linux/kernel/rcu/tree_exp.h
Uladzislau Rezki (Sony) 988f569ae0 rcu: Reduce synchronize_rcu() latency
A call to a synchronize_rcu() can be optimized from a latency
point of view. Workloads which depend on this can benefit of it.

The delay of wakeme_after_rcu() callback, which unblocks a waiter,
depends on several factors:

- how fast a process of offloading is started. Combination of:
    - !CONFIG_RCU_NOCB_CPU/CONFIG_RCU_NOCB_CPU;
    - !CONFIG_RCU_LAZY/CONFIG_RCU_LAZY;
    - other.
- when started, invoking path is interrupted due to:
    - time limit;
    - need_resched();
    - if limit is reached.
- where in a nocb list it is located;
- how fast previous callbacks completed;

Example:

1. On our embedded devices i can easily trigger the scenario when
it is a last in the list out of ~3600 callbacks:

<snip>
  <...>-29      [001] d..1. 21950.145313: rcu_batch_start: rcu_preempt CBs=3613 bl=28
...
  <...>-29      [001] ..... 21950.152578: rcu_invoke_callback: rcu_preempt rhp=00000000b2d6dee8 func=__free_vm_area_struct.cfi_jt
  <...>-29      [001] ..... 21950.152579: rcu_invoke_callback: rcu_preempt rhp=00000000a446f607 func=__free_vm_area_struct.cfi_jt
  <...>-29      [001] ..... 21950.152580: rcu_invoke_callback: rcu_preempt rhp=00000000a5cab03b func=__free_vm_area_struct.cfi_jt
  <...>-29      [001] ..... 21950.152581: rcu_invoke_callback: rcu_preempt rhp=0000000013b7e5ee func=__free_vm_area_struct.cfi_jt
  <...>-29      [001] ..... 21950.152582: rcu_invoke_callback: rcu_preempt rhp=000000000a8ca6f9 func=__free_vm_area_struct.cfi_jt
  <...>-29      [001] ..... 21950.152583: rcu_invoke_callback: rcu_preempt rhp=000000008f162ca8 func=wakeme_after_rcu.cfi_jt
  <...>-29      [001] d..1. 21950.152625: rcu_batch_end: rcu_preempt CBs-invoked=3612 idle=....
<snip>

2. We use cpuset/cgroup to classify tasks and assign them into
different cgroups. For example "backgrond" group which binds tasks
only to little CPUs or "foreground" which makes use of all CPUs.
Tasks can be migrated between groups by a request if an acceleration
is needed.

See below an example how "surfaceflinger" task gets migrated.
Initially it is located in the "system-background" cgroup which
allows to run only on little cores. In order to speed it up it
can be temporary moved into "foreground" cgroup which allows
to use big/all CPUs:

cgroup_attach_task():
 -> cgroup_migrate_execute()
   -> cpuset_can_attach()
     -> percpu_down_write()
       -> rcu_sync_enter()
         -> synchronize_rcu()
   -> now move tasks to the new cgroup.
 -> cgroup_migrate_finish()

<snip>
         rcuop/1-29      [000] .....  7030.528570: rcu_invoke_callback: rcu_preempt rhp=00000000461605e0 func=wakeme_after_rcu.cfi_jt
    PERFD-SERVER-1855    [000] d..1.  7030.530293: cgroup_attach_task: dst_root=3 dst_id=22 dst_level=1 dst_path=/foreground pid=1900 comm=surfaceflinger
   TimerDispatch-2768    [002] d..5.  7030.537542: sched_migrate_task: comm=surfaceflinger pid=1900 prio=98 orig_cpu=0 dest_cpu=4
<snip>

"Boosting a task" depends on synchronize_rcu() latency:

- first trace shows a completion of synchronize_rcu();
- second shows attaching a task to a new group;
- last shows a final step when migration occurs.

3. To address this drawback, maintain a separate track that consists
of synchronize_rcu() callers only. After completion of a grace period
users are deferred to a dedicated worker to process requests.

4. This patch reduces the latency of synchronize_rcu() approximately
by ~30-40% on synthetic tests. The real test case, camera launch time,
shows(time is in milliseconds):

1-run 542 vs 489 improvement 9%
2-run 540 vs 466 improvement 13%
3-run 518 vs 468 improvement 9%
4-run 531 vs 457 improvement 13%
5-run 548 vs 475 improvement 13%
6-run 509 vs 484 improvement 4%

Synthetic test(no "noise" from other callbacks):
Hardware: x86_64 64 CPUs, 64GB of memory
Linux-6.6

- 10K tasks(simultaneous);
- each task does(1000 loops)
     synchronize_rcu();
     kfree(p);

default: CONFIG_RCU_NOCB_CPU: takes 54 seconds to complete all users;
patch: CONFIG_RCU_NOCB_CPU: takes 35 seconds to complete all users.

Running 60K gives approximately same results on my setup. Please note
it is without any interaction with another type of callbacks, otherwise
it will impact a lot a default case.

5. By default it is disabled. To enable this perform one of the
below sequence:

echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp
or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1"

Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Co-developed-by: Neeraj Upadhyay (AMD) <neeraj.iitr10@gmail.com>
Signed-off-by: Neeraj Upadhyay (AMD) <neeraj.iitr10@gmail.com>
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
2024-04-15 19:47:49 +02:00

1098 lines
34 KiB
C

/* SPDX-License-Identifier: GPL-2.0+ */
/*
* RCU expedited grace periods
*
* Copyright IBM Corporation, 2016
*
* Authors: Paul E. McKenney <paulmck@linux.ibm.com>
*/
#include <linux/lockdep.h>
static void rcu_exp_handler(void *unused);
static int rcu_print_task_exp_stall(struct rcu_node *rnp);
static void rcu_exp_print_detail_task_stall_rnp(struct rcu_node *rnp);
/*
* Record the start of an expedited grace period.
*/
static void rcu_exp_gp_seq_start(void)
{
rcu_seq_start(&rcu_state.expedited_sequence);
rcu_poll_gp_seq_start_unlocked(&rcu_state.gp_seq_polled_exp_snap);
}
/*
* Return the value that the expedited-grace-period counter will have
* at the end of the current grace period.
*/
static __maybe_unused unsigned long rcu_exp_gp_seq_endval(void)
{
return rcu_seq_endval(&rcu_state.expedited_sequence);
}
/*
* Record the end of an expedited grace period.
*/
static void rcu_exp_gp_seq_end(void)
{
rcu_poll_gp_seq_end_unlocked(&rcu_state.gp_seq_polled_exp_snap);
rcu_seq_end(&rcu_state.expedited_sequence);
smp_mb(); /* Ensure that consecutive grace periods serialize. */
}
/*
* Take a snapshot of the expedited-grace-period counter, which is the
* earliest value that will indicate that a full grace period has
* elapsed since the current time.
*/
static unsigned long rcu_exp_gp_seq_snap(void)
{
unsigned long s;
smp_mb(); /* Caller's modifications seen first by other CPUs. */
s = rcu_seq_snap(&rcu_state.expedited_sequence);
trace_rcu_exp_grace_period(rcu_state.name, s, TPS("snap"));
return s;
}
/*
* Given a counter snapshot from rcu_exp_gp_seq_snap(), return true
* if a full expedited grace period has elapsed since that snapshot
* was taken.
*/
static bool rcu_exp_gp_seq_done(unsigned long s)
{
return rcu_seq_done(&rcu_state.expedited_sequence, s);
}
/*
* Reset the ->expmaskinit values in the rcu_node tree to reflect any
* recent CPU-online activity. Note that these masks are not cleared
* when CPUs go offline, so they reflect the union of all CPUs that have
* ever been online. This means that this function normally takes its
* no-work-to-do fastpath.
*/
static void sync_exp_reset_tree_hotplug(void)
{
bool done;
unsigned long flags;
unsigned long mask;
unsigned long oldmask;
int ncpus = smp_load_acquire(&rcu_state.ncpus); /* Order vs. locking. */
struct rcu_node *rnp;
struct rcu_node *rnp_up;
/* If no new CPUs onlined since last time, nothing to do. */
if (likely(ncpus == rcu_state.ncpus_snap))
return;
rcu_state.ncpus_snap = ncpus;
/*
* Each pass through the following loop propagates newly onlined
* CPUs for the current rcu_node structure up the rcu_node tree.
*/
rcu_for_each_leaf_node(rnp) {
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if (rnp->expmaskinit == rnp->expmaskinitnext) {
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
continue; /* No new CPUs, nothing to do. */
}
/* Update this node's mask, track old value for propagation. */
oldmask = rnp->expmaskinit;
rnp->expmaskinit = rnp->expmaskinitnext;
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
/* If was already nonzero, nothing to propagate. */
if (oldmask)
continue;
/* Propagate the new CPU up the tree. */
mask = rnp->grpmask;
rnp_up = rnp->parent;
done = false;
while (rnp_up) {
raw_spin_lock_irqsave_rcu_node(rnp_up, flags);
if (rnp_up->expmaskinit)
done = true;
rnp_up->expmaskinit |= mask;
raw_spin_unlock_irqrestore_rcu_node(rnp_up, flags);
if (done)
break;
mask = rnp_up->grpmask;
rnp_up = rnp_up->parent;
}
}
}
/*
* Reset the ->expmask values in the rcu_node tree in preparation for
* a new expedited grace period.
*/
static void __maybe_unused sync_exp_reset_tree(void)
{
unsigned long flags;
struct rcu_node *rnp;
sync_exp_reset_tree_hotplug();
rcu_for_each_node_breadth_first(rnp) {
raw_spin_lock_irqsave_rcu_node(rnp, flags);
WARN_ON_ONCE(rnp->expmask);
WRITE_ONCE(rnp->expmask, rnp->expmaskinit);
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
}
}
/*
* Return non-zero if there is no RCU expedited grace period in progress
* for the specified rcu_node structure, in other words, if all CPUs and
* tasks covered by the specified rcu_node structure have done their bit
* for the current expedited grace period.
*/
static bool sync_rcu_exp_done(struct rcu_node *rnp)
{
raw_lockdep_assert_held_rcu_node(rnp);
return READ_ONCE(rnp->exp_tasks) == NULL &&
