks_rcu_qs(); - } - - // Re-enable CPU hotplug now that the holdout list scan has completed. - cpus_read_unlock(); - - if (needreport) { - if (*firstreport) - pr_err("INFO: rcu_tasks_trace detected stalls? (Late IPI?)\n"); - show_stalled_ipi_trace(); - } -} - -static void rcu_tasks_trace_empty_fn(void *unused) -{ -} - -/* Wait for grace period to complete and provide ordering. */ -static void rcu_tasks_trace_postgp(struct rcu_tasks *rtp) -{ - int cpu; - - // Wait for any lingering IPI handlers to complete. Note that - // if a CPU has gone offline or transitioned to userspace in the - // meantime, all IPI handlers should have been drained beforehand. - // Yes, this assumes that CPUs process IPIs in order. If that ever - // changes, there will need to be a recheck and/or timed wait. - for_each_online_cpu(cpu) - if (WARN_ON_ONCE(smp_load_acquire(per_cpu_ptr(&trc_ipi_to_cpu, cpu)))) - smp_call_function_single(cpu, rcu_tasks_trace_empty_fn, NULL, 1); - - smp_mb(); // Caller's code must be ordered after wakeup. - // Pairs with pretty much every ordering primitive. -} - -/* Report any needed quiescent state for this exiting task. */ -static void exit_tasks_rcu_finish_trace(struct task_struct *t) -{ - union rcu_special trs = READ_ONCE(t->trc_reader_special); - - rcu_trc_cmpxchg_need_qs(t, 0, TRC_NEED_QS_CHECKED); - WARN_ON_ONCE(READ_ONCE(t->trc_reader_nesting)); - if (WARN_ON_ONCE(rcu_ld_need_qs(t) & TRC_NEED_QS || trs.b.blocked)) - rcu_read_unlock_trace_special(t); - else - WRITE_ONCE(t->trc_reader_nesting, 0); -} - -/** - * call_rcu_tasks_trace() - Queue a callback trace task-based grace period - * @rhp: 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 trace rcu-tasks - * grace period elapses, in other words after all currently executing - * trace rcu-tasks read-side critical sections have completed. These - * read-side critical sections are delimited by calls to rcu_read_lock_trace() - * and rcu_read_unlock_trace(). - * - * See the description of call_rcu() for more detailed information on - * memory ordering guarantees. - */ -void call_rcu_tasks_trace(struct rcu_head *rhp, rcu_callback_t func) -{ - call_rcu_tasks_generic(rhp, func, &rcu_tasks_trace); -} -EXPORT_SYMBOL_GPL(call_rcu_tasks_trace); - -/** - * synchronize_rcu_tasks_trace - wait for a trace rcu-tasks grace period - * - * Control will return to the caller some time after a trace rcu-tasks - * grace period has elapsed, in other words after all currently executing - * trace rcu-tasks read-side critical sections have elapsed. These read-side - * critical sections are delimited by calls to rcu_read_lock_trace() - * and rcu_read_unlock_trace(). - * - * 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_trace() function is not - * (yet) intended for heavy use from multiple CPUs. - * - * See the description of synchronize_rcu() for more detailed information - * on memory ordering guarantees. - */ -void synchronize_rcu_tasks_trace(void) -{ - RCU_LOCKDEP_WARN(lock_is_held(&rcu_trace_lock_map), "Illegal synchronize_rcu_tasks_trace() in RCU Tasks Trace read-side critical section"); - synchronize_rcu_tasks_generic(&rcu_tasks_trace); -} -EXPORT_SYMBOL_GPL(synchronize_rcu_tasks_trace); - -/** - * rcu_barrier_tasks_trace - Wait for in-flight call_rcu_tasks_trace() 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_trace(void) -{ - rcu_barrier_tasks_generic(&rcu_tasks_trace); -} -EXPORT_SYMBOL_GPL(rcu_barrier_tasks_trace); - -int rcu_tasks_trace_lazy_ms = -1; -module_param(rcu_tasks_trace_lazy_ms, int, 0444); - -static int __init rcu_spawn_tasks_trace_kthread(void) -{ - if (IS_ENABLED(CONFIG_TASKS_TRACE_RCU_READ_MB)) { - rcu_tasks_trace.gp_sleep = HZ / 10; - rcu_tasks_trace.init_fract = HZ / 10; - } else { - rcu_tasks_trace.gp_sleep = HZ / 200; - if (rcu_tasks_trace.gp_sleep <= 0) - rcu_tasks_trace.gp_sleep = 1; - rcu_tasks_trace.init_fract = HZ / 200; - if (rcu_tasks_trace.init_fract <= 0) - rcu_tasks_trace.init_fract = 1; - } - if (rcu_tasks_trace_lazy_ms >= 0) - rcu_tasks_trace.lazy_jiffies = msecs_to_jiffies(rcu_tasks_trace_lazy_ms); - rcu_tasks_trace.pregp_func = rcu_tasks_trace_pregp_step; - rcu_tasks_trace.postscan_func = rcu_tasks_trace_postscan; - rcu_tasks_trace.holdouts_func = check_all_holdout_tasks_trace; - rcu_tasks_trace.postgp_func = rcu_tasks_trace_postgp; - rcu_spawn_tasks_kthread_generic(&rcu_tasks_trace); - return 0; -} - -#if !defined(CONFIG_TINY_RCU) -void show_rcu_tasks_trace_gp_kthread(void) -{ - char buf[64]; - - snprintf(buf, sizeof(buf), "N%lu h:%lu/%lu/%lu", - data_race(n_trc_holdouts), - data_race(n_heavy_reader_ofl_updates), - data_race(n_heavy_reader_updates), - data_race(n_heavy_reader_attempts)); - show_rcu_tasks_generic_gp_kthread(&rcu_tasks_trace, buf); -} -EXPORT_SYMBOL_GPL(show_rcu_tasks_trace_gp_kthread); - -void rcu_tasks_trace_torture_stats_print(char *tt, char *tf) -{ - rcu_tasks_torture_stats_print_generic(&rcu_tasks_trace, tt, tf, ""); -} -EXPORT_SYMBOL_GPL(rcu_tasks_trace_torture_stats_print); -#endif // !defined(CONFIG_TINY_RCU) - -struct task_struct *get_rcu_tasks_trace_gp_kthread(void) -{ - return rcu_tasks_trace.kthread_ptr; -} -EXPORT_SYMBOL_GPL(get_rcu_tasks_trace_gp_kthread); - -void rcu_tasks_trace_get_gp_data(int *flags, unsigned long *gp_seq) -{ - *flags = 0; - *gp_seq = rcu_seq_current(&rcu_tasks_trace.tasks_gp_seq); -} -EXPORT_SYMBOL_GPL(rcu_tasks_trace_get_gp_data); - -#else /* #ifdef CONFIG_TASKS_TRACE_RCU */ -static void exit_tasks_rcu_finish_trace(struct task_struct *t) { } -#endif /* #else #ifdef CONFIG_TASKS_TRACE_RCU */ - #ifndef CONFIG_TINY_RCU void show_rcu_tasks_gp_kthreads(void) { show_rcu_tasks_classic_gp_kthread(); show_rcu_tasks_rude_gp_kthread(); - show_rcu_tasks_trace_gp_kthread(); } #endif /* #ifndef CONFIG_TINY_RCU */ @@ -2250,13 +1571,9 @@ void __init tasks_cblist_init_generic(void) #ifdef CONFIG_TASKS_RUDE_RCU cblist_init_generic(&rcu_tasks_rude); #endif - -#ifdef CONFIG_TASKS_TRACE_RCU - cblist_init_generic(&rcu_tasks_trace); -#endif } -void __init rcu_init_tasks_generic(void) +static int __init rcu_init_tasks_generic(void) { #ifdef CONFIG_TASKS_RCU rcu_spawn_tasks_kthread(); @@ -2266,14 +1583,26 @@ void __init rcu_init_tasks_generic(void) rcu_spawn_tasks_rude_kthread(); #endif -#ifdef CONFIG_TASKS_TRACE_RCU - rcu_spawn_tasks_trace_kthread(); -#endif - // Run the self-tests. rcu_tasks_initiate_self_tests(); + + return 0; } +core_initcall(rcu_init_tasks_generic); #else /* #ifdef CONFIG_TASKS_RCU_GENERIC */ static inline void rcu_tasks_bootup_oddness(void) {} #endif /* #else #ifdef CONFIG_TASKS_RCU_GENERIC */ + +#ifdef CONFIG_TASKS_TRACE_RCU + +//////////////////////////////////////////////////////////////////////// +// +// Tracing variant of Tasks RCU. This variant is designed to be used +// to protect tracing hooks, including those of BPF. This variant +// is implemented via a straightforward mapping onto SRCU-fast. + +DEFINE_SRCU_FAST(rcu_tasks_trace_srcu_struct); +EXPORT_SYMBOL_GPL(rcu_tasks_trace_srcu_struct); + +#endif /* #else #ifdef CONFIG_TASKS_TRACE_RCU */ diff --git a/kernel/rcu/tiny.c b/kernel/rcu/tiny.c index 4b3f31911465..585cade21010 100644 --- a/kernel/rcu/tiny.c +++ b/kernel/rcu/tiny.c @@ -70,12 +70,10 @@ void rcu_qs(void) */ void rcu_sched_clock_irq(int user) { - if (user) { + if (user) rcu_qs(); - } else if (rcu_ctrlblk.donetail != rcu_ctrlblk.curtail) { - set_tsk_need_resched(current); - set_preempt_need_resched(); - } + else if (rcu_ctrlblk.donetail != rcu_ctrlblk.curtail) + set_need_resched_current(); } /* @@ -85,15 +83,8 @@ void rcu_sched_clock_irq(int user) static inline bool rcu_reclaim_tiny(struct rcu_head *head) { rcu_callback_t f; - unsigned long offset = (unsigned long)head->func; rcu_lock_acquire(&rcu_callback_map); - if (__is_kvfree_rcu_offset(offset)) { - trace_rcu_invoke_kvfree_callback("", head, offset); - kvfree((void *)head - offset); - rcu_lock_release(&rcu_callback_map); - return true; - } trace_rcu_invoke_callback("", head); f = head->func; @@ -159,10 +150,6 @@ void synchronize_rcu(void) } EXPORT_SYMBOL_GPL(synchronize_rcu); -static void tiny_rcu_leak_callback(struct rcu_head *rhp) -{ -} - /* * Post an RCU callback to be invoked after the end of an RCU grace * period. But since we have but one CPU, that would be after any @@ -178,9 +165,6 @@ void call_rcu(struct rcu_head *head, rcu_callback_t func) pr_err("%s(): Double-freed CB %p->%pS()!!! ", __func__, head, head->func); mem_dump_obj(head); } - - if (!