motorhead/qemu
1
0
Fork 0
mirror of https://github.com/Motorhead1991/qemu.git synced 2025-07-30 13:53:54 -06:00
Code Issues Projects Releases Packages Wiki Activity Actions

qemu-thread: Use futex for QemuEvent on Windows

Browse source 
Use the futex-based implementation of QemuEvent on Windows to
remove code duplication and remove the overhead of event object
construction and destruction.

Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Link: https://lore.kernel.org/r/20250526-event-v4-6-5b784cc8e1de@daynix.com
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This commit is contained in:
Akihiko Odaki 2025-05-26 14:29:13 +09:00 • committed by Paolo Bonzini
parent d1895f4c17
commit 69e10db83e
7 changed files with 182 additions and 315 deletions
Download patch file Download diff file Expand all files Collapse all files

9
include/qemu/thread-posix.h
View file

@ -32,15 +32,6 @@ struct QemuSemaphore {
unsigned int count; unsigned int count;
}; };
struct QemuEvent {
#ifndef CONFIG_LINUX
pthread_mutex_t lock;
pthread_cond_t cond;
#endif
unsigned value;
bool initialized;
};
struct QemuThread { struct QemuThread {
pthread_t thread; pthread_t thread;
}; };

6
include/qemu/thread-win32.h
View file

@ -28,12 +28,6 @@ struct QemuSemaphore {
bool initialized; bool initialized;
}; };
struct QemuEvent {
int value;
HANDLE event;
bool initialized;
};
typedef struct QemuThreadData QemuThreadData; typedef struct QemuThreadData QemuThreadData;
struct QemuThread { struct QemuThread {
QemuThreadData *data; QemuThreadData *data;

11
include/qemu/thread.h
View file

@ -3,13 +3,22 @@
#include "qemu/processor.h" #include "qemu/processor.h"
#include "qemu/atomic.h" #include "qemu/atomic.h"
#include "qemu/futex.h"
typedef struct QemuCond QemuCond; typedef struct QemuCond QemuCond;
typedef struct QemuSemaphore QemuSemaphore; typedef struct QemuSemaphore QemuSemaphore;
typedef struct QemuEvent QemuEvent;
typedef struct QemuLockCnt QemuLockCnt; typedef struct QemuLockCnt QemuLockCnt;
typedef struct QemuThread QemuThread; typedef struct QemuThread QemuThread;
typedef struct QemuEvent {
#ifndef HAVE_FUTEX
pthread_mutex_t lock;
pthread_cond_t cond;
#endif
unsigned value;
bool initialized;
} QemuEvent;
#ifdef _WIN32 #ifdef _WIN32
#include "qemu/thread-win32.h" #include "qemu/thread-win32.h"
#else #else

