2001-04-26 16:45:43 +00:00
|
|
|
/*
|
2011-04-08 13:03:26 -07:00
|
|
|
Simple DirectMedia Layer
|
|
|
|
Copyright (C) 1997-2011 Sam Lantinga <slouken@libsdl.org>
|
|
|
|
|
|
|
|
This software is provided 'as-is', without any express or implied
|
|
|
|
warranty. In no event will the authors be held liable for any damages
|
|
|
|
arising from the use of this software.
|
|
|
|
|
|
|
|
Permission is granted to anyone to use this software for any purpose,
|
|
|
|
including commercial applications, and to alter it and redistribute it
|
|
|
|
freely, subject to the following restrictions:
|
|
|
|
|
|
|
|
1. The origin of this software must not be misrepresented; you must not
|
|
|
|
claim that you wrote the original software. If you use this software
|
|
|
|
in a product, an acknowledgment in the product documentation would be
|
|
|
|
appreciated but is not required.
|
|
|
|
2. Altered source versions must be plainly marked as such, and must not be
|
|
|
|
misrepresented as being the original software.
|
|
|
|
3. This notice may not be removed or altered from any source distribution.
|
2001-04-26 16:45:43 +00:00
|
|
|
*/
|
2006-02-21 08:46:50 +00:00
|
|
|
#include "SDL_config.h"
|
2001-04-26 16:45:43 +00:00
|
|
|
|
2011-10-31 05:56:58 -04:00
|
|
|
#if SDL_THREAD_WINDOWS
|
|
|
|
|
2001-04-26 16:45:43 +00:00
|
|
|
/* Mutex functions using the Win32 API */
|
|
|
|
|
2011-01-24 21:20:30 -08:00
|
|
|
#include "../../core/windows/SDL_windows.h"
|
2001-04-26 16:45:43 +00:00
|
|
|
|
|
|
|
#include "SDL_mutex.h"
|
|
|
|
|
|
|
|
|
2006-07-10 21:04:37 +00:00
|
|
|
struct SDL_mutex
|
|
|
|
{
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
CRITICAL_SECTION cs;
|
2001-04-26 16:45:43 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
/* Create a mutex */
|
2006-07-10 21:04:37 +00:00
|
|
|
SDL_mutex *
|
|
|
|
SDL_CreateMutex(void)
|
2001-04-26 16:45:43 +00:00
|
|
|
{
|
2006-07-10 21:04:37 +00:00
|
|
|
SDL_mutex *mutex;
|
2001-04-26 16:45:43 +00:00
|
|
|
|
2006-07-10 21:04:37 +00:00
|
|
|
/* Allocate mutex memory */
|
|
|
|
mutex = (SDL_mutex *) SDL_malloc(sizeof(*mutex));
|
|
|
|
if (mutex) {
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
/* Initialize */
|
2011-02-17 09:26:15 -08:00
|
|
|
#ifdef _WIN32_WCE
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
InitializeCriticalSection(&mutex->cs);
|
2011-02-17 09:26:15 -08:00
|
|
|
#else
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
/* On SMP systems, a non-zero spin count generally helps performance */
|
2011-02-17 09:26:15 -08:00
|
|
|
InitializeCriticalSectionAndSpinCount(&mutex->cs, 2000);
|
|
|
|
#endif
|
2006-07-10 21:04:37 +00:00
|
|
|
} else {
|
|
|
|
SDL_OutOfMemory();
|
|
|
|
}
|
|
|
|
return (mutex);
|
2001-04-26 16:45:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Free the mutex */
|
2006-07-10 21:04:37 +00:00
|
|
|
void
|
|
|
|
SDL_DestroyMutex(SDL_mutex * mutex)
|
2001-04-26 16:45:43 +00:00
|
|
|
{
|
2006-07-10 21:04:37 +00:00
|
|
|
if (mutex) {
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
DeleteCriticalSection(&mutex->cs);
|
2006-07-10 21:04:37 +00:00
|
|
|
SDL_free(mutex);
|
|
|
|
}
|
2001-04-26 16:45:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Lock the mutex */
|
2006-07-10 21:04:37 +00:00
|
|
|
int
|
|
|
|
SDL_mutexP(SDL_mutex * mutex)
|
2001-04-26 16:45:43 +00:00
|
|
|
{
|
2006-07-10 21:04:37 +00:00
|
|
|
if (mutex == NULL) {
|
|
|
|
SDL_SetError("Passed a NULL mutex");
|
|
|
|
return -1;
|
|
|
|
}
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
|
|
|
|
EnterCriticalSection(&mutex->cs);
|
2006-07-10 21:04:37 +00:00
|
|
|
return (0);
|
2001-04-26 16:45:43 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Unlock the mutex */
|
2006-07-10 21:04:37 +00:00
|
|
|
int
|
|
|
|
SDL_mutexV(SDL_mutex * mutex)
|
2001-04-26 16:45:43 +00:00
|
|
|
{
|
2006-07-10 21:04:37 +00:00
|
|
|
if (mutex == NULL) {
|
|
|
|
SDL_SetError("Passed a NULL mutex");
|
|
|
|
return -1;
|
|
|
|
}
|
Fixed bug 1128
Patrick Baggett 2011-02-16 22:58:33 PST
This enhancement is for both x86/x64 Windows.
The SDL implementation of mutexes uses the Win32 API interprocess
synchronization primitive called a "Mutex". This implementation is subpar
because it has a much higher overhead than an intraprocess mutex. The exact
technical details are below, but my tests have shown that for reasonably high
contention (10 threads on 4 physical cores), it has 13x higher overhead than
the Win32 CriticalSection API.
If this enhancement is accepted, I will write a patch to implement SDL mutexes
using the critical section API, which should dramatically reduce overhead and
improve scalability.
-- Tech details --
Normally, Win32 Mutexes are used across process boundaries to synchronize
separate processes. In order to lock or unlock them, a user->kernel space
transition is necessary, even in the uncontented case on a single CPU machine.
Win32 CriticalSection objects can only be used within the same process virtual
address space and thus to lock one, does not require a user->kernel space
transition for the uncontended case, and additionally may spin a short while
before going into kernel wait. This small spin allows a thread to obtain the
lock if the mutex is released shortly after the thread starts spinning, in
effect bypassing the overhead of user->kernel space transition which has higher
overhead than the spinning itself.
2011-02-17 09:13:12 -08:00
|
|
|
|
|
|
|
LeaveCriticalSection(&mutex->cs);
|
2006-07-10 21:04:37 +00:00
|
|
|
return (0);
|
2001-04-26 16:45:43 +00:00
|
|
|
}
|
2006-07-10 21:04:37 +00:00
|
|
|
|
2011-10-31 05:56:58 -04:00
|
|
|
#endif /* SDL_THREAD_WINDOWS */
|
|
|
|
|
2006-07-10 21:04:37 +00:00
|
|
|
/* vi: set ts=4 sw=4 expandtab: */
|