__is_kvfree_rcu_offset((unsigned long)head->func)) - WRITE_ONCE(head->func, tiny_rcu_leak_callback); return; } @@ -246,15 +230,18 @@ bool poll_state_synchronize_rcu(unsigned long oldstate) } EXPORT_SYMBOL_GPL(poll_state_synchronize_rcu); -#ifdef CONFIG_KASAN_GENERIC -void kvfree_call_rcu(struct rcu_head *head, void *ptr) +#if IS_ENABLED(CONFIG_RCU_TORTURE_TEST) +unsigned long long rcutorture_gather_gp_seqs(void) { - if (head) - kasan_record_aux_stack(ptr); + return READ_ONCE(rcu_ctrlblk.gp_seq) & 0xffffULL; +} +EXPORT_SYMBOL_GPL(rcutorture_gather_gp_seqs); - __kvfree_call_rcu(head, ptr); +void rcutorture_format_gp_seqs(unsigned long long seqs, char *cp, size_t len) +{ + snprintf(cp, len, "g%04llx", seqs & 0xffffULL); } -EXPORT_SYMBOL_GPL(kvfree_call_rcu); +EXPORT_SYMBOL_GPL(rcutorture_format_gp_seqs); #endif void __init rcu_init(void) diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 475f31deed14..55df6d37145e 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -80,6 +80,15 @@ static void rcu_sr_normal_gp_cleanup_work(struct work_struct *); static DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_data, rcu_data) = { .gpwrap = true, }; + +int rcu_get_gpwrap_count(int cpu) +{ + struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu); + + return READ_ONCE(rdp->gpwrap_count); +} +EXPORT_SYMBOL_GPL(rcu_get_gpwrap_count); + static struct rcu_state rcu_state = { .level = { &rcu_state.node[0] }, .gp_state = RCU_GP_IDLE, @@ -151,7 +160,7 @@ static void rcu_report_qs_rnp(unsigned long mask, struct rcu_node *rnp, unsigned long gps, unsigned long flags); static void invoke_rcu_core(void); static void rcu_report_exp_rdp(struct rcu_data *rdp); -static void sync_sched_exp_online_cleanup(int cpu); +static void rcu_report_qs_rdp(struct rcu_data *rdp); static void check_cb_ovld_locked(struct rcu_data *rdp, struct rcu_node *rnp); static bool rcu_rdp_is_offloaded(struct rcu_data *rdp); static bool rcu_rdp_cpu_online(struct rcu_data *rdp); @@ -368,7 +377,7 @@ EXPORT_SYMBOL_GPL(rcu_momentary_eqs); */ static int rcu_is_cpu_rrupt_from_idle(void) { - long nesting; + long nmi_nesting = ct_nmi_nesting(); /* * Usually called from the tick; but also used from smp_function_call() @@ -380,21 +389,28 @@ static int rcu_is_cpu_rrupt_from_idle(void) /* Check for counter underflows */ RCU_LOCKDEP_WARN(ct_nesting() < 0, "RCU nesting counter underflow!"); - RCU_LOCKDEP_WARN(ct_nmi_nesting() <= 0, - "RCU nmi_nesting counter underflow/zero!"); - /* Are we at first interrupt nesting level? */ - nesting = ct_nmi_nesting(); - if (nesting > 1) + /* Non-idle interrupt or nested idle interrupt */ + if (nmi_nesting > 1) return false; /* - * If we're not in an interrupt, we must be in the idle task! + * Non nested idle interrupt (interrupting section where RCU + * wasn't watching). */ - WARN_ON_ONCE(!nesting && !is_idle_task(current)); + if (nmi_nesting == 1) + return true; + + /* Not in an interrupt */ + if (!nmi_nesting) { + RCU_LOCKDEP_WARN(!in_task() || !is_idle_task(current), + "RCU nmi_nesting counter not in idle task!"); + return !rcu_is_watching_curr_cpu(); + } + + RCU_LOCKDEP_WARN(1, "RCU nmi_nesting counter underflow/zero!"); - /* Does CPU appear to be idle from an RCU standpoint? */ - return ct_nesting() == 0; + return false; } #define DEFAULT_RCU_BLIMIT (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) ? 1000 : 10) @@ -538,7 +554,27 @@ void rcutorture_get_gp_data(int *flags, unsigned long *gp_seq) } EXPORT_SYMBOL_GPL(rcutorture_get_gp_data); -#if defined(CONFIG_NO_HZ_FULL) && (!defined(CONFIG_GENERIC_ENTRY) || !defined(CONFIG_KVM_XFER_TO_GUEST_WORK)) +/* Gather grace-period sequence numbers for rcutorture diagnostics. */ +unsigned long long rcutorture_gather_gp_seqs(void) +{ + return ((READ_ONCE(rcu_state.gp_seq) & 0xffffULL) << 40) | + ((READ_ONCE(rcu_state.expedited_sequence) & 0xffffffULL) << 16) | + (READ_ONCE(rcu_state.gp_seq_polled) & 0xffffULL); +} +EXPORT_SYMBOL_GPL(rcutorture_gather_gp_seqs); + +/* Format grace-period sequence numbers for rcutorture diagnostics. */ +void rcutorture_format_gp_seqs(unsigned long long seqs, char *cp, size_t len) +{ + unsigned int egp = (seqs >> 16) & 0xffffffULL; + unsigned int ggp = (seqs >> 40) & 0xffffULL; + unsigned int pgp = seqs & 0xffffULL; + + snprintf(cp, len, "g%04x:e%06x:p%04x", ggp, egp, pgp); +} +EXPORT_SYMBOL_GPL(rcutorture_format_gp_seqs); + +#if defined(CONFIG_NO_HZ_FULL) && (!defined(CONFIG_GENERIC_ENTRY) || !defined(CONFIG_VIRT_XFER_TO_GUEST_WORK)) /* * An empty function that will trigger a reschedule on * IRQ tail once IRQs get re-enabled on userspace/guest resume. @@ -567,7 +603,7 @@ noinstr void rcu_irq_work_resched(void) if (IS_ENABLED(CONFIG_GENERIC_ENTRY) && !(current->flags & PF_VCPU)) return; - if (IS_ENABLED(CONFIG_KVM_XFER_TO_GUEST_WORK) && (current->flags & PF_VCPU)) + if (IS_ENABLED(CONFIG_VIRT_XFER_TO_GUEST_WORK) && (current->flags & PF_VCPU)) return; instrumentation_begin(); @@ -576,7 +612,7 @@ noinstr void rcu_irq_work_resched(void) } instrumentation_end(); } -#endif /* #if defined(CONFIG_NO_HZ_FULL) && (!defined(CONFIG_GENERIC_ENTRY) || !defined(CONFIG_KVM_XFER_TO_GUEST_WORK)) */ +#endif /* #if defined(CONFIG_NO_HZ_FULL) && (!defined(CONFIG_GENERIC_ENTRY) || !defined(CONFIG_VIRT_XFER_TO_GUEST_WORK)) */ #ifdef CONFIG_PROVE_RCU /** @@ -737,6 +773,25 @@ void rcu_request_urgent_qs_task(struct task_struct *t) smp_store_release(per_cpu_ptr(&rcu_data.rcu_urgent_qs, cpu), true); } +static unsigned long seq_gpwrap_lag = ULONG_MAX / 4; + +/** + * rcu_set_gpwrap_lag - Set RCU GP sequence overflow lag value. + * @lag_gps: Set overflow lag to this many grace period worth of counters + * which is used by rcutorture to quickly force a gpwrap situation. + * @lag_gps = 0 means we reset it back to the boot-time value. + */ +void rcu_set_gpwrap_lag(unsigned long lag_gps) +{ + unsigned long lag_seq_count; + + lag_seq_count = (lag_gps == 0) + ? ULONG_MAX / 4 + : lag_gps << RCU_SEQ_CTR_SHIFT; + WRITE_ONCE(seq_gpwrap_lag, lag_seq_count); +} +EXPORT_SYMBOL_GPL(rcu_set_gpwrap_lag); + /* * When trying to report a quiescent state on behalf of some other CPU, * it is our responsibility to check for and handle potential overflow @@ -747,9 +802,11 @@ void rcu_request_urgent_qs_task(struct task_struct *t) static void rcu_gpnum_ovf(struct rcu_node *rnp, struct rcu_data *rdp) { raw_lockdep_assert_held_rcu_node(rnp); - if (ULONG_CMP_LT(rcu_seq_current(&rdp->gp_seq) + ULONG_MAX / 4, - rnp->gp_seq)) + if (ULONG_CMP_LT(rcu_seq_current(&rdp->gp_seq) + seq_gpwrap_lag, + rnp->gp_seq)) { WRITE_ONCE(rdp->gpwrap, true); + WRITE_ONCE(rdp->gpwrap_count, READ_ONCE(rdp->gpwrap_count) + 1); + } if (ULONG_CMP_LT(rdp->rcu_iw_gp_seq + ULONG_MAX / 4, rnp->gp_seq)) rdp->rcu_iw_gp_seq = rnp->gp_seq + ULONG_MAX / 4; } @@ -781,6 +838,10 @@ static int rcu_watching_snap_save(struct rcu_data *rdp) return 0; } +#ifndef arch_irq_stat_cpu +#define arch_irq_stat_cpu(cpu) 0 +#endif + /* * Returns positive if the specified CPU has passed through a quiescent state * by virtue of being in or having passed through an dynticks idle state since @@ -916,9 +977,9 @@ static int rcu_watching_snap_recheck(struct rcu_data *rdp) rsrp->cputime_irq = kcpustat_field(kcsp, CPUTIME_IRQ, cpu); rsrp->cputime_softirq = kcpustat_field(kcsp, CPUTIME_SOFTIRQ, cpu); rsrp->cputime_system = kcpustat_field(kcsp, CPUTIME_SYSTEM, cpu); - rsrp->nr_hardirqs = kstat_cpu_irqs_sum(rdp->cpu); - rsrp->nr_softirqs = kstat_cpu_softirqs_sum(rdp->cpu); - rsrp->nr_csw = nr_context_switches_cpu(rdp->cpu); + rsrp->nr_hardirqs = kstat_cpu_irqs_sum(cpu) + arch_irq_stat_cpu(cpu); + rsrp->nr_softirqs = kstat_cpu_softirqs_sum(cpu); + rsrp->nr_csw = nr_context_switches_cpu(cpu); rsrp->jiffies = jiffies; rsrp->gp_seq = rdp->gp_seq; } @@ -1040,38 +1101,6 @@ static bool rcu_future_gp_cleanup(struct rcu_node *rnp) return needmore; } -static void swake_up_one_online_ipi(void *arg) -{ - struct swait_queue_head *wqh = arg; - - swake_up_one(wqh); -} - -static void swake_up_one_online(struct swait_queue_head *wqh) -{ - int cpu = get_cpu(); - - /* - * If called from rcutree_report_cpu_starting(), wake up - * is dangerous that late in the CPU-down hotplug process. The - * scheduler might queue an ignored hrtimer. Defer the wake up - * to an online CPU instead. - */ - if (unlikely(cpu_is_offline(cpu))) { - int target; - - target = cpumask_any_and(housekeeping_cpumask(HK_TYPE_RCU), - cpu_online_mask); - - smp_call_function_single(target, swake_up_one_online_ipi, - wqh, 0); - put_cpu(); - } else { - put_cpu(); - swake_up_one(wqh); - } -} - /* * Awaken the grace-period kthread. Don't do a self-awaken (unless in an * interrupt or softirq handler, in which case we just might immediately @@ -1096,7 +1125,7 @@ static void rcu_gp_kthread_wake(void) return; WRITE_ONCE(rcu_state.gp_wake_time, jiffies); WRITE_ONCE(rcu_state.gp_wake_seq, READ_ONCE(rcu_state.gp_seq)); - swake_up_one_online(&rcu_state.gp_wq); + swake_up_one(&rcu_state.gp_wq); } /* @@ -1254,7 +1283,7 @@ static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) /* Handle the ends of any preceding grace periods first. */ if (rcu_seq_completed_gp(rdp->gp_seq, rnp->gp_seq) || - unlikely(READ_ONCE(rdp->gpwrap))) { + unlikely(rdp->gpwrap)) { if (!offloaded) ret = rcu_advance_cbs(rnp, rdp); /* Advance CBs. */ rdp->core_needs_qs = false; @@ -1268,7 +1297,7 @@ static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) /* Now handle the beginnings of any new-to-this-CPU grace periods. */ if (rcu_seq_new_gp(rdp->gp_seq, rnp->gp_seq) || - unlikely(READ_ONCE(rdp->gpwrap))) { + unlikely(rdp->gpwrap)) { /* * If the current grace period is waiting for this CPU, * set up to detect a quiescent state, otherwise don't @@ -1283,7 +1312,7 @@ static bool __note_gp_changes(struct rcu_node *rnp, struct rcu_data *rdp) rdp->gp_seq = rnp->gp_seq; /* Remember new grace-period state. */ if (ULONG_CMP_LT(rdp->gp_seq_needed, rnp->gp_seq_needed) || rdp->gpwrap) WRITE_ONCE(rdp->gp_seq_needed, rnp->gp_seq_needed); - if (IS_ENABLED(CONFIG_PROVE_RCU) && READ_ONCE(rdp->gpwrap)) + if (IS_ENABLED(CONFIG_PROVE_RCU) && rdp->gpwrap) WRITE_ONCE(rdp->last_sched_clock, jiffies); WRITE_ONCE(rdp->gpwrap, false); rcu_gpnum_ovf(rnp, rdp); @@ -1603,8 +1632,10 @@ static void rcu_sr_put_wait_head(struct llist_node *node) atomic_set_release(&sr_wn->inuse, 0); } -/* Disabled by default. */ -static int rcu_normal_wake_from_gp; +/* Enable rcu_normal_wake_from_gp automatically on small systems. */ +#define WAKE_FROM_GP_CPU_THRESHOLD 16 + +static int rcu_normal_wake_from_gp = -1; module_param(rcu_normal_wake_from_gp, int, 0644); static struct workqueue_struct *sync_wq; @@ -1612,12 +1643,10 @@ static void rcu_sr_normal_complete(struct llist_node *node) { struct rcu_synchronize *rs = container_of( (struct rcu_head *) node, struct rcu_synchronize, head); - unsigned long oldstate = (unsigned long) rs->head.func; WARN_ONCE(IS_ENABLED(CONFIG_PROVE_RCU) && - !poll_state_synchronize_rcu(oldstate), - "A full grace period is not passed yet: %lu", - rcu_seq_diff(get_state_synchronize_rcu(), oldstate)); + !poll_state_synchronize_rcu_full(&rs->oldstate), + "A full grace period is not passed yet!