READ_ONCE(rnp->expmask) == 0;
}
/*
* Like sync_rcu_exp_done(), but where the caller does not hold the
* rcu_node's ->lock.
*/
static bool sync_rcu_exp_done_unlocked(struct rcu_node *rnp)
{
unsigned long flags;
bool ret;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
ret = sync_rcu_exp_done(rnp);
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return ret;
}
/*
* Report the exit from RCU read-side critical section for the last task
* that queued itself during or before the current expedited preemptible-RCU
* grace period. This event is reported either to the rcu_node structure on
* which the task was queued or to one of that rcu_node structure's ancestors,
* recursively up the tree. (Calm down, calm down, we do the recursion
* iteratively!)
*/
static void __rcu_report_exp_rnp(struct rcu_node *rnp,
bool wake, unsigned long flags)
__releases(rnp->lock)
{
unsigned long mask;
raw_lockdep_assert_held_rcu_node(rnp);
for (;;) {
if (!sync_rcu_exp_done(rnp)) {
if (!rnp->expmask)
rcu_initiate_boost(rnp, flags);
else
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
break;
}
if (rnp->parent == NULL) {
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
if (wake)
swake_up_one_online(&rcu_state.expedited_wq);
break;
}
mask = rnp->grpmask;
raw_spin_unlock_rcu_node(rnp); /* irqs remain disabled */
rnp = rnp->parent;
raw_spin_lock_rcu_node(rnp); /* irqs already disabled */
WARN_ON_ONCE(!(rnp->expmask & mask));
WRITE_ONCE(rnp->expmask, rnp->expmask & ~mask);
}
}
/*
* Report expedited quiescent state for specified node. This is a
* lock-acquisition wrapper function for __rcu_report_exp_rnp().
*/
static void __maybe_unused rcu_report_exp_rnp(struct rcu_node *rnp, bool wake)
{
unsigned long flags;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
__rcu_report_exp_rnp(rnp, wake, flags);
}
/*
* Report expedited quiescent state for multiple CPUs, all covered by the
* specified leaf rcu_node structure.
*/
static void rcu_report_exp_cpu_mult(struct rcu_node *rnp,
unsigned long mask, bool wake)
{
int cpu;
unsigned long flags;
struct rcu_data *rdp;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if (!(rnp->expmask & mask)) {
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return;
}
WRITE_ONCE(rnp->expmask, rnp->expmask & ~mask);
for_each_leaf_node_cpu_mask(rnp, cpu, mask) {
rdp = per_cpu_ptr(&rcu_data, cpu);
if (!IS_ENABLED(CONFIG_NO_HZ_FULL) || !rdp->rcu_forced_tick_exp)
continue;
rdp->rcu_forced_tick_exp = false;
tick_dep_clear_cpu(cpu, TICK_DEP_BIT_RCU_EXP);
}
__rcu_report_exp_rnp(rnp, wake, flags); /* Releases rnp->lock. */
}
/*
* Report expedited quiescent state for specified rcu_data (CPU).
*/
static void rcu_report_exp_rdp(struct rcu_data *rdp)
{
WRITE_ONCE(rdp->cpu_no_qs.b.exp, false);
rcu_report_exp_cpu_mult(rdp->mynode, rdp->grpmask, true);
}
/* Common code for work-done checking. */
static bool sync_exp_work_done(unsigned long s)
{
if (rcu_exp_gp_seq_done(s)) {
trace_rcu_exp_grace_period(rcu_state.name, s, TPS("done"));
smp_mb(); /* Ensure test happens before caller kfree(). */
return true;
}
return false;
}
/*
* Funnel-lock acquisition for expedited grace periods. Returns true
* if some other task completed an expedited grace period that this task
* can piggy-back on, and with no mutex held. Otherwise, returns false
* with the mutex held, indicating that the caller must actually do the
* expedited grace period.
*/
static bool exp_funnel_lock(unsigned long s)
{
struct rcu_data *rdp = per_cpu_ptr(&rcu_data, raw_smp_processor_id());
struct rcu_node *rnp = rdp->mynode;
struct rcu_node *rnp_root = rcu_get_root();
/* Low-contention fastpath. */
if (ULONG_CMP_LT(READ_ONCE(rnp->exp_seq_rq), s) &&
(rnp == rnp_root ||
ULONG_CMP_LT(READ_ONCE(rnp_root->exp_seq_rq), s)) &&
mutex_trylock(&rcu_state.exp_mutex))
goto fastpath;
/*
* Each pass through the following loop works its way up
* the rcu_node tree, returning if others have done the work or
* otherwise falls through to acquire ->exp_mutex. The mapping
* from CPU to rcu_node structure can be inexact, as it is just
* promoting locality and is not strictly needed for correctness.
*/
for (; rnp != NULL; rnp = rnp->parent) {
if (sync_exp_work_done(s))
return true;
/* Work not done, either wait here or go up. */
spin_lock(&rnp->exp_lock);
if (ULONG_CMP_GE(rnp->exp_seq_rq, s)) {
/* Someone else doing GP, so wait for them. */
spin_unlock(&rnp->exp_lock);
trace_rcu_exp_funnel_lock(rcu_state.name, rnp->level,
rnp->grplo, rnp->grphi,
TPS("wait"));
wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
sync_exp_work_done(s));
return true;
}
WRITE_ONCE(rnp->exp_seq_rq, s); /* Followers can wait on us. */
spin_unlock(&rnp->exp_lock);
trace_rcu_exp_funnel_lock(rcu_state.name, rnp->level,
rnp->grplo, rnp->grphi, TPS("nxtlvl"));
}
mutex_lock(&rcu_state.exp_mutex);
fastpath:
if (sync_exp_work_done(s)) {
mutex_unlock(&rcu_state.exp_mutex);
return true;
}
rcu_exp_gp_seq_start();
trace_rcu_exp_grace_period(rcu_state.name, s, TPS("start"));
return false;
}
/*
* Select the CPUs within the specified rcu_node that the upcoming
* expedited grace period needs to wait for.
*/
static void __sync_rcu_exp_select_node_cpus(struct rcu_exp_work *rewp)
{
int cpu;
unsigned long flags;
unsigned long mask_ofl_test;
unsigned long mask_ofl_ipi;
int ret;
struct rcu_node *rnp = container_of(rewp, struct rcu_node, rew);
raw_spin_lock_irqsave_rcu_node(rnp, flags);
/* Each pass checks a CPU for identity, offline, and idle. */
mask_ofl_test = 0;
for_each_leaf_node_cpu_mask(rnp, cpu, rnp->expmask) {
struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu);
unsigned long mask = rdp->grpmask;
int snap;
if (raw_smp_processor_id() == cpu ||
!(rnp->qsmaskinitnext & mask)) {
mask_ofl_test |= mask;
} else {
snap = rcu_dynticks_snap(cpu);
if (rcu_dynticks_in_eqs(snap))
mask_ofl_test |= mask;
else
rdp->exp_dynticks_snap = snap;
}
}
mask_ofl_ipi = rnp->expmask & ~mask_ofl_test;
/*
* Need to wait for any blocked tasks as well. Note that
* additional blocking tasks will also block the expedited GP
* until such time as the ->expmask bits are cleared.
*/
if (rcu_preempt_has_tasks(rnp))
WRITE_ONCE(rnp->exp_tasks, rnp->blkd_tasks.next);
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
/* IPI the remaining CPUs for expedited quiescent state. */
for_each_leaf_node_cpu_mask(rnp, cpu, mask_ofl_ipi) {
struct rcu_data *rdp = per_cpu_ptr(&rcu_data, cpu);
unsigned long mask = rdp->grpmask;
retry_ipi:
if (rcu_dynticks_in_eqs_since(rdp, rdp->exp_dynticks_snap)) {
mask_ofl_test |= mask;
continue;
}
if (get_cpu() == cpu) {
mask_ofl_test |= mask;
put_cpu();
continue;
}
ret = smp_call_function_single(cpu, rcu_exp_handler, NULL, 0);
put_cpu();
/* The CPU will report the QS in response to the IPI. */
if (!ret)
continue;
/* Failed, raced with CPU hotplug operation. */
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if ((rnp->qsmaskinitnext & mask) &&
(rnp->expmask & mask)) {
/* Online, so delay for a bit and try again. */
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
trace_rcu_exp_grace_period(rcu_state.name, rcu_exp_gp_seq_endval(), TPS("selectofl"));
schedule_timeout_idle(1);
goto retry_ipi;
}
/* CPU really is offline, so we must report its QS. */
if (rnp->expmask & mask)
mask_ofl_test |= mask;
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
}
/* Report quiescent states for those that went offline. */
if (mask_ofl_test)
rcu_report_exp_cpu_mult(rnp, mask_ofl_test, false);
}
static void rcu_exp_sel_wait_wake(unsigned long s);
static void sync_rcu_exp_select_node_cpus(struct kthread_work *wp)
{
struct rcu_exp_work *rewp =
container_of(wp, struct rcu_exp_work, rew_work);
__sync_rcu_exp_select_node_cpus(rewp);
}
static inline bool rcu_exp_worker_started(void)
{
return !!READ_ONCE(rcu_exp_gp_kworker);
}
static inline bool rcu_exp_par_worker_started(struct rcu_node *rnp)
{
return !!READ_ONCE(rnp->exp_kworker);
}
static inline void sync_rcu_exp_select_cpus_queue_work(struct rcu_node *rnp)
{
kthread_init_work(&rnp->rew.rew_work, sync_rcu_exp_select_node_cpus);
/*
* Use rcu_exp_par_gp_kworker, because flushing a work item from
* another work item on the same kthread worker can result in
* deadlock.
*/
kthread_queue_work(READ_ONCE(rnp->exp_kworker), &rnp->rew.rew_work);
}
static inline void sync_rcu_exp_select_cpus_flush_work(struct rcu_node *rnp)
{
kthread_flush_work(&rnp->rew.rew_work);
}
/*
* Work-queue handler to drive an expedited grace period forward.
*/
static void wait_rcu_exp_gp(struct kthread_work *wp)
{
struct rcu_exp_work *rewp;
rewp = container_of(wp, struct rcu_exp_work, rew_work);
rcu_exp_sel_wait_wake(rewp->rew_s);
}
static inline void synchronize_rcu_expedited_queue_work(struct rcu_exp_work *rew)
{
kthread_init_work(&rew->rew_work, wait_rcu_exp_gp);
kthread_queue_work(rcu_exp_gp_kworker, &rew->rew_work);
}
/*
* Select the nodes that the upcoming expedited grace period needs
* to wait for.
*/
static void sync_rcu_exp_select_cpus(void)
{
struct rcu_node *rnp;
trace_rcu_exp_grace_period(rcu_state.name, rcu_exp_gp_seq_endval(), TPS("reset"));
sync_exp_reset_tree();
trace_rcu_exp_grace_period(rcu_state.name, rcu_exp_gp_seq_endval(), TPS("select"));
/* Schedule work for each leaf rcu_node structure. */
rcu_for_each_leaf_node(rnp) {
rnp->exp_need_flush = false;
if (!READ_ONCE(rnp->expmask))
continue; /* Avoid early boot non-existent wq. */
if (!rcu_exp_par_worker_started(rnp) ||
rcu_scheduler_active != RCU_SCHEDULER_RUNNING ||
rcu_is_last_leaf_node(rnp)) {
/* No worker started yet or last leaf, do direct call. */
sync_rcu_exp_select_node_cpus(&rnp->rew.rew_work);
continue;
}
sync_rcu_exp_select_cpus_queue_work(rnp);
rnp->exp_need_flush = true;
}
/* Wait for jobs (if any) to complete. */
rcu_for_each_leaf_node(rnp)
if (rnp->exp_need_flush)
sync_rcu_exp_select_cpus_flush_work(rnp);
}
/*
* Wait for the expedited grace period to elapse, within time limit.
* If the time limit is exceeded without the grace period elapsing,
* return false, otherwise return true.
*/
static bool synchronize_rcu_expedited_wait_once(long tlimit)
{
int t;
struct rcu_node *rnp_root = rcu_get_root();
t = swait_event_timeout_exclusive(rcu_state.expedited_wq,
sync_rcu_exp_done_unlocked(rnp_root),
tlimit);
// Workqueues should not be signaled.
if (t > 0 || sync_rcu_exp_done_unlocked(rnp_root))
return true;
WARN_ON(t < 0); /* workqueues should not be signaled. */
return false;
}
/*
* Wait for the expedited grace period to elapse, issuing any needed
* RCU CPU stall warnings along the way.
*/
static void synchronize_rcu_expedited_wait(void)
{
int cpu;
unsigned long j;
unsigned long jiffies_stall;
unsigned long jiffies_start;
unsigned long mask;
int ndetected;
struct rcu_data *rdp;
struct rcu_node *rnp;
struct rcu_node *rnp_root = rcu_get_root();
unsigned long flags;
trace_rcu_exp_grace_period(rcu_state.name, rcu_exp_gp_seq_endval(), TPS("startwait"));
jiffies_stall = rcu_exp_jiffies_till_stall_check();
jiffies_start = jiffies;
if (tick_nohz_full_enabled() && rcu_inkernel_boot_has_ended()) {
if (synchronize_rcu_expedited_wait_once(1))
return;
rcu_for_each_leaf_node(rnp) {
raw_spin_lock_irqsave_rcu_node(rnp, flags);
mask = READ_ONCE(rnp->expmask);
for_each_leaf_node_cpu_mask(rnp, cpu, mask) {
rdp = per_cpu_ptr(&rcu_data, cpu);
if (rdp->rcu_forced_tick_exp)
continue;
rdp->rcu_forced_tick_exp = true;
if (cpu_online(cpu))
tick_dep_set_cpu(cpu, TICK_DEP_BIT_RCU_EXP);
}
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
}
j = READ_ONCE(jiffies_till_first_fqs);
if (synchronize_rcu_expedited_wait_once(j + HZ))
return;
}
for (;;) {
unsigned long j;
if (synchronize_rcu_expedited_wait_once(jiffies_stall))
return;
if (rcu_stall_is_suppressed())
continue;
j = jiffies;
rcu_stall_notifier_call_chain(RCU_STALL_NOTIFY_EXP, (void *)(j - jiffies_start));
trace_rcu_stall_warning(rcu_state.name, TPS("ExpeditedStall"));
pr_err("INFO: %s detected expedited stalls on CPUs/tasks: {",
rcu_state.name);
ndetected = 0;
rcu_for_each_leaf_node(rnp) {
ndetected += rcu_print_task_exp_stall(rnp);
for_each_leaf_node_possible_cpu(rnp, cpu) {
struct rcu_data *rdp;
mask = leaf_node_cpu_bit(rnp, cpu);
if (!(READ_ONCE(rnp->expmask) & mask))
continue;
ndetected++;
rdp = per_cpu_ptr(&rcu_data, cpu);
pr_cont(" %d-%c%c%c%c", cpu,
"O."[!!cpu_online(cpu)],
"o."[!!(rdp->grpmask & rnp->expmaskinit)],
"N."[!!(rdp->grpmask & rnp->expmaskinitnext)],
"D."[!!data_race(rdp->cpu_no_qs.b.exp)]);
}
}
pr_cont(" } %lu jiffies s: %lu root: %#lx/%c\n",
j - jiffies_start, rcu_state.expedited_sequence,
data_race(rnp_root->expmask),
".T"[!!data_race(rnp_root->exp_tasks)]);
if (ndetected) {
pr_err("blocking rcu_node structures (internal RCU debug):");
rcu_for_each_node_breadth_first(rnp) {
if (rnp == rnp_root)
continue; /* printed unconditionally */
if (sync_rcu_exp_done_unlocked(rnp))
continue;
pr_cont(" l=%u:%d-%d:%#lx/%c",
rnp->level, rnp->grplo, rnp->grphi,
data_race(rnp->expmask),
".T"[!!data_race(rnp->exp_tasks)]);
}
pr_cont("\n");
}
rcu_for_each_leaf_node(rnp) {
for_each_leaf_node_possible_cpu(rnp, cpu) {
mask = leaf_node_cpu_bit(rnp, cpu);
if (!(READ_ONCE(rnp->expmask) & mask))
continue;
preempt_disable(); // For smp_processor_id() in dump_cpu_task().
dump_cpu_task(cpu);
preempt_enable();
}
rcu_exp_print_detail_task_stall_rnp(rnp);
}
jiffies_stall = 3 * rcu_exp_jiffies_till_stall_check() + 3;
panic_on_rcu_stall();
}
}
/*
* Wait for the current expedited grace period to complete, and then
* wake up everyone who piggybacked on the just-completed expedited
* grace period. Also update all the ->exp_seq_rq counters as needed
* in order to avoid counter-wrap problems.
*/
static void rcu_exp_wait_wake(unsigned long s)
{
struct rcu_node *rnp;
synchronize_rcu_expedited_wait();
// Switch over to wakeup mode, allowing the next GP to proceed.
// End the previous grace period only after acquiring the mutex
// to ensure that only one GP runs concurrently with wakeups.
mutex_lock(&rcu_state.exp_wake_mutex);
rcu_exp_gp_seq_end();
trace_rcu_exp_grace_period(rcu_state.name, s, TPS("end"));
rcu_for_each_node_breadth_first(rnp) {
if (ULONG_CMP_LT(READ_ONCE(rnp->exp_seq_rq), s)) {
spin_lock(&rnp->exp_lock);
/* Recheck, avoid hang in case someone just arrived. */
if (ULONG_CMP_LT(rnp->exp_seq_rq, s))
WRITE_ONCE(rnp->exp_seq_rq, s);
spin_unlock(&rnp->exp_lock);
}
smp_mb(); /* All above changes before wakeup. */
wake_up_all(&rnp->exp_wq[rcu_seq_ctr(s) & 0x3]);
}
