mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-05 12:06:42 +07:00
46c498c2cd
commit 14e568e78
(stop_machine: Use smpboot threads) introduced the
following regression:
Before this commit the stopper enabled bit was set in the online
notifier.
CPU0 CPU1
cpu_up
cpu online
hotplug_notifier(ONLINE)
stopper(CPU1)->enabled = true;
...
stop_machine()
The conversion to smpboot threads moved the enablement to the wakeup
path of the parked thread. The majority of users seem to have the
following working order:
CPU0 CPU1
cpu_up
cpu online
unpark_threads()
wakeup(stopper[CPU1])
....
stopper thread runs
stopper(CPU1)->enabled = true;
stop_machine()
But Konrad and Sander have observed:
CPU0 CPU1
cpu_up
cpu online
unpark_threads()
wakeup(stopper[CPU1])
....
stop_machine()
stopper thread runs
stopper(CPU1)->enabled = true;
Now the stop machinery kicks CPU0 into the stop loop, where it gets
stuck forever because the queue code saw stopper(CPU1)->enabled ==
false, so CPU0 waits for CPU1 to enter stomp_machine, but the CPU1
stopper work got discarded due to enabled == false.
Add a pre_unpark function to the smpboot thread descriptor and call it
before waking the thread.
This fixes the problem at hand, but the stop_machine code should be
more robust. The stopper->enabled flag smells fishy at best.
Thanks to Konrad for going through a loop of debug patches and
providing the information to decode this issue.
Reported-and-tested-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Reported-and-tested-by: Sander Eikelenboom <linux@eikelenboom.it>
Cc: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1302261843240.22263@ionos
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
545 lines
15 KiB
C
545 lines
15 KiB
C
/*
|
|
* kernel/stop_machine.c
|
|
*
|
|
* Copyright (C) 2008, 2005 IBM Corporation.
|
|
* Copyright (C) 2008, 2005 Rusty Russell rusty@rustcorp.com.au
|
|
* Copyright (C) 2010 SUSE Linux Products GmbH
|
|
* Copyright (C) 2010 Tejun Heo <tj@kernel.org>
|
|
*
|
|
* This file is released under the GPLv2 and any later version.
|
|
*/
|
|
#include <linux/completion.h>
|
|
#include <linux/cpu.h>
|
|
#include <linux/init.h>
|
|
#include <linux/kthread.h>
|
|
#include <linux/export.h>
|
|
#include <linux/percpu.h>
|
|
#include <linux/sched.h>
|
|
#include <linux/stop_machine.h>
|
|
#include <linux/interrupt.h>
|
|
#include <linux/kallsyms.h>
|
|
#include <linux/smpboot.h>
|
|
#include <linux/atomic.h>
|
|
|
|
/*
|
|
* Structure to determine completion condition and record errors. May
|
|
* be shared by works on different cpus.
