Merge pull request #425 from Subv/coretiming

Ported the CoreTiming namespace from PPSSPP
This commit is contained in:
bunnei 2015-01-07 15:30:46 -05:00
commit 3eca33c6a4
6 changed files with 372 additions and 410 deletions

View File

@ -22,6 +22,7 @@ static std::shared_ptr<Logger> global_logger;
SUB(Common, Memory) \
CLS(Core) \
SUB(Core, ARM11) \
SUB(Core, Timing) \
CLS(Config) \
CLS(Debug) \
SUB(Debug, Emulated) \

View File

@ -41,6 +41,7 @@ enum class Class : ClassType {
Common_Memory, ///< Memory mapping and management functions
Core, ///< LLE emulation core
Core_ARM11, ///< ARM11 CPU core
Core_Timing, ///< CoreTiming functions
Config, ///< Emulator configuration (including commandline)
Debug, ///< Debugging tools
Debug_Emulated, ///< Debug messages from the emulated programs

View File

@ -103,6 +103,8 @@ public:
return num_instructions;
}
s64 down_count; ///< A decreasing counter of remaining cycles before the next event, decreased by the cpu run loop
protected:
/**

View File

@ -9,6 +9,8 @@
#include "core/arm/dyncom/arm_dyncom.h"
#include "core/arm/dyncom/arm_dyncom_interpreter.h"
#include "core/core_timing.h"
const static cpu_config_t s_arm11_cpu_info = {
"armv6", "arm11", 0x0007b000, 0x0007f000, NONCACHE
};
@ -77,6 +79,9 @@ u64 ARM_DynCom::GetTicks() const {
void ARM_DynCom::AddTicks(u64 ticks) {
this->ticks += ticks;
down_count -= ticks;
if (down_count < 0)
CoreTiming::Advance();
}
void ARM_DynCom::ExecuteInstructions(int num_instructions) {
@ -85,7 +90,8 @@ void ARM_DynCom::ExecuteInstructions(int num_instructions) {
// Dyncom only breaks on instruction dispatch. This only happens on every instruction when
// executing one instruction at a time. Otherwise, if a block is being executed, more
// instructions may actually be executed than specified.
ticks += InterpreterMainLoop(state.get());
unsigned ticks_executed = InterpreterMainLoop(state.get());
AddTicks(ticks_executed);
}
void ARM_DynCom::SaveContext(ThreadContext& ctx) {

View File

@ -1,16 +1,14 @@
// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
// Copyright (c) 2012- PPSSPP Project / Dolphin Project.
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#include <vector>
#include <cstdio>
#include <atomic>
#include <cstdio>
#include <mutex>
#include <vector>
#include "common/chunk_file.h"
#include "common/msg_handler.h"
#include "common/string_util.h"
#include "common/log.h"
#include "core/core.h"
#include "core/core_timing.h"
@ -22,16 +20,15 @@ int g_clock_rate_arm11 = 268123480;
namespace CoreTiming
{
struct EventType
{
EventType() {}
EventType(TimedCallback cb, const char *n)
EventType(TimedCallback cb, const char* n)
: callback(cