From d65d411c9259a2499081e1e7ed91088232666b57 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 25 Jul 2023 12:08:50 +0100 Subject: treewide: context_tracking: Rename CONTEXT_* into CT_STATE_* Context tracking state related symbols currently use a mix of the CONTEXT_ (e.g. CONTEXT_KERNEL) and CT_SATE_ (e.g. CT_STATE_MASK) prefixes. Clean up the naming and make the ctx_state enum use the CT_STATE_ prefix. Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Acked-by: Frederic Weisbecker Acked-by: Thomas Gleixner Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking.h | 16 ++++++++-------- include/linux/context_tracking_state.h | 20 ++++++++++---------- include/linux/entry-common.h | 2 +- 3 files changed, 19 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking.h b/include/linux/context_tracking.h index 6e76b9dba00e..28fcfa184903 100644 --- a/include/linux/context_tracking.h +++ b/include/linux/context_tracking.h @@ -26,26 +26,26 @@ extern void user_exit_callable(void); static inline void user_enter(void) { if (context_tracking_enabled()) - ct_user_enter(CONTEXT_USER); + ct_user_enter(CT_STATE_USER); } static inline void user_exit(void) { if (context_tracking_enabled()) - ct_user_exit(CONTEXT_USER); + ct_user_exit(CT_STATE_USER); } /* Called with interrupts disabled. */ static __always_inline void user_enter_irqoff(void) { if (context_tracking_enabled()) - __ct_user_enter(CONTEXT_USER); + __ct_user_enter(CT_STATE_USER); } static __always_inline void user_exit_irqoff(void) { if (context_tracking_enabled()) - __ct_user_exit(CONTEXT_USER); + __ct_user_exit(CT_STATE_USER); } static inline enum ctx_state exception_enter(void) @@ -57,7 +57,7 @@ static inline enum ctx_state exception_enter(void) return 0; prev_ctx = __ct_state(); - if (prev_ctx != CONTEXT_KERNEL) + if (prev_ctx != CT_STATE_KERNEL) ct_user_exit(prev_ctx); return prev_ctx; @@ -67,7 +67,7 @@ static inline void exception_exit(enum ctx_state prev_ctx) { if (!IS_ENABLED(CONFIG_HAVE_CONTEXT_TRACKING_USER_OFFSTACK) && context_tracking_enabled()) { - if (prev_ctx != CONTEXT_KERNEL) + if (prev_ctx != CT_STATE_KERNEL) ct_user_enter(prev_ctx); } } @@ -75,7 +75,7 @@ static inline void exception_exit(enum ctx_state prev_ctx) static __always_inline bool context_tracking_guest_enter(void) { if (context_tracking_enabled()) - __ct_user_enter(CONTEXT_GUEST); + __ct_user_enter(CT_STATE_GUEST); return context_tracking_enabled_this_cpu(); } @@ -83,7 +83,7 @@ static __always_inline bool context_tracking_guest_enter(void) static __always_inline void context_tracking_guest_exit(void) { if (context_tracking_enabled()) - __ct_user_exit(CONTEXT_GUEST); + __ct_user_exit(CT_STATE_GUEST); } #define CT_WARN_ON(cond) WARN_ON(context_tracking_enabled() && (cond)) diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index bbff5f7f8803..f1c53125edee 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -10,18 +10,18 @@ #define DYNTICK_IRQ_NONIDLE ((LONG_MAX / 2) + 1) enum ctx_state { - CONTEXT_DISABLED = -1, /* returned by ct_state() if unknown */ - CONTEXT_KERNEL = 0, - CONTEXT_IDLE = 1, - CONTEXT_USER = 2, - CONTEXT_GUEST = 3, - CONTEXT_MAX = 4, + CT_STATE_DISABLED = -1, /* returned by ct_state() if unknown */ + CT_STATE_KERNEL = 0, + CT_STATE_IDLE = 1, + CT_STATE_USER = 2, + CT_STATE_GUEST = 3, + CT_STATE_MAX = 4, }; /* Even value for idle, else odd. */ -#define RCU_DYNTICKS_IDX CONTEXT_MAX +#define RCU_DYNTICKS_IDX CT_STATE_MAX -#define CT_STATE_MASK (CONTEXT_MAX - 1) +#define CT_STATE_MASK (CT_STATE_MAX - 1) #define CT_DYNTICKS_MASK (~CT_STATE_MASK) struct context_tracking { @@ -123,14 +123,14 @@ static inline bool context_tracking_enabled_this_cpu(void) * * Returns the current cpu's context tracking state if context tracking * is enabled. If context tracking is disabled, returns - * CONTEXT_DISABLED. This should be used primarily for debugging. + * CT_STATE_DISABLED. This should be used primarily for debugging. */ static __always_inline int ct_state(void) { int ret; if (!context_tracking_enabled()) - return CONTEXT_DISABLED; + return CT_STATE_DISABLED; preempt_disable(); ret = __ct_state(); diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h index b0fb775a600d..1e50cdb83ae5 100644 --- a/include/linux/entry-common.h +++ b/include/linux/entry-common.h @@ -108,7 +108,7 @@ static __always_inline void enter_from_user_mode(struct pt_regs *regs) arch_enter_from_user_mode(regs); lockdep_hardirqs_off(CALLER_ADDR0); - CT_WARN_ON(__ct_state() != CONTEXT_USER); + CT_WARN_ON(__ct_state() != CT_STATE_USER); user_exit_irqoff(); instrumentation_begin(); -- cgit From 4aa35e0db6d758e8f952ed30d7ac3117c376562c Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 25 Jul 2023 12:19:01 +0100 Subject: context_tracking, rcu: Rename RCU_DYNTICKS_IDX into CT_RCU_WATCHING The symbols relating to the CT_STATE part of context_tracking.state are now all prefixed with CT_STATE. The RCU dynticks counter part of that atomic variable still involves symbols with different prefixes, align them all to be prefixed with CT_RCU_WATCHING. Suggested-by: "Paul E. McKenney" Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Acked-by: Thomas Gleixner Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking.h | 6 +++--- include/linux/context_tracking_state.h | 12 ++++++------ kernel/context_tracking.c | 22 +++++++++++----------- kernel/rcu/tree.c | 12 ++++++------ 4 files changed, 26 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking.h b/include/linux/context_tracking.h index 28fcfa184903..a6c36780cc3b 100644 --- a/include/linux/context_tracking.h +++ b/include/linux/context_tracking.h @@ -119,7 +119,7 @@ extern void ct_idle_exit(void); */ static __always_inline bool rcu_dynticks_curr_cpu_in_eqs(void) { - return !(raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & RCU_DYNTICKS_IDX); + return !(raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING); } /* @@ -142,7 +142,7 @@ static __always_inline bool warn_rcu_enter(void) preempt_disable_notrace(); if (rcu_dynticks_curr_cpu_in_eqs()) { ret = true; - ct_state_inc(RCU_DYNTICKS_IDX); + ct_state_inc(CT_RCU_WATCHING); } return ret; @@ -151,7 +151,7 @@ static __always_inline bool warn_rcu_enter(void) static __always_inline void warn_rcu_exit(bool rcu) { if (rcu) - ct_state_inc(RCU_DYNTICKS_IDX); + ct_state_inc(CT_RCU_WATCHING); preempt_enable_notrace(); } diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index f1c53125edee..94d6a935af3b 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -18,11 +18,11 @@ enum ctx_state { CT_STATE_MAX = 4, }; -/* Even value for idle, else odd. */ -#define RCU_DYNTICKS_IDX CT_STATE_MAX +/* Odd value for watching, else even. */ +#define CT_RCU_WATCHING CT_STATE_MAX #define CT_STATE_MASK (CT_STATE_MAX - 1) -#define CT_DYNTICKS_MASK (~CT_STATE_MASK) +#define CT_RCU_WATCHING_MASK (~CT_STATE_MASK) struct context_tracking { #ifdef CONFIG_CONTEXT_TRACKING_USER @@ -58,21 +58,21 @@ static __always_inline int __ct_state(void) #ifdef CONFIG_CONTEXT_TRACKING_IDLE static __always_inline int ct_dynticks(void) { - return atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_DYNTICKS_MASK; + return atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING_MASK; } static __always_inline int ct_dynticks_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); - return atomic_read(&ct->state) & CT_DYNTICKS_MASK; + return atomic_read(&ct->state) & CT_RCU_WATCHING_MASK; } static __always_inline int ct_dynticks_cpu_acquire(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); - return atomic_read_acquire(&ct->state) & CT_DYNTICKS_MASK; + return atomic_read_acquire(&ct->state) & CT_RCU_WATCHING_MASK; } static __always_inline long ct_dynticks_nesting(void) diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 4bb5751af994..b2589bc59e18 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -31,7 +31,7 @@ DEFINE_PER_CPU(struct context_tracking, context_tracking) = { .dynticks_nesting = 1, .dynticks_nmi_nesting = DYNTICK_IRQ_NONIDLE, #endif - .state = ATOMIC_INIT(RCU_DYNTICKS_IDX), + .state = ATOMIC_INIT(CT_RCU_WATCHING), }; EXPORT_SYMBOL_GPL(context_tracking); @@ -90,7 +90,7 @@ static noinstr void ct_kernel_exit_state(int offset) rcu_dynticks_task_trace_enter(); // Before ->dynticks update! seq = ct_state_inc(offset); // RCU is no longer watching. Better be in extended quiescent state! - WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && (seq & RCU_DYNTICKS_IDX)); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && (seq & CT_RCU_WATCHING)); } /* @@ -110,7 +110,7 @@ static noinstr void ct_kernel_enter_state(int offset) seq = ct_state_inc(offset); // RCU is now watching. Better not be in an extended quiescent state! rcu_dynticks_task_trace_exit(); // After ->dynticks update! - WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !(seq & RCU_DYNTICKS_IDX)); + WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !(seq & CT_RCU_WATCHING)); } /* @@ -236,7 +236,7 @@ void noinstr ct_nmi_exit(void) instrumentation_end(); // RCU is watching here ... - ct_kernel_exit_state(RCU_DYNTICKS_IDX); + ct_kernel_exit_state(CT_RCU_WATCHING); // ... but is no longer watching here. if (!in_nmi()) @@ -277,7 +277,7 @@ void noinstr ct_nmi_enter(void) rcu_dynticks_task_exit(); // RCU is not watching here ... - ct_kernel_enter_state(RCU_DYNTICKS_IDX); + ct_kernel_enter_state(CT_RCU_WATCHING); // ... but is watching here. instrumentation_begin(); @@ -317,7 +317,7 @@ void noinstr ct_nmi_enter(void) void noinstr ct_idle_enter(void) { WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !raw_irqs_disabled()); - ct_kernel_exit(false, RCU_DYNTICKS_IDX + CT_STATE_IDLE); + ct_kernel_exit(false, CT_RCU_WATCHING + CT_STATE_IDLE); } EXPORT_SYMBOL_GPL(ct_idle_enter); @@ -335,7 +335,7 @@ void noinstr ct_idle_exit(void) unsigned long flags; raw_local_irq_save(flags); - ct_kernel_enter(false, RCU_DYNTICKS_IDX - CT_STATE_IDLE); + ct_kernel_enter(false, CT_RCU_WATCHING - CT_STATE_IDLE); raw_local_irq_restore(flags); } EXPORT_SYMBOL_GPL(ct_idle_exit); @@ -504,7 +504,7 @@ void noinstr __ct_user_enter(enum ctx_state state) * CPU doesn't need to maintain the tick for RCU maintenance purposes * when the CPU runs in userspace. */ - ct_kernel_exit(true, RCU_DYNTICKS_IDX + state); + ct_kernel_exit(true, CT_RCU_WATCHING + state); /* * Special case if we only track user <-> kernel transitions for tickless @@ -534,7 +534,7 @@ void noinstr __ct_user_enter(enum ctx_state state) /* * Tracking for vtime and RCU EQS. Make sure we don't race * with NMIs. OTOH we don't care about ordering here since - * RCU only requires RCU_DYNTICKS_IDX increments to be fully + * RCU only requires CT_RCU_WATCHING increments to be fully * ordered. */ raw_atomic_add(state, &ct->state); @@ -620,7 +620,7 @@ void noinstr __ct_user_exit(enum ctx_state state) * Exit RCU idle mode while entering the kernel because it can * run a RCU read side critical section anytime. */ - ct_kernel_enter(true, RCU_DYNTICKS_IDX - state); + ct_kernel_enter(true, CT_RCU_WATCHING - state); if (state == CT_STATE_USER) { instrumentation_begin(); vtime_user_exit(current); @@ -644,7 +644,7 @@ void noinstr __ct_user_exit(enum ctx_state state) /* * Tracking for vtime and RCU EQS. Make sure we don't race * with NMIs. OTOH we don't care about ordering here since - * RCU only requires RCU_DYNTICKS_IDX increments to be fully + * RCU only requires CT_RCU_WATCHING increments to be fully * ordered. */ raw_atomic_sub(state, &ct->state); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index e641cc681901..04f87c44385c 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -294,9 +294,9 @@ void rcu_softirq_qs(void) */ static void rcu_dynticks_eqs_online(void) { - if (ct_dynticks() & RCU_DYNTICKS_IDX) + if (ct_dynticks() & CT_RCU_WATCHING) return; - ct_state_inc(RCU_DYNTICKS_IDX); + ct_state_inc(CT_RCU_WATCHING); } /* @@ -305,7 +305,7 @@ static void rcu_dynticks_eqs_online(void) */ static bool rcu_dynticks_in_eqs(int snap) { - return !(snap & RCU_DYNTICKS_IDX); + return !(snap & CT_RCU_WATCHING); } /* @@ -335,7 +335,7 @@ bool rcu_dynticks_zero_in_eqs(int cpu, int *vp) int snap; // If not quiescent, force back to earlier extended quiescent state. - snap = ct_dynticks_cpu(cpu) & ~RCU_DYNTICKS_IDX; + snap = ct_dynticks_cpu(cpu) & ~CT_RCU_WATCHING; smp_rmb(); // Order ->dynticks and *vp reads. if (READ_ONCE(*vp)) return false; // Non-zero, so report failure; @@ -361,9 +361,9 @@ notrace void rcu_momentary_dyntick_idle(void) int seq; raw_cpu_write(rcu_data.rcu_need_heavy_qs, false); - seq = ct_state_inc(2 * RCU_DYNTICKS_IDX); + seq = ct_state_inc(2 * CT_RCU_WATCHING); /* It is illegal to call this from idle state. */ - WARN_ON_ONCE(!(seq & RCU_DYNTICKS_IDX)); + WARN_ON_ONCE(!(seq & CT_RCU_WATCHING)); rcu_preempt_deferred_qs(current); } EXPORT_SYMBOL_GPL(rcu_momentary_dyntick_idle); -- cgit From a4a7921ec08d016f9d692752e2f82f66369c7ffa Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 10:45:56 +0200 Subject: context_tracking, rcu: Rename ct_dynticks() into ct_rcu_watching() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/context_tracking.c | 10 +++++----- kernel/rcu/tree.c | 2 +- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 94d6a935af3b..cb90d8c17810 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -56,7 +56,7 @@ static __always_inline int __ct_state(void) #endif #ifdef CONFIG_CONTEXT_TRACKING_IDLE -static __always_inline int ct_dynticks(void) +static __always_inline int ct_rcu_watching(void) { return atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING_MASK; } diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index b2589bc59e18..868ae0bcd4be 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -137,7 +137,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) instrumentation_begin(); lockdep_assert_irqs_disabled(); - trace_rcu_dyntick(TPS("Start"), ct_dynticks_nesting(), 0, ct_dynticks()); + trace_rcu_dyntick(TPS("Start"), ct_dynticks_nesting(), 0, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); rcu_preempt_deferred_qs(current); @@ -182,7 +182,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) // instrumentation for the noinstr ct_kernel_enter_state() instrument_atomic_write(&ct->state, sizeof(ct->state)); - trace_rcu_dyntick(TPS("End"), ct_dynticks_nesting(), 1, ct_dynticks()); + trace_rcu_dyntick(TPS("End"), ct_dynticks_nesting(), 1, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); WRITE_ONCE(ct->dynticks_nesting, 1); WARN_ON_ONCE(ct_dynticks_nmi_nesting()); @@ -220,7 +220,7 @@ void noinstr ct_nmi_exit(void) */ if (ct_dynticks_nmi_nesting() != 1) { trace_rcu_dyntick(TPS("--="), ct_dynticks_nmi_nesting(), ct_dynticks_nmi_nesting() - 2, - ct_dynticks()); + ct_rcu_watching()); WRITE_ONCE(ct->dynticks_nmi_nesting, /* No store tearing. */ ct_dynticks_nmi_nesting() - 2); instrumentation_end(); @@ -228,7 +228,7 @@ void noinstr ct_nmi_exit(void) } /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */ - trace_rcu_dyntick(TPS("Startirq"), ct_dynticks_nmi_nesting(), 0, ct_dynticks()); + trace_rcu_dyntick(TPS("Startirq"), ct_dynticks_nmi_nesting(), 0, ct_rcu_watching()); WRITE_ONCE(ct->dynticks_nmi_nesting, 0); /* Avoid store tearing. */ // instrumentation for the noinstr ct_kernel_exit_state() @@ -296,7 +296,7 @@ void noinstr ct_nmi_enter(void) trace_rcu_dyntick(incby == 1 ? TPS("Endirq") : TPS("++="), ct_dynticks_nmi_nesting(), - ct_dynticks_nmi_nesting() + incby, ct_dynticks()); + ct_dynticks_nmi_nesting() + incby, ct_rcu_watching()); instrumentation_end(); WRITE_ONCE(ct->dynticks_nmi_nesting, /* Prevent store tearing. */ ct_dynticks_nmi_nesting() + incby); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 04f87c44385c..3a2f0325aa43 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -294,7 +294,7 @@ void rcu_softirq_qs(void) */ static void rcu_dynticks_eqs_online(void) { - if (ct_dynticks() & CT_RCU_WATCHING) + if (ct_rcu_watching() & CT_RCU_WATCHING) return; ct_state_inc(CT_RCU_WATCHING); } -- cgit From a9fde9d1a5dd42edadf61cf4e64aa234b4c5bd3f Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 10:47:14 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_cpu() into ct_rcu_watching_cpu() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/rcu/tree.c | 10 +++++----- kernel/rcu/tree_stall.h | 4 ++-- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index cb90d8c17810..ad5a06a42b4a 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -61,7 +61,7 @@ static __always_inline int ct_rcu_watching(void) return atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING_MASK; } -static __always_inline int ct_dynticks_cpu(int cpu) +static __always_inline int ct_rcu_watching_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 3a2f0325aa43..e7f612e9f7e5 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -335,14 +335,14 @@ bool rcu_dynticks_zero_in_eqs(int cpu, int *vp) int snap; // If not quiescent, force back to earlier extended quiescent state. - snap = ct_dynticks_cpu(cpu) & ~CT_RCU_WATCHING; - smp_rmb(); // Order ->dynticks and *vp reads. + snap = ct_rcu_watching_cpu(cpu) & ~CT_RCU_WATCHING; + smp_rmb(); // Order CT state and *vp reads. if (READ_ONCE(*vp)) return false; // Non-zero, so report failure; - smp_rmb(); // Order *vp read and ->dynticks re-read. + smp_rmb(); // Order *vp read and CT state re-read. // If still in the same extended quiescent state, we are good! - return snap == ct_dynticks_cpu(cpu); + return snap == ct_rcu_watching_cpu(cpu); } /* @@ -4805,7 +4805,7 @@ rcu_boot_init_percpu_data(int cpu) rdp->grpmask = leaf_node_cpu_bit(rdp->mynode, cpu); INIT_WORK(&rdp->strict_work, strict_work_handler); WARN_ON_ONCE(ct->dynticks_nesting != 1); - WARN_ON_ONCE(rcu_dynticks_in_eqs(ct_dynticks_cpu(cpu))); + WARN_ON_ONCE(rcu_dynticks_in_eqs(ct_rcu_watching_cpu(cpu))); rdp->barrier_seq_snap = rcu_state.barrier_sequence; rdp->rcu_ofl_gp_seq = rcu_state.gp_seq; rdp->rcu_ofl_gp_state = RCU_GP_CLEANED; diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h index 4b0e9d7c4c68..d65974448e81 100644 --- a/kernel/rcu/tree_stall.h +++ b/kernel/rcu/tree_stall.h @@ -501,7 +501,7 @@ static void print_cpu_stall_info(int cpu) } delta = rcu_seq_ctr(rdp->mynode->gp_seq - rdp->rcu_iw_gp_seq); falsepositive = rcu_is_gp_kthread_starving(NULL) && - rcu_dynticks_in_eqs(ct_dynticks_cpu(cpu)); + rcu_dynticks_in_eqs(ct_rcu_watching_cpu(cpu)); rcuc_starved = rcu_is_rcuc_kthread_starving(rdp, &j); if (rcuc_starved) // Print signed value, as negative values indicate a probable bug. @@ -515,7 +515,7 @@ static void print_cpu_stall_info(int cpu) rdp->rcu_iw_pending ? (int)min(delta, 9UL) + '0' : "!."[!delta], ticks_value, ticks_title, - ct_dynticks_cpu(cpu) & 0xffff, + ct_rcu_watching_cpu(cpu) & 0xffff, ct_dynticks_nesting_cpu(cpu), ct_dynticks_nmi_nesting_cpu(cpu), rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu), data_race(rcu_state.n_force_qs) - rcu_state.n_force_qs_gpstart, -- cgit From 125716c393889f9035567d4d78da239483f63ece Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 14:47:12 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_cpu_acquire() into ct_rcu_watching_cpu_acquire() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst | 2 +- include/linux/context_tracking_state.h | 2 +- kernel/rcu/tree.c | 4 ++-- kernel/rcu/tree_exp.h | 2 +- 4 files changed, 5 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst b/Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst index 728b1e690c64..2d7036ad7476 100644 --- a/Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst +++ b/Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.rst @@ -149,7 +149,7 @@ This case is handled by calls to the strongly ordered ``atomic_add_return()`` read-modify-write atomic operation that is invoked within ``rcu_dynticks_eqs_enter()`` at idle-entry time and within ``rcu_dynticks_eqs_exit()`` at idle-exit time. -The grace-period kthread invokes first ``ct_dynticks_cpu_acquire()`` +The grace-period kthread invokes first ``ct_rcu_watching_cpu_acquire()`` (preceded by a full memory barrier) and ``rcu_dynticks_in_eqs_since()`` (both of which rely on acquire semantics) to detect idle CPUs. diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index ad5a06a42b4a..ad6570ffeff3 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -68,7 +68,7 @@ static __always_inline int ct_rcu_watching_cpu(int cpu) return atomic_read(&ct->state) & CT_RCU_WATCHING_MASK; } -static __always_inline int ct_dynticks_cpu_acquire(int cpu) +static __always_inline int ct_rcu_watching_cpu_acquire(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index e7f612e9f7e5..45a9f3667c2a 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -323,7 +323,7 @@ static bool rcu_dynticks_in_eqs_since(struct rcu_data *rdp, int snap) * performed by the remote CPU prior to entering idle and therefore can * rely solely on acquire semantics. */ - return snap != ct_dynticks_cpu_acquire(rdp->cpu); + return snap != ct_rcu_watching_cpu_acquire(rdp->cpu); } /* @@ -782,7 +782,7 @@ static int dyntick_save_progress_counter(struct rcu_data *rdp) * Ordering between remote CPU's pre idle accesses and post grace period * updater's accesses is enforced by the below acquire semantic. */ - rdp->dynticks_snap = ct_dynticks_cpu_acquire(rdp->cpu); + rdp->dynticks_snap = ct_rcu_watching_cpu_acquire(rdp->cpu); if (rcu_dynticks_in_eqs(rdp->dynticks_snap)) { trace_rcu_fqs(rcu_state.name, rdp->gp_seq, rdp->cpu, TPS("dti")); rcu_gpnum_ovf(rdp->mynode, rdp); diff --git a/kernel/rcu/tree_exp.h b/kernel/rcu/tree_exp.h index 4acd29d16fdb..daa87fec703f 100644 --- a/kernel/rcu/tree_exp.h +++ b/kernel/rcu/tree_exp.h @@ -376,7 +376,7 @@ static void __sync_rcu_exp_select_node_cpus(struct rcu_exp_work *rewp) * post grace period updater's accesses is enforced by the * below acquire semantic. */ - snap = ct_dynticks_cpu_acquire(cpu); + snap = ct_rcu_watching_cpu_acquire(cpu); if (rcu_dynticks_in_eqs(snap)) mask_ofl_test |= mask; else -- cgit From 7aeba709a048d870c15940af8b620b16281c3b9e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 30 May 2024 15:45:42 +0200 Subject: rcu/nocb: Introduce RCU_NOCB_LOCKDEP_WARN() Checking for races against concurrent (de-)offloading implies the creation of !CONFIG_RCU_NOCB_CPU stubs to check if each relevant lock is held. For now this only implies the nocb_lock but more are to be expected. Create instead a NOCB specific version of RCU_LOCKDEP_WARN() to avoid the proliferation of stubs. Signed-off-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Reviewed-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rcupdate.h | 7 +++++++ kernel/rcu/tree_nocb.h | 14 -------------- kernel/rcu/tree_plugin.h | 4 ++-- 3 files changed, 9 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 13f6f00aecf9..d48d3c237305 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -144,11 +144,18 @@ void rcu_init_nohz(void); int rcu_nocb_cpu_offload(int cpu); int rcu_nocb_cpu_deoffload(int cpu); void rcu_nocb_flush_deferred_wakeup(void); + +#define RCU_NOCB_LOCKDEP_WARN(c, s) RCU_LOCKDEP_WARN(c, s) + #else /* #ifdef CONFIG_RCU_NOCB_CPU */ + static inline void rcu_init_nohz(void) { } static inline int rcu_nocb_cpu_offload(int cpu) { return -EINVAL; } static inline int rcu_nocb_cpu_deoffload(int cpu) { return 0; } static inline void rcu_nocb_flush_deferred_wakeup(void) { } + +#define RCU_NOCB_LOCKDEP_WARN(c, s) + #endif /* #else #ifdef CONFIG_RCU_NOCB_CPU */ /* diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 3ce30841119a..f4112fc663a7 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -16,10 +16,6 @@ #ifdef CONFIG_RCU_NOCB_CPU static cpumask_var_t rcu_nocb_mask; /* CPUs to have callbacks offloaded. */ static bool __read_mostly rcu_nocb_poll; /* Offload kthread are to poll. */ -static inline int rcu_lockdep_is_held_nocb(struct rcu_data *rdp) -{ - return lockdep_is_held(&rdp->nocb_lock); -} static inline bool rcu_current_is_nocb_kthread(struct rcu_data *rdp) { @@ -1653,16 +1649,6 @@ static void show_rcu_nocb_state(struct rcu_data *rdp) #else /* #ifdef CONFIG_RCU_NOCB_CPU */ -static inline int rcu_lockdep_is_held_nocb(struct rcu_data *rdp) -{ - return 0; -} - -static inline bool rcu_current_is_nocb_kthread(struct rcu_data *rdp) -{ - return false; -} - /* No ->nocb_lock to acquire. */ static void rcu_nocb_lock(struct rcu_data *rdp) { diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index c569da65b421..f752b2a1d887 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -24,10 +24,10 @@ static bool rcu_rdp_is_offloaded(struct rcu_data *rdp) * timers have their own means of synchronization against the * offloaded state updaters. */ - RCU_LOCKDEP_WARN( + RCU_NOCB_LOCKDEP_WARN( !(lockdep_is_held(&rcu_state.barrier_mutex) || (IS_ENABLED(CONFIG_HOTPLUG_CPU) && lockdep_is_cpus_held()) || - rcu_lockdep_is_held_nocb(rdp) || + lockdep_is_held(&rdp->nocb_lock) || (!(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preemptible()) && rdp == this_cpu_ptr(&rcu_data)) || rcu_current_is_nocb_kthread(rdp)), -- cgit From bae6076ebbd14cc1f1bd4de65c2db21d3ab109d7 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 30 May 2024 15:45:50 +0200 Subject: rcu/nocb: Remove SEGCBLIST_RCU_CORE RCU core can't be running anymore while in the middle of (de-)offloading since this sort of transition now only applies to offline CPUs. The SEGCBLIST_RCU_CORE state can therefore be removed. Signed-off-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Reviewed-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rcu_segcblist.h | 9 ++++----- kernel/rcu/rcu_segcblist.h | 9 --------- kernel/rcu/tree.c | 3 --- kernel/rcu/tree_nocb.h | 9 --------- 4 files changed, 4 insertions(+), 26 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcu_segcblist.h b/include/linux/rcu_segcblist.h index ba95c06675e1..5469c54cd778 100644 --- a/include/linux/rcu_segcblist.h +++ b/include/linux/rcu_segcblist.h @@ -185,11 +185,10 @@ struct rcu_cblist { * ---------------------------------------------------------------------------- */ #define SEGCBLIST_ENABLED BIT(0) -#define SEGCBLIST_RCU_CORE BIT(1) -#define SEGCBLIST_LOCKING BIT(2) -#define SEGCBLIST_KTHREAD_CB BIT(3) -#define SEGCBLIST_KTHREAD_GP BIT(4) -#define SEGCBLIST_OFFLOADED BIT(5) +#define SEGCBLIST_LOCKING BIT(1) +#define SEGCBLIST_KTHREAD_CB BIT(2) +#define SEGCBLIST_KTHREAD_GP BIT(3) +#define SEGCBLIST_OFFLOADED BIT(4) struct rcu_segcblist { struct rcu_head *head; diff --git a/kernel/rcu/rcu_segcblist.h b/kernel/rcu/rcu_segcblist.h index 4fe877f5f654..7a0962dfee86 100644 --- a/kernel/rcu/rcu_segcblist.h +++ b/kernel/rcu/rcu_segcblist.h @@ -95,15 +95,6 @@ static inline bool rcu_segcblist_is_offloaded(struct rcu_segcblist *rsclp) return false; } -static inline bool rcu_segcblist_completely_offloaded(struct rcu_segcblist *rsclp) -{ - if (IS_ENABLED(CONFIG_RCU_NOCB_CPU) && - !rcu_segcblist_test_flags(rsclp, SEGCBLIST_RCU_CORE)) - return true; - - return false; -} - /* * Are all segments following the specified segment of the specified * rcu_segcblist structure empty of callbacks? (The specified diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 1a272c678533..82e831b969e4 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -79,9 +79,6 @@ static void rcu_sr_normal_gp_cleanup_work(struct work_struct *); static DEFINE_PER_CPU_SHARED_ALIGNED(struct rcu_data, rcu_data) = { .gpwrap = true, -#ifdef CONFIG_RCU_NOCB_CPU - .cblist.flags = SEGCBLIST_RCU_CORE, -#endif }; static struct rcu_state rcu_state = { .level = { &rcu_state.node[0] }, diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index af44e75eb0cd..24daf606de0c 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -1060,7 +1060,6 @@ static int rcu_nocb_rdp_deoffload(struct rcu_data *rdp) WARN_ON_ONCE(rcu_cblist_n_cbs(&rdp->nocb_bypass)); WARN_ON_ONCE(rcu_segcblist_n_cbs(&rdp->cblist)); - rcu_segcblist_set_flags(cblist, SEGCBLIST_RCU_CORE); wake_gp = rdp_offload_toggle(rdp, false, flags); mutex_lock(&rdp_gp->nocb_gp_kthread_mutex); @@ -1168,13 +1167,6 @@ static int rcu_nocb_rdp_offload(struct rcu_data *rdp) swait_event_exclusive(rdp->nocb_state_wq, rcu_segcblist_test_flags(cblist, SEGCBLIST_KTHREAD_GP)); - /* - * All kthreads are ready to work, we can finally enable nocb bypass. - */ - rcu_nocb_lock_irqsave(rdp, flags); - rcu_segcblist_clear_flags(cblist, SEGCBLIST_RCU_CORE); - rcu_nocb_unlock_irqrestore(rdp, flags); - return 0; } @@ -1350,7 +1342,6 @@ void __init rcu_init_nohz(void) rcu_segcblist_init(&rdp->cblist); rcu_segcblist_offload(&rdp->cblist, true); rcu_segcblist_set_flags(&rdp->cblist, SEGCBLIST_KTHREAD_GP); - rcu_segcblist_clear_flags(&rdp->cblist, SEGCBLIST_RCU_CORE); } rcu_organize_nocb_kthreads(); } -- cgit From 91e43b9044a4f9b6976dc28b3f1446b733cc68de Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 30 May 2024 15:45:51 +0200 Subject: rcu/nocb: Remove SEGCBLIST_KTHREAD_CB This state excerpt from the (de-)offloading state machine was used to implement an ad-hoc kthread parking of rcuo kthreads. This code has been removed and therefore the related state can be erased as well. Signed-off-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Reviewed-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rcu_segcblist.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcu_segcblist.h b/include/linux/rcu_segcblist.h index 5469c54cd778..1ef1bb54853d 100644 --- a/include/linux/rcu_segcblist.h +++ b/include/linux/rcu_segcblist.h @@ -186,9 +186,8 @@ struct rcu_cblist { */ #define SEGCBLIST_ENABLED BIT(0) #define SEGCBLIST_LOCKING BIT(1) -#define SEGCBLIST_KTHREAD_CB BIT(2) -#define SEGCBLIST_KTHREAD_GP BIT(3) -#define SEGCBLIST_OFFLOADED BIT(4) +#define SEGCBLIST_KTHREAD_GP BIT(2) +#define SEGCBLIST_OFFLOADED BIT(3) struct rcu_segcblist { struct rcu_head *head; -- cgit From 9aed3b51fd6186582a95abd9fa67782982540749 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 2 Jul 2024 18:49:35 -0700 Subject: rcu: Better define "atomic" for list replacement The kernel-doc headers for list_replace_rcu() and hlist_replace_rcu() claim that the replacement is atomic, which it is, but only for readers. Avoid confusion by making it clear that the atomic nature of these functions applies only to readers, not to concurrent updaters. Signed-off-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rculist.h | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rculist.h b/include/linux/rculist.h index 3dc1e58865f7..14dfa6008467 100644 --- a/include/linux/rculist.h +++ b/include/linux/rculist.h @@ -191,7 +191,10 @@ static inline void hlist_del_init_rcu(struct hlist_node *n) * @old : the element to be replaced * @new : the new element to insert * - * The @old entry will be replaced with the @new entry atomically. + * The @old entry will be replaced with the @new entry atomically from + * the perspective of concurrent readers. It is the caller's responsibility + * to synchronize with concurrent updaters, if any. + * * Note: @old should not be empty. */ static inline void list_replace_rcu(struct list_head *old, @@ -519,7 +522,9 @@ static inline void hlist_del_rcu(struct hlist_node *n) * @old : the element to be replaced * @new : the new element to insert * - * The @old entry will be replaced with the @new entry atomically. + * The @old entry will be replaced with the @new entry atomically from + * the perspective of concurrent readers. It is the caller's responsibility + * to synchronize with concurrent updaters, if any. */ static inline void hlist_replace_rcu(struct hlist_node *old, struct hlist_node *new) -- cgit From bf66471987b4bd537bc800353ca7d39bfd1d1022 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 10:51:10 +0200 Subject: context_tracking, rcu: Rename struct context_tracking .dynticks_nesting into .nesting The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. [ neeraj.upadhyay: Fix htmldocs build error reported by Stephen Rothwell ] Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- Documentation/RCU/Design/Data-Structures/Data-Structures.rst | 10 +++++----- include/linux/context_tracking_state.h | 6 +++--- include/trace/events/rcu.h | 2 +- kernel/context_tracking.c | 10 +++++----- kernel/rcu/tree.c | 8 ++++---- 5 files changed, 18 insertions(+), 18 deletions(-) (limited to 'include/linux') diff --git a/Documentation/RCU/Design/Data-Structures/Data-Structures.rst b/Documentation/RCU/Design/Data-Structures/Data-Structures.rst index b34990c7c377..69860bbec202 100644 --- a/Documentation/RCU/Design/Data-Structures/Data-Structures.rst +++ b/Documentation/RCU/Design/Data-Structures/Data-Structures.rst @@ -935,7 +935,7 @@ This portion of the rcu_data structure is declared as follows: :: - 1 long dynticks_nesting; + 1 long nesting; 2 long dynticks_nmi_nesting; 3 atomic_t dynticks; 4 bool rcu_need_heavy_qs; @@ -945,7 +945,7 @@ These fields in the rcu_data structure maintain the per-CPU dyntick-idle state for the corresponding CPU. The fields may be accessed only from the corresponding CPU (and from tracing) unless otherwise stated. -The ``->dynticks_nesting`` field counts the nesting depth of process +The ``->nesting`` field counts the nesting depth of process execution, so that in normal circumstances this counter has value zero or one. NMIs, irqs, and tracers are counted by the ``->dynticks_nmi_nesting`` field. Because NMIs cannot be masked, changes @@ -960,9 +960,9 @@ process-level transitions. However, it turns out that when running in non-idle kernel context, the Linux kernel is fully capable of entering interrupt handlers that never exit and perhaps also vice versa. Therefore, whenever the -``->dynticks_nesting`` field is incremented up from zero, the +``->nesting`` field is incremented up from zero, the ``->dynticks_nmi_nesting`` field is set to a large positive number, and -whenever the ``->dynticks_nesting`` field is decremented down to zero, +whenever the ``->nesting`` field is decremented down to zero, the ``->dynticks_nmi_nesting`` field is set to zero. Assuming that the number of misnested interrupts is not sufficient to overflow the counter, this approach corrects the ``->dynticks_nmi_nesting`` field @@ -992,7 +992,7 @@ code. +-----------------------------------------------------------------------+ | **Quick Quiz**: | +-----------------------------------------------------------------------+ -| Why not simply combine the ``->dynticks_nesting`` and | +| Why not simply combine the ``->nesting`` and | | ``->dynticks_nmi_nesting`` counters into a single counter that just | | counts the number of reasons that the corresponding CPU is non-idle? | +-----------------------------------------------------------------------+ diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index ad6570ffeff3..65290e7677e6 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -39,7 +39,7 @@ struct context_tracking { atomic_t state; #endif #ifdef CONFIG_CONTEXT_TRACKING_IDLE - long dynticks_nesting; /* Track process nesting level. */ + long nesting; /* Track process nesting level. */ long dynticks_nmi_nesting; /* Track irq/NMI nesting level. */ #endif }; @@ -77,14 +77,14 @@ static __always_inline int ct_rcu_watching_cpu_acquire(int cpu) static __always_inline long ct_dynticks_nesting(void) { - return __this_cpu_read(context_tracking.dynticks_nesting); + return __this_cpu_read(context_tracking.nesting); } static __always_inline long ct_dynticks_nesting_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); - return ct->dynticks_nesting; + return ct->nesting; } static __always_inline long ct_dynticks_nmi_nesting(void) diff --git a/include/trace/events/rcu.h b/include/trace/events/rcu.h index 31b3e0d3e65f..4066b6d51e46 100644 --- a/include/trace/events/rcu.h +++ b/include/trace/events/rcu.h @@ -469,7 +469,7 @@ TRACE_EVENT(rcu_stall_warning, * polarity: "Start", "End", "StillNonIdle" for entering, exiting or still not * being in dyntick-idle mode. * context: "USER" or "IDLE" or "IRQ". - * NMIs nested in IRQs are inferred with dynticks_nesting > 1 in IRQ context. + * NMIs nested in IRQs are inferred with nesting > 1 in IRQ context. * * These events also take a pair of numbers, which indicate the nesting * depth before and after the event of interest, and a third number that is diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 868ae0bcd4be..5cfdfc03b401 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -28,7 +28,7 @@ DEFINE_PER_CPU(struct context_tracking, context_tracking) = { #ifdef CONFIG_CONTEXT_TRACKING_IDLE - .dynticks_nesting = 1, + .nesting = 1, .dynticks_nmi_nesting = DYNTICK_IRQ_NONIDLE, #endif .state = ATOMIC_INIT(CT_RCU_WATCHING), @@ -131,7 +131,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) ct_dynticks_nesting() == 0); if (ct_dynticks_nesting() != 1) { // RCU will still be watching, so just do accounting and leave. - ct->dynticks_nesting--; + ct->nesting--; return; } @@ -145,7 +145,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) instrument_atomic_write(&ct->state, sizeof(ct->state)); instrumentation_end(); - WRITE_ONCE(ct->dynticks_nesting, 0); /* Avoid irq-access tearing. */ + WRITE_ONCE(ct->nesting, 0); /* Avoid irq-access tearing. */ // RCU is watching here ... ct_kernel_exit_state(offset); // ... but is no longer watching here. @@ -170,7 +170,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && oldval < 0); if (oldval) { // RCU was already watching, so just do accounting and leave. - ct->dynticks_nesting++; + ct->nesting++; return; } rcu_dynticks_task_exit(); @@ -184,7 +184,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) trace_rcu_dyntick(TPS("End"), ct_dynticks_nesting(), 1, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); - WRITE_ONCE(ct->dynticks_nesting, 1); + WRITE_ONCE(ct->nesting, 1); WARN_ON_ONCE(ct_dynticks_nmi_nesting()); WRITE_ONCE(ct->dynticks_nmi_nesting, DYNTICK_IRQ_NONIDLE); instrumentation_end(); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 45a9f3667c2a..0d5a539acd58 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -389,7 +389,7 @@ static int rcu_is_cpu_rrupt_from_idle(void) /* Check for counter underflows */ RCU_LOCKDEP_WARN(ct_dynticks_nesting() < 0, - "RCU dynticks_nesting counter underflow!"); + "RCU nesting counter underflow!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() <= 0, "RCU dynticks_nmi_nesting counter underflow/zero!"); @@ -597,7 +597,7 @@ void rcu_irq_exit_check_preempt(void) lockdep_assert_irqs_disabled(); RCU_LOCKDEP_WARN(ct_dynticks_nesting() <= 0, - "RCU dynticks_nesting counter underflow/zero!"); + "RCU nesting counter underflow/zero!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE, "Bad RCU dynticks_nmi_nesting counter\n"); @@ -4804,7 +4804,7 @@ rcu_boot_init_percpu_data(int cpu) /* Set up local state, ensuring consistent view of global state. */ rdp->grpmask = leaf_node_cpu_bit(rdp->mynode, cpu); INIT_WORK(&rdp->strict_work, strict_work_handler); - WARN_ON_ONCE(ct->dynticks_nesting != 1); + WARN_ON_ONCE(ct->nesting != 1); WARN_ON_ONCE(rcu_dynticks_in_eqs(ct_rcu_watching_cpu(cpu))); rdp->barrier_seq_snap = rcu_state.barrier_sequence; rdp->rcu_ofl_gp_seq = rcu_state.gp_seq; @@ -4898,7 +4898,7 @@ int rcutree_prepare_cpu(unsigned int cpu) rdp->qlen_last_fqs_check = 0; rdp->n_force_qs_snap = READ_ONCE(rcu_state.n_force_qs); rdp->blimit = blimit; - ct->dynticks_nesting = 1; /* CPU not up, no tearing. */ + ct->nesting = 1; /* CPU not up, no tearing. */ raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled. */ /* -- cgit From 1089c0078b69bce0c2d540e3cc43470ba9e5dcfd Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 14:45:30 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_nesting() into ct_nesting() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/context_tracking.c | 10 +++++----- kernel/rcu/tree.c | 6 +++--- 3 files changed, 9 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 65290e7677e6..586c1ff22c2e 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -75,7 +75,7 @@ static __always_inline int ct_rcu_watching_cpu_acquire(int cpu) return atomic_read_acquire(&ct->state) & CT_RCU_WATCHING_MASK; } -static __always_inline long ct_dynticks_nesting(void) +static __always_inline long ct_nesting(void) { return __this_cpu_read(context_tracking.nesting); } diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 5cfdfc03b401..a951bde0bbcb 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -128,8 +128,8 @@ static void noinstr ct_kernel_exit(bool user, int offset) WARN_ON_ONCE(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE); WRITE_ONCE(ct->dynticks_nmi_nesting, 0); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && - ct_dynticks_nesting() == 0); - if (ct_dynticks_nesting() != 1) { + ct_nesting() == 0); + if (ct_nesting() != 1) { // RCU will still be watching, so just do accounting and leave. ct->nesting--; return; @@ -137,7 +137,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) instrumentation_begin(); lockdep_assert_irqs_disabled(); - trace_rcu_dyntick(TPS("Start"), ct_dynticks_nesting(), 0, ct_rcu_watching()); + trace_rcu_dyntick(TPS("Start"), ct_nesting(), 0, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); rcu_preempt_deferred_qs(current); @@ -166,7 +166,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) long oldval; WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !raw_irqs_disabled()); - oldval = ct_dynticks_nesting(); + oldval = ct_nesting(); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && oldval < 0); if (oldval) { // RCU was already watching, so just do accounting and leave. @@ -182,7 +182,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) // instrumentation for the noinstr ct_kernel_enter_state() instrument_atomic_write(&ct->state, sizeof(ct->state)); - trace_rcu_dyntick(TPS("End"), ct_dynticks_nesting(), 1, ct_rcu_watching()); + trace_rcu_dyntick(TPS("End"), ct_nesting(), 1, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); WRITE_ONCE(ct->nesting, 1); WARN_ON_ONCE(ct_dynticks_nmi_nesting()); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 0d5a539acd58..cf69a234080f 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -388,7 +388,7 @@ static int rcu_is_cpu_rrupt_from_idle(void) lockdep_assert_irqs_disabled(); /* Check for counter underflows */ - RCU_LOCKDEP_WARN(ct_dynticks_nesting() < 0, + RCU_LOCKDEP_WARN(ct_nesting() < 0, "RCU nesting counter underflow!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() <= 0, "RCU dynticks_nmi_nesting counter underflow/zero!"); @@ -404,7 +404,7 @@ static int rcu_is_cpu_rrupt_from_idle(void) WARN_ON_ONCE(!nesting && !is_idle_task(current)); /* Does CPU appear to be idle from an RCU standpoint? */ - return ct_dynticks_nesting() == 0; + return ct_nesting() == 0; } #define DEFAULT_RCU_BLIMIT (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) ? 1000 : 10) @@ -596,7 +596,7 @@ void rcu_irq_exit_check_preempt(void) { lockdep_assert_irqs_disabled(); - RCU_LOCKDEP_WARN(ct_dynticks_nesting() <= 0, + RCU_LOCKDEP_WARN(ct_nesting() <= 0, "RCU nesting counter underflow/zero!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE, -- cgit From bca9455da531a3c2520c28ff349d0dab4e471d28 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 14:48:26 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_nesting_cpu() into ct_nesting_cpu() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, and the 'dynticks' prefix can be dropped without losing any meaning. Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/rcu/tree_stall.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 586c1ff22c2e..fd42d8120ac2 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -80,7 +80,7 @@ static __always_inline long ct_nesting(void) return __this_cpu_read(context_tracking.nesting); } -static __always_inline long ct_dynticks_nesting_cpu(int cpu) +static __always_inline long ct_nesting_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h index d65974448e81..59b1d84a4749 100644 --- a/kernel/rcu/tree_stall.h +++ b/kernel/rcu/tree_stall.h @@ -516,7 +516,7 @@ static void print_cpu_stall_info(int cpu) "!."[!delta], ticks_value, ticks_title, ct_rcu_watching_cpu(cpu) & 0xffff, - ct_dynticks_nesting_cpu(cpu), ct_dynticks_nmi_nesting_cpu(cpu), + ct_nesting_cpu(cpu), ct_dynticks_nmi_nesting_cpu(cpu), rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu), data_race(rcu_state.n_force_qs) - rcu_state.n_force_qs_gpstart, rcuc_starved ? buf : "", -- cgit From dc5fface4b30508933b515a4985401c8c8f6bcfd Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 10:52:03 +0200 Subject: context_tracking, rcu: Rename struct context_tracking .dynticks_nmi_nesting into .nmi_nesting The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, and the 'dynticks' prefix can be dropped without losing any meaning. [ neeraj.upadhyay: Fix htmldocs build error reported by Stephen Rothwell ] Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- .../RCU/Design/Data-Structures/Data-Structures.rst | 14 +++++----- include/linux/context_tracking_state.h | 6 ++--- kernel/context_tracking.c | 30 +++++++++++----------- kernel/rcu/tree.c | 4 +-- 4 files changed, 27 insertions(+), 27 deletions(-) (limited to 'include/linux') diff --git a/Documentation/RCU/Design/Data-Structures/Data-Structures.rst b/Documentation/RCU/Design/Data-Structures/Data-Structures.rst index 69860bbec202..28d9f9e1783f 100644 --- a/Documentation/RCU/Design/Data-Structures/Data-Structures.rst +++ b/Documentation/RCU/Design/Data-Structures/Data-Structures.rst @@ -936,7 +936,7 @@ This portion of the rcu_data structure is declared as follows: :: 1 long nesting; - 2 long dynticks_nmi_nesting; + 2 long nmi_nesting; 3 atomic_t dynticks; 4 bool rcu_need_heavy_qs; 5 bool rcu_urgent_qs; @@ -948,11 +948,11 @@ the corresponding CPU (and from tracing) unless otherwise stated. The ``->nesting`` field counts the nesting depth of process execution, so that in normal circumstances this counter has value zero or one. NMIs, irqs, and tracers are counted by the -``->dynticks_nmi_nesting`` field. Because NMIs cannot be masked, changes +``->nmi_nesting`` field. Because NMIs cannot be masked, changes to this variable have to be undertaken carefully using an algorithm provided by Andy Lutomirski. The initial transition from idle adds one, and nested transitions add two, so that a nesting level of five is -represented by a ``->dynticks_nmi_nesting`` value of nine. This counter +represented by a ``->nmi_nesting`` value of nine. This counter can therefore be thought of as counting the number of reasons why this CPU cannot be permitted to enter dyntick-idle mode, aside from process-level transitions. @@ -961,11 +961,11 @@ However, it turns out that when running in non-idle kernel context, the Linux kernel is fully capable of entering interrupt handlers that never exit and perhaps also vice versa. Therefore, whenever the ``->nesting`` field is incremented up from zero, the -``->dynticks_nmi_nesting`` field is set to a large positive number, and +``->nmi_nesting`` field is set to a large positive number, and whenever the ``->nesting`` field is decremented down to zero, -the ``->dynticks_nmi_nesting`` field is set to zero. Assuming that +the ``->nmi_nesting`` field is set to zero. Assuming that the number of misnested interrupts is not sufficient to overflow the -counter, this approach corrects the ``->dynticks_nmi_nesting`` field +counter, this approach corrects the ``->nmi_nesting`` field every time the corresponding CPU enters the idle loop from process context. @@ -993,7 +993,7 @@ code. | **Quick Quiz**: | +-----------------------------------------------------------------------+ | Why not simply combine the ``->nesting`` and | -| ``->dynticks_nmi_nesting`` counters into a single counter that just | +| ``->nmi_nesting`` counters into a single counter that just | | counts the number of reasons that the corresponding CPU is non-idle? | +-----------------------------------------------------------------------+ | **Answer**: | diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index fd42d8120ac2..12d00adf29e1 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -40,7 +40,7 @@ struct context_tracking { #endif #ifdef CONFIG_CONTEXT_TRACKING_IDLE long nesting; /* Track process nesting level. */ - long dynticks_nmi_nesting; /* Track irq/NMI nesting level. */ + long nmi_nesting; /* Track irq/NMI nesting level. */ #endif }; @@ -89,14 +89,14 @@ static __always_inline long ct_nesting_cpu(int cpu) static __always_inline long ct_dynticks_nmi_nesting(void) { - return __this_cpu_read(context_tracking.dynticks_nmi_nesting); + return __this_cpu_read(context_tracking.nmi_nesting); } static __always_inline long ct_dynticks_nmi_nesting_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); - return ct->dynticks_nmi_nesting; + return ct->nmi_nesting; } #endif /* #ifdef CONFIG_CONTEXT_TRACKING_IDLE */ diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index a951bde0bbcb..ae94215aa132 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -29,7 +29,7 @@ DEFINE_PER_CPU(struct context_tracking, context_tracking) = { #ifdef CONFIG_CONTEXT_TRACKING_IDLE .nesting = 1, - .dynticks_nmi_nesting = DYNTICK_IRQ_NONIDLE, + .nmi_nesting = DYNTICK_IRQ_NONIDLE, #endif .state = ATOMIC_INIT(CT_RCU_WATCHING), }; @@ -117,7 +117,7 @@ static noinstr void ct_kernel_enter_state(int offset) * Enter an RCU extended quiescent state, which can be either the * idle loop or adaptive-tickless usermode execution. * - * We crowbar the ->dynticks_nmi_nesting field to zero to allow for + * We crowbar the ->nmi_nesting field to zero to allow for * the possibility of usermode upcalls having messed up our count * of interrupt nesting level during the prior busy period. */ @@ -126,7 +126,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) struct context_tracking *ct = this_cpu_ptr(&context_tracking); WARN_ON_ONCE(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE); - WRITE_ONCE(ct->dynticks_nmi_nesting, 0); + WRITE_ONCE(ct->nmi_nesting, 0); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && ct_nesting() == 0); if (ct_nesting() != 1) { @@ -156,7 +156,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) * Exit an RCU extended quiescent state, which can be either the * idle loop or adaptive-tickless usermode execution. * - * We crowbar the ->dynticks_nmi_nesting field to DYNTICK_IRQ_NONIDLE to + * We crowbar the ->nmi_nesting field to DYNTICK_IRQ_NONIDLE to * allow for the possibility of usermode upcalls messing up our count of * interrupt nesting level during the busy period that is just now starting. */ @@ -186,7 +186,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); WRITE_ONCE(ct->nesting, 1); WARN_ON_ONCE(ct_dynticks_nmi_nesting()); - WRITE_ONCE(ct->dynticks_nmi_nesting, DYNTICK_IRQ_NONIDLE); + WRITE_ONCE(ct->nmi_nesting, DYNTICK_IRQ_NONIDLE); instrumentation_end(); } @@ -194,7 +194,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) * ct_nmi_exit - inform RCU of exit from NMI context * * If we are returning from the outermost NMI handler that interrupted an - * RCU-idle period, update ct->state and ct->dynticks_nmi_nesting + * RCU-idle period, update ct->state and ct->nmi_nesting * to let the RCU grace-period handling know that the CPU is back to * being RCU-idle. * @@ -207,7 +207,7 @@ void noinstr ct_nmi_exit(void) instrumentation_begin(); /* - * Check for ->dynticks_nmi_nesting underflow and bad ->dynticks. + * Check for ->nmi_nesting underflow and bad ->dynticks. * (We are exiting an NMI handler, so RCU better be paying attention * to us!) */ @@ -221,7 +221,7 @@ void noinstr ct_nmi_exit(void) if (ct_dynticks_nmi_nesting() != 1) { trace_rcu_dyntick(TPS("--="), ct_dynticks_nmi_nesting(), ct_dynticks_nmi_nesting() - 2, ct_rcu_watching()); - WRITE_ONCE(ct->dynticks_nmi_nesting, /* No store tearing. */ + WRITE_ONCE(ct->nmi_nesting, /* No store tearing. */ ct_dynticks_nmi_nesting() - 2); instrumentation_end(); return; @@ -229,7 +229,7 @@ void noinstr ct_nmi_exit(void) /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */ trace_rcu_dyntick(TPS("Startirq"), ct_dynticks_nmi_nesting(), 0, ct_rcu_watching()); - WRITE_ONCE(ct->dynticks_nmi_nesting, 0); /* Avoid store tearing. */ + WRITE_ONCE(ct->nmi_nesting, 0); /* Avoid store tearing. */ // instrumentation for the noinstr ct_kernel_exit_state() instrument_atomic_write(&ct->state, sizeof(ct->state)); @@ -247,7 +247,7 @@ void noinstr ct_nmi_exit(void) * ct_nmi_enter - inform RCU of entry to NMI context * * If the CPU was idle from RCU's viewpoint, update ct->state and - * ct->dynticks_nmi_nesting to let the RCU grace-period handling know + * ct->nmi_nesting to let the RCU grace-period handling know * that the CPU is active. This implementation permits nested NMIs, as * long as the nesting level does not overflow an int. (You will probably * run out of stack space first.) @@ -264,10 +264,10 @@ void noinstr ct_nmi_enter(void) WARN_ON_ONCE(ct_dynticks_nmi_nesting() < 0); /* - * If idle from RCU viewpoint, atomically increment ->dynticks - * to mark non-idle and increment ->dynticks_nmi_nesting by one. - * Otherwise, increment ->dynticks_nmi_nesting by two. This means - * if ->dynticks_nmi_nesting is equal to one, we are guaranteed + * If idle from RCU viewpoint, atomically increment CT state + * to mark non-idle and increment ->nmi_nesting by one. + * Otherwise, increment ->nmi_nesting by two. This means + * if ->nmi_nesting is equal to one, we are guaranteed * to be in the outermost NMI handler that interrupted an RCU-idle * period (observation due to Andy Lutomirski). */ @@ -298,7 +298,7 @@ void noinstr ct_nmi_enter(void) ct_dynticks_nmi_nesting(), ct_dynticks_nmi_nesting() + incby, ct_rcu_watching()); instrumentation_end(); - WRITE_ONCE(ct->dynticks_nmi_nesting, /* Prevent store tearing. */ + WRITE_ONCE(ct->nmi_nesting, /* Prevent store tearing. */ ct_dynticks_nmi_nesting() + incby); barrier(); } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index cf69a234080f..934f6b34a551 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -391,7 +391,7 @@ static int rcu_is_cpu_rrupt_from_idle(void) RCU_LOCKDEP_WARN(ct_nesting() < 0, "RCU nesting counter underflow!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() <= 0, - "RCU dynticks_nmi_nesting counter underflow/zero!"); + "RCU nmi_nesting counter underflow/zero!"); /* Are we at first interrupt nesting level? */ nesting = ct_dynticks_nmi_nesting(); @@ -600,7 +600,7 @@ void rcu_irq_exit_check_preempt(void) "RCU nesting counter underflow/zero!"); RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE, - "Bad RCU dynticks_nmi_nesting counter\n"); + "Bad RCU nmi_nesting counter\n"); RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), "RCU in extended quiescent state!"); } -- cgit From 8375cb260d7e43610934af63748d1a51201449f8 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 14:46:14 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_nmi_nesting() into ct_nmi_nesting() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, and the 'dynticks' prefix can be dropped without losing any meaning. Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/context_tracking.c | 24 ++++++++++++------------ kernel/rcu/tree.c | 6 +++--- 3 files changed, 16 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 12d00adf29e1..8f32fe599c5c 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -87,7 +87,7 @@ static __always_inline long ct_nesting_cpu(int cpu) return ct->nesting; } -static __always_inline long ct_dynticks_nmi_nesting(void) +static __always_inline long ct_nmi_nesting(void) { return __this_cpu_read(context_tracking.nmi_nesting); } diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index ae94215aa132..115843eeb030 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -125,7 +125,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) { struct context_tracking *ct = this_cpu_ptr(&context_tracking); - WARN_ON_ONCE(ct_dynticks_nmi_nesting() != DYNTICK_IRQ_NONIDLE); + WARN_ON_ONCE(ct_nmi_nesting() != DYNTICK_IRQ_NONIDLE); WRITE_ONCE(ct->nmi_nesting, 0); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && ct_nesting() == 0); @@ -185,7 +185,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) trace_rcu_dyntick(TPS("End"), ct_nesting(), 1, ct_rcu_watching()); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); WRITE_ONCE(ct->nesting, 1); - WARN_ON_ONCE(ct_dynticks_nmi_nesting()); + WARN_ON_ONCE(ct_nmi_nesting()); WRITE_ONCE(ct->nmi_nesting, DYNTICK_IRQ_NONIDLE); instrumentation_end(); } @@ -207,28 +207,28 @@ void noinstr ct_nmi_exit(void) instrumentation_begin(); /* - * Check for ->nmi_nesting underflow and bad ->dynticks. + * Check for ->nmi_nesting underflow and bad CT state. * (We are exiting an NMI handler, so RCU better be paying attention * to us!) */ - WARN_ON_ONCE(ct_dynticks_nmi_nesting() <= 0); + WARN_ON_ONCE(ct_nmi_nesting() <= 0); WARN_ON_ONCE(rcu_dynticks_curr_cpu_in_eqs()); /* * If the nesting level is not 1, the CPU wasn't RCU-idle, so * leave it in non-RCU-idle state. */ - if (ct_dynticks_nmi_nesting() != 1) { - trace_rcu_dyntick(TPS("--="), ct_dynticks_nmi_nesting(), ct_dynticks_nmi_nesting() - 2, + if (ct_nmi_nesting() != 1) { + trace_rcu_dyntick(TPS("--="), ct_nmi_nesting(), ct_nmi_nesting() - 2, ct_rcu_watching()); WRITE_ONCE(ct->nmi_nesting, /* No store tearing. */ - ct_dynticks_nmi_nesting() - 2); + ct_nmi_nesting() - 2); instrumentation_end(); return; } /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */ - trace_rcu_dyntick(TPS("Startirq"), ct_dynticks_nmi_nesting(), 0, ct_rcu_watching()); + trace_rcu_dyntick(TPS("Startirq"), ct_nmi_nesting(), 0, ct_rcu_watching()); WRITE_ONCE(ct->nmi_nesting, 0); /* Avoid store tearing. */ // instrumentation for the noinstr ct_kernel_exit_state() @@ -261,7 +261,7 @@ void noinstr ct_nmi_enter(void) struct context_tracking *ct = this_cpu_ptr(&context_tracking); /* Complain about underflow. */ - WARN_ON_ONCE(ct_dynticks_nmi_nesting() < 0); + WARN_ON_ONCE(ct_nmi_nesting() < 0); /* * If idle from RCU viewpoint, atomically increment CT state @@ -295,11 +295,11 @@ void noinstr ct_nmi_enter(void) } trace_rcu_dyntick(incby == 1 ? TPS("Endirq") : TPS("++="), - ct_dynticks_nmi_nesting(), - ct_dynticks_nmi_nesting() + incby, ct_rcu_watching()); + ct_nmi_nesting(), + ct_nmi_nesting() + incby, ct_rcu_watching()); instrumentation_end(); WRITE_ONCE(ct->nmi_nesting, /* Prevent store tearing. */ - ct_dynticks_nmi_nesting() + incby); + ct_nmi_nesting() + incby); barrier(); } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 934f6b34a551..14cc314eedad 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -390,11 +390,11 @@ 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_dynticks_nmi_nesting() <= 0, + RCU_LOCKDEP_WARN(ct_nmi_nesting() <= 0, "RCU nmi_nesting counter underflow/zero!"); /* Are we at first interrupt nesting level? */ - nesting = ct_dynticks_nmi_nesting(); + nesting = ct_nmi_nesting(); if (nesting > 1) return false; @@ -598,7 +598,7 @@ void rcu_irq_exit_check_preempt(void) RCU_LOCKDEP_WARN(ct_nesting() <= 0, "RCU nesting counter underflow/zero!"); - RCU_LOCKDEP_WARN(ct_dynticks_nmi_nesting() != + RCU_LOCKDEP_WARN(ct_nmi_nesting() != DYNTICK_IRQ_NONIDLE, "Bad RCU nmi_nesting counter\n"); RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), -- cgit From 2ef2890b7a94fbbfa8fdf0a90499ae1df476b8e8 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 14:49:13 +0200 Subject: context_tracking, rcu: Rename ct_dynticks_nmi_nesting_cpu() into ct_nmi_nesting_cpu() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, and the 'dynticks' prefix can be dropped without losing any meaning. Suggested-by: Frederic Weisbecker Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/rcu/tree_stall.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 8f32fe599c5c..34fd504e53a8 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -92,7 +92,7 @@ static __always_inline long ct_nmi_nesting(void) return __this_cpu_read(context_tracking.nmi_nesting); } -static __always_inline long ct_dynticks_nmi_nesting_cpu(int cpu) +static __always_inline long ct_nmi_nesting_cpu(int cpu) { struct context_tracking *ct = per_cpu_ptr(&context_tracking, cpu); diff --git a/kernel/rcu/tree_stall.h b/kernel/rcu/tree_stall.h index 59b1d84a4749..ec49f0155bec 100644 --- a/kernel/rcu/tree_stall.h +++ b/kernel/rcu/tree_stall.h @@ -516,7 +516,7 @@ static void print_cpu_stall_info(int cpu) "!."[!delta], ticks_value, ticks_title, ct_rcu_watching_cpu(cpu) & 0xffff, - ct_nesting_cpu(cpu), ct_dynticks_nmi_nesting_cpu(cpu), + ct_nesting_cpu(cpu), ct_nmi_nesting_cpu(cpu), rdp->softirq_snap, kstat_softirqs_cpu(RCU_SOFTIRQ, cpu), data_race(rcu_state.n_force_qs) - rcu_state.n_force_qs_gpstart, rcuc_starved ? buf : "", -- cgit From e1de438336222dbe27f2d4baa8c01074fcac2bef Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Mon, 29 Apr 2024 15:52:32 +0200 Subject: context_tracking, rcu: Rename DYNTICK_IRQ_NONIDLE into CT_NESTING_IRQ_NONIDLE The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, and the 'dynticks' prefix can be dropped without losing any meaning. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- kernel/context_tracking.c | 8 ++++---- kernel/rcu/tree.c | 2 +- 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 34fd504e53a8..0dbda59c9f37 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -7,7 +7,7 @@ #include /* Offset to allow distinguishing irq vs. task-based idle entry/exit. */ -#define DYNTICK_IRQ_NONIDLE ((LONG_MAX / 2) + 1) +#define CT_NESTING_IRQ_NONIDLE ((LONG_MAX / 2) + 1) enum ctx_state { CT_STATE_DISABLED = -1, /* returned by ct_state() if unknown */ diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 115843eeb030..8262f57a4363 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -29,7 +29,7 @@ DEFINE_PER_CPU(struct context_tracking, context_tracking) = { #ifdef CONFIG_CONTEXT_TRACKING_IDLE .nesting = 1, - .nmi_nesting = DYNTICK_IRQ_NONIDLE, + .nmi_nesting = CT_NESTING_IRQ_NONIDLE, #endif .state = ATOMIC_INIT(CT_RCU_WATCHING), }; @@ -125,7 +125,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) { struct context_tracking *ct = this_cpu_ptr(&context_tracking); - WARN_ON_ONCE(ct_nmi_nesting() != DYNTICK_IRQ_NONIDLE); + WARN_ON_ONCE(ct_nmi_nesting() != CT_NESTING_IRQ_NONIDLE); WRITE_ONCE(ct->nmi_nesting, 0); WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && ct_nesting() == 0); @@ -156,7 +156,7 @@ static void noinstr ct_kernel_exit(bool user, int offset) * Exit an RCU extended quiescent state, which can be either the * idle loop or adaptive-tickless usermode execution. * - * We crowbar the ->nmi_nesting field to DYNTICK_IRQ_NONIDLE to + * We crowbar the ->nmi_nesting field to CT_NESTING_IRQ_NONIDLE to * allow for the possibility of usermode upcalls messing up our count of * interrupt nesting level during the busy period that is just now starting. */ @@ -186,7 +186,7 @@ static void noinstr ct_kernel_enter(bool user, int offset) WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) && !user && !is_idle_task(current)); WRITE_ONCE(ct->nesting, 1); WARN_ON_ONCE(ct_nmi_nesting()); - WRITE_ONCE(ct->nmi_nesting, DYNTICK_IRQ_NONIDLE); + WRITE_ONCE(ct->nmi_nesting, CT_NESTING_IRQ_NONIDLE); instrumentation_end(); } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 14cc314eedad..4778c873f4ac 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -599,7 +599,7 @@ void rcu_irq_exit_check_preempt(void) RCU_LOCKDEP_WARN(ct_nesting() <= 0, "RCU nesting counter underflow/zero!"); RCU_LOCKDEP_WARN(ct_nmi_nesting() != - DYNTICK_IRQ_NONIDLE, + CT_NESTING_IRQ_NONIDLE, "Bad RCU nmi_nesting counter\n"); RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), "RCU in extended quiescent state!"); -- cgit From 29bc83e4d90546aa794a9584786086b141a6ba4d Mon Sep 17 00:00:00 2001 From: JP Kobryn Date: Mon, 15 Jul 2024 16:23:24 -0700 Subject: srcu: faster gp seq wrap-around Using a higher value for the initial gp sequence counters allows for wrapping to occur faster. It can help with surfacing any issues that may be happening as a result of the wrap around. Signed-off-by: JP Kobryn Tested-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rcupdate.h | 3 +++ include/linux/srcutree.h | 15 ++++++++++++++- kernel/rcu/rcu.h | 3 --- kernel/rcu/srcutree.c | 7 ++++--- 4 files changed, 21 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 13f6f00aecf9..8d56db70d417 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -34,6 +34,9 @@ #define ULONG_CMP_GE(a, b) (ULONG_MAX / 2 >= (a) - (b)) #define ULONG_CMP_LT(a, b) (ULONG_MAX / 2 < (a) - (b)) +#define RCU_SEQ_CTR_SHIFT 2 +#define RCU_SEQ_STATE_MASK ((1 << RCU_SEQ_CTR_SHIFT) - 1) + /* Exported common interfaces */ void call_rcu(struct rcu_head *head, rcu_callback_t func); void rcu_barrier_tasks(void); diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h index 8f3f72480e78..ed57598394de 100644 --- a/include/linux/srcutree.h +++ b/include/linux/srcutree.h @@ -129,10 +129,23 @@ struct srcu_struct { #define SRCU_STATE_SCAN1 1 #define SRCU_STATE_SCAN2 2 +/* + * Values for initializing gp sequence fields. Higher values allow wrap arounds to + * occur earlier. + * The second value with state is useful in the case of static initialization of + * srcu_usage where srcu_gp_seq_needed is expected to have some state value in its + * lower bits (or else it will appear to be already initialized within + * the call check_init_srcu_struct()). + */ +#define SRCU_GP_SEQ_INITIAL_VAL ((0UL - 100UL) << RCU_SEQ_CTR_SHIFT) +#define SRCU_GP_SEQ_INITIAL_VAL_WITH_STATE (SRCU_GP_SEQ_INITIAL_VAL - 1) + #define __SRCU_USAGE_INIT(name) \ { \ .lock = __SPIN_LOCK_UNLOCKED(name.lock), \ - .srcu_gp_seq_needed = -1UL, \ + .srcu_gp_seq = SRCU_GP_SEQ_INITIAL_VAL, \ + .srcu_gp_seq_needed = SRCU_GP_SEQ_INITIAL_VAL_WITH_STATE, \ + .srcu_gp_seq_needed_exp = SRCU_GP_SEQ_INITIAL_VAL, \ .work = __DELAYED_WORK_INITIALIZER(name.work, NULL, 0), \ } diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index 38238e595a61..2bfed9855d67 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -54,9 +54,6 @@ * grace-period sequence number. */ -#define RCU_SEQ_CTR_SHIFT 2 -#define RCU_SEQ_STATE_MASK ((1 << RCU_SEQ_CTR_SHIFT) - 1) - /* Low-order bit definition for polled grace-period APIs. */ #define RCU_GET_STATE_COMPLETED 0x1 diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index b24db425f16d..6fd9c914ce64 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -247,7 +247,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) 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 = 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); @@ -258,7 +258,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) if (!ssp->sda) goto err_free_sup; init_srcu_struct_data(ssp); - ssp->srcu_sup->srcu_gp_seq_needed_exp = 0; + 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)) @@ -266,7 +266,8 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) 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, 0); /* Init done. */ + smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, + SRCU_GP_SEQ_INITIAL_VAL); /* Init done. */ return 0; err_free_sda: -- cgit From 7945b741d1fc071a621366c512a060ea08848955 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 3 Jul 2024 19:47:47 -0700 Subject: rcu-tasks: Remove RCU Tasks Rude asynchronous APIs The call_rcu_tasks_rude() and rcu_barrier_tasks_rude() APIs are currently unused. This commit therefore removes their definitions and boot-time self-tests. Signed-off-by: Paul E. McKenney Cc: Peter Zijlstra Signed-off-by: Neeraj Upadhyay --- include/linux/rcupdate.h | 2 -- kernel/rcu/tasks.h | 55 +++++++++++++++--------------------------------- 2 files changed, 17 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 13f6f00aecf9..31e679c7110e 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -37,7 +37,6 @@ /* Exported common interfaces */ void call_rcu(struct rcu_head *head, rcu_callback_t func); void rcu_barrier_tasks(void); -void rcu_barrier_tasks_rude(void); void synchronize_rcu(void); struct rcu_gp_oldstate; @@ -202,7 +201,6 @@ do { \ } while (0) # ifdef CONFIG_TASKS_RUDE_RCU -void call_rcu_tasks_rude(struct rcu_head *head, rcu_callback_t func); void synchronize_rcu_tasks_rude(void); # endif diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h index ba3440a45b6d..4bc038bcc016 100644 --- a/kernel/rcu/tasks.h +++ b/kernel/rcu/tasks.h @@ -396,7 +396,7 @@ static void rcu_barrier_tasks_generic_cb(struct rcu_head *rhp) // Wait for all in-flight callbacks for the specified RCU Tasks flavor. // Operates in a manner similar to rcu_barrier(). -static void rcu_barrier_tasks_generic(struct rcu_tasks *rtp) +static void __maybe_unused rcu_barrier_tasks_generic(struct rcu_tasks *rtp) { int cpu; unsigned long flags; @@ -1244,13 +1244,12 @@ void exit_tasks_rcu_finish(void) { exit_tasks_rcu_finish_trace(current); } //////////////////////////////////////////////////////////////////////// // -// "Rude" variant of Tasks RCU, inspired by Steve Rostedt's trick of -// passing an empty function to schedule_on_each_cpu(). This approach -// provides an asynchronous call_rcu_tasks_rude() API and batching of -// concurrent calls to the synchronous synchronize_rcu_tasks_rude() API. -// This invokes schedule_on_each_cpu() in order to send IPIs far and wide -// and induces otherwise unnecessary context switches on all online CPUs, -// whether idle or not. +// "Rude" variant of Tasks RCU, inspired by Steve Rostedt's +// trick of passing an empty function to schedule_on_each_cpu(). +// This approach provides batching of concurrent calls to the synchronous +// synchronize_rcu_tasks_rude() API. This invokes schedule_on_each_cpu() +// in order to send IPIs far and wide and induces otherwise unnecessary +// context switches on all online CPUs, whether idle or not. // // Callback handling is provided by the rcu_tasks_kthread() function. // @@ -1268,11 +1267,11 @@ static void rcu_tasks_rude_wait_gp(struct rcu_tasks *rtp) schedule_on_each_cpu(rcu_tasks_be_rude); } -void call_rcu_tasks_rude(struct rcu_head *rhp, rcu_callback_t func); +static void call_rcu_tasks_rude(struct rcu_head *rhp, rcu_callback_t func); DEFINE_RCU_TASKS(rcu_tasks_rude, rcu_tasks_rude_wait_gp, call_rcu_tasks_rude, "RCU Tasks Rude"); -/** +/* * call_rcu_tasks_rude() - Queue a callback rude 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 @@ -1289,12 +1288,14 @@ DEFINE_RCU_TASKS(rcu_tasks_rude, rcu_tasks_rude_wait_gp, call_rcu_tasks_rude, * * See the description of call_rcu() for more detailed information on * memory ordering guarantees. + * + * This is no longer exported, and is instead reserved for use by + * synchronize_rcu_tasks_rude(). */ -void call_rcu_tasks_rude(struct rcu_head *rhp, rcu_callback_t func) +static void call_rcu_tasks_rude(struct rcu_head *rhp, rcu_callback_t func) { call_rcu_tasks_generic(rhp, func, &rcu_tasks_rude); } -EXPORT_SYMBOL_GPL(call_rcu_tasks_rude); /** * synchronize_rcu_tasks_rude - wait for a rude rcu-tasks grace period @@ -1320,26 +1321,9 @@ void synchronize_rcu_tasks_rude(void) } EXPORT_SYMBOL_GPL(synchronize_rcu_tasks_rude); -/** - * rcu_barrier_tasks_rude - Wait for in-flight call_rcu_tasks_rude() 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_rude(void) -{ - rcu_barrier_tasks_generic(&rcu_tasks_rude); -} -EXPORT_SYMBOL_GPL(rcu_barrier_tasks_rude); - -int rcu_tasks_rude_lazy_ms = -1; -module_param(rcu_tasks_rude_lazy_ms, int, 0444); - static int __init rcu_spawn_tasks_rude_kthread(void) { rcu_tasks_rude.gp_sleep = HZ / 10; - if (rcu_tasks_rude_lazy_ms >= 0) - rcu_tasks_rude.lazy_jiffies = msecs_to_jiffies(rcu_tasks_rude_lazy_ms); rcu_spawn_tasks_kthread_generic(&rcu_tasks_rude); return 0; } @@ -2069,11 +2053,6 @@ static struct rcu_tasks_test_desc tests[] = { /* If not defined, the test is skipped. */ .notrun = IS_ENABLED(CONFIG_TASKS_RCU), }, - { - .name = "call_rcu_tasks_rude()", - /* If not defined, the test is skipped. */ - .notrun = IS_ENABLED(CONFIG_TASKS_RUDE_RCU), - }, { .name = "call_rcu_tasks_trace()", /* If not defined, the test is skipped. */ @@ -2081,6 +2060,7 @@ static struct rcu_tasks_test_desc tests[] = { } }; +#if defined(CONFIG_TASKS_RCU) || defined(CONFIG_TASKS_TRACE_RCU) static void test_rcu_tasks_callback(struct rcu_head *rhp) { struct rcu_tasks_test_desc *rttd = @@ -2090,6 +2070,7 @@ static void test_rcu_tasks_callback(struct rcu_head *rhp) rttd->notrun = false; } +#endif // #if defined(CONFIG_TASKS_RCU) || defined(CONFIG_TASKS_TRACE_RCU) static void rcu_tasks_initiate_self_tests(void) { @@ -2102,16 +2083,14 @@ static void rcu_tasks_initiate_self_tests(void) #ifdef CONFIG_TASKS_RUDE_RCU pr_info("Running RCU Tasks Rude wait API self tests\n"); - tests[1].runstart = jiffies; synchronize_rcu_tasks_rude(); - call_rcu_tasks_rude(&tests[1].rh, test_rcu_tasks_callback); #endif #ifdef CONFIG_TASKS_TRACE_RCU pr_info("Running RCU Tasks Trace wait API self tests\n"); - tests[2].runstart = jiffies; + tests[1].runstart = jiffies; synchronize_rcu_tasks_trace(); - call_rcu_tasks_trace(&tests[2].rh, test_rcu_tasks_callback); + call_rcu_tasks_trace(&tests[1].rh, test_rcu_tasks_callback); #endif } -- cgit From fe91cf39db0939ac8d523b5a1c31840f7cbe205c Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 1 Aug 2024 17:34:25 -0700 Subject: rcu/tasks: Add detailed grace-period and barrier diagnostics This commit adds rcu_tasks_torture_stats_print(), rcu_tasks_trace_torture_stats_print(), and rcu_tasks_rude_torture_stats_print() functions that provide detailed diagnostics on grace-period, callback, and barrier state. Signed-off-by: "Paul E. McKenney" Signed-off-by: Neeraj Upadhyay --- include/linux/rcupdate.h | 3 +++ kernel/rcu/tasks.h | 66 ++++++++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 67 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 31e679c7110e..17463e95b6ef 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -164,6 +164,7 @@ static inline void rcu_nocb_flush_deferred_wakeup(void) { } } while (0) void call_rcu_tasks(struct rcu_head *head, rcu_callback_t func); void synchronize_rcu_tasks(void); +void rcu_tasks_torture_stats_print(char *tt, char *tf); # else # define rcu_tasks_classic_qs(t, preempt) do { } while (0) # define call_rcu_tasks call_rcu @@ -190,6 +191,7 @@ void rcu_tasks_trace_qs_blkd(struct task_struct *t); rcu_tasks_trace_qs_blkd(t); \ } \ } while (0) +void rcu_tasks_trace_torture_stats_print(char *tt, char *tf); # else # define rcu_tasks_trace_qs(t) do { } while (0) # endif @@ -202,6 +204,7 @@ do { \ # ifdef CONFIG_TASKS_RUDE_RCU void synchronize_rcu_tasks_rude(void); +void rcu_tasks_rude_torture_stats_print(char *tt, char *tf); # endif #define rcu_note_voluntary_context_switch(t) rcu_tasks_qs(t, false) diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h index d44abcd656d6..5f6d80ce1e47 100644 --- a/kernel/rcu/tasks.h +++ b/kernel/rcu/tasks.h @@ -714,9 +714,7 @@ static void __init rcu_tasks_bootup_oddness(void) #endif /* #ifdef CONFIG_TASKS_TRACE_RCU */ } -#endif /* #ifndef CONFIG_TINY_RCU */ -#ifndef CONFIG_TINY_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) { @@ -750,6 +748,52 @@ static void show_rcu_tasks_generic_gp_kthread(struct rcu_tasks *rtp, char *s) rtp->lazy_jiffies, s); } + +/* Dump out more rcutorture-relevant state common to all RCU-tasks flavors. */ +static void rcu_tasks_torture_stats_print_generic(struct rcu_tasks *rtp, char *tt, + char *tf, char *tst) +{ + cpumask_var_t cm; + int cpu; + bool gotcb = false; + unsigned long j = jiffies; + + pr_alert("%s%s Tasks%s RCU g%ld gp_start %lu gp_jiffies %lu gp_state %d (%s).\n", + tt, tf, tst, data_race(rtp->tasks_gp_seq), + j - data_race(rtp->gp_start), j - data_race(rtp->gp_jiffies), + data_race(rtp->gp_state), tasks_gp_state_getname(rtp)); + pr_alert("\tEnqueue shift %d limit %d Dequeue limit %d gpseq %lu.\n", + data_race(rtp->percpu_enqueue_shift), + data_race(rtp->percpu_enqueue_lim), + data_race(rtp->percpu_dequeue_lim), + data_race(rtp->percpu_dequeue_gpseq)); + (void)zalloc_cpumask_var(&cm, GFP_KERNEL); + pr_alert("\tCallback counts:"); + for_each_possible_cpu(cpu) { + long n; + struct rcu_tasks_percpu *rtpcp = per_cpu_ptr(rtp->rtpcpu, cpu); + + if (cpumask_available(cm) && !rcu_barrier_cb_is_done(&rtpcp->barrier_q_head)) + cpumask_set_cpu(cpu, cm); + n = rcu_segcblist_n_cbs(&rtpcp->cblist); + if (!n) + continue; + pr_cont(" %d:%ld", cpu, n); + gotcb = true; + } + if (gotcb) + pr_cont(".\n"); + else + pr_cont(" (none).\n"); + pr_alert("\tBarrier seq %lu count %d holdout CPUs ", + data_race(rtp->barrier_q_seq), atomic_read(&rtp->barrier_q_count)); + if (cpumask_available(cm) && !cpumask_empty(cm)) + pr_cont(" %*pbl.\n", cpumask_pr_args(cm)); + else + pr_cont("(none).