|
|
*/
|
|
struct cpu_stop_done {
|
|
atomic_t nr_todo; /* nr left to execute */
|
|
bool executed; /* actually executed? */
|
|
int ret; /* collected return value */
|
|
struct completion completion; /* fired if nr_todo reaches 0 */
|
|
};
|
|
|
|
/* the actual stopper, one per every possible cpu, enabled on online cpus */
|
|
struct cpu_stopper {
|
|
spinlock_t lock;
|
|
bool enabled; /* is this stopper enabled? */
|
|
struct list_head works; /* list of pending works */
|
|
};
|
|
|
|
static DEFINE_PER_CPU(struct cpu_stopper, cpu_stopper);
|
|
static DEFINE_PER_CPU(struct task_struct *, cpu_stopper_task);
|
|
static bool stop_machine_initialized = false;
|
|
|
|
static void cpu_stop_init_done(struct cpu_stop_done *done, unsigned int nr_todo)
|
|
{
|
|
memset(done, 0, sizeof(*done));
|
|
atomic_set(&done->nr_todo, nr_todo);
|
|
init_completion(&done->completion);
|
|
}
|
|
|
|
/* signal completion unless @done is NULL */
|
|
static void cpu_stop_signal_done(struct cpu_stop_done *done, bool executed)
|
|
{
|
|
if (done) {
|
|
if (executed)
|
|
done->executed = true;
|
|
if (atomic_dec_and_test(&done->nr_todo))
|
|
complete(&done->completion);
|
|
}
|
|
}
|
|
|
|
/* queue @work to @stopper. if offline, @work is completed immediately */
|
|
static void cpu_stop_queue_work(unsigned int cpu, struct cpu_stop_work *work)
|
|
{
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
struct task_struct *p = per_cpu(cpu_stopper_task, cpu);
|
|
|
|
unsigned long flags;
|
|
|
|
spin_lock_irqsave(&stopper->lock, flags);
|
|
|
|
if (stopper->enabled) {
|
|
list_add_tail(&work->list, &stopper->works);
|
|
wake_up_process(p);
|
|
} else
|
|
cpu_stop_signal_done(work->done, false);
|
|
|
|
spin_unlock_irqrestore(&stopper->lock, flags);
|
|
}
|
|
|
|
/**
|
|
* stop_one_cpu - stop a cpu
|
|
* @cpu: cpu to stop
|
|
* @fn: function to execute
|
|
* @arg: argument to @fn
|
|
*
|
|
* Execute @fn(@arg) on @cpu. @fn is run in a process context with
|
|
* the highest priority preempting any task on the cpu and
|
|
* monopolizing it. This function returns after the execution is
|
|
* complete.
|
|
*
|
|
* This function doesn't guarantee @cpu stays online till @fn
|
|
* completes. If @cpu goes down in the middle, execution may happen
|
|
* partially or fully on different cpus. @fn should either be ready
|
|
* for that or the caller should ensure that @cpu stays online until
|
|
* this function completes.
|
|
*
|
|
* CONTEXT:
|
|
* Might sleep.
|
|
*
|
|
* RETURNS:
|
|
* -ENOENT if @fn(@arg) was not executed because @cpu was offline;
|
|
* otherwise, the return value of @fn.
|
|
*/
|
|
int stop_one_cpu(unsigned int cpu, cpu_stop_fn_t fn, void *arg)
|
|
{
|
|
struct cpu_stop_done done;
|
|
struct cpu_stop_work work = { .fn = fn, .arg = arg, .done = &done };
|
|
|
|
cpu_stop_init_done(&done, 1);
|
|
cpu_stop_queue_work(cpu, &work);
|
|
wait_for_completion(&done.completion);
|
|
return done.executed ? done.ret : -ENOENT;
|
|
}
|
|
|
|
/**
|
|
* stop_one_cpu_nowait - stop a cpu but don't wait for completion
|
|
* @cpu: cpu to stop
|
|
* @fn: function to execute
|
|
* @arg: argument to @fn
|
|
*
|
|
* Similar to stop_one_cpu() but doesn't wait for completion. The
|
|
* caller is responsible for ensuring @work_buf is currently unused
|
|
* and will remain untouched until stopper starts executing @fn.
|
|
*
|
|
* CONTEXT:
|
|
* Don't care.
|
|
*/
|
|
void stop_one_cpu_nowait(unsigned int cpu, cpu_stop_fn_t fn, void *arg,
|
|
struct cpu_stop_work *work_buf)
|
|
{
|
|
*work_buf = (struct cpu_stop_work){ .fn = fn, .arg = arg, };
|
|
cpu_stop_queue_work(cpu, work_buf);
|
|
}
|
|
|
|
/* static data for stop_cpus */
|
|
static DEFINE_MUTEX(stop_cpus_mutex);
|
|
static DEFINE_PER_CPU(struct cpu_stop_work, stop_cpus_work);
|
|
|
|
static void queue_stop_cpus_work(const struct cpumask *cpumask,
|
|
cpu_stop_fn_t fn, void *arg,
|
|
struct cpu_stop_done *done)
|
|
{
|
|
struct cpu_stop_work *work;
|
|
unsigned int cpu;
|
|
|
|
/* initialize works and done */
|
|
for_each_cpu(cpu, cpumask) {
|
|
work = &per_cpu(stop_cpus_work, cpu);
|
|
work->fn = fn;
|
|
work->arg = arg;
|
|
work->done = done;
|
|
}
|
|
|
|
/*
|
|
* Disable preemption while queueing to avoid getting
|
|
* preempted by a stopper which might wait for other stoppers
|
|
* to enter @fn which can lead to deadlock.