171
util/event.c Normal file
View file

@ -0,0 +1,171 @@
/* SPDX-License-Identifier: GPL-2.0-or-later */
#include "qemu/osdep.h"
#include "qemu/thread.h"
/*
* Valid transitions:
* - FREE -> SET (qemu_event_set)
* - BUSY -> SET (qemu_event_set)
* - SET -> FREE (qemu_event_reset)
* - FREE -> BUSY (qemu_event_wait)
*
* With futex, the waking and blocking operations follow
* BUSY -> SET and FREE -> BUSY, respectively.
*
* Without futex, BUSY -> SET and FREE -> BUSY never happen. Instead, the waking
* operation follows FREE -> SET and the blocking operation will happen in
* qemu_event_wait() if the event is not SET.
*
* SET->BUSY does not happen (it can be observed from the outside but
* it really is SET->FREE->BUSY).
*
* busy->free provably cannot happen; to enforce it, the set->free transition
* is done with an OR, which becomes a no-op if the event has concurrently
* transitioned to free or busy.
*/
#define EV_SET 0
#define EV_FREE 1
#define EV_BUSY -1
void qemu_event_init(QemuEvent *ev, bool init)
{
#ifndef HAVE_FUTEX
pthread_mutex_init(&ev->lock, NULL);
pthread_cond_init(&ev->cond, NULL);
#endif
ev->value = (init ? EV_SET : EV_FREE);
ev->initialized = true;
}
void qemu_event_destroy(QemuEvent *ev)
{
assert(ev->initialized);
ev->initialized = false;
#ifndef HAVE_FUTEX
pthread_mutex_destroy(&ev->lock);
pthread_cond_destroy(&ev->cond);
#endif
}
void qemu_event_set(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef HAVE_FUTEX
/*
* Pairs with both qemu_event_reset() and qemu_event_wait().
*
* qemu_event_set has release semantics, but because it *loads*
* ev->value we need a full memory barrier here.
*/
smp_mb();
if (qatomic_read(&ev->value) != EV_SET) {
int old = qatomic_xchg(&ev->value, EV_SET);
/* Pairs with memory barrier in kernel futex_wait system call. */
smp_mb__after_rmw();
if (old == EV_BUSY) {
/* There were waiters, wake them up. */
qemu_futex_wake_all(ev);
}
}
#else
pthread_mutex_lock(&ev->lock);
/* Pairs with qemu_event_reset()'s load acquire. */
qatomic_store_release(&ev->value, EV_SET);
pthread_cond_broadcast(&ev->cond);
pthread_mutex_unlock(&ev->lock);
#endif
}
void qemu_event_reset(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef HAVE_FUTEX
/*
* If there was a concurrent reset (or even reset+wait),
* do nothing. Otherwise change EV_SET->EV_FREE.
*/
qatomic_or(&ev->value, EV_FREE);
/*
* Order reset before checking the condition in the caller.
* Pairs with the first memory barrier in qemu_event_set().
*/
smp_mb__after_rmw();
#else
/*
* If futexes are not available, there are no EV_FREE->EV_BUSY
* transitions because wakeups are done entirely through the
* condition variable. Since qatomic_set() only writes EV_FREE,
* the load seems useless but in reality, the acquire synchronizes
* with qemu_event_set()'s store release: if qemu_event_reset()
* sees EV_SET here, then the caller will certainly see a
* successful condition and skip qemu_event_wait():
*
* done = 1; if (done == 0)
* qemu_event_set() { qemu_event_reset() {
* lock();
* ev->value = EV_SET -----> load ev->value
* ev->value = old value | EV_FREE
* cond_broadcast()
* unlock(); }
* } if (done == 0)
* // qemu_event_wait() not called
*/
qatomic_set(&ev->value, qatomic_load_acquire(&ev->value) | EV_FREE);
#endif
}
void qemu_event_wait(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef HAVE_FUTEX
while (true) {
/*
* qemu_event_wait must synchronize with qemu_event_set even if it does
* not go down the slow path, so this load-acquire is needed that
* synchronizes with the first memory barrier in qemu_event_set().
*/
unsigned value = qatomic_load_acquire(&ev->value);
if (value == EV_SET) {
break;
}
if (value == EV_FREE) {
/*
* Leave the event reset and tell qemu_event_set that there are
* waiters. No need to retry, because there cannot be a concurrent
* busy->free transition. After the CAS, the event will be either
* set or busy.
*
* This cmpxchg doesn't have particular ordering requirements if it
* succeeds (moving the store earlier can only cause
* qemu_event_set() to issue _more_ wakeups), the failing case needs
* acquire semantics like the load above.
*/
if (qatomic_cmpxchg(&ev->value, EV_FREE, EV_BUSY) == EV_SET) {
break;
}
}
/*
* This is the final check for a concurrent set, so it does need
* a smp_mb() pairing with the second barrier of qemu_event_set().
* The barrier is inside the FUTEX_WAIT system call.
*/
qemu_futex_wait(ev, EV_BUSY);
}
#else
pthread_mutex_lock(&ev->lock);
while (qatomic_read(&ev->value) != EV_SET) {
pthread_cond_wait(&ev->cond, &ev->lock);
}
pthread_mutex_unlock(&ev->lock);
#endif
}

1
util/meson.build
View file

@ -35,6 +35,7 @@ if glib_has_gslice
endif endif
util_ss.add(files('defer-call.c')) util_ss.add(files('defer-call.c'))
util_ss.add(files('envlist.c', 'path.c', 'module.c')) util_ss.add(files('envlist.c', 'path.c', 'module.c'))
util_ss.add(files('event.c'))
util_ss.add(files('host-utils.c')) util_ss.add(files('host-utils.c'))
util_ss.add(files('bitmap.c', 'bitops.c')) util_ss.add(files('bitmap.c', 'bitops.c'))
util_ss.add(files('fifo8.c')) util_ss.add(files('fifo8.c'))