\n"); + free_cpumask_var(cm); +} + #endif // #ifndef CONFIG_TINY_RCU static void exit_tasks_rcu_finish_trace(struct task_struct *t); @@ -1201,6 +1245,12 @@ void show_rcu_tasks_classic_gp_kthread(void) show_rcu_tasks_generic_gp_kthread(&rcu_tasks, ""); } EXPORT_SYMBOL_GPL(show_rcu_tasks_classic_gp_kthread); + +void rcu_tasks_torture_stats_print(char *tt, char *tf) +{ + rcu_tasks_torture_stats_print_generic(&rcu_tasks, tt, tf, ""); +} +EXPORT_SYMBOL_GPL(rcu_tasks_torture_stats_print); #endif // !defined(CONFIG_TINY_RCU) struct task_struct *get_rcu_tasks_gp_kthread(void) @@ -1361,6 +1411,12 @@ void show_rcu_tasks_rude_gp_kthread(void) show_rcu_tasks_generic_gp_kthread(&rcu_tasks_rude, ""); } EXPORT_SYMBOL_GPL(show_rcu_tasks_rude_gp_kthread); + +void rcu_tasks_rude_torture_stats_print(char *tt, char *tf) +{ + rcu_tasks_torture_stats_print_generic(&rcu_tasks_rude, tt, tf, ""); +} +EXPORT_SYMBOL_GPL(rcu_tasks_rude_torture_stats_print); #endif // !defined(CONFIG_TINY_RCU) struct task_struct *get_rcu_tasks_rude_gp_kthread(void) @@ -2038,6 +2094,12 @@ void show_rcu_tasks_trace_gp_kthread(void) 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) -- cgit From ac9d45544cd571decca395715d0b0a3b617d02f4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Mon, 1 Jul 2024 13:33:58 -0700 Subject: locking/csd_lock: Provide an indication of ongoing CSD-lock stall If a CSD-lock stall goes on long enough, it will cause an RCU CPU stall warning. This additional warning provides much additional console-log traffic and little additional information. Therefore, provide a new csd_lock_is_stuck() function that returns true if there is an ongoing CSD-lock stall. This function will be used by the RCU CPU stall warnings to provide a one-line indication of the stall when this function returns true. [ neeraj.upadhyay: Apply Rik van Riel feedback. ] [ neeraj.upadhyay: Apply kernel test robot feedback. ] Signed-off-by: Paul E. McKenney Cc: Imran Khan Cc: Ingo Molnar Cc: Leonardo Bras Cc: "Peter Zijlstra (Intel)" Cc: Rik van Riel Signed-off-by: Neeraj Upadhyay --- include/linux/smp.h | 6 ++++++ kernel/smp.c | 16 ++++++++++++++++ lib/Kconfig.debug | 1 + 3 files changed, 23 insertions(+) (limited to 'include/linux') diff --git a/include/linux/smp.h b/include/linux/smp.h index fcd61dfe2af3..3871bd32018f 100644 --- a/include/linux/smp.h +++ b/include/linux/smp.h @@ -294,4 +294,10 @@ int smpcfd_prepare_cpu(unsigned int cpu); int smpcfd_dead_cpu(unsigned int cpu); int smpcfd_dying_cpu(unsigned int cpu); +#ifdef CONFIG_CSD_LOCK_WAIT_DEBUG +bool csd_lock_is_stuck(void); +#else +static inline bool csd_lock_is_stuck(void) { return false; } +#endif + #endif /* __LINUX_SMP_H */ diff --git a/kernel/smp.c b/kernel/smp.c index e87953729230..202cda4d2a55 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -208,6 +208,19 @@ static int csd_lock_wait_getcpu(call_single_data_t *csd) return -1; } +static atomic_t n_csd_lock_stuck; + +/** + * csd_lock_is_stuck - Has a CSD-lock acquisition been stuck too long? + * + * Returns @true if a CSD-lock acquisition is stuck and has been stuck + * long enough for a "non-responsive CSD lock" message to be printed. + */ +bool csd_lock_is_stuck(void) +{ + return !!atomic_read(&n_csd_lock_stuck); +} + /* * Complain if too much time spent waiting. Note that only * the CSD_TYPE_SYNC/ASYNC types provide the destination CPU, @@ -229,6 +242,7 @@ static bool csd_lock_wait_toolong(call_single_data_t *csd, u64 ts0, u64 *ts1, in cpu = csd_lock_wait_getcpu(csd); pr_alert("csd: CSD lock (#%d) got unstuck on CPU#%02d, CPU#%02d released the lock.\n", *bug_id, raw_smp_processor_id(), cpu); + atomic_dec(&n_csd_lock_stuck); return true; } @@ -252,6 +266,8 @@ static bool csd_lock_wait_toolong(call_single_data_t *csd, u64 ts0, u64 *ts1, in pr_alert("csd: %s non-responsive CSD lock (#%d) on CPU#%d, waiting %lld ns for CPU#%02d %pS(%ps).\n", firsttime ? "Detected" : "Continued", *bug_id, raw_smp_processor_id(), (s64)ts_delta, cpu, csd->func, csd->info); + if (firsttime) + atomic_inc(&n_csd_lock_stuck); /* * If the CSD lock is still stuck after 5 minutes, it is unlikely * to become unstuck. Use a signed comparison to avoid triggering diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index a30c03a66172..4e5f61cba8e4 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -1614,6 +1614,7 @@ config SCF_TORTURE_TEST config CSD_LOCK_WAIT_DEBUG bool "Debugging for csd_lock_wait(), called from smp_call_function*()" depends on DEBUG_KERNEL + depends on SMP depends on 64BIT default n help -- cgit From fda70207135b60dd1a7c8f16cc036bb1cc344490 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Tue, 16 Apr 2024 11:01:23 +0200 Subject: context_tracking, rcu: Rename rcu_dynticks_curr_cpu_in_eqs() into rcu_is_watching_curr_cpu() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, reflect that change in the related helpers. Note that "watching" is the opposite of "in EQS", so the negation is lifted out of the helper and into the callsites. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking.h | 12 ++++++++---- kernel/context_tracking.c | 6 +++--- kernel/rcu/tree.c | 6 +++--- 3 files changed, 14 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking.h b/include/linux/context_tracking.h index a6c36780cc3b..d53092ffa9db 100644 --- a/include/linux/context_tracking.h +++ b/include/linux/context_tracking.h @@ -113,13 +113,17 @@ extern void ct_idle_enter(void); extern void ct_idle_exit(void); /* - * Is the current CPU in an extended quiescent state? + * Is RCU watching the current CPU (IOW, it is not in an extended quiescent state)? + * + * Note that this returns the actual boolean data (watching / not watching), + * whereas ct_rcu_watching() returns the RCU_WATCHING subvariable of + * context_tracking.state. * * No ordering, as we are sampling CPU-local information. */ -static __always_inline bool rcu_dynticks_curr_cpu_in_eqs(void) +static __always_inline bool rcu_is_watching_curr_cpu(void) { - return !(raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING); + return raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_RCU_WATCHING; } /* @@ -140,7 +144,7 @@ static __always_inline bool warn_rcu_enter(void) * lots of the actual reporting also relies on RCU. */ preempt_disable_notrace(); - if (rcu_dynticks_curr_cpu_in_eqs()) { + if (!rcu_is_watching_curr_cpu()) { ret = true; ct_state_inc(CT_RCU_WATCHING); } diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index 00c8f128885a..31c6d0c1b79d 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -212,7 +212,7 @@ void noinstr ct_nmi_exit(void) * to us!) */ WARN_ON_ONCE(ct_nmi_nesting() <= 0); - WARN_ON_ONCE(rcu_dynticks_curr_cpu_in_eqs()); + WARN_ON_ONCE(!rcu_is_watching_curr_cpu()); /* * If the nesting level is not 1, the CPU wasn't RCU-idle, so @@ -271,7 +271,7 @@ void noinstr ct_nmi_enter(void) * to be in the outermost NMI handler that interrupted an RCU-idle * period (observation due to Andy Lutomirski). */ - if (rcu_dynticks_curr_cpu_in_eqs()) { + if (!rcu_is_watching_curr_cpu()) { if (!in_nmi()) rcu_task_enter(); @@ -281,7 +281,7 @@ void noinstr ct_nmi_enter(void) // ... but is watching here. instrumentation_begin(); - // instrumentation for the noinstr rcu_dynticks_curr_cpu_in_eqs() + // instrumentation for the noinstr rcu_is_watching_curr_cpu() instrument_atomic_read(&ct->state, sizeof(ct->state)); // instrumentation for the noinstr ct_kernel_enter_state() instrument_atomic_write(&ct->state, sizeof(ct->state)); diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 4778c873f4ac..68eca7d3fdd1 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -601,7 +601,7 @@ void rcu_irq_exit_check_preempt(void) RCU_LOCKDEP_WARN(ct_nmi_nesting() != CT_NESTING_IRQ_NONIDLE, "Bad RCU nmi_nesting counter\n"); - RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), + RCU_LOCKDEP_WARN(!rcu_is_watching_curr_cpu(), "RCU in extended quiescent state!"); } #endif /* #ifdef CONFIG_PROVE_RCU */ @@ -641,7 +641,7 @@ void __rcu_irq_enter_check_tick(void) if (in_nmi()) return; - RCU_LOCKDEP_WARN(rcu_dynticks_curr_cpu_in_eqs(), + RCU_LOCKDEP_WARN(!rcu_is_watching_curr_cpu(), "Illegal rcu_irq_enter_check_tick() from extended quiescent state"); if (!tick_nohz_full_cpu(rdp->cpu) || @@ -723,7 +723,7 @@ notrace bool rcu_is_watching(void) bool ret; preempt_disable_notrace(); - ret = !rcu_dynticks_curr_cpu_in_eqs(); + ret = rcu_is_watching_curr_cpu(); preempt_enable_notrace(); return ret; } -- cgit From 32a9f26e5e266f0116cc8536e5e4544523ddd334 Mon Sep 17 00:00:00 2001 From: Valentin Schneider Date: Mon, 29 Apr 2024 16:43:36 +0200 Subject: rcu: Rename rcu_momentary_dyntick_idle() into rcu_momentary_eqs() The context_tracking.state RCU_DYNTICKS subvariable has been renamed to RCU_WATCHING, replace "dyntick_idle" into "eqs" to drop the dyntick reference. Signed-off-by: Valentin Schneider Reviewed-by: Frederic Weisbecker Signed-off-by: Neeraj Upadhyay --- include/linux/rcutiny.h | 2 +- include/linux/rcutree.h | 2 +- kernel/rcu/rcutorture.c | 4 ++-- kernel/rcu/tree.c | 4 ++-- kernel/rcu/tree_nocb.h | 2 +- kernel/rcu/tree_plugin.h | 6 +++--- kernel/stop_machine.c | 2 +- kernel/trace/trace_osnoise.c | 4 ++-- 8 files changed, 13 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h index d9ac7b136aea..cf2b5a188f78 100644 --- a/include/linux/rcutiny.h +++ b/include/linux/rcutiny.h @@ -158,7 +158,7 @@ void rcu_scheduler_starting(void); static inline void rcu_end_inkernel_boot(void) { } static inline bool rcu_inkernel_boot_has_ended(void) { return true; } static inline bool rcu_is_watching(void) { return true; } -static inline void rcu_momentary_dyntick_idle(void) { } +static inline void rcu_momentary_eqs(void) { } static inline void kfree_rcu_scheduler_running(void) { } static inline bool rcu_gp_might_be_stalled(void) { return false; } diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h index 254244202ea9..7dbde2b6f714 100644 --- a/include/linux/rcutree.h +++ b/include/linux/rcutree.h @@ -37,7 +37,7 @@ void synchronize_rcu_expedited(void); void kvfree_call_rcu(struct rcu_head *head, void *ptr); void rcu_barrier(void); -void rcu_momentary_dyntick_idle(void); +void rcu_momentary_eqs(void); void kfree_rcu_scheduler_running(void); bool rcu_gp_might_be_stalled(void); diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c index 08bf7c669dd3..ef757c3e84a2 100644 --- a/kernel/rcu/rcutorture.c +++ b/kernel/rcu/rcutorture.c @@ -2680,7 +2680,7 @@ static unsigned long rcu_torture_fwd_prog_cbfree(struct rcu_fwd *rfp) rcu_torture_fwd_prog_cond_resched(freed); if (tick_nohz_full_enabled()) { local_irq_save(flags); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); local_irq_restore(flags); } } @@ -2830,7 +2830,7 @@ static void rcu_torture_fwd_prog_cr(struct rcu_fwd *rfp) rcu_torture_fwd_prog_cond_resched(n_launders + n_max_cbs); if (tick_nohz_full_enabled()) { local_irq_save(flags); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); local_irq_restore(flags); } } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index b53b354417ae..e0f6fc272906 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -367,7 +367,7 @@ bool rcu_watching_zero_in_eqs(int cpu, int *vp) * * The caller must have disabled interrupts and must not be idle. */ -notrace void rcu_momentary_dyntick_idle(void) +notrace void rcu_momentary_eqs(void) { int seq; @@ -377,7 +377,7 @@ notrace void rcu_momentary_dyntick_idle(void) WARN_ON_ONCE(!(seq & CT_RCU_WATCHING)); rcu_preempt_deferred_qs(current); } -EXPORT_SYMBOL_GPL(rcu_momentary_dyntick_idle); +EXPORT_SYMBOL_GPL(rcu_momentary_eqs); /** * rcu_is_cpu_rrupt_from_idle - see if 'interrupted' from idle diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 3ce30841119a..e629646ac6f5 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -917,7 +917,7 @@ static void nocb_cb_wait(struct rcu_data *rdp) WARN_ON_ONCE(!rcu_rdp_is_offloaded(rdp)); local_irq_save(flags); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); local_irq_restore(flags); /* * Disable BH to provide the expected environment. Also, when diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index c569da65b421..354fbf2e411c 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -869,7 +869,7 @@ static void rcu_qs(void) /* * Register an urgently needed quiescent state. If there is an - * emergency, invoke rcu_momentary_dyntick_idle() to do a heavy-weight + * emergency, invoke rcu_momentary_eqs() to do a heavy-weight * dyntick-idle quiescent state visible to other CPUs, which will in * some cases serve for expedited as well as normal grace periods. * Either way, register a lightweight quiescent state. @@ -889,7 +889,7 @@ void rcu_all_qs(void) this_cpu_write(rcu_data.rcu_urgent_qs, false); if (unlikely(raw_cpu_read(rcu_data.rcu_need_heavy_qs))) { local_irq_save(flags); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); local_irq_restore(flags); } rcu_qs(); @@ -909,7 +909,7 @@ void rcu_note_context_switch(bool preempt) goto out; this_cpu_write(rcu_data.rcu_urgent_qs, false); if (unlikely(raw_cpu_read(rcu_data.rcu_need_heavy_qs))) - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); out: rcu_tasks_qs(current, preempt); trace_rcu_utilization(TPS("End context switch")); diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index cedb17ba158a..da821ce258ea 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -251,7 +251,7 @@ static int multi_cpu_stop(void *data) */ touch_nmi_watchdog(); } - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); } while (curstate != MULTI_STOP_EXIT); local_irq_restore(flags); diff --git a/kernel/trace/trace_osnoise.c b/kernel/trace/trace_osnoise.c index 66a871553d4a..ad86b1ba2e43 100644 --- a/kernel/trace/trace_osnoise.c +++ b/kernel/trace/trace_osnoise.c @@ -1535,7 +1535,7 @@ static int run_osnoise(void) * This will eventually cause unwarranted noise as PREEMPT_RCU * will force preemption as the means of ending the current * grace period. We avoid this problem by calling - * rcu_momentary_dyntick_idle(), which performs a zero duration + * rcu_momentary_eqs(), which performs a zero duration * EQS allowing PREEMPT_RCU to end the current grace period. * This call shouldn't be wrapped inside an RCU critical * section. @@ -1547,7 +1547,7 @@ static int run_osnoise(void) if (!disable_irq) local_irq_disable(); - rcu_momentary_dyntick_idle(); + rcu_momentary_eqs(); if (!disable_irq) local_irq_enable(); -- cgit From 4040b1139904b3f72b56862fc9a8d3e9abb69ffb Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Mon, 17 Jun 2024 08:51:14 -0700 Subject: context_tracking: Tag context_tracking_enabled_this_cpu() __always_inline Force context_tracking_enabled_this_cpu() to be inlined so that invoking it from guest_context_enter_irqoff(), which KVM uses in non-instrumentable code, doesn't unexpectedly leave a noinstr section. vmlinux.o: warning: objtool: vmx_vcpu_enter_exit+0x1c7: call to context_tracking_enabled_this_cpu() leaves .noinstr.text section vmlinux.o: warning: objtool: svm_vcpu_enter_exit+0x83: call to context_tracking_enabled_this_cpu() leaves .noinstr.text section Note, the CONFIG_CONTEXT_TRACKING_USER=n stub is already __always_inline. Signed-off-by: Sean Christopherson Signed-off-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/context_tracking_state.