|
|
*/
|
|
preempt_disable();
|
|
for_each_cpu(cpu, cpumask)
|
|
cpu_stop_queue_work(cpu, &per_cpu(stop_cpus_work, cpu));
|
|
preempt_enable();
|
|
}
|
|
|
|
static int __stop_cpus(const struct cpumask *cpumask,
|
|
cpu_stop_fn_t fn, void *arg)
|
|
{
|
|
struct cpu_stop_done done;
|
|
|
|
cpu_stop_init_done(&done, cpumask_weight(cpumask));
|
|
queue_stop_cpus_work(cpumask, fn, arg, &done);
|
|
wait_for_completion(&done.completion);
|
|
return done.executed ? done.ret : -ENOENT;
|
|
}
|
|
|
|
/**
|
|
* stop_cpus - stop multiple cpus
|
|
* @cpumask: cpus to stop
|
|
* @fn: function to execute
|
|
* @arg: argument to @fn
|
|
*
|
|
* Execute @fn(@arg) on online cpus in @cpumask. On each target cpu,
|
|
* @fn is run in a process context with the highest priority
|
|
* preempting any task on the cpu and monopolizing it. This function
|
|
* returns after all executions are complete.
|
|
*
|
|
* This function doesn't guarantee the cpus in @cpumask stay online
|
|
* till @fn completes. If some cpus go down in the middle, execution
|
|
* on the cpu may happen partially or fully on different cpus. @fn
|
|
* should either be ready for that or the caller should ensure that
|
|
* the cpus stay online until this function completes.
|
|
*
|
|
* All stop_cpus() calls are serialized making it safe for @fn to wait
|
|
* for all cpus to start executing it.
|
|
*
|
|
* CONTEXT:
|
|
* Might sleep.
|
|
*
|
|
* RETURNS:
|
|
* -ENOENT if @fn(@arg) was not executed at all because all cpus in
|
|
* @cpumask were offline; otherwise, 0 if all executions of @fn
|
|
* returned 0, any non zero return value if any returned non zero.
|
|
*/
|
|
int stop_cpus(const struct cpumask *cpumask, cpu_stop_fn_t fn, void *arg)
|
|
{
|
|
int ret;
|
|
|
|
/* static works are used, process one request at a time */
|
|
mutex_lock(&stop_cpus_mutex);
|
|
ret = __stop_cpus(cpumask, fn, arg);
|
|
mutex_unlock(&stop_cpus_mutex);
|
|
return ret;
|
|
}
|
|
|
|
/**
|
|
* try_stop_cpus - try to stop multiple cpus
|
|
* @cpumask: cpus to stop
|
|
* @fn: function to execute
|
|
* @arg: argument to @fn
|
|
*
|
|
* Identical to stop_cpus() except that it fails with -EAGAIN if
|
|
* someone else is already using the facility.
|
|
*
|
|
* CONTEXT:
|
|
* Might sleep.
|
|
*
|
|
* RETURNS:
|
|
* -EAGAIN if someone else is already stopping cpus, -ENOENT if
|
|
* @fn(@arg) was not executed at all because all cpus in @cpumask were
|
|
* offline; otherwise, 0 if all executions of @fn returned 0, any non
|
|
* zero return value if any returned non zero.