\n"); /* Finally. */ complete(&rs->completion); @@ -1780,6 +1809,7 @@ static noinline_for_stack bool rcu_gp_init(void) struct rcu_data *rdp; struct rcu_node *rnp = rcu_get_root(); bool start_new_poll; + unsigned long old_gp_seq; WRITE_ONCE(rcu_state.gp_activity, jiffies); raw_spin_lock_irq_rcu_node(rnp); @@ -1801,10 +1831,31 @@ static noinline_for_stack bool rcu_gp_init(void) /* Advance to a new grace period and initialize state. */ record_gp_stall_check_time(); + /* + * A new wait segment must be started before gp_seq advanced, so + * that previous gp waiters won't observe the new gp_seq. + */ + start_new_poll = rcu_sr_normal_gp_init(); /* Record GP times before starting GP, hence rcu_seq_start(). */ + old_gp_seq = rcu_state.gp_seq; + /* + * Critical ordering: rcu_seq_start() must happen BEFORE the CPU hotplug + * scan below. Otherwise we risk a race where a newly onlining CPU could + * be missed by the current grace period, potentially leading to + * use-after-free errors. For a detailed explanation of this race, see + * Documentation/RCU/Design/Requirements/Requirements.rst in the + * "Hotplug CPU" section. + * + * Also note that the root rnp's gp_seq is kept separate from, and lags, + * the rcu_state's gp_seq, for a reason. See the Quick-Quiz on + * Single-node systems for more details (in Data-Structures.rst). + */ rcu_seq_start(&rcu_state.gp_seq); + /* Ensure that rcu_seq_done_exact() guardband doesn't give false positives. */ + WARN_ON_ONCE(IS_ENABLED(CONFIG_PROVE_RCU) && + rcu_seq_done_exact(&old_gp_seq, rcu_seq_snap(&rcu_state.gp_seq))); + ASSERT_EXCLUSIVE_WRITER(rcu_state.gp_seq); - start_new_poll = rcu_sr_normal_gp_init(); trace_rcu_grace_period(rcu_state.name, rcu_state.gp_seq, TPS("start")); rcu_poll_gp_seq_start(&rcu_state.gp_seq_polled_snap); raw_spin_unlock_irq_rcu_node(rnp); @@ -1835,6 +1886,10 @@ static noinline_for_stack bool rcu_gp_init(void) /* Exclude CPU hotplug operations. */ rcu_for_each_leaf_node(rnp) { local_irq_disable(); + /* + * Serialize with CPU offline. See Requirements.rst > Hotplug CPU > + * Concurrent Quiescent State Reporting for Offline CPUs. + */ arch_spin_lock(&rcu_state.ofl_lock); raw_spin_lock_rcu_node(rnp); if (rnp->qsmaskinit == rnp->qsmaskinitnext && @@ -1909,7 +1964,12 @@ static noinline_for_stack bool rcu_gp_init(void) trace_rcu_grace_period_init(rcu_state.name, rnp->gp_seq, rnp->level, rnp->grplo, rnp->grphi, rnp->qsmask); - /* Quiescent states for tasks on any now-offline CPUs. */ + /* + * Quiescent states for tasks on any now-offline CPUs. Since we + * released the ofl and rnp lock before this loop, CPUs might + * have gone offline and we have to report QS on their behalf. + * See Requirements.rst > Hotplug CPU > Concurrent QS Reporting. + */ mask = rnp->qsmask & ~rnp->qsmaskinitnext; rnp->rcu_gp_init_mask = mask; if ((mask || rnp->wait_blkd_tasks) && rcu_is_leaf_node(rnp)) @@ -1924,6 +1984,17 @@ static noinline_for_stack bool rcu_gp_init(void) if (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD)) on_each_cpu(rcu_strict_gp_boundary, NULL, 0); + /* + * Immediately report QS for the GP kthread's CPU. The GP kthread + * cannot be in an RCU read-side critical section while running + * the FQS scan. This eliminates the need for a second FQS wait + * when all CPUs are idle. + */ + preempt_disable(); + rcu_qs(); + rcu_report_qs_rdp(this_cpu_ptr(&rcu_data)); + preempt_enable(); + return true; } @@ -2637,10 +2708,8 @@ void rcu_sched_clock_irq(int user) /* The load-acquire pairs with the store-release setting to true. */ if (smp_load_acquire(this_cpu_ptr(&rcu_data.rcu_urgent_qs))) { /* Idle and userspace execution already are quiescent states. */ - if (!rcu_is_cpu_rrupt_from_idle() && !user) { - set_tsk_need_resched(current); - set_preempt_need_resched(); - } + if (!rcu_is_cpu_rrupt_from_idle() && !user) + set_need_resched_current(); __this_cpu_write(rcu_data.rcu_urgent_qs, false); } rcu_flavor_sched_clock_irq(user); @@ -2765,7 +2834,6 @@ static void strict_work_handler(struct work_struct *work) /* Perform RCU core processing work for the current CPU. */ static __latent_entropy void rcu_core(void) { - unsigned long flags; struct rcu_data *rdp = raw_cpu_ptr(&rcu_data); struct rcu_node *rnp = rdp->mynode; @@ -2778,8 +2846,8 @@ static __latent_entropy void rcu_core(void) if (IS_ENABLED(CONFIG_PREEMPT_COUNT) && (!(preempt_count() & PREEMPT_MASK))) { rcu_preempt_deferred_qs(current); } else if (rcu_preempt_need_deferred_qs(current)) { - set_tsk_need_resched(current); - set_preempt_need_resched(); + guard(irqsave)(); + set_need_resched_current(); } /* Update RCU state based on any recent quiescent states. */ @@ -2788,10 +2856,9 @@ static __latent_entropy void rcu_core(void) /* No grace period and unregistered callbacks? */ if (!rcu_gp_in_progress() && rcu_segcblist_is_enabled(&rdp->cblist) && !rcu_rdp_is_offloaded(rdp)) { - local_irq_save(flags); + guard(irqsave)(); if (!rcu_segcblist_restempty(&rdp->cblist, RCU_NEXT_READY_TAIL)) rcu_accelerate_cbs_unlocked(rnp, rdp); - local_irq_restore(flags); } rcu_check_gp_start_stall(rnp, rdp, rcu_jiffies_till_stall_check()); @@ -2931,13 +2998,8 @@ static int __init rcu_spawn_core_kthreads(void) static void rcutree_enqueue(struct rcu_data *rdp, struct rcu_head *head, rcu_callback_t func) { rcu_segcblist_enqueue(&rdp->cblist, head); - if (__is_kvfree_rcu_offset((unsigned long)func)) - trace_rcu_kvfree_callback(rcu_state.name, head, - (unsigned long)func, - rcu_segcblist_n_cbs(&rdp->cblist)); - else - trace_rcu_callback(rcu_state.name, head, - rcu_segcblist_n_cbs(&rdp->cblist)); + trace_rcu_callback(rcu_state.name, head, + rcu_segcblist_n_cbs(&rdp->cblist)); trace_rcu_segcb_stats(&rdp->cblist, TPS("SegCBQueued")); } @@ -3047,6 +3109,10 @@ __call_rcu_common(struct rcu_head *head, rcu_callback_t func, bool lazy_in) /* Misaligned rcu_head! */ WARN_ON_ONCE((unsigned long)head & (sizeof(void *) - 1)); + /* Avoid NULL dereference if callback is NULL. */ + if (WARN_ON_ONCE(!func)) + return; + if (debug_rcu_head_queue(head)) { /* * Probable double call_rcu(), so leak the callback. @@ -3107,7 +3173,7 @@ module_param(enable_rcu_lazy, bool, 0444); * critical sections have completed. * * Use this API instead of call_rcu() if you don't want the callback to be - * invoked after very long periods of time, which can happen on systems without + * delayed for very long periods of time, which can happen on systems without * memory pressure and on systems which are lightly loaded or mostly idle. * This function will cause callbacks to be invoked sooner than later at the * expense of extra power. Other than that, this function is identical to, and @@ -3138,6 +3204,12 @@ EXPORT_SYMBOL_GPL(call_rcu_hurry); * might well execute concurrently with RCU read-side critical sections * that started after call_rcu() was invoked. * + * It is perfectly legal to repost an RCU callback, potentially with + * a different callback function, from within its callback function. + * The specified function will be invoked after another full grace period + * has elapsed. This use case is similar in form to the common practice + * of reposting a timer from within its own handler. + * * RCU read-side critical sections are delimited by rcu_read_lock() * and rcu_read_unlock(), and may be nested. In addition, but only in * v5.0 and later, regions of code across which interrupts, preemption, @@ -3166,6 +3238,13 @@ EXPORT_SYMBOL_GPL(call_rcu_hurry); * * Implementation of these memory-ordering guarantees is described here: * Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst. + * + * Specific to call_rcu() (as opposed to the other call_rcu*() functions), + * in kernels built with CONFIG_RCU_LAZY=y, call_rcu() might delay for many + * seconds before starting the grace period needed by the corresponding + * callback. This delay can significantly improve energy-efficiency + * on low-utilization battery-powered devices. To avoid this delay, + * in latency-sensitive kernel code, use call_rcu_hurry(). */ void call_rcu(struct rcu_head *head, rcu_callback_t func) { @@ -3201,7 +3280,7 @@ static void synchronize_rcu_normal(void) trace_rcu_sr_normal(rcu_state.name, &rs.head, TPS("request")); - if (!READ_ONCE(rcu_normal_wake_from_gp)) { + if (READ_ONCE(rcu_normal_wake_from_gp) < 1) { wait_rcu_gp(call_rcu_hurry); goto trace_complete_out; } @@ -3214,7 +3293,7 @@ static void synchronize_rcu_normal(void) * snapshot before adding a request. */ if (IS_ENABLED(CONFIG_PROVE_RCU)) - rs.head.func = (void *) get_state_synchronize_rcu(); + get_state_synchronize_rcu_full(&rs.oldstate); rcu_sr_normal_add_req(&rs); @@ -3357,14 +3436,17 @@ EXPORT_SYMBOL_GPL(get_state_synchronize_rcu); */ void get_state_synchronize_rcu_full(struct rcu_gp_oldstate *rgosp) { - struct rcu_node *rnp = rcu_get_root(); - /* * Any prior manipulation of RCU-protected data must happen * before the loads from ->gp_seq and ->expedited_sequence. */ smp_mb(); /* ^^^ */ - rgosp->rgos_norm = rcu_seq_snap(&rnp->gp_seq); + + // Yes, rcu_state.gp_seq, not rnp_root->gp_seq, the latter's use + // in poll_state_synchronize_rcu_full() notwithstanding. Use of + // the latter here would result in too-short grace periods due to + // interactions with newly onlined CPUs. + rgosp->rgos_norm = rcu_seq_snap(&rcu_state.gp_seq); rgosp->rgos_exp = rcu_seq_snap(&rcu_state.expedited_sequence); } EXPORT_SYMBOL_GPL(get_state_synchronize_rcu_full); @@ -3699,7 +3781,7 @@ static void rcu_barrier_entrain(struct rcu_data *rdp) } rcu_nocb_unlock(rdp); if (wake_nocb) - wake_nocb_gp(rdp, false); + wake_nocb_gp(rdp); smp_store_release(&rdp->barrier_seq_snap, gseq); } @@ -3726,6 +3808,11 @@ static void rcu_barrier_handler(void *cpu_in) * to complete. For example, if there are no RCU callbacks queued anywhere * in the system, then rcu_barrier() is within its rights to return * immediately, without waiting for anything, much less an RCU grace period. + * In fact, rcu_barrier() will normally not result in any RCU grace periods + * beyond those that were already destined to be executed. + * + * In kernels built with CONFIG_RCU_LAZY=y, this function also hurries all + * pending lazy RCU callbacks. */ void rcu_barrier(void) { @@ -3942,7 +4029,7 @@ bool rcu_cpu_online(int cpu) * RCU on an offline processor during initial boot, hence the check for * rcu_scheduler_fully_active. */ -bool rcu_lockdep_current_cpu_online(void) +bool notrace rcu_lockdep_current_cpu_online(void) { struct rcu_data *rdp; bool ret = false; @@ -4188,6 +4275,8 @@ int rcutree_prepare_cpu(unsigned int cpu) rdp->rcu_iw_gp_seq = rdp->gp_seq - 1; trace_rcu_grace_period(rcu_state.name, rdp->gp_seq, TPS("cpuonl")); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); + + rcu_preempt_deferred_qs_init(rdp); rcu_spawn_rnp_kthreads(rnp); rcu_spawn_cpu_nocb_kthread(cpu); ASSERT_EXCLUSIVE_WRITER(rcu_state.n_online_cpus); @@ -4223,7 +4312,6 @@ int rcutree_online_cpu(unsigned int cpu) raw_spin_unlock_irqrestore_rcu_node(rnp, flags); if (rcu_scheduler_active == RCU_SCHEDULER_INACTIVE) return 0; /* Too early in boot for scheduler work. */ - sync_sched_exp_online_cleanup(cpu); // Stop-machine done, so allow nohz_full to disable tick. tick_dep_clear(TICK_DEP_BIT_RCU); @@ -4313,6 +4401,12 @@ void rcutree_report_cpu_dead(void) * may introduce a new READ-side while it is actually off the QS masks. */ lockdep_assert_irqs_disabled(); + /* + * CPUHP_AP_SMPCFD_DYING was the last call for rcu_exp_handler() execution. + * The requested QS must have been reported on the last context switch + * from stop machine to idle. + */ + WARN_ON_ONCE(rdp->cpu_no_qs.b.exp); // Do any dangling deferred wakeups. do_nocb_deferred_wakeup(rdp); @@ -4320,6 +4414,13 @@ void rcutree_report_cpu_dead(void) /* Remove outgoing CPU from mask in the leaf rcu_node structure. */ mask = rdp->grpmask; + + /* + * Hold the ofl_lock and rnp lock to avoid races between CPU going + * offline and doing a QS report (as below), versus rcu_gp_init(). + * See Requirements.rst > Hotplug CPU > Concurrent QS Reporting section + * for more details. + */ arch_spin_lock(&rcu_state.ofl_lock); raw_spin_lock_irqsave_rcu_node(rnp, flags); /* Enforce GP memory-order guarantee. */ rdp->rcu_ofl_gp_seq = READ_ONCE(rcu_state.gp_seq); @@ -4330,6 +4431,7 @@ void rcutree_report_cpu_dead(void) rcu_report_qs_rnp(mask, rnp, rnp->gp_seq, flags); raw_spin_lock_irqsave_rcu_node(rnp, flags); } + /* Clear from ->qsmaskinitnext to mark offline. */ WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext & ~mask); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); arch_spin_unlock(&rcu_state.ofl_lock); @@ -4796,12 +4898,18 @@ void __init rcu_init(void) rcutree_online_cpu(cpu); /* Create workqueue for Tree SRCU and for expedited GPs. */ - rcu_gp_wq = alloc_workqueue("rcu_gp", WQ_MEM_RECLAIM, 0); + rcu_gp_wq = alloc_workqueue("rcu_gp", WQ_MEM_RECLAIM | WQ_PERCPU, 0); WARN_ON(!rcu_gp_wq); - sync_wq = alloc_workqueue("sync_wq", WQ_MEM_RECLAIM, 0); + sync_wq = alloc_workqueue("sync_wq", WQ_MEM_RECLAIM | WQ_UNBOUND, 0); WARN_ON(!sync_wq); + /* Respect if explicitly disabled via a boot parameter. */ + if (rcu_normal_wake_from_gp < 0) { + if (num_possible_cpus() <= WAKE_FROM_GP_CPU_THRESHOLD) + rcu_normal_wake_from_gp = 1; + } + /* Fill in default value for rcutree.qovld boot parameter. */ /* -After- the rcu_node ->lock fields are initialized! */ if (qovld < 0) diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index a9a811d9d7a3..7dfc57e9adb1 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -168,12 +168,23 @@ struct rcu_snap_record { u64 cputime_irq; /* Accumulated cputime of hard irqs */ u64 cputime_softirq;/* Accumulated cputime of soft irqs */ u64 cputime_system; /* Accumulated cputime of kernel tasks */ - unsigned long nr_hardirqs; /* Accumulated number of hard irqs */ + u64 nr_hardirqs; /* Accumulated number of hard irqs */ unsigned int nr_softirqs; /* Accumulated number of soft irqs */ unsigned long long nr_csw; /* Accumulated number of task switches */ unsigned long jiffies; /* Track jiffies value */ }; +/* + * An IRQ work (deferred_qs_iw) is used by RCU to get the scheduler's attention. + * to report quiescent states at the soonest possible time. + * The request can be in one of the following states: + * - DEFER_QS_IDLE: An IRQ work is yet to be scheduled. + * - DEFER_QS_PENDING: An IRQ work was scheduled but either not yet run, or it + * ran and we still haven't reported a quiescent state. + */ +#define DEFER_QS_IDLE 0 +#define DEFER_QS_PENDING 1 + /* Per-CPU data for read-copy update. */ struct rcu_data { /* 1) quiescent-state and grace-period handling : */ @@ -183,6 +194,7 @@ struct rcu_data { bool core_needs_qs; /* Core waits for quiescent state. */ bool beenonline; /* CPU online at least once. */ bool gpwrap; /* Possible ->gp_seq wrap. */ + unsigned int gpwrap_count; /* Count of GP sequence wrap. */ bool cpu_started; /* RCU watching this onlining CPU. */ struct rcu_node *mynode; /* This CPU's leaf of hierarchy */ unsigned long grpmask; /* Mask to apply to leaf qsmask. */ @@ -191,7 +203,7 @@ struct rcu_data { /* during and after the last grace */ /* period it is aware of. */ struct irq_work defer_qs_iw; /* Obtain later scheduler attention. */ - bool defer_qs_iw_pending; /* Scheduler attention pending? */ + int defer_qs_pending; /* irqwork or softirq pending? */ struct work_struct strict_work; /* Schedule readers for strict GPs. */ /* 2) batch handling */ @@ -289,7 +301,6 @@ struct rcu_data { #define RCU_NOCB_WAKE_BYPASS 1 #define RCU_NOCB_WAKE_LAZY 2 #define RCU_NOCB_WAKE 3 -#define RCU_NOCB_WAKE_FORCE 4 #define RCU_JIFFIES_TILL_FORCE_QS (1 + (HZ > 250) + (HZ > 500)) /* For jiffies_till_first_fqs and */ @@ -476,6 +487,7 @@ static int rcu_print_task_exp_stall(struct rcu_node *rnp); static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp); static void rcu_flavor_sched_clock_irq(int user); static void dump_blkd_tasks(struct rcu_node *rnp, int ncheck); +static void rcu_preempt_deferred_qs_init(struct rcu_data *rdp); static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags); static void rcu_preempt_boost_start_gp(struct rcu_node *rnp); static bool rcu_is_callbacks_kthread(struct rcu_data *rdp); @@ -487,7 +499,7 @@ static void zero_cpu_stall_ticks(struct rcu_data *rdp); static struct swait_queue_head *rcu_nocb_gp_get(struct rcu_node *rnp); static void rcu_nocb_gp_cleanup(struct swait_queue_head *sq); static void rcu_init_one_nocb(struct rcu_node *rnp); -static bool wake_nocb_gp(struct rcu_data *rdp, bool force); +static bool wake_nocb_gp(struct rcu_data *rdp); static bool rcu_nocb_flush_bypass(struct rcu_data *rdp, struct rcu_head *rhp, unsigned long j, bool lazy); static void call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *head, diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h index 77efed89c79e..82cada459e5d 100644 --- a/kernel/rcu/tree_exp.h +++ b/kernel/rcu/tree_exp.h @@ -141,6 +141,13 @@ static void __maybe_unused sync_exp_reset_tree(void) raw_spin_lock_irqsave_rcu_node(rnp, flags); WARN_ON_ONCE(rnp->expmask); WRITE_ONCE(rnp->expmask, rnp->expmaskinit); + /* + * Need to wait for any blocked tasks as well. Note that + * additional blocking tasks will also block the expedited GP + * until such time as the ->expmask bits are cleared. + */ + if (rcu_is_leaf_node(rnp) && rcu_preempt_has_tasks(rnp)) + WRITE_ONCE(rnp->exp_tasks, rnp->blkd_tasks.next); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } } @@ -200,7 +207,7 @@ static void __rcu_report_exp_rnp(struct rcu_node *rnp, if (rnp->parent == NULL) { raw_spin_unlock_irqrestore_rcu_node(rnp, flags); if (wake) - swake_up_one_online(&rcu_state.expedited_wq); + swake_up_one(&rcu_state.expedited_wq); break; } @@ -230,17 +237,19 @@ static void __maybe_unused rcu_report_exp_rnp(struct rcu_node *rnp, bool wake) * specified leaf rcu_node structure, which is acquired by the caller. */ static void rcu_report_exp_cpu_mult(struct rcu_node *rnp, unsigned long flags, - unsigned long mask, bool wake) + unsigned long mask_in, bool wake) __releases(rnp->lock) { int cpu; + unsigned long mask; struct rcu_data *rdp; raw_lockdep_assert_held_rcu_node(rnp); - if (!