trace_rcu_exp_grace_period(rcu_state.name, s, TPS("endwake"));
mutex_unlock(&rcu_state.exp_wake_mutex);
}
/*
* Common code to drive an expedited grace period forward, used by
* workqueues and mid-boot-time tasks.
*/
static void rcu_exp_sel_wait_wake(unsigned long s)
{
/* Initialize the rcu_node tree in preparation for the wait. */
sync_rcu_exp_select_cpus();
/* Wait and clean up, including waking everyone. */
rcu_exp_wait_wake(s);
}
#ifdef CONFIG_PREEMPT_RCU
/*
* Remote handler for smp_call_function_single(). If there is an
* RCU read-side critical section in effect, request that the
* next rcu_read_unlock() record the quiescent state up the
* ->expmask fields in the rcu_node tree. Otherwise, immediately
* report the quiescent state.
*/
static void rcu_exp_handler(void *unused)
{
int depth = rcu_preempt_depth();
unsigned long flags;
struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
struct rcu_node *rnp = rdp->mynode;
struct task_struct *t = current;
/*
* First, the common case of not being in an RCU read-side
* critical section. If also enabled or idle, immediately
* report the quiescent state, otherwise defer.
*/
if (!depth) {
if (!(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK)) ||
rcu_is_cpu_rrupt_from_idle()) {
rcu_report_exp_rdp(rdp);
} else {
WRITE_ONCE(rdp->cpu_no_qs.b.exp, true);
set_tsk_need_resched(t);
set_preempt_need_resched();
}
return;
}
/*
* Second, the less-common case of being in an RCU read-side
* critical section. In this case we can count on a future
* rcu_read_unlock(). However, this rcu_read_unlock() might
* execute on some other CPU, but in that case there will be
* a future context switch. Either way, if the expedited
* grace period is still waiting on this CPU, set ->deferred_qs
* so that the eventual quiescent state will be reported.
* Note that there is a large group of race conditions that
* can have caused this quiescent state to already have been
* reported, so we really do need to check ->expmask.
*/
if (depth > 0) {
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if (rnp->expmask & rdp->grpmask) {
WRITE_ONCE(rdp->cpu_no_qs.b.exp, true);
t->rcu_read_unlock_special.b.exp_hint = true;
}
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return;
}
// Finally, negative nesting depth should not happen.
WARN_ON_ONCE(1);
}
/* PREEMPTION=y, so no PREEMPTION=n expedited grace period to clean up after. */
static void sync_sched_exp_online_cleanup(int cpu)
{
}
/*
* Scan the current list of tasks blocked within RCU read-side critical
* sections, printing out the tid of each that is blocking the current
* expedited grace period.
*/
static int rcu_print_task_exp_stall(struct rcu_node *rnp)
{
unsigned long flags;
int ndetected = 0;
struct task_struct *t;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if (!rnp->exp_tasks) {
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return 0;
}
t = list_entry(rnp->exp_tasks->prev,
struct task_struct, rcu_node_entry);
list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) {
pr_cont(" P%d", t->pid);
ndetected++;
}
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return ndetected;
}
/*
* Scan the current list of tasks blocked within RCU read-side critical
* sections, dumping the stack of each that is blocking the current
* expedited grace period.
*/
static void rcu_exp_print_detail_task_stall_rnp(struct rcu_node *rnp)
{
unsigned long flags;
struct task_struct *t;
if (!rcu_exp_stall_task_details)
return;
raw_spin_lock_irqsave_rcu_node(rnp, flags);
if (!READ_ONCE(rnp->exp_tasks)) {
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
return;
}
t = list_entry(rnp->exp_tasks->prev,
struct task_struct, rcu_node_entry);
list_for_each_entry_continue(t, &rnp->blkd_tasks, rcu_node_entry) {
/*
* We could be printing a lot while holding a spinlock.
* Avoid triggering hard lockup.
*/
touch_nmi_watchdog();
sched_show_task(t);
}
raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
}
#else /* #ifdef CONFIG_PREEMPT_RCU */
/* Request an expedited quiescent state. */
static void rcu_exp_need_qs(void)
{
__this_cpu_write(rcu_data.cpu_no_qs.b.exp, true);
/* Store .exp before .rcu_urgent_qs. */
smp_store_release(this_cpu_ptr(&rcu_data.rcu_urgent_qs), true);
set_tsk_need_resched(current);
set_preempt_need_resched();
}
/* Invoked on each online non-idle CPU for expedited quiescent state. */
static void rcu_exp_handler(void *unused)
{
struct rcu_data *rdp = this_cpu_ptr(&rcu_data);
struct rcu_node *rnp = rdp->mynode;
bool preempt_bh_enabled = !(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK));
if (!(READ_ONCE(rnp->expmask) & rdp->grpmask) ||
__this_cpu_read(rcu_data.cpu_no_qs.b.exp))
return;
if (rcu_is_cpu_rrupt_from_idle() ||
(IS_ENABLED(CONFIG_PREEMPT_COUNT) && preempt_bh_enabled)) {
rcu_report_exp_rdp(this_cpu_ptr(&rcu_data));
