pcsx2/common/Darwin/DarwinThreads.cpp

280 lines
7.5 KiB
C++

// SPDX-FileCopyrightText: 2002-2024 PCSX2 Dev Team
// SPDX-License-Identifier: GPL-3.0+
#include "common/Threading.h"
#include "common/Assertions.h"
#include <cstdio>
#include <cassert> // assert
#include <sched.h>
#include <sys/time.h> // gettimeofday()
#include <pthread.h>
#include <unistd.h>
#include <mach/mach.h>
#include <mach/mach_error.h> // mach_error_string()
#include <mach/mach_init.h>
#include <mach/mach_port.h>
#include <mach/mach_time.h> // mach_absolute_time()
#include <mach/semaphore.h> // semaphore_*()
#include <mach/task.h> // semaphore_create() and semaphore_destroy()
#include <mach/thread_act.h>
// Note: assuming multicore is safer because it forces the interlocked routines to use
// the LOCK prefix. The prefix works on single core CPUs fine (but is slow), but not
// having the LOCK prefix is very bad indeed.
__forceinline void Threading::Timeslice()
{
sched_yield();
}
// For use in spin/wait loops, acts as a hint to Intel CPUs and should, in theory
// improve performance and reduce cpu power consumption.
__forceinline void Threading::SpinWait()
{
// If this doesn't compile you can just comment it out (it only serves as a
// performance hint and isn't required).
#if defined(_M_X86)
__asm__("pause");
#elif defined(_M_ARM64)
__asm__ __volatile__("isb");
#endif
}
__forceinline void Threading::EnableHiresScheduler()
{
// Darwin has customizable schedulers, see xnu/osfmk/man. Not
// implemented yet though (and not sure if useful for pcsx2).
}
__forceinline void Threading::DisableHiresScheduler()
{
// see EnableHiresScheduler()
}
// Just like on Windows, this is not really the number of ticks per second,
// but just a factor by which one has to divide GetThreadCpuTime() or
// pxThread::GetCpuTime() if one wants to receive a value in seconds. NOTE:
// doing this will of course yield precision loss.
u64 Threading::GetThreadTicksPerSecond()
{
return 1000000; // the *CpuTime() functions return values in microseconds
}
// gets the CPU time used by the current thread (both system and user), in
// microseconds, returns 0 on failure
static u64 getthreadtime(thread_port_t thread)
{
mach_msg_type_number_t count = THREAD_BASIC_INFO_COUNT;
thread_basic_info_data_t info;
kern_return_t kr = thread_info(thread, THREAD_BASIC_INFO,
(thread_info_t)&info, &count);
if (kr != KERN_SUCCESS)
{
return 0;
}
// add system and user time
return (u64)info.user_time.seconds * (u64)1e6 +
(u64)info.user_time.microseconds +
(u64)info.system_time.seconds * (u64)1e6 +
(u64)info.system_time.microseconds;
}
// Returns the current timestamp (not relative to a real world clock) in microseconds
u64 Threading::GetThreadCpuTime()
{
// we could also use mach_thread_self() and mach_port_deallocate(), but
// that calls upon mach traps (kinda like system calls). Unless I missed
// something in the COMMPAGE (like Linux vDSO) which makes overrides it
// to be user-space instead. In contract,
// pthread_mach_thread_np(pthread_self()) is entirely in user-space.
u64 us = getthreadtime(pthread_mach_thread_np(pthread_self()));
return us;
}
// --------------------------------------------------------------------------------------
// Semaphore Implementation for Darwin/OSX
//
// Sadly, Darwin/OSX needs its own implementation of Semaphores instead of
// relying on phtreads, because OSX unnamed semaphore (the best kind)
// support is very poor.
//
// This implementation makes use of Mach primitives instead. These are also
// what Grand Central Dispatch (GCD) is based on, as far as I understand:
// http://newosxbook.com/articles/GCD.html.
//
// --------------------------------------------------------------------------------------
static void MACH_CHECK(kern_return_t mach_retval)
{
if (mach_retval != KERN_SUCCESS)
{
fprintf(stderr, "mach error: %s", mach_error_string(mach_retval));
assert(mach_retval == KERN_SUCCESS);
}
}
Threading::KernelSemaphore::KernelSemaphore()
{
MACH_CHECK(semaphore_create(mach_task_self(), &m_sema, SYNC_POLICY_FIFO, 0));
}
Threading::KernelSemaphore::~KernelSemaphore()
{
MACH_CHECK(semaphore_destroy(mach_task_self(), m_sema));
}
void Threading::KernelSemaphore::Post()
{
MACH_CHECK(semaphore_signal(m_sema));
}
void Threading::KernelSemaphore::Wait()
{
MACH_CHECK(semaphore_wait(m_sema));
}
bool Threading::KernelSemaphore::TryWait()
{
mach_timespec_t time = {};
kern_return_t res = semaphore_timedwait(m_sema, time);
if (res == KERN_OPERATION_TIMED_OUT)
return false;
MACH_CHECK(res);
return true;
}
Threading::ThreadHandle::ThreadHandle() = default;
Threading::ThreadHandle::ThreadHandle(const ThreadHandle& handle)
: m_native_handle(handle.m_native_handle)
{
}
Threading::ThreadHandle::ThreadHandle(ThreadHandle&& handle)
: m_native_handle(handle.m_native_handle)
{
handle.m_native_handle = nullptr;
}
Threading::ThreadHandle::~ThreadHandle() = default;
Threading::ThreadHandle Threading::ThreadHandle::GetForCallingThread()
{
ThreadHandle ret;
ret.m_native_handle = pthread_self();
return ret;
}
Threading::ThreadHandle& Threading::ThreadHandle::operator=(ThreadHandle&& handle)
{
m_native_handle = handle.m_native_handle;
handle.m_native_handle = nullptr;
return *this;
}
Threading::ThreadHandle& Threading::ThreadHandle::operator=(const ThreadHandle& handle)
{
m_native_handle = handle.m_native_handle;
return *this;
}
u64 Threading::ThreadHandle::GetCPUTime() const
{
return getthreadtime(pthread_mach_thread_np((pthread_t)m_native_handle));
}
bool Threading::ThreadHandle::SetAffinity(u64 processor_mask) const
{
// Doesn't appear to be possible to set affinity.
return false;
}
Threading::Thread::Thread() = default;
Threading::Thread::Thread(Thread&& thread)
: ThreadHandle(thread)
, m_stack_size(thread.m_stack_size)
{
thread.m_stack_size = 0;
}
Threading::Thread::Thread(EntryPoint func)
: ThreadHandle()
{
if (!Start(std::move(func)))
pxFailRel("Failed to start implicitly started thread.");
}
Threading::Thread::~Thread()
{
pxAssertRel(!m_native_handle, "Thread should be detached or joined at destruction");
}
void Threading::Thread::SetStackSize(u32 size)
{
pxAssertRel(!m_native_handle, "Can't change the stack size on a started thread");
m_stack_size = size;
}
void* Threading::Thread::ThreadProc(void* param)
{
std::unique_ptr<EntryPoint> entry(static_cast<EntryPoint*>(param));
(*entry.get())();
return nullptr;
}
bool Threading::Thread::Start(EntryPoint func)
{
pxAssertRel(!m_native_handle, "Can't start an already-started thread");
std::unique_ptr<EntryPoint> func_clone(std::make_unique<EntryPoint>(std::move(func)));
pthread_attr_t attrs;
bool has_attributes = false;
if (m_stack_size != 0)
{
has_attributes = true;
pthread_attr_init(&attrs);
}
if (m_stack_size != 0)
pthread_attr_setstacksize(&attrs, m_stack_size);
pthread_t handle;
const int res = pthread_create(&handle, has_attributes ? &attrs : nullptr, ThreadProc, func_clone.get());
if (res != 0)
return false;
// thread started, it'll release the memory
m_native_handle = (void*)handle;
func_clone.release();
return true;
}
void Threading::Thread::Detach()
{
pxAssertRel(m_native_handle, "Can't detach without a thread");
pthread_detach((pthread_t)m_native_handle);
m_native_handle = nullptr;
}
void Threading::Thread::Join()
{
pxAssertRel(m_native_handle, "Can't join without a thread");
void* retval;
const int res = pthread_join((pthread_t)m_native_handle, &retval);
if (res != 0)
pxFailRel("pthread_join() for thread join failed");
m_native_handle = nullptr;
}
// name can be up to 16 bytes
void Threading::SetNameOfCurrentThread(const char* name)
{
pthread_setname_np(name);
}