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index 0dbda59c9f37..7b8433d5a8ef 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -113,7 +113,7 @@ static __always_inline bool context_tracking_enabled_cpu(int cpu) return context_tracking_enabled() && per_cpu(context_tracking.active, cpu); } -static inline bool context_tracking_enabled_this_cpu(void) +static __always_inline bool context_tracking_enabled_this_cpu(void) { return context_tracking_enabled() && __this_cpu_read(context_tracking.active); } -- cgit From 1fcb932c8b5ce86219d7dedcd63659351a43291c Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 4 Jul 2024 00:56:40 +0200 Subject: rcu/nocb: Simplify (de-)offloading state machine Now that the (de-)offloading process can only apply to offline CPUs, there is no more concurrency between rcu_core and nocb kthreads. Also the mutation now happens on empty queues. Therefore the state machine can be reduced to a single bit called SEGCBLIST_OFFLOADED. Simplify the transition as follows: * Upon offloading: queue the rdp to be added to the rcuog list and wait for the rcuog kthread to set the SEGCBLIST_OFFLOADED bit. Unpark rcuo kthread. * Upon de-offloading: Park rcuo kthread. Queue the rdp to be removed from the rcuog list and wait for the rcuog kthread to clear the SEGCBLIST_OFFLOADED bit. Signed-off-by: Frederic Weisbecker Signed-off-by: Paul E. McKenney Reviewed-by: Paul E. McKenney Signed-off-by: Neeraj Upadhyay --- include/linux/rcu_segcblist.h | 4 +- kernel/rcu/rcu_segcblist.c | 11 ---- kernel/rcu/rcu_segcblist.h | 2 +- kernel/rcu/tree_nocb.h | 138 ++++++++++++++++++++---------------------- 4 files changed, 68 insertions(+), 87 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcu_segcblist.h b/include/linux/rcu_segcblist.h index 1ef1bb54853d..2fdc2208f1ca 100644 --- a/include/linux/rcu_segcblist.h +++ b/include/linux/rcu_segcblist.h @@ -185,9 +185,7 @@ struct rcu_cblist { * ---------------------------------------------------------------------------- */ #define SEGCBLIST_ENABLED BIT(0) -#define SEGCBLIST_LOCKING BIT(1) -#define SEGCBLIST_KTHREAD_GP BIT(2) -#define SEGCBLIST_OFFLOADED BIT(3) +#define SEGCBLIST_OFFLOADED BIT(1) struct rcu_segcblist { struct rcu_head *head; diff --git a/kernel/rcu/rcu_segcblist.c b/kernel/rcu/rcu_segcblist.c index 1693ea22ef1b..298a2c573f02 100644 --- a/kernel/rcu/rcu_segcblist.c +++ b/kernel/rcu/rcu_segcblist.c @@ -260,17 +260,6 @@ void rcu_segcblist_disable(struct rcu_segcblist *rsclp) rcu_segcblist_clear_flags(rsclp, SEGCBLIST_ENABLED); } -/* - * Mark the specified rcu_segcblist structure as offloaded (or not) - */ -void rcu_segcblist_offload(struct rcu_segcblist *rsclp, bool offload) -{ - if (offload) - rcu_segcblist_set_flags(rsclp, SEGCBLIST_LOCKING | SEGCBLIST_OFFLOADED); - else - rcu_segcblist_clear_flags(rsclp, SEGCBLIST_OFFLOADED); -} - /* * Does the specified rcu_segcblist structure contain callbacks that * are ready to be invoked? diff --git a/kernel/rcu/rcu_segcblist.h b/kernel/rcu/rcu_segcblist.h index 7a0962dfee86..259904075636 100644 --- a/kernel/rcu/rcu_segcblist.h +++ b/kernel/rcu/rcu_segcblist.h @@ -89,7 +89,7 @@ static inline bool rcu_segcblist_is_enabled(struct rcu_segcblist *rsclp) static inline bool rcu_segcblist_is_offloaded(struct rcu_segcblist *rsclp) { if (IS_ENABLED(CONFIG_RCU_NOCB_CPU) && - rcu_segcblist_test_flags(rsclp, SEGCBLIST_LOCKING)) + rcu_segcblist_test_flags(rsclp, SEGCBLIST_OFFLOADED)) return true; return false; diff --git a/kernel/rcu/tree_nocb.h b/kernel/rcu/tree_nocb.h index 24daf606de0c..857024ee0349 100644 --- a/kernel/rcu/tree_nocb.h +++ b/kernel/rcu/tree_nocb.h @@ -604,37 +604,33 @@ static void call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *head, } } -static int nocb_gp_toggle_rdp(struct rcu_data *rdp) +static void nocb_gp_toggle_rdp(struct rcu_data *rdp_gp, struct rcu_data *rdp) { struct rcu_segcblist *cblist = &rdp->cblist; unsigned long flags; - int ret; - rcu_nocb_lock_irqsave(rdp, flags); - if (rcu_segcblist_test_flags(cblist, SEGCBLIST_OFFLOADED) && - !rcu_segcblist_test_flags(cblist, SEGCBLIST_KTHREAD_GP)) { + /* + * Locking orders future de-offloaded callbacks enqueue against previous + * handling of this rdp. Ie: Make sure rcuog is done with this rdp before + * deoffloaded callbacks can be enqueued. + */ + raw_spin_lock_irqsave(&rdp->nocb_lock, flags); + if (!rcu_segcblist_test_flags(cblist, SEGCBLIST_OFFLOADED)) { /* * Offloading. Set our flag and notify the offload worker. * We will handle this rdp until it ever gets de-offloaded. */ - rcu_segcblist_set_flags(cblist, SEGCBLIST_KTHREAD_GP); - ret = 1; - } else if (!rcu_segcblist_test_flags(cblist, SEGCBLIST_OFFLOADED) && - rcu_segcblist_test_flags(cblist, SEGCBLIST_KTHREAD_GP)) { + list_add_tail(&rdp->nocb_entry_rdp, &rdp_gp->nocb_head_rdp); + rcu_segcblist_set_flags(cblist, SEGCBLIST_OFFLOADED); + } else { /* * De-offloading. Clear our flag and notify the de-offload worker. * We will ignore this rdp until it ever gets re-offloaded. */ - rcu_segcblist_clear_flags(cblist, SEGCBLIST_KTHREAD_GP); - ret = 0; - } else { - WARN_ON_ONCE(1); - ret = -1; + list_del(&rdp->nocb_entry_rdp); + rcu_segcblist_clear_flags(cblist, SEGCBLIST_OFFLOADED); } - - rcu_nocb_unlock_irqrestore(rdp, flags); - - return ret; + raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); } static void nocb_gp_sleep(struct rcu_data *my_rdp, int cpu) @@ -841,14 +837,7 @@ static void nocb_gp_wait(struct rcu_data *my_rdp) } if (rdp_toggling) { - int ret; - - ret = nocb_gp_toggle_rdp(rdp_toggling); - if (ret == 1) - list_add_tail(&rdp_toggling->nocb_entry_rdp, &my_rdp->nocb_head_rdp); - else if (ret == 0) - list_del(&rdp_toggling->nocb_entry_rdp); - + nocb_gp_toggle_rdp(my_rdp, rdp_toggling); swake_up_one(&rdp_toggling->nocb_state_wq); } @@ -1018,16 +1007,11 @@ void rcu_nocb_flush_deferred_wakeup(void) } EXPORT_SYMBOL_GPL(rcu_nocb_flush_deferred_wakeup); -static int rdp_offload_toggle(struct rcu_data *rdp, - bool offload, unsigned long flags) - __releases(rdp->nocb_lock) +static int rcu_nocb_queue_toggle_rdp(struct rcu_data *rdp) { - struct rcu_segcblist *cblist = &rdp->cblist; struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; bool wake_gp = false; - - rcu_segcblist_offload(cblist, offload); - rcu_nocb_unlock_irqrestore(rdp, flags); + unsigned long flags; raw_spin_lock_irqsave(&rdp_gp->nocb_gp_lock, flags); // Queue this rdp for add/del to/from the list to iterate on rcuog @@ -1041,9 +1025,25 @@ static int rdp_offload_toggle(struct rcu_data *rdp, return wake_gp; } +static bool rcu_nocb_rdp_deoffload_wait_cond(struct rcu_data *rdp) +{ + unsigned long flags; + bool ret; + + /* + * Locking makes sure rcuog is done handling this rdp before deoffloaded + * enqueue can happen. Also it keeps the SEGCBLIST_OFFLOADED flag stable + * while the ->nocb_lock is held. + */ + raw_spin_lock_irqsave(&rdp->nocb_lock, flags); + ret = !rcu_segcblist_test_flags(&rdp->cblist, SEGCBLIST_OFFLOADED); + raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); + + return ret; +} + static int rcu_nocb_rdp_deoffload(struct rcu_data *rdp) { - struct rcu_segcblist *cblist = &rdp->cblist; unsigned long flags; int wake_gp; struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; @@ -1056,51 +1056,42 @@ static int rcu_nocb_rdp_deoffload(struct rcu_data *rdp) /* Flush all callbacks from segcblist and bypass */ rcu_barrier(); + /* + * Make sure the rcuoc kthread isn't in the middle of a nocb locked + * sequence while offloading is deactivated, along with nocb locking. + */ + if (rdp->nocb_cb_kthread) + kthread_park(rdp->nocb_cb_kthread); + rcu_nocb_lock_irqsave(rdp, flags); WARN_ON_ONCE(rcu_cblist_n_cbs(&rdp->nocb_bypass)); WARN_ON_ONCE(rcu_segcblist_n_cbs(&rdp->cblist)); + rcu_nocb_unlock_irqrestore(rdp, flags); - wake_gp = rdp_offload_toggle(rdp, false, flags); + wake_gp = rcu_nocb_queue_toggle_rdp(rdp); mutex_lock(&rdp_gp->nocb_gp_kthread_mutex); + if (rdp_gp->nocb_gp_kthread) { if (wake_gp) wake_up_process(rdp_gp->nocb_gp_kthread); swait_event_exclusive(rdp->nocb_state_wq, - !rcu_segcblist_test_flags(cblist, - SEGCBLIST_KTHREAD_GP)); - if (rdp->nocb_cb_kthread) - kthread_park(rdp->nocb_cb_kthread); + rcu_nocb_rdp_deoffload_wait_cond(rdp)); } else { /* * No kthread to clear the flags for us or remove the rdp from the nocb list * to iterate. Do it here instead. Locking doesn't look stricly necessary * but we stick to paranoia in this rare path. */ - rcu_nocb_lock_irqsave(rdp, flags); - rcu_segcblist_clear_flags(&rdp->cblist, SEGCBLIST_KTHREAD_GP); - rcu_nocb_unlock_irqrestore(rdp, flags); + raw_spin_lock_irqsave(&rdp->nocb_lock, flags); + rcu_segcblist_clear_flags(&rdp->cblist, SEGCBLIST_OFFLOADED); + raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); list_del(&rdp->nocb_entry_rdp); } - mutex_unlock(&rdp_gp->nocb_gp_kthread_mutex); - /* - * Lock one last time to acquire latest callback updates from kthreads - * so we can later handle callbacks locally without locking. - */ - rcu_nocb_lock_irqsave(rdp, flags); - /* - * Theoretically we could clear SEGCBLIST_LOCKING after the nocb - * lock is released but how about being paranoid for once? - */ - rcu_segcblist_clear_flags(cblist, SEGCBLIST_LOCKING); - /* - * Without SEGCBLIST_LOCKING, we can't use - * rcu_nocb_unlock_irqrestore() anymore. - */ - raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); + mutex_unlock(&rdp_gp->nocb_gp_kthread_mutex); return 0; } @@ -1129,10 +1120,20 @@ int rcu_nocb_cpu_deoffload(int cpu) } EXPORT_SYMBOL_GPL(rcu_nocb_cpu_deoffload); -static int rcu_nocb_rdp_offload(struct rcu_data *rdp) +static bool rcu_nocb_rdp_offload_wait_cond(struct rcu_data *rdp) { - struct rcu_segcblist *cblist = &rdp->cblist; unsigned long flags; + bool ret; + + raw_spin_lock_irqsave(&rdp->nocb_lock, flags); + ret = rcu_segcblist_test_flags(&rdp->cblist, SEGCBLIST_OFFLOADED); + raw_spin_unlock_irqrestore(&rdp->nocb_lock, flags); + + return ret; +} + +static int rcu_nocb_rdp_offload(struct rcu_data *rdp) +{ int wake_gp; struct rcu_data *rdp_gp = rdp->nocb_gp_rdp; @@ -1152,20 +1153,14 @@ static int rcu_nocb_rdp_offload(struct rcu_data *rdp) WARN_ON_ONCE(rcu_cblist_n_cbs(&rdp->nocb_bypass)); WARN_ON_ONCE(rcu_segcblist_n_cbs(&rdp->cblist)); - /* - * Can't use rcu_nocb_lock_irqsave() before SEGCBLIST_LOCKING - * is set. - */ - raw_spin_lock_irqsave(&rdp->nocb_lock, flags); - - wake_gp = rdp_offload_toggle(rdp, true, flags); + wake_gp = rcu_nocb_queue_toggle_rdp(rdp); if (wake_gp) wake_up_process(rdp_gp->nocb_gp_kthread); - kthread_unpark(rdp->nocb_cb_kthread); - swait_event_exclusive(rdp->nocb_state_wq, - rcu_segcblist_test_flags(cblist, SEGCBLIST_KTHREAD_GP)); + rcu_nocb_rdp_offload_wait_cond(rdp)); + + kthread_unpark(rdp->nocb_cb_kthread); return 0; } @@ -1340,8 +1335,7 @@ void __init rcu_init_nohz(void) rdp = per_cpu_ptr(&rcu_data, cpu); if (rcu_segcblist_empty(&rdp->cblist)) rcu_segcblist_init(&rdp->cblist); - rcu_segcblist_offload(&rdp->cblist, true); - rcu_segcblist_set_flags(&rdp->cblist, SEGCBLIST_KTHREAD_GP); + rcu_segcblist_set_flags(&rdp->cblist, SEGCBLIST_OFFLOADED); } rcu_organize_nocb_kthreads(); } -- cgit