(rnp->expmask & mask)) { + if (!(rnp->expmask & mask_in)) { raw_spin_unlock_irqrestore_rcu_node(rnp, flags); return; } + mask = mask_in & rnp->expmask; WRITE_ONCE(rnp->expmask, rnp->expmask & ~mask); for_each_leaf_node_cpu_mask(rnp, cpu, mask) { rdp = per_cpu_ptr(&rcu_data, cpu); @@ -391,13 +400,6 @@ static void __sync_rcu_exp_select_node_cpus(struct rcu_exp_work *rewp) } mask_ofl_ipi = rnp->expmask & ~mask_ofl_test; - /* - * Need to wait for any blocked tasks as well. Note that - * additional blocking tasks will also block the expedited GP - * until such time as the ->expmask bits are cleared. - */ - if (rcu_preempt_has_tasks(rnp)) - WRITE_ONCE(rnp->exp_tasks, rnp->blkd_tasks.next); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); /* IPI the remaining CPUs for expedited quiescent state. */ @@ -587,7 +589,12 @@ static void synchronize_rcu_expedited_stall(unsigned long jiffies_start, unsigne pr_cont(" } %lu jiffies s: %lu root: %#lx/%c\n", j - jiffies_start, rcu_state.expedited_sequence, data_race(rnp_root->expmask), ".T"[!!data_race(rnp_root->exp_tasks)]); - if (ndetected) { + if (!ndetected) { + // This is invoked from the grace-period worker, so + // a new grace period cannot have started. And if this + // worker were stalled, we would not get here. ;-) + pr_err("INFO: Expedited stall ended before state dump start\n"); + } else { pr_err("blocking rcu_node structures (internal RCU debug):"); rcu_for_each_node_breadth_first(rnp) { if (rnp == rnp_root) @@ -727,8 +734,7 @@ static void rcu_exp_need_qs(void) __this_cpu_write(rcu_data.cpu_no_qs.b.exp, true); /* Store .exp before .rcu_urgent_qs. */ smp_store_release(this_cpu_ptr(&rcu_data.rcu_urgent_qs), true); - set_tsk_need_resched(current); - set_preempt_need_resched(); + set_need_resched_current(); } #ifdef CONFIG_PREEMPT_RCU @@ -749,12 +755,8 @@ static void rcu_exp_handler(void *unused) struct task_struct *t = current; /* - * First, is there no need for a quiescent state from this CPU, - * or is this CPU already looking for a quiescent state for the - * current grace period? If either is the case, just leave. - * However, this should not happen due to the preemptible - * sync_sched_exp_online_cleanup() implementation being a no-op, - * so warn if this does happen. + * WARN if the CPU is unexpectedly already looking for a + * QS or has already reported one. */ ASSERT_EXCLUSIVE_WRITER_SCOPED(rdp->cpu_no_qs.b.exp); if (WARN_ON_ONCE(!(READ_ONCE(rnp->expmask) & rdp->grpmask) || @@ -801,11 +803,6 @@ static void rcu_exp_handler(void *unused) WARN_ON_ONCE(1); } -/* PREEMPTION=y, so no PREEMPTION=n expedited grace period to clean up after. */ -static void sync_sched_exp_online_cleanup(int cpu) -{ -} - /* * Scan the current list of tasks blocked within RCU read-side critical * sections, printing out the tid of each that is blocking the current @@ -883,38 +880,6 @@ static void rcu_exp_handler(void *unused) rcu_exp_need_qs(); } -/* Send IPI for expedited cleanup if needed at end of CPU-hotplug operation. */ -static void sync_sched_exp_online_cleanup(int cpu) -{ - unsigned long flags; - int my_cpu; - struct rcu_data *rdp; - int ret; - struct rcu_node *rnp; - - rdp = per_cpu_ptr(&rcu_data, cpu); - rnp = rdp->mynode; - my_cpu = get_cpu(); - /* Quiescent state either not needed or already requested, leave. */ - if (!(READ_ONCE(rnp->expmask) & rdp->grpmask) || - READ_ONCE(rdp->cpu_no_qs.b.exp)) { - put_cpu(); - return; - } - /* Quiescent state needed on current CPU, so set it up locally. */ - if (my_cpu == cpu) { - local_irq_save(flags); - rcu_exp_need_qs(); - local_irq_restore(flags); - put_cpu(); - return; - } - /* Quiescent state needed on some other CPU, send IPI. */ - ret = smp_call_function_single(cpu, rcu_exp_handler, NULL, 0); - put_cpu(); - WARN_ON_ONCE(ret); -} - /* * Because preemptible RCU does not exist, we never have to check for * tasks blocked within RCU read-side critical sections that are diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 2605dd234a13..1047b30cd46b 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -190,9 +190,18 @@ static void rcu_init_one_nocb(struct rcu_node *rnp) init_swait_queue_head(&rnp->nocb_gp_wq[1]); } +/* Clear any pending deferred wakeup timer (nocb_gp_lock must be held). */ +static void nocb_defer_wakeup_cancel(struct rcu_data *rdp_gp) +{ + if (rdp_gp->nocb_defer_wakeup > RCU_NOCB_WAKE_NOT) { + WRITE_ONCE(rdp_gp->nocb_defer_wakeup, RCU_NOCB_WAKE_NOT); + timer_delete(&rdp_gp->nocb_timer); + } +} + static bool __wake_nocb_gp(struct rcu_data *rdp_gp, struct rcu_data *rdp, - bool force, unsigned long flags) + unsigned long flags) __releases(rdp_gp->nocb_gp_lock) { bool needwake = false; @@ -204,19 +213,16 @@ static bool __wake_nocb_gp(struct rcu_data *rdp_gp, return false; } - if (rdp_gp->nocb_defer_wakeup > RCU_NOCB_WAKE_NOT) { - WRITE_ONCE(rdp_gp->nocb_defer_wakeup, RCU_NOCB_WAKE_NOT); - del_timer(&rdp_gp->nocb_timer); - } + nocb_defer_wakeup_cancel(rdp_gp); - if (force || READ_ONCE(rdp_gp->nocb_gp_sleep)) { + if (READ_ONCE(rdp_gp->nocb_gp_sleep)) { WRITE_ONCE(rdp_gp->nocb_gp_sleep, false); needwake = true; } raw_spin_unlock_irqrestore(&rdp_gp->nocb_gp_lock, flags); if (needwake) { trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("DoWake")); - swake_up_one_online(&rdp_gp->nocb_gp_wq); + swake_up_one(&rdp_gp->nocb_gp_wq); } return needwake; @@ -225,13 +231,13 @@ static bool __wake_nocb_gp(struct rcu_data *rdp_gp, /* * Kick the GP kthread for this NOCB group. */ -static bool wake_nocb_gp(struct rcu_data *rdp, bool force) +static bool wake_nocb_gp(struct rcu_data *rdp) { unsigned long flags; struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; raw_spin_lock_irqsave(&rdp_gp->nocb_gp_lock, flags); - return __wake_nocb_gp(rdp_gp, rdp, force, flags); + return __wake_nocb_gp(rdp_gp, rdp, flags); } #ifdef CONFIG_RCU_LAZY @@ -276,7 +282,7 @@ static void wake_nocb_gp_defer(struct rcu_data *rdp, int waketype, * callback storms, no need to wake up too early. */ if (waketype == RCU_NOCB_WAKE_LAZY && - rdp->nocb_defer_wakeup == RCU_NOCB_WAKE_NOT) { + rdp_gp->nocb_defer_wakeup == RCU_NOCB_WAKE_NOT) { mod_timer(&rdp_gp->nocb_timer, jiffies + rcu_get_jiffies_lazy_flush()); WRITE_ONCE(rdp_gp->nocb_defer_wakeup, waketype); } else if (waketype == RCU_NOCB_WAKE_BYPASS) { @@ -373,6 +379,38 @@ static void rcu_nocb_try_flush_bypass(struct rcu_data *rdp, unsigned long j) } /* + * Determine if the bypass queue needs to be flushed based on time and size. + * For lazy-only bypass queues, use the lazy flush timeout; otherwise flush + * based on jiffy advancement. The flush_faster controls flush aggressiveness. + */ +static bool nocb_bypass_needs_flush(struct rcu_data *rdp, long bypass_ncbs, + long lazy_ncbs, unsigned long j, + bool flush_faster) +{ + bool bypass_is_lazy; + unsigned long bypass_first; + unsigned long flush_timeout; + long qhimark_thresh; + + if (!bypass_ncbs) + return false; + + qhimark_thresh = flush_faster ? qhimark : 2 * qhimark; + if (bypass_ncbs >= qhimark_thresh) + return true; + + bypass_first = READ_ONCE(rdp->nocb_bypass_first); + bypass_is_lazy = (bypass_ncbs == lazy_ncbs); + + if (bypass_is_lazy) + flush_timeout = rcu_get_jiffies_lazy_flush(); + else + flush_timeout = flush_faster ? 