170
util/qemu-thread-posix.c
View file

@ -317,176 +317,6 @@ void qemu_sem_wait(QemuSemaphore *sem)
qemu_mutex_unlock(&sem->mutex); qemu_mutex_unlock(&sem->mutex);
} }
#ifdef CONFIG_LINUX
#include "qemu/futex.h"
#endif
/* Valid transitions:
* - FREE -> SET (qemu_event_set)
* - BUSY -> SET (qemu_event_set)
* - SET -> FREE (qemu_event_reset)
* - FREE -> BUSY (qemu_event_wait)
*
* With futex, the waking and blocking operations follow
* BUSY -> SET and FREE -> BUSY, respectively.
*
* Without futex, BUSY -> SET and FREE -> BUSY never happen. Instead, the waking
* operation follows FREE -> SET and the blocking operation will happen in
* qemu_event_wait() if the event is not SET.
*
* SET->BUSY does not happen (it can be observed from the outside but
* it really is SET->FREE->BUSY).
*
* busy->free provably cannot happen; to enforce it, the set->free transition
* is done with an OR, which becomes a no-op if the event has concurrently
* transitioned to free or busy.
*/
#define EV_SET 0
#define EV_FREE 1
#define EV_BUSY -1
void qemu_event_init(QemuEvent *ev, bool init)
{
#ifndef CONFIG_LINUX
pthread_mutex_init(&ev->lock, NULL);
pthread_cond_init(&ev->cond, NULL);
#endif
ev->value = (init ? EV_SET : EV_FREE);
ev->initialized = true;
}
void qemu_event_destroy(QemuEvent *ev)
{
assert(ev->initialized);
ev->initialized = false;
#ifndef CONFIG_LINUX
pthread_mutex_destroy(&ev->lock);
pthread_cond_destroy(&ev->cond);
#endif
}
void qemu_event_set(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef CONFIG_LINUX
/*
* Pairs with both qemu_event_reset() and qemu_event_wait().
*
* qemu_event_set has release semantics, but because it *loads*
* ev->value we need a full memory barrier here.
*/
smp_mb();
if (qatomic_read(&ev->value) != EV_SET) {
int old = qatomic_xchg(&ev->value, EV_SET);
/* Pairs with memory barrier in kernel futex_wait system call. */
smp_mb__after_rmw();
if (old == EV_BUSY) {
/* There were waiters, wake them up. */
qemu_futex_wake_all(ev);
}
}
#else
pthread_mutex_lock(&ev->lock);
/* Pairs with qemu_event_reset()'s load acquire. */
qatomic_store_release(&ev->value, EV_SET);
pthread_cond_broadcast(&ev->cond);
pthread_mutex_unlock(&ev->lock);
#endif
}
void qemu_event_reset(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef CONFIG_LINUX
/*
* If there was a concurrent reset (or even reset+wait),
* do nothing. Otherwise change EV_SET->EV_FREE.
*/
qatomic_or(&ev->value, EV_FREE);
/*
* Order reset before checking the condition in the caller.
* Pairs with the first memory barrier in qemu_event_set().
*/
smp_mb__after_rmw();
#else
/*
* If futexes are not available, there are no EV_FREE->EV_BUSY
* transitions because wakeups are done entirely through the
* condition variable. Since qatomic_set() only writes EV_FREE,
* the load seems useless but in reality, the acquire synchronizes
* with qemu_event_set()'s store release: if qemu_event_reset()
* sees EV_SET here, then the caller will certainly see a
* successful condition and skip qemu_event_wait():
*
* done = 1; if (done == 0)
* qemu_event_set() { qemu_event_reset() {
* lock();
* ev->value = EV_SET -----> load ev->value
* ev->value = old value | EV_FREE
* cond_broadcast()
* unlock(); }
* } if (done == 0)
* // qemu_event_wait() not called
*/
qatomic_set(&ev->value, qatomic_load_acquire(&ev->value) | EV_FREE);
#endif
}
void qemu_event_wait(QemuEvent *ev)
{
assert(ev->initialized);
#ifdef CONFIG_LINUX
while (true) {
/*
* qemu_event_wait must synchronize with qemu_event_set even if it does
* not go down the slow path, so this load-acquire is needed that
* synchronizes with the first memory barrier in qemu_event_set().
*/
unsigned value = qatomic_load_acquire(&ev->value);
if (value == EV_SET) {
break;
}
if (value == EV_FREE) {
/*
* Leave the event reset and tell qemu_event_set that there are
* waiters. No need to retry, because there cannot be a concurrent
* busy->free transition. After the CAS, the event will be either
* set or busy.
*
* This cmpxchg doesn't have particular ordering requirements if it
* succeeds (moving the store earlier can only cause qemu_event_set()
* to issue _more_ wakeups), the failing case needs acquire semantics
* like the load above.
*/
if (qatomic_cmpxchg(&ev->value, EV_FREE, EV_BUSY) == EV_SET) {
break;
}
}
/*
* This is the final check for a concurrent set, so it does need
* a smp_mb() pairing with the second barrier of qemu_event_set().
* The barrier is inside the FUTEX_WAIT system call.
*/
qemu_futex_wait(ev, EV_BUSY);
}
#else
pthread_mutex_lock(&ev->lock);
while (qatomic_read(&ev->value) != EV_SET) {
pthread_cond_wait(&ev->cond, &ev->lock);
}
pthread_mutex_unlock(&ev->lock);
#endif
}
static __thread NotifierList thread_exit; static __thread NotifierList thread_exit;
/* /*

129
util/qemu-thread-win32.c
View file

@ -231,135 +231,6 @@ void qemu_sem_wait(QemuSemaphore *sem)
} }
} }
/* Wrap a Win32 manual-reset event with a fast userspace path. The idea
* is to reset the Win32 event lazily, as part of a test-reset-test-wait
* sequence. Such a sequence is, indeed, how QemuEvents are used by
* RCU and other subsystems!
*
* Valid transitions:
* - free->set, when setting the event
* - busy->set, when setting the event, followed by SetEvent
* - set->free, when resetting the event
* - free->busy, when waiting
*
* set->busy does not happen (it can be observed from the outside but
* it really is set->free->busy).
*
* busy->free provably cannot happen; to enforce it, the set->free transition
* is done with an OR, which becomes a no-op if the event has concurrently
* transitioned to free or busy (and is faster than cmpxchg).
*/
#define EV_SET 0
#define EV_FREE 1
#define EV_BUSY -1
void qemu_event_init(QemuEvent *ev, bool init)
{
/* Manual reset. */
ev->event = CreateEvent(NULL, TRUE, TRUE, NULL);
ev->value = (init ? EV_SET : EV_FREE);
ev->initialized = true;
}
void qemu_event_destroy(QemuEvent *ev)
{
assert(ev->initialized);
ev->initialized = false;
CloseHandle(ev->event);
}
void qemu_event_set(QemuEvent *ev)
{
assert(ev->initialized);
/*
* Pairs with both qemu_event_reset() and qemu_event_wait().
*
* qemu_event_set has release semantics, but because it *loads*
* ev->value we need a full memory barrier here.
*/
smp_mb();
if (qatomic_read(&ev->value) != EV_SET) {
int old = qatomic_xchg(&ev->value, EV_SET);
/* Pairs with memory barrier after ResetEvent. */
smp_mb__after_rmw();
if (old == EV_BUSY) {
/* There were waiters, wake them up. */
SetEvent(ev->event);
}
}
}
void qemu_event_reset(QemuEvent *ev)
{
assert(ev->initialized);
/*
* If there was a concurrent reset (or even reset+wait),
* do nothing. Otherwise change EV_SET->EV_FREE.
*/
qatomic_or(&ev->value, EV_FREE);
/*
* Order reset before checking the condition in the caller.
* Pairs with the first memory barrier in qemu_event_set().
*/
smp_mb__after_rmw();
}
void qemu_event_wait(QemuEvent *ev)
{
unsigned value;
assert(ev->initialized);
/*
* qemu_event_wait must synchronize with qemu_event_set even if it does
* not go down the slow path, so this load-acquire is needed that
* synchronizes with the first memory barrier in qemu_event_set().
*
* If we do go down the slow path, there is no requirement at all: we
* might miss a qemu_event_set() here but ultimately the memory barrier in
* qemu_futex_wait() will ensure the check is done correctly.
*/
value = qatomic_load_acquire(&ev->value);
if (value != EV_SET) {
if (value == EV_FREE) {
/*
* Here the underlying kernel event is reset, but qemu_event_set is
* not yet going to call SetEvent. However, there will be another
* check for EV_SET below when setting EV_BUSY. At that point it
* is safe to call WaitForSingleObject.
*/
ResetEvent(ev->event);
/*
* It is not clear whether ResetEvent provides this barrier; kernel
* APIs (KeResetEvent/KeClearEvent) do not. Better safe than sorry!
*/
smp_mb();
/*
* Leave the event reset and tell qemu_event_set that there are
* waiters. No need to retry, because there cannot be a concurrent
* busy->free transition. After the CAS, the event will be either
* set or busy.
*/
if (qatomic_cmpxchg(&ev->value, EV_FREE, EV_BUSY) == EV_SET) {
return;
}
}
/*
* ev->value is now EV_BUSY. Since we didn't observe EV_SET,
* qemu_event_set() must observe EV_BUSY and call SetEvent().
*/
WaitForSingleObject(ev->event, INFINITE);
}
}
struct QemuThreadData { struct QemuThreadData {
/* Passed to win32_start_routine. */ /* Passed to win32_start_routine. */
void *(*start_routine)(void *); void *(*start_routine)(void *);