xemu/util/thread-pool.c
Nicolas Saenz Julienne 71ad4713cc util/event-loop-base: Introduce options to set the thread pool size
The thread pool regulates itself: when idle, it kills threads until
empty, when in demand, it creates new threads until full. This behaviour
doesn't play well with latency sensitive workloads where the price of
creating a new thread is too high. For example, when paired with qemu's
'-mlock', or using safety features like SafeStack, creating a new thread
has been measured take multiple milliseconds.

In order to mitigate this let's introduce a new 'EventLoopBase'
property to set the thread pool size. The threads will be created during
the pool's initialization or upon updating the property's value, remain
available during its lifetime regardless of demand, and destroyed upon
freeing it. A properly characterized workload will then be able to
configure the pool to avoid any latency spikes.

Signed-off-by: Nicolas Saenz Julienne <nsaenzju@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Markus Armbruster <armbru@redhat.com>
Message-id: 20220425075723.20019-4-nsaenzju@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2022-05-09 10:43:23 +01:00

400 lines
11 KiB
C

/*
* QEMU block layer thread pool
*
* Copyright IBM, Corp. 2008
* Copyright Red Hat, Inc. 2012
*
* Authors:
* Anthony Liguori <aliguori@us.ibm.com>
* Paolo Bonzini <pbonzini@redhat.com>
*
* This work is licensed under the terms of the GNU GPL, version 2. See
* the COPYING file in the top-level directory.
*
* Contributions after 2012-01-13 are licensed under the terms of the
* GNU GPL, version 2 or (at your option) any later version.
*/
#include "qemu/osdep.h"
#include "qemu/queue.h"
#include "qemu/thread.h"
#include "qemu/coroutine.h"
#include "trace.h"
#include "block/thread-pool.h"
#include "qemu/main-loop.h"
static void do_spawn_thread(ThreadPool *pool);
typedef struct ThreadPoolElement ThreadPoolElement;
enum ThreadState {
THREAD_QUEUED,
THREAD_ACTIVE,
THREAD_DONE,
};
struct ThreadPoolElement {
BlockAIOCB common;
ThreadPool *pool;
ThreadPoolFunc *func;
void *arg;
/* Moving state out of THREAD_QUEUED is protected by lock. After
* that, only the worker thread can write to it. Reads and writes
* of state and ret are ordered with memory barriers.
*/
enum ThreadState state;
int ret;
/* Access to this list is protected by lock. */
QTAILQ_ENTRY(ThreadPoolElement) reqs;
/* Access to this list is protected by the global mutex. */
QLIST_ENTRY(ThreadPoolElement) all;
};
struct ThreadPool {
AioContext *ctx;
QEMUBH *completion_bh;
QemuMutex lock;
QemuCond worker_stopped;
QemuSemaphore sem;
QEMUBH *new_thread_bh;
/* The following variables are only accessed from one AioContext. */
QLIST_HEAD(, ThreadPoolElement) head;
/* The following variables are protected by lock. */
QTAILQ_HEAD(, ThreadPoolElement) request_list;
int cur_threads;
int idle_threads;
int new_threads; /* backlog of threads we need to create */
int pending_threads; /* threads created but not running yet */
bool stopping;
int min_threads;
int max_threads;
};
static inline bool back_to_sleep(ThreadPool *pool, int ret)
{
/*
* The semaphore timed out, we should exit the loop except when:
* - There is work to do, we raced with the signal.
* - The max threads threshold just changed, we raced with the signal.
* - The thread pool forces a minimum number of readily available threads.