0 : 1; + + return time_after(j, bypass_first + flush_timeout); +} + +/* * See whether it is appropriate to use the ->nocb_bypass list in order * to control contention on ->nocb_lock. A limited number of direct * enqueues are permitted into ->cblist per jiffy. If ->nocb_bypass @@ -398,7 +436,8 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp, unsigned long cur_gp_seq; unsigned long j = jiffies; long ncbs = rcu_cblist_n_cbs(&rdp->nocb_bypass); - bool bypass_is_lazy = (ncbs == READ_ONCE(rdp->lazy_len)); + long lazy_len = READ_ONCE(rdp->lazy_len); + bool bypass_is_lazy = (ncbs == lazy_len); lockdep_assert_irqs_disabled(); @@ -450,10 +489,7 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp, // If ->nocb_bypass has been used too long or is too full, // flush ->nocb_bypass to ->cblist. - if ((ncbs && !bypass_is_lazy && j != READ_ONCE(rdp->nocb_bypass_first)) || - (ncbs && bypass_is_lazy && - (time_after(j, READ_ONCE(rdp->nocb_bypass_first) + rcu_get_jiffies_lazy_flush()))) || - ncbs >= qhimark) { + if (nocb_bypass_needs_flush(rdp, ncbs, lazy_len, j, true)) { rcu_nocb_lock(rdp); *was_alldone = !rcu_segcblist_pend_cbs(&rdp->cblist); @@ -518,22 +554,17 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp, } /* - * Awaken the no-CBs grace-period kthread if needed, either due to it - * legitimately being asleep or due to overload conditions. - * - * If warranted, also wake up the kthread servicing this CPUs queues. + * Awaken the no-CBs grace-period kthread if needed due to it legitimately + * being asleep. */ static void __call_rcu_nocb_wake(struct rcu_data *rdp, bool was_alldone, unsigned long flags) __releases(rdp->nocb_lock) { long bypass_len; - unsigned long cur_gp_seq; - unsigned long j; long lazy_len; long len; struct task_struct *t; - struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; // If we are being polled or there is no kthread, just leave. t = READ_ONCE(rdp->nocb_gp_kthread); @@ -549,53 +580,26 @@ static void __call_rcu_nocb_wake(struct rcu_data *rdp, bool was_alldone, lazy_len = READ_ONCE(rdp->lazy_len); if (was_alldone) { rdp->qlen_last_fqs_check = len; + rcu_nocb_unlock(rdp); // Only lazy CBs in bypass list if (lazy_len && bypass_len == lazy_len) { - rcu_nocb_unlock(rdp); wake_nocb_gp_defer(rdp, RCU_NOCB_WAKE_LAZY, TPS("WakeLazy")); - } else if (!irqs_disabled_flags(flags) && cpu_online(rdp->cpu)) { + } else if (!irqs_disabled_flags(flags)) { /* ... if queue was empty ... */ - rcu_nocb_unlock(rdp); - wake_nocb_gp(rdp, false); + wake_nocb_gp(rdp); trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("WakeEmpty")); } else { - /* - * Don't do the wake-up upfront on fragile paths. - * Also offline CPUs can't call swake_up_one_online() from - * (soft-)IRQs. Rely on the final deferred wake-up from - * rcutree_report_cpu_dead() - */ - rcu_nocb_unlock(rdp); wake_nocb_gp_defer(rdp, RCU_NOCB_WAKE, TPS("WakeEmptyIsDeferred")); } - } else if (len > rdp->qlen_last_fqs_check + qhimark) { - /* ... or if many callbacks queued. */ - rdp->qlen_last_fqs_check = len; - j = jiffies; - if (j != rdp->nocb_gp_adv_time && - rcu_segcblist_nextgp(&rdp->cblist, &cur_gp_seq) && - rcu_seq_done(&rdp->mynode->gp_seq, cur_gp_seq)) { - rcu_advance_cbs_nowake(rdp->mynode, rdp); - rdp->nocb_gp_adv_time = j; - } - smp_mb(); /* Enqueue before timer_pending(). */ - if ((rdp->nocb_cb_sleep || - !rcu_segcblist_ready_cbs(&rdp->cblist)) && - !timer_pending(&rdp_gp->nocb_timer)) { - rcu_nocb_unlock(rdp); - wake_nocb_gp_defer(rdp, RCU_NOCB_WAKE_FORCE, - TPS("WakeOvfIsDeferred")); - } else { - rcu_nocb_unlock(rdp); - trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("WakeNot")); - } - } else { - rcu_nocb_unlock(rdp); - trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("WakeNot")); + + return; } + + rcu_nocb_unlock(rdp); + trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("WakeNot")); } static void call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *head, @@ -699,15 +703,8 @@ static void nocb_gp_wait(struct rcu_data *my_rdp) bypass_ncbs = rcu_cblist_n_cbs(&rdp->nocb_bypass); lazy_ncbs = READ_ONCE(rdp->lazy_len); - if (bypass_ncbs && (lazy_ncbs == bypass_ncbs) && - (time_after(j, READ_ONCE(rdp->nocb_bypass_first) + rcu_get_jiffies_lazy_flush()) || - bypass_ncbs > 2 * qhimark)) { - flush_bypass = true; - } else if (bypass_ncbs && (lazy_ncbs != bypass_ncbs) && - (time_after(j, READ_ONCE(rdp->nocb_bypass_first) + 1) || - bypass_ncbs > 2 * qhimark)) { - flush_bypass = true; - } else if (!bypass_ncbs && rcu_segcblist_empty(&rdp->cblist)) { + flush_bypass = nocb_bypass_needs_flush(rdp, bypass_ncbs, lazy_ncbs, j, false); + if (!flush_bypass && !bypass_ncbs && rcu_segcblist_empty(&rdp->cblist)) { rcu_nocb_unlock_irqrestore(rdp, flags); continue; /* No callbacks here, try next. */ } @@ -820,10 +817,7 @@ static void nocb_gp_wait(struct rcu_data *my_rdp) if (rdp_toggling) my_rdp->nocb_toggling_rdp = NULL; - if (my_rdp->nocb_defer_wakeup > RCU_NOCB_WAKE_NOT) { - WRITE_ONCE(my_rdp->nocb_defer_wakeup, RCU_NOCB_WAKE_NOT); - del_timer(&my_rdp->nocb_timer); - } + nocb_defer_wakeup_cancel(my_rdp); WRITE_ONCE(my_rdp->nocb_gp_sleep, true); raw_spin_unlock_irqrestore(&my_rdp->nocb_gp_lock, flags); } else { @@ -972,7 +966,6 @@ static bool do_nocb_deferred_wakeup_common(struct rcu_data *rdp_gp, unsigned long flags) __releases(rdp_gp->nocb_gp_lock) { - int ndw; int ret; if (!rcu_nocb_need_deferred_wakeup(rdp_gp, level)) { @@ -980,8 +973,7 @@ static bool do_nocb_deferred_wakeup_common(struct rcu_data *rdp_gp, return false; } - ndw = rdp_gp->nocb_defer_wakeup; - ret = __wake_nocb_gp(rdp_gp, rdp, ndw == RCU_NOCB_WAKE_FORCE, flags); + ret = __wake_nocb_gp(rdp_gp, rdp, flags); trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("DeferredWake")); return ret; @@ -991,13 +983,12 @@ static bool do_nocb_deferred_wakeup_common(struct rcu_data *rdp_gp, static void do_nocb_deferred_wakeup_timer(struct timer_list *t) { unsigned long flags; - struct rcu_data *rdp = from_timer(rdp, t, nocb_timer); + struct rcu_data *rdp = timer_container_of(rdp, t, nocb_timer); WARN_ON_ONCE(rdp->nocb_gp_rdp != rdp); trace_rcu_nocb_wake(rcu_state.name, rdp->cpu, TPS("Timer")); raw_spin_lock_irqsave(&rdp->nocb_gp_lock, flags); - smp_mb__after_spinlock(); /* Timer expire before wakeup. */ do_nocb_deferred_wakeup_common(rdp, rdp, RCU_NOCB_WAKE_BYPASS, flags); } @@ -1113,30 +1104,6 @@ static int rcu_nocb_rdp_deoffload(struct rcu_data *rdp) return 0; } -int rcu_nocb_cpu_deoffload(int cpu) -{ - struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu); - int ret = 0; - - cpus_read_lock(); - mutex_lock(&rcu_state.nocb_mutex); - if (rcu_rdp_is_offloaded(rdp)) { - if (!cpu_online(cpu)) { - ret = rcu_nocb_rdp_deoffload(rdp); - if (!ret) - cpumask_clear_cpu(cpu, rcu_nocb_mask); - } else { - pr_info("NOCB: Cannot CB-deoffload online CPU %d\n", rdp->cpu); - ret = -EINVAL; - } - } - mutex_unlock(&rcu_state.nocb_mutex); - cpus_read_unlock(); - - return ret; -} -EXPORT_SYMBOL_GPL(rcu_nocb_cpu_deoffload); - static bool rcu_nocb_rdp_offload_wait_cond(struct rcu_data *rdp) { unsigned long flags; @@ -1152,7 +1119,6 @@ static bool rcu_nocb_rdp_offload_wait_cond(struct rcu_data *rdp) static int rcu_nocb_rdp_offload(struct rcu_data *rdp) { int wake_gp; - struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; WARN_ON_ONCE(cpu_online(rdp->cpu)); /* @@ -1162,7 +1128,7 @@ static int rcu_nocb_rdp_offload(struct rcu_data *rdp) if (!rdp->nocb_gp_rdp) return -EINVAL; - if (WARN_ON_ONCE(!rdp_gp->nocb_gp_kthread)) + if (WARN_ON_ONCE(!rdp->nocb_gp_kthread)) return -EINVAL; pr_info("Offloading %d\n", rdp->cpu); @@ -1172,7 +1138,7 @@ static int rcu_nocb_rdp_offload(struct rcu_data *rdp) wake_gp = rcu_nocb_queue_toggle_rdp(rdp); if (wake_gp) - wake_up_process(rdp_gp->nocb_gp_kthread); + wake_up_process(rdp->nocb_gp_kthread); swait_event_exclusive(rdp->nocb_state_wq, rcu_nocb_rdp_offload_wait_cond(rdp)); @@ -1182,28 +1148,52 @@ static int rcu_nocb_rdp_offload(struct rcu_data *rdp) return 0; } -int rcu_nocb_cpu_offload(int cpu) +/* Common helper for CPU offload/deoffload operations. */ +static int rcu_nocb_cpu_toggle_offload(int cpu, bool offload) { struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu); int ret = 0; cpus_read_lock(); mutex_lock(&rcu_state.nocb_mutex); - if (!rcu_rdp_is_offloaded(rdp)) { - if (!cpu_online(cpu)) { - ret = rcu_nocb_rdp_offload(rdp); - if (!ret) - cpumask_set_cpu(cpu, rcu_nocb_mask); - } else { - pr_info("NOCB: Cannot CB-offload online CPU %d\n", rdp->cpu); - ret = -EINVAL; - } + + /* Already in desired state, nothing to do. */ + if (rcu_rdp_is_offloaded(rdp) == offload) + goto out_unlock; + + if (cpu_online(cpu)) { + pr_info("NOCB: Cannot CB-%soffload online CPU %d\n", + offload ? "" : "de", rdp->cpu); + ret = -EINVAL; + goto out_unlock; } + + if (offload) { + ret = rcu_nocb_rdp_offload(rdp); + if (!ret) + cpumask_set_cpu(cpu, rcu_nocb_mask); + } else { + ret = rcu_nocb_rdp_deoffload(rdp); + if (!ret) + cpumask_clear_cpu(cpu, rcu_nocb_mask); + } + +out_unlock: mutex_unlock(&rcu_state.nocb_mutex); cpus_read_unlock(); - return ret; } + +int rcu_nocb_cpu_deoffload(int cpu) +{ + return rcu_nocb_cpu_toggle_offload(cpu, false /* de-offload */); +} +EXPORT_SYMBOL_GPL(rcu_nocb_cpu_deoffload); + +int rcu_nocb_cpu_offload(int cpu) +{ + return rcu_nocb_cpu_toggle_offload(cpu, true /* offload */); +} EXPORT_SYMBOL_GPL(rcu_nocb_cpu_offload); #ifdef CONFIG_RCU_LAZY @@ -1279,7 +1269,7 @@ lazy_rcu_shrink_scan(struct shrinker *shrink, struct shrink_control *sc) } rcu_nocb_try_flush_bypass(rdp, jiffies); rcu_nocb_unlock_irqrestore(rdp, flags); - wake_nocb_gp(rdp, false); + wake_nocb_gp(rdp); sc->nr_to_scan -= _count; count += _count; if (sc->nr_to_scan <= 0) @@ -1557,8 +1547,11 @@ static void show_rcu_nocb_gp_state(struct rcu_data *rdp) /* Dump out nocb kthread state for the specified rcu_data structure. */ static void show_rcu_nocb_state(struct rcu_data *rdp) { - char bufw[20]; - char bufr[20]; + char bufd[22]; + char bufw[45]; + char bufr[45]; + char bufn[22]; + char bufb[22]; struct rcu_data *nocb_next_rdp; struct rcu_segcblist *rsclp = &rdp->cblist; bool waslocked; @@ -1567,14 +1560,21 @@ static void show_rcu_nocb_state(struct rcu_data *rdp) if (rdp->nocb_gp_rdp == rdp) show_rcu_nocb_gp_state(rdp); + if (!rcu_segcblist_is_offloaded(&rdp->cblist)) + return; + nocb_next_rdp = list_next_or_null_rcu(&rdp->nocb_gp_rdp->nocb_head_rdp, &rdp->nocb_entry_rdp, typeof(*rdp), nocb_entry_rdp); - sprintf(bufw, "%ld", rsclp->gp_seq[RCU_WAIT_TAIL]); - sprintf(bufr, "%ld", rsclp->gp_seq[RCU_NEXT_READY_TAIL]); - pr_info(" CB %d^%d->%d %c%c%c%c%c F%ld L%ld C%d %c%c%s%c%s%c%c q%ld %c CPU %d%s\n", + sprintf(bufd, "%ld", rsclp->seglen[RCU_DONE_TAIL]); + sprintf(bufw, "%ld(%ld)", rsclp->seglen[RCU_WAIT_TAIL], rsclp->gp_seq[RCU_WAIT_TAIL]); + sprintf(bufr, "%ld(%ld)", rsclp->seglen[RCU_NEXT_READY_TAIL], + rsclp->gp_seq[RCU_NEXT_READY_TAIL]); + sprintf(bufn, "%ld", rsclp->seglen[RCU_NEXT_TAIL]); + sprintf(bufb, "%ld", rcu_cblist_n_cbs(&rdp->nocb_bypass)); + pr_info(" CB %d^%d->%d %c%c%c%c%c F%ld L%ld C%d %c%s%c%s%c%s%c%s%c%s q%ld %c CPU %d%s\n", rdp->cpu, rdp->nocb_gp_rdp->cpu, nocb_next_rdp ? nocb_next_rdp->cpu : -1, "kK"[!!rdp->nocb_cb_kthread], @@ -1586,12 +1586,15 @@ static void show_rcu_nocb_state(struct rcu_data *rdp) jiffies - rdp->nocb_nobypass_last, rdp->nocb_nobypass_count, ".D"[rcu_segcblist_ready_cbs(rsclp)], + rcu_segcblist_segempty(rsclp, RCU_DONE_TAIL) ? "" : bufd, ".W"[!rcu_segcblist_segempty(rsclp, RCU_WAIT_TAIL)], rcu_segcblist_segempty(rsclp, RCU_WAIT_TAIL) ? "" : bufw, ".R"[!rcu_segcblist_segempty(rsclp, RCU_NEXT_READY_TAIL)], rcu_segcblist_segempty(rsclp, RCU_NEXT_READY_TAIL) ? "" : bufr, ".N"[!rcu_segcblist_segempty(rsclp, RCU_NEXT_TAIL)], + rcu_segcblist_segempty(rsclp, RCU_NEXT_TAIL) ? "" : bufn, ".B"[!!rcu_cblist_n_cbs(&rdp->nocb_bypass)], + !rcu_cblist_n_cbs(&rdp->nocb_bypass) ? "" : bufb, rcu_segcblist_n_cbs(&rdp->cblist), rdp->nocb_cb_kthread ? task_state_to_char(rdp->nocb_cb_kthread) : '.', rdp->nocb_cb_kthread ? (int)task_cpu(rdp->nocb_cb_kthread) : -1, @@ -1651,7 +1654,7 @@ static void rcu_init_one_nocb(struct rcu_node *rnp) { } -static bool wake_nocb_gp(struct rcu_data *rdp, bool force) +static bool wake_nocb_gp(struct rcu_data *rdp) { return false; } diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 3600152b858e..95ad967adcf3 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -29,7 +29,7 @@ static bool rcu_rdp_is_offloaded(struct rcu_data *rdp) (IS_ENABLED(CONFIG_HOTPLUG_CPU) && lockdep_is_cpus_held()) || lockdep_is_held(&rdp->nocb_lock) || lockdep_is_held(&rcu_state.nocb_mutex) || - (!(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preemptible()) && + ((!(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preemptible()) || softirq_count()) && rdp == this_cpu_ptr(&rcu_data)) || rcu_current_is_nocb_kthread(rdp)), "Unsafe read of RCU_NOCB offloaded state" @@ -486,13 +486,16 @@ rcu_preempt_deferred_qs_irqrestore(struct task_struct *t, unsigned long flags) struct rcu_node *rnp; union rcu_special special; + rdp = this_cpu_ptr(&rcu_data); + if (rdp->defer_qs_pending == DEFER_QS_PENDING) + rdp->defer_qs_pending = DEFER_QS_IDLE; + /* * If RCU core is waiting for this CPU to exit its critical section, * report the fact that it has exited. Because irqs are disabled, * t->rcu_read_unlock_special cannot change. */ special = t->rcu_read_unlock_special; - rdp = this_cpu_ptr(&rcu_data); if (!special.s && !rdp->cpu_no_qs.b.exp) { local_irq_restore(flags); return; @@ -534,7 +537,6 @@ rcu_preempt_deferred_qs_irqrestore(struct task_struct *t, unsigned long flags) WARN_ON_ONCE(rnp->completedqs == rnp->gp_seq && (!empty_norm || rnp->qsmask)); empty_exp = sync_rcu_exp_done(rnp); - smp_mb(); /* ensure expedited fastpath sees end of RCU c-s. */ np = rcu_next_node_entry(t, rnp); list_del_init(&t->rcu_node_entry); t->rcu_blocked_node = NULL; @@ -626,8 +628,93 @@ static void rcu_preempt_deferred_qs_handler(struct irq_work *iwp) { struct rcu_data *rdp; + lockdep_assert_irqs_disabled(); rdp = container_of(iwp, struct rcu_data, defer_qs_iw); - rdp->defer_qs_iw_pending = false; + + /* + * If the IRQ work handler happens to run in the middle of RCU read-side + * critical section, it could be ineffective in getting the scheduler's + * attention to report a deferred quiescent state (the whole point of the + * IRQ work). For this reason, requeue the IRQ work. + * + * Basically, we want to avoid following situation: + * 1. rcu_read_unlock() queues IRQ work (state -> DEFER_QS_PENDING) + * 2. CPU enters new rcu_read_lock() + * 3. IRQ work runs but cannot report QS due to rcu_preempt_depth() > 0 + * 4. rcu_read_unlock() does not re-queue work (state still PENDING) + * 5. Deferred QS reporting does not happen. + */ + if (rcu_preempt_depth() > 0) + WRITE_ONCE(rdp->defer_qs_pending, DEFER_QS_IDLE); +} + +/* + * Check if expedited grace period processing during unlock is needed. + * + * This function determines whether expedited handling is required based on: + * 1. Task blocking an expedited grace period (based on a heuristic, could be + * false-positive, see below.) + * 2. CPU participating in an expedited grace period + * 3. Strict grace period mode requiring expedited handling + * 4. RCU priority deboosting needs when interrupts were disabled + * + * @t: The task being checked + * @rdp: The per-CPU RCU data + * @rnp: The RCU node for this CPU + * @irqs_were_disabled: Whether interrupts were disabled before rcu_read_unlock() + * + * Returns true if expedited processing of the rcu_read_unlock() is needed. + */ +static bool rcu_unlock_needs_exp_handling(struct task_struct *t, + struct rcu_data *rdp, + struct rcu_node *rnp, + bool irqs_were_disabled) +{ + /* + * Check if this task is blocking an expedited grace period. If the + * task was preempted within an RCU read-side critical section and is + * on the expedited grace period blockers list (exp_tasks), we need + * expedited handling to unblock the expedited GP. This is not an exact + * check because 't' might not be on the exp_tasks list at all - its + * just a fast heuristic that can be false-positive sometimes. + */ + if (t->rcu_blocked_node && READ_ONCE(t->rcu_blocked_node->exp_tasks)) + return true; + + /* + * Check if this CPU is participating in an expedited grace period. + * The expmask bitmap tracks which CPUs need to check in for the + * current expedited GP. If our CPU's bit is set, we need expedited + * handling to help complete the expedited GP. + */ + if (rdp->grpmask & READ_ONCE(rnp->expmask)) + return true; + + /* + * In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, all grace periods + * are treated as short for testing purposes even if that means + * disturbing the system more. Check if either: + * - This CPU has not yet reported a quiescent state, or + * - This task was preempted within an RCU critical section + * In either case, require expedited handling for strict GP mode. + */ + if (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) && + ((rdp->grpmask & READ_ONCE(rnp->qsmask)) || t->rcu_blocked_node)) + return true; + + /* + * RCU priority boosting case: If a task is subject to RCU priority + * boosting and exits an RCU read-side critical section with interrupts + * disabled, we need expedited handling to ensure timely deboosting. + * Without this, a low-priority task could incorrectly run at high + * real-time priority for an extended period degrading real-time + * responsiveness. This applies to all CONFIG_RCU_BOOST=y kernels, + * not just to PREEMPT_RT. + */ + if (IS_ENABLED(CONFIG_RCU_BOOST) && irqs_were_disabled && t->rcu_blocked_node) + return true; + + return false; } /* @@ -649,41 +736,33 @@ static void rcu_read_unlock_special(struct task_struct *t) local_irq_save(flags); irqs_were_disabled = irqs_disabled_flags(flags); if (preempt_bh_were_disabled || irqs_were_disabled) { - bool expboost; // Expedited GP in flight or possible boosting. + bool needs_exp; // Expedited handling needed. struct rcu_data *rdp = this_cpu_ptr(&rcu_data); struct rcu_node *rnp = rdp->mynode; - expboost = (t->rcu_blocked_node && READ_ONCE(t->rcu_blocked_node->exp_tasks)) || - (rdp->grpmask & READ_ONCE(rnp->expmask)) || - (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) && - ((rdp->grpmask & READ_ONCE(rnp->qsmask)) || t->rcu_blocked_node)) || - (IS_ENABLED(CONFIG_RCU_BOOST) && irqs_were_disabled && - t->rcu_blocked_node); + needs_exp = rcu_unlock_needs_exp_handling(t, rdp, rnp, irqs_were_disabled); + // Need to defer quiescent state until everything is enabled. - if (use_softirq && (in_hardirq() || (expboost && !irqs_were_disabled))) { + if (use_softirq && (in_hardirq() || (needs_exp && !irqs_were_disabled))) { // Using softirq, safe to awaken, and either the // wakeup is free or there is either an expedited // GP in flight or a potential need to deboost. - raise_softirq_irqoff(RCU_SOFTIRQ); + if (rdp->defer_qs_pending != DEFER_QS_PENDING) { + rdp->defer_qs_pending = DEFER_QS_PENDING; + raise_softirq_irqoff(RCU_SOFTIRQ); + } } else { // Enabling BH or preempt does reschedule, so... // Also if no expediting and no possible deboosting, // slow is OK. Plus nohz_full CPUs eventually get // tick enabled. - set_tsk_need_resched(current); - set_preempt_need_resched(); + set_need_resched_current(); if (IS_ENABLED(CONFIG_IRQ_WORK) && irqs_were_disabled && - expboost && !rdp->defer_qs_iw_pending && cpu_online(rdp->cpu)) { + needs_exp && rdp->defer_qs_pending != DEFER_QS_PENDING && + cpu_online(rdp->cpu)) { // Get scheduler to re-evaluate and call hooks. // If !IRQ_WORK, FQS scan will eventually IPI. - if (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) && - IS_ENABLED(CONFIG_PREEMPT_RT)) - rdp->defer_qs_iw = IRQ_WORK_INIT_HARD( - rcu_preempt_deferred_qs_handler); - else - init_irq_work(&rdp->defer_qs_iw, - rcu_preempt_deferred_qs_handler); - rdp->defer_qs_iw_pending = true; + rdp->defer_qs_pending = DEFER_QS_PENDING; irq_work_queue_on(&rdp->defer_qs_iw, rdp->cpu); } } @@ -736,10 +815,8 @@ static void rcu_flavor_sched_clock_irq(int user) if (rcu_preempt_depth() > 0 || (preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK))) { /* No QS, force context switch if deferred. */ - if (rcu_preempt_need_deferred_qs(t)) { - set_tsk_need_resched(t); - set_preempt_need_resched(); - } + if (rcu_preempt_need_deferred_qs(t)) + set_need_resched_current(); } else if (rcu_preempt_need_deferred_qs(t)) { rcu_preempt_deferred_qs(t); /* Report deferred QS. */ return; @@ -822,6 +899,10 @@ dump_blkd_tasks(struct rcu_node *rnp, int ncheck) } } +static void rcu_preempt_deferred_qs_init(struct rcu_data *rdp) +{ + rdp->defer_qs_iw = IRQ_WORK_INIT_HARD(rcu_preempt_deferred_qs_handler); +} #else /* #ifdef CONFIG_PREEMPT_RCU */ /* @@ -833,8 +914,17 @@ void rcu_read_unlock_strict(void) { struct rcu_data *rdp; - if (irqs_disabled() || preempt_count() || !rcu_state.gp_kthread) + if (irqs_disabled() || in_atomic_preempt_off() || !rcu_state.gp_kthread) return; + + /* + * rcu_report_qs_rdp() can only be invoked with a stable rdp and + * from the local CPU. + * + * The in_atomic_preempt_off() check ensures that we come here holding + * the last preempt_count (which will get dropped once we return to + * __rcu_read_unlock(). + */ rdp = this_cpu_ptr(&rcu_data); rdp->cpu_no_qs.b.norm = false; rcu_report_qs_rdp(rdp); @@ -975,13 +1065,16 @@ static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp) */ static void rcu_flavor_sched_clock_irq(int user) { - if (user || rcu_is_cpu_rrupt_from_idle()) { + if (user || rcu_is_cpu_rrupt_from_idle() || + (IS_ENABLED(CONFIG_PREEMPT_COUNT) && + (preempt_count() == HARDIRQ_OFFSET))) { /* * Get here if this CPU took its interrupt from user - * mode or from the idle loop, and if this is not a - * nested interrupt. In this case, the CPU is in - * a quiescent state, so note it. + * mode, from the idle loop without this being a nested + * interrupt, or while not holding the task preempt count + * (with PREEMPT_COUNT=y). In this case, the CPU is in a + * quiescent state, so note it. * * No memory barrier is required here because rcu_qs() * references only CPU-local variables that other CPUs @@ -1009,6 +1102,8 @@ dump_blkd_tasks(struct rcu_node *rnp, int ncheck) WARN_ON_ONCE(!list_empty(&rnp->blkd_tasks)); } +static void rcu_preempt_deferred_qs_init(struct rcu_data *rdp) { } + #endif /* #else #ifdef CONFIG_PREEMPT_RCU */ /* diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h index 925fcdad5dea..b67532cb8770 100644 --- a/kernel/rcu/tree_stall.h +++ b/kernel/rcu/tree_stall.h @@ -17,8 +17,59 @@ // Controlling CPU stall warnings, including delay calculation. /* panic() on RCU Stall sysctl. */ -int sysctl_panic_on_rcu_stall __read_mostly; -int sysctl_max_rcu_stall_to_panic __read_mostly; +static int sysctl_panic_on_rcu_stall __read_mostly; +static int sysctl_max_rcu_stall_to_panic __read_mostly; + +static const struct ctl_table rcu_stall_sysctl_table[] = { + { + .procname = "panic_on_rcu_stall", + .data = &sysctl_panic_on_rcu_stall, + .maxlen = sizeof(sysctl_panic_on_rcu_stall), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, + { + .procname = "max_rcu_stall_to_panic", + .data = &sysctl_max_rcu_stall_to_panic, + .maxlen = sizeof(sysctl_max_rcu_stall_to_panic), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ONE, + .extra2 = SYSCTL_INT_MAX, + }, +}; + +static int __init init_rcu_stall_sysctl(void) +{ + register_sysctl_init("kernel", rcu_stall_sysctl_table); + return 0; +} + +subsys_initcall(init_rcu_stall_sysctl); + +#ifdef CONFIG_SYSFS + +static unsigned int rcu_stall_count; + +static ssize_t rcu_stall_count_show(struct kobject *kobj, struct kobj_attribute *attr, + char *page) +{ + return sysfs_emit(page, "%u\n", rcu_stall_count); +} + +static struct kobj_attribute rcu_stall_count_attr = __ATTR_RO(rcu_stall_count); + +static __init int kernel_rcu_stall_sysfs_init(void) +{ + sysfs_add_file_to_group(kernel_kobj, &rcu_stall_count_attr.attr, NULL); + return 0; +} + +late_initcall(kernel_rcu_stall_sysfs_init); + +#endif // CONFIG_SYSFS #ifdef CONFIG_PROVE_RCU #define RCU_STALL_DELAY_DELTA (5 * HZ) @@ -112,6 +163,13 @@ static void panic_on_rcu_stall(void) { static int cpu_stall; + /* + * Attempt to kick out the BPF scheduler if it's installed and defer + * the panic to give the system a chance to recover. + */ + if (scx_rcu_cpu_stall()) + return; + if (++cpu_stall < sysctl_max_rcu_stall_to_panic) return; @@ -435,8 +493,8 @@ static void print_cpu_stat_info(int cpu) rsr.cputime_system = kcpustat_field(kcsp, CPUTIME_SYSTEM, cpu); pr_err("\t hardirqs softirqs csw/system\n"); - pr_err("\t number: %8ld %10d %12lld\n", - kstat_cpu_irqs_sum(cpu) - rsrp->nr_hardirqs, + pr_err("\t number: %8lld %10d %12lld\n", + kstat_cpu_irqs_sum(cpu) + arch_irq_stat_cpu(cpu) - rsrp->nr_hardirqs, kstat_cpu_softirqs_sum(cpu) - rsrp->nr_softirqs, nr_context_switches_cpu(cpu) - rsrp->nr_csw); pr_err("\tcputime: %8lld %10lld %12lld ==> %d(ms)\n", @@ -705,8 +763,7 @@ static void print_cpu_stall(unsigned long gp_seq, unsigned long gps) * progress and it could be we're stuck in kernel space without context * switches for an entirely unreasonable amount of time. */ - set_tsk_need_resched(current); - set_preempt_need_resched(); + set_need_resched_current(); } static bool csd_lock_suppress_rcu_stall; @@ -784,6 +841,10 @@ static void check_cpu_stall(struct rcu_data *rdp) if (kvm_check_and_clear_guest_paused()) return; +#ifdef CONFIG_SYSFS + ++rcu_stall_count; +#endif + rcu_stall_notifier_call_chain(RCU_STALL_NOTIFY_NORM, (void *)j - gps); if (READ_ONCE(csd_lock_suppress_rcu_stall) && csd_lock_is_stuck()) { pr_err("INFO: %s detected stall, but suppressed full report due to a stuck CSD-lock.\n", rcu_state.name); @@ -927,8 +988,7 @@ void show_rcu_gp_kthreads(void) for_each_possible_cpu(cpu) { rdp = per_cpu_ptr(&rcu_data, cpu); cbs += data_race(READ_ONCE(rdp->n_cbs_invoked)); - if (rcu_segcblist_is_offloaded(&rdp->cblist)) - show_rcu_nocb_state(rdp); + show_rcu_nocb_state(rdp); } pr_info("RCU callbacks invoked since boot: %lu\n", cbs); show_rcu_tasks_gp_kthreads(); diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c index c912b594ba98..b62735a67884 100644 --- a/kernel/rcu/update.c +++ b/kernel/rcu/update.c @@ -117,7 +117,7 @@ static bool rcu_read_lock_held_common(bool *ret) return false; } -int rcu_read_lock_sched_held(void) +int notrace rcu_read_lock_sched_held(void) { bool ret; @@ -342,7 +342,7 @@ EXPORT_SYMBOL_GPL(debug_lockdep_rcu_enabled); * 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) +int notrace rcu_read_lock_held(void) { bool ret; @@ -367,7 +367,7 @@ EXPORT_SYMBOL_GPL(rcu_read_lock_held); * Note that rcu_read_lock_bh() is disallowed if the CPU is either idle or * offline from an RCU perspective, so check for those as well. */ -int rcu_read_lock_bh_held(void) +int notrace rcu_read_lock_bh_held(void) { bool ret; @@ -377,7 +377,7 @@ int rcu_read_lock_bh_held(void) } EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held); -int rcu_read_lock_any_held(void) +int notrace rcu_read_lock_any_held(void) { bool ret; @@ -538,6 +538,28 @@ long torture_sched_setaffinity(pid_t pid, const struct cpumask *in_mask, bool do EXPORT_SYMBOL_GPL(torture_sched_setaffinity); #endif +#if IS_ENABLED(CONFIG_TRIVIAL_PREEMPT_RCU) +// Trivial and stupid grace-period wait. Defined here so that lockdep +// kernels can find tasklist_lock. +void synchronize_rcu_trivial_preempt(void) +{ + struct task_struct *g; + struct task_struct *t; + + smp_mb(); // Order prior accesses before grace-period start. + rcu_read_lock(); // Protect task list. + for_each_process_thread(g, t) { + if (t == current) + continue; // Don't deadlock on ourselves! + // Order later rcu_read_lock() on other tasks after QS. + while (smp_load_acquire(&t->rcu_trivial_preempt_nesting)) + continue; + } + rcu_read_unlock(); +} +EXPORT_SYMBOL_GPL(synchronize_rcu_trivial_preempt); +#endif // #if IS_ENABLED(CONFIG_TRIVIAL_PREEMPT_RCU) + int rcu_cpu_stall_notifiers __read_mostly; // !0 = provide stall notifiers (rarely useful) EXPORT_SYMBOL_GPL(rcu_cpu_stall_notifiers); @@ -614,7 +636,7 @@ static void early_boot_test_call_rcu(void) call_rcu(&head, test_callback); early_srcu_cookie = start_poll_synchronize_srcu(&early_srcu); call_srcu(&early_srcu, &shead, test_callback); - rhp = kmalloc(sizeof(*rhp), GFP_KERNEL); + rhp = kmalloc_obj(*rhp); if (!WARN_ON_ONCE(!rhp)) kfree_rcu(rhp, rh); } |