return;
}
rcu_exp_need_qs();
}
/* Send IPI for expedited cleanup if needed at end of CPU-hotplug operation. */
static void sync_sched_exp_online_cleanup(int cpu)
{
unsigned long flags;
int my_cpu;
struct rcu_data *rdp;
int ret;
struct rcu_node *rnp;
rdp = per_cpu_ptr(&rcu_data, cpu);
rnp = rdp->mynode;
my_cpu = get_cpu();
/* Quiescent state either not needed or already requested, leave. */
if (!(READ_ONCE(rnp->expmask) & rdp->grpmask) ||
READ_ONCE(rdp->cpu_no_qs.b.exp)) {
put_cpu();
return;
}
/* Quiescent state needed on current CPU, so set it up locally. */
if (my_cpu == cpu) {
local_irq_save(flags);
rcu_exp_need_qs();
local_irq_restore(flags);
put_cpu();
return;
}
/* Quiescent state needed on some other CPU, send IPI. */
ret = smp_call_function_single(cpu, rcu_exp_handler, NULL, 0);
put_cpu();
WARN_ON_ONCE(ret);
}
/*
* Because preemptible RCU does not exist, we never have to check for
* tasks blocked within RCU read-side critical sections that are
* blocking the current expedited grace period.
*/
static int rcu_print_task_exp_stall(struct rcu_node *rnp)
{
return 0;
}
/*
* Because preemptible RCU does not exist, we never have to print out
* tasks blocked within RCU read-side critical sections that are blocking
* the current expedited grace period.
*/
static void rcu_exp_print_detail_task_stall_rnp(struct rcu_node *rnp)
{
}
#endif /* #else #ifdef CONFIG_PREEMPT_RCU */
/**
* synchronize_rcu_expedited - Brute-force RCU grace period
*
* Wait for an RCU grace period, but expedite it. The basic idea is to
* IPI all non-idle non-nohz online CPUs. The IPI handler checks whether
* the CPU is in an RCU critical section, and if so, it sets a flag that
* causes the outermost rcu_read_unlock() to report the quiescent state
* for RCU-preempt or asks the scheduler for help for RCU-sched. On the
* other hand, if the CPU is not in an RCU read-side critical section,
* the IPI handler reports the quiescent state immediately.
*
* Although this is a great improvement over previous expedited
* implementations, it is still unfriendly to real-time workloads, so is
* thus not recommended for any sort of common-case code. In fact, if
* you are using synchronize_rcu_expedited() in a loop, please restructure
* your code to batch your updates, and then use a single synchronize_rcu()
* instead.
*
* This has the same semantics as (but is more brutal than) synchronize_rcu().
*/
void synchronize_rcu_expedited(void)
{
unsigned long flags;
struct rcu_exp_work rew;
struct rcu_node *rnp;
unsigned long s;
RCU_LOCKDEP_WARN(lock_is_held(&rcu_bh_lock_map) ||
lock_is_held(&rcu_lock_map) ||
lock_is_held(&rcu_sched_lock_map),
"Illegal synchronize_rcu_expedited() in RCU read-side critical section");
/* Is the state is such that the call is a grace period? */
if (rcu_blocking_is_gp()) {
// Note well that this code runs with !PREEMPT && !SMP.
// In addition, all code that advances grace periods runs
// at process level. Therefore, this expedited GP overlaps
// with other expedited GPs only by being fully nested within
// them, which allows reuse of ->gp_seq_polled_exp_snap.
rcu_poll_gp_seq_start_unlocked(&rcu_state.gp_seq_polled_exp_snap);
rcu_poll_gp_seq_end_unlocked(&rcu_state.gp_seq_polled_exp_snap);
local_irq_save(flags);
WARN_ON_ONCE(num_online_cpus() > 1);
rcu_state.expedited_sequence += (1 << RCU_SEQ_CTR_SHIFT);
local_irq_restore(flags);
return; // Context allows vacuous grace periods.
}
/* If expedited grace periods are prohibited, fall back to normal. */
if (rcu_gp_is_normal()) {
synchronize_rcu_normal();
return;
}
/* Take a snapshot of the sequence number. */
s = rcu_exp_gp_seq_snap();
if (exp_funnel_lock(s))
return; /* Someone else did our work for us. */
/* Ensure that load happens before action based on it. */
if (unlikely((rcu_scheduler_active == RCU_SCHEDULER_INIT) || !rcu_exp_worker_started())) {
/* Direct call during scheduler init and early_initcalls(). */
rcu_exp_sel_wait_wake(s);
} else {
/* Marshall arguments & schedule the expedited grace period. */
rew.rew_s = s;
synchronize_rcu_expedited_queue_work(&rew);
}
/* Wait for expedited grace period to complete. */
rnp = rcu_get_root();
wait_event(rnp->exp_wq[rcu_seq_ctr(s) & 0x3],
sync_exp_work_done(s));
smp_mb(); /* Work actions happen before return. */
/* Let the next expedited grace period start. */
mutex_unlock(&rcu_state.exp_mutex);
}
EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
/*
* Ensure that start_poll_synchronize_rcu_expedited() has the expedited
* RCU grace periods that it needs.
*/
static void sync_rcu_do_polled_gp(struct work_struct *wp)
{
unsigned long flags;
int i = 0;
struct rcu_node *rnp = container_of(wp, struct rcu_node, exp_poll_wq);
unsigned long s;
raw_spin_lock_irqsave(&rnp->exp_poll_lock, flags);
s = rnp->exp_seq_poll_rq;
rnp->exp_seq_poll_rq = RCU_GET_STATE_COMPLETED;
raw_spin_unlock_irqrestore(&rnp->exp_poll_lock, flags);
if (s == RCU_GET_STATE_COMPLETED)
return;
while (!poll_state_synchronize_rcu(s)) {
synchronize_rcu_expedited();
if (i == 10 || i == 20)
pr_info("%s: i = %d s = %lx gp_seq_polled = %lx\n", __func__, i, s, READ_ONCE(rcu_state.gp_seq_polled));
i++;
}
raw_spin_lock_irqsave(&rnp->exp_poll_lock, flags);
s = rnp->exp_seq_poll_rq;
if (poll_state_synchronize_rcu(s))
rnp->exp_seq_poll_rq = RCU_GET_STATE_COMPLETED;
raw_spin_unlock_irqrestore(&rnp->exp_poll_lock, flags);
}
/**
* start_poll_synchronize_rcu_expedited - Snapshot current RCU state and start expedited grace period
*
* Returns a cookie to pass to a call to cond_synchronize_rcu(),
* cond_synchronize_rcu_expedited(), or poll_state_synchronize_rcu(),
* allowing them to determine whether or not any sort of grace period has
* elapsed in the meantime. If the needed expedited grace period is not
* already slated to start, initiates that grace period.
*/
unsigned long start_poll_synchronize_rcu_expedited(void)
{
unsigned long flags;
struct rcu_data *rdp;
struct rcu_node *rnp;
unsigned long s;
s = get_state_synchronize_rcu();
rdp = per_cpu_ptr(&rcu_data, raw_smp_processor_id());
rnp = rdp->mynode;
if (rcu_init_invoked())
raw_spin_lock_irqsave(&rnp->exp_poll_lock, flags);
if (!poll_state_synchronize_rcu(s)) {
if (rcu_init_invoked()) {
rnp->exp_seq_poll_rq = s;
queue_work(rcu_gp_wq, &rnp->exp_poll_wq);
}
}
if (rcu_init_invoked())
raw_spin_unlock_irqrestore(&rnp->exp_poll_lock, flags);
return s;
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu_expedited);
/**
* start_poll_synchronize_rcu_expedited_full - Take a full snapshot and start expedited grace period
* @rgosp: Place to put snapshot of grace-period state
*
* Places the normal and expedited grace-period states in rgosp. This
* state value can be passed to a later call to cond_synchronize_rcu_full()
* or poll_state_synchronize_rcu_full() to determine whether or not a
* grace period (whether normal or expedited) has elapsed in the meantime.
* If the needed expedited grace period is not already slated to start,
* initiates that grace period.
*/
void start_poll_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
get_state_synchronize_rcu_full(rgosp);
(void)start_poll_synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(start_poll_synchronize_rcu_expedited_full);
/**
* cond_synchronize_rcu_expedited - Conditionally wait for an expedited RCU grace period
*
* @oldstate: value from get_state_synchronize_rcu(), start_poll_synchronize_rcu(), or start_poll_synchronize_rcu_expedited()
*
* If any type of full RCU grace period has elapsed since the earlier
* call to get_state_synchronize_rcu(), start_poll_synchronize_rcu(),
* or start_poll_synchronize_rcu_expedited(), just return. Otherwise,
* invoke synchronize_rcu_expedited() to wait for a full grace period.
*
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited for
* more than 2 billion grace periods (and way more on a 64-bit system!),
* so waiting for a couple of additional grace periods should be just fine.
*
* This function provides the same memory-ordering guarantees that
* would be provided by a synchronize_rcu() that was invoked at the call
* to the function that provided @oldstate and that returned at the end
* of this function.
*/
void cond_synchronize_rcu_expedited(unsigned long oldstate)
{
if (!poll_state_synchronize_rcu(oldstate))
synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu_expedited);
/**
* cond_synchronize_rcu_expedited_full - Conditionally wait for an expedited RCU grace period
* @rgosp: value from get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(), or start_poll_synchronize_rcu_expedited_full()
*
* If a full RCU grace period has elapsed since the call to
* get_state_synchronize_rcu_full(), start_poll_synchronize_rcu_full(),
* or start_poll_synchronize_rcu_expedited_full() from which @rgosp was
* obtained, just return. Otherwise, invoke synchronize_rcu_expedited()
* to wait for a full grace period.
*
* Yes, this function does not take counter wrap into account.
* But counter wrap is harmless. If the counter wraps, we have waited for
* more than 2 billion grace periods (and way more on a 64-bit system!),
* so waiting for a couple of additional grace periods should be just fine.
*
* This function provides the same memory-ordering guarantees that
* would be provided by a synchronize_rcu() that was invoked at the call
* to the function that provided @rgosp and that returned at the end of
* this function.
*/
void cond_synchronize_rcu_expedited_full(struct rcu_gp_oldstate *rgosp)
{
if (!poll_state_synchronize_rcu_full(rgosp))
synchronize_rcu_expedited();
}
EXPORT_SYMBOL_GPL(cond_synchronize_rcu_expedited_full);