*/
if (ret == -1 && (!QTAILQ_EMPTY(&pool->request_list) ||
pool->cur_threads > pool->max_threads ||
pool->cur_threads <= pool->min_threads)) {
return true;
}
return false;
}
static void *worker_thread(void *opaque)
{
ThreadPool *pool = opaque;
qemu_mutex_lock(&pool->lock);
pool->pending_threads--;
do_spawn_thread(pool);
while (!pool->stopping) {
ThreadPoolElement *req;
int ret;
do {
pool->idle_threads++;
qemu_mutex_unlock(&pool->lock);
ret = qemu_sem_timedwait(&pool->sem, 10000);
qemu_mutex_lock(&pool->lock);
pool->idle_threads--;
} while (back_to_sleep(pool, ret));
if (ret == -1 || pool->stopping ||
pool->cur_threads > pool->max_threads) {
break;
}
req = QTAILQ_FIRST(&pool->request_list);
QTAILQ_REMOVE(&pool->request_list, req, reqs);
req->state = THREAD_ACTIVE;
qemu_mutex_unlock(&pool->lock);
ret = req->func(req->arg);
req->ret = ret;
/* Write ret before state. */
smp_wmb();
req->state = THREAD_DONE;
qemu_mutex_lock(&pool->lock);
qemu_bh_schedule(pool->completion_bh);
}
pool->cur_threads--;
qemu_cond_signal(&pool->worker_stopped);
qemu_mutex_unlock(&pool->lock);
return NULL;
}
static void do_spawn_thread(ThreadPool *pool)
{
QemuThread t;
/* Runs with lock taken. */
if (!pool->new_threads) {
return;
}
pool->new_threads--;
pool->pending_threads++;
qemu_thread_create(&t, "worker", worker_thread, pool, QEMU_THREAD_DETACHED);
}
static void spawn_thread_bh_fn(void *opaque)
{
ThreadPool *pool = opaque;
qemu_mutex_lock(&pool->lock);
do_spawn_thread(pool);
qemu_mutex_unlock(&pool->lock);
}
static void spawn_thread(ThreadPool *pool)
{
pool->cur_threads++;
pool->new_threads++;
/* If there are threads being created, they will spawn new workers, so
* we don't spend time creating many threads in a loop holding a mutex or
* starving the current vcpu.
*
* If there are no idle threads, ask the main thread to create one, so we
* inherit the correct affinity instead of the vcpu affinity.
*/
if (!pool->pending_threads) {
qemu_bh_schedule(pool->new_thread_bh);
}
}
static void thread_pool_completion_bh(void *opaque)
{
ThreadPool *pool = opaque;
ThreadPoolElement *elem, *next;
aio_context_acquire(pool->ctx);
restart:
QLIST_FOREACH_SAFE(elem, &pool->head, all, next) {
if (elem->state != THREAD_DONE) {
continue;
}
trace_thread_pool_complete(pool, elem, elem->common.opaque,
elem->ret);
QLIST_REMOVE(elem, all);
if (elem->common.cb) {
/* Read state before ret. */
smp_rmb();
/* Schedule ourselves in case elem->common.cb() calls aio_poll() to
* wait for another request that completed at the same time.
*/
qemu_bh_schedule(pool->completion_bh);
aio_context_release(pool->ctx);
elem->common.cb(elem->common.opaque, elem->ret);
aio_context_acquire(pool->ctx);
/* We can safely cancel the completion_bh here regardless of someone
* else having scheduled it meanwhile because we reenter the
* completion function anyway (goto restart).
*/
qemu_bh_cancel(pool->completion_bh);
qemu_aio_unref(elem);
goto restart;
} else {
qemu_aio_unref(elem);
}
}
aio_context_release(pool->ctx);
}
static void thread_pool_cancel(BlockAIOCB *acb)
{
ThreadPoolElement *elem = (ThreadPoolElement *)acb;
ThreadPool *pool = elem->pool;
trace_thread_pool_cancel(elem, elem->common.opaque);
QEMU_LOCK_GUARD(&pool->lock);
if (elem->state == THREAD_QUEUED &&
/* No thread has yet started working on elem. we can try to "steal"
* the item from the worker if we can get a signal from the
* semaphore. Because this is non-blocking, we can do it with
* the lock taken and ensure that elem will remain THREAD_QUEUED.