|
|
*/
|
|
int try_stop_cpus(const struct cpumask *cpumask, cpu_stop_fn_t fn, void *arg)
|
|
{
|
|
int ret;
|
|
|
|
/* static works are used, process one request at a time */
|
|
if (!mutex_trylock(&stop_cpus_mutex))
|
|
return -EAGAIN;
|
|
ret = __stop_cpus(cpumask, fn, arg);
|
|
mutex_unlock(&stop_cpus_mutex);
|
|
return ret;
|
|
}
|
|
|
|
static int cpu_stop_should_run(unsigned int cpu)
|
|
{
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
unsigned long flags;
|
|
int run;
|
|
|
|
spin_lock_irqsave(&stopper->lock, flags);
|
|
run = !list_empty(&stopper->works);
|
|
spin_unlock_irqrestore(&stopper->lock, flags);
|
|
return run;
|
|
}
|
|
|
|
static void cpu_stopper_thread(unsigned int cpu)
|
|
{
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
struct cpu_stop_work *work;
|
|
int ret;
|
|
|
|
repeat:
|
|
work = NULL;
|
|
spin_lock_irq(&stopper->lock);
|
|
if (!list_empty(&stopper->works)) {
|
|
work = list_first_entry(&stopper->works,
|
|
struct cpu_stop_work, list);
|
|
list_del_init(&work->list);
|
|
}
|
|
spin_unlock_irq(&stopper->lock);
|
|
|
|
if (work) {
|
|
cpu_stop_fn_t fn = work->fn;
|
|
void *arg = work->arg;
|
|
struct cpu_stop_done *done = work->done;
|
|
char ksym_buf[KSYM_NAME_LEN] __maybe_unused;
|
|
|
|
/* cpu stop callbacks are not allowed to sleep */
|
|
preempt_disable();
|
|
|
|
ret = fn(arg);
|
|
if (ret)
|
|
done->ret = ret;
|
|
|
|
/* restore preemption and check it's still balanced */
|
|
preempt_enable();
|
|
WARN_ONCE(preempt_count(),
|
|
"cpu_stop: %s(%p) leaked preempt count\n",
|
|
kallsyms_lookup((unsigned long)fn, NULL, NULL, NULL,
|
|
ksym_buf), arg);
|
|
|
|
cpu_stop_signal_done(done, true);
|
|
goto repeat;
|
|
}
|
|
}
|
|
|
|
extern void sched_set_stop_task(int cpu, struct task_struct *stop);
|
|
|
|
static void cpu_stop_create(unsigned int cpu)
|
|
{
|
|
sched_set_stop_task(cpu, per_cpu(cpu_stopper_task, cpu));
|
|
}
|
|
|
|
static void cpu_stop_park(unsigned int cpu)
|
|
{
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
struct cpu_stop_work *work;
|
|
unsigned long flags;
|
|
|
|
/* drain remaining works */
|
|
spin_lock_irqsave(&stopper->lock, flags);
|
|
list_for_each_entry(work, &stopper->works, list)
|
|
cpu_stop_signal_done(work->done, false);
|
|
stopper->enabled = false;
|
|
spin_unlock_irqrestore(&stopper->lock, flags);
|
|
}
|
|
|
|
static void cpu_stop_unpark(unsigned int cpu)
|
|
{
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
|
|
spin_lock_irq(&stopper->lock);
|
|
stopper->enabled = true;
|
|
spin_unlock_irq(&stopper->lock);
|
|
}
|
|
|
|
static struct smp_hotplug_thread cpu_stop_threads = {
|
|
.store = &cpu_stopper_task,
|
|
.thread_should_run = cpu_stop_should_run,
|
|
.thread_fn = cpu_stopper_thread,
|
|
.thread_comm = "migration/%u",
|
|
.create = cpu_stop_create,
|
|
.setup = cpu_stop_unpark,
|
|
.park = cpu_stop_park,
|
|
.pre_unpark = cpu_stop_unpark,
|
|
.selfparking = true,
|
|
};
|
|
|
|
static int __init cpu_stop_init(void)
|
|
{
|
|
unsigned int cpu;
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
|
|
|
|
spin_lock_init(&stopper->lock);
|
|
INIT_LIST_HEAD(&stopper->works);
|
|
}
|
|
|
|
BUG_ON(smpboot_register_percpu_thread(&cpu_stop_threads));
|
|
stop_machine_initialized = true;
|
|
return 0;
|
|
}
|
|
early_initcall(cpu_stop_init);
|
|
|
|
#ifdef CONFIG_STOP_MACHINE
|
|
|
|
/* This controls the threads on each CPU. */
|
|
enum stopmachine_state {
|
|
/* Dummy starting state for thread. */
|
|
STOPMACHINE_NONE,
|
|
/* Awaiting everyone to be scheduled. */
|
|
STOPMACHINE_PREPARE,
|
|
/* Disable interrupts. */
|
|
STOPMACHINE_DISABLE_IRQ,
|
|
/* Run the function */
|
|
STOPMACHINE_RUN,
|
|
/* Exit */
|
|
STOPMACHINE_EXIT,
|
|
};
|
|
|
|
struct stop_machine_data {
|
|
int (*fn)(void *);
|
|
void *data;
|
|
/* Like num_online_cpus(), but hotplug cpu uses us, so we need this. */
|
|
unsigned int num_threads;
|
|
const struct cpumask *active_cpus;
|
|
|
|
enum stopmachine_state state;
|
|
atomic_t thread_ack;
|
|
};
|
|
|
|
static void set_state(struct stop_machine_data *smdata,
|
|
enum stopmachine_state newstate)
|
|
{
|
|
/* Reset ack counter. */
|
|
atomic_set(&smdata->thread_ack, smdata->num_threads);
|
|
smp_wmb();
|
|
smdata->state = newstate;
|
|
}
|
|
|
|
/* Last one to ack a state moves to the next state. */
|
|
static void ack_state(struct stop_machine_data *smdata)
|
|
{
|
|
if (atomic_dec_and_test(&smdata->thread_ack))
|
|
set_state(smdata, smdata->state + 1);
|
|
}
|
|
|
|
/* This is the cpu_stop function which stops the CPU. */
|
|
static int stop_machine_cpu_stop(void *data)
|
|
{
|
|
struct stop_machine_data *smdata = data;
|
|
enum stopmachine_state curstate = STOPMACHINE_NONE;
|
|
int cpu = smp_processor_id(), err = 0;
|
|
unsigned long flags;
|
|
bool is_active;
|
|
|
|
/*
|
|
* When called from stop_machine_from_inactive_cpu(), irq might
|
|
* already be disabled. Save the state and restore it on exit.
|
|
*/
|
|
local_save_flags(flags);
|
|
|
|
if (!smdata->active_cpus)
|
|
is_active = cpu == cpumask_first(cpu_online_mask);
|
|
else
|
|
is_active = cpumask_test_cpu(cpu, smdata->active_cpus);
|
|
|
|
/* Simple state machine */
|
|
do {
|
|
/* Chill out and ensure we re-read stopmachine_state. */
|
|
cpu_relax();
|
|
if (smdata->state != curstate) {
|
|
curstate = smdata->state;
|
|
switch (curstate) {
|
|
case STOPMACHINE_DISABLE_IRQ:
|
|
local_irq_disable();
|
|
hard_irq_disable();
|
|
break;
|
|
case STOPMACHINE_RUN:
|
|
if (is_active)
|
|
err = smdata->fn(smdata->data);
|
|
break;
|
|
default:
|
|
break;
|
|
}
|
|
ack_state(smdata);
|
|
}
|
|
} while (curstate != STOPMACHINE_EXIT);
|
|
|
|
local_irq_restore(flags);
|
|
return err;
|
|
}
|
|
|
|
int __stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
|
|
{
|
|
struct stop_machine_data smdata = { .fn = fn, .data = data,
|
|
.num_threads = num_online_cpus(),
|
|
.active_cpus = cpus };
|
|
|
|
if (!stop_machine_initialized) {
|
|
/*
|
|
* Handle the case where stop_machine() is called
|
|
* early in boot before stop_machine() has been
|
|
* initialized.
|
|
*/
|
|
unsigned long flags;
|
|
int ret;
|
|
|
|
WARN_ON_ONCE(smdata.num_threads != 1);
|
|
|
|
local_irq_save(flags);
|
|
hard_irq_disable();
|
|
ret = (*fn)(data);
|
|
local_irq_restore(flags);
|
|
|
|
return ret;
|
|
}
|
|
|
|
/* Set the initial state and stop all online cpus. */
|
|
set_state(&smdata, STOPMACHINE_PREPARE);
|
|
return stop_cpus(cpu_online_mask, stop_machine_cpu_stop, &smdata);
|
|
}
|
|
|
|
int stop_machine(int (*fn)(void *), void *data, const struct cpumask *cpus)
|
|
{
|
|
int ret;
|
|
|
|
/* No CPUs can come up or down during this. */
|
|
get_online_cpus();
|
|
ret = __stop_machine(fn, data, cpus);
|
|
put_online_cpus();
|
|
return ret;
|
|
}
|
|
EXPORT_SYMBOL_GPL(stop_machine);
|
|
|
|
/**
|
|
* stop_machine_from_inactive_cpu - stop_machine() from inactive CPU
|
|
* @fn: the function to run
|
|
* @data: the data ptr for the @fn()
|
|
* @cpus: the cpus to run the @fn() on (NULL = any online cpu)
|
|
*
|
|
* This is identical to stop_machine() but can be called from a CPU which
|
|
* is not active. The local CPU is in the process of hotplug (so no other
|
|
* CPU hotplug can start) and not marked active and doesn't have enough
|
|
* context to sleep.
|
|
*
|
|
* This function provides stop_machine() functionality for such state by
|
|
* using busy-wait for synchronization and executing @fn directly for local
|
|
* CPU.
|
|
*
|
|
* CONTEXT:
|
|
* Local CPU is inactive. Temporarily stops all active CPUs.
|
|
*
|
|
* RETURNS:
|
|
* 0 if all executions of @fn returned 0, any non zero return value if any
|
|
* returned non zero.
|
|
*/
|
|
int stop_machine_from_inactive_cpu(int (*fn)(void *), void *data,
|
|
const struct cpumask *cpus)
|
|
{
|
|
struct stop_machine_data smdata = { .fn = fn, .data = data,
|
|
.active_cpus = cpus };
|
|
struct cpu_stop_done done;
|
|
int ret;
|
|
|
|
/* Local CPU must be inactive and CPU hotplug in progress. */
|
|
BUG_ON(cpu_active(raw_smp_processor_id()));
|
|
smdata.num_threads = num_active_cpus() + 1; /* +1 for local */
|
|
|
|
/* No proper task established and can't sleep - busy wait for lock. */
|
|
while (!mutex_trylock(&stop_cpus_mutex))
|
|
cpu_relax();
|
|
|
|
/* Schedule work on other CPUs and execute directly for local CPU */
|
|
set_state(&smdata, STOPMACHINE_PREPARE);
|
|
cpu_stop_init_done(&done, num_active_cpus());
|
|
queue_stop_cpus_work(cpu_active_mask, stop_machine_cpu_stop, &smdata,
|
|
&done);
|
|
ret = stop_machine_cpu_stop(&smdata);
|
|
|
|
/* Busy wait for completion. */
|
|
while (!completion_done(&done.completion))
|
|
cpu_relax();
|
|
|
|
mutex_unlock(&stop_cpus_mutex);
|
|
return ret ?: done.ret;
|
|
}
|
|
|
|
#endif /* CONFIG_STOP_MACHINE */
|