*/
qemu_sem_timedwait(&pool->sem, 0) == 0) {
QTAILQ_REMOVE(&pool->request_list, elem, reqs);
qemu_bh_schedule(pool->completion_bh);
elem->state = THREAD_DONE;
elem->ret = -ECANCELED;
}
}
static AioContext *thread_pool_get_aio_context(BlockAIOCB *acb)
{
ThreadPoolElement *elem = (ThreadPoolElement *)acb;
ThreadPool *pool = elem->pool;
return pool->ctx;
}
static const AIOCBInfo thread_pool_aiocb_info = {
.aiocb_size = sizeof(ThreadPoolElement),
.cancel_async = thread_pool_cancel,
.get_aio_context = thread_pool_get_aio_context,
};
BlockAIOCB *thread_pool_submit_aio(ThreadPool *pool,
ThreadPoolFunc *func, void *arg,
BlockCompletionFunc *cb, void *opaque)
{
ThreadPoolElement *req;
req = qemu_aio_get(&thread_pool_aiocb_info, NULL, cb, opaque);
req->func = func;
req->arg = arg;
req->state = THREAD_QUEUED;
req->pool = pool;
QLIST_INSERT_HEAD(&pool->head, req, all);
trace_thread_pool_submit(pool, req, arg);
qemu_mutex_lock(&pool->lock);
if (pool->idle_threads == 0 && pool->cur_threads < pool->max_threads) {
spawn_thread(pool);
}
QTAILQ_INSERT_TAIL(&pool->request_list, req, reqs);
qemu_mutex_unlock(&pool->lock);
qemu_sem_post(&pool->sem);
return &req->common;
}
typedef struct ThreadPoolCo {
Coroutine *co;
int ret;
} ThreadPoolCo;
static void thread_pool_co_cb(void *opaque, int ret)
{
ThreadPoolCo *co = opaque;
co->ret = ret;
aio_co_wake(co->co);
}
int coroutine_fn thread_pool_submit_co(ThreadPool *pool, ThreadPoolFunc *func,
void *arg)
{
ThreadPoolCo tpc = { .co = qemu_coroutine_self(), .ret = -EINPROGRESS };
assert(qemu_in_coroutine());
thread_pool_submit_aio(pool, func, arg, thread_pool_co_cb, &tpc);
qemu_coroutine_yield();
return tpc.ret;
}
void thread_pool_submit(ThreadPool *pool, ThreadPoolFunc *func, void *arg)
{
thread_pool_submit_aio(pool, func, arg, NULL, NULL);
}
void thread_pool_update_params(ThreadPool *pool, AioContext *ctx)
{
qemu_mutex_lock(&pool->lock);
pool->min_threads = ctx->thread_pool_min;
pool->max_threads = ctx->thread_pool_max;
/*
* We either have to:
* - Increase the number available of threads until over the min_threads
* threshold.
* - Decrease the number of available threads until under the max_threads
* threshold.
* - Do nothing. The current number of threads fall in between the min and
* max thresholds. We'll let the pool manage itself.
*/
for (int i = pool->cur_threads; i < pool->min_threads; i++) {
spawn_thread(pool);
}
for (int i = pool->cur_threads; i > pool->max_threads; i--) {
qemu_sem_post(&pool->sem);
}
qemu_mutex_unlock(&pool->lock);
}
static void thread_pool_init_one(ThreadPool *pool, AioContext *ctx)
{
if (!ctx) {
ctx = qemu_get_aio_context();
}
memset(pool, 0, sizeof(*pool));
pool->ctx = ctx;
pool->completion_bh = aio_bh_new(ctx, thread_pool_completion_bh, pool);
qemu_mutex_init(&pool->lock);
qemu_cond_init(&pool->worker_stopped);
qemu_sem_init(&pool->sem, 0);
pool->new_thread_bh = aio_bh_new(ctx, spawn_thread_bh_fn, pool);
QLIST_INIT(&pool->head);
QTAILQ_INIT(&pool->request_list);
thread_pool_update_params(pool, ctx);
}
ThreadPool *thread_pool_new(AioContext *ctx)
{
ThreadPool *pool = g_new(ThreadPool, 1);
thread_pool_init_one(pool, ctx);
return pool;
}
void thread_pool_free(ThreadPool *pool)
{
if (!pool) {
return;
}
assert(QLIST_EMPTY(&pool->head));
qemu_mutex_lock(&pool->lock);
/* Stop new threads from spawning */
qemu_bh_delete(pool->new_thread_bh);
pool->cur_threads -= pool->new_threads;
pool->new_threads = 0;
/* Wait for worker threads to terminate */
pool->stopping = true;
while (pool->cur_threads > 0) {
qemu_sem_post(&pool->sem);
qemu_cond_wait(&pool->worker_stopped, &pool->lock);
}
qemu_mutex_unlock(&pool->lock);
qemu_bh_delete(pool->completion_bh);
qemu_sem_destroy(&pool->sem);
qemu_cond_destroy(&pool->worker_stopped);
qemu_mutex_destroy(&pool->lock);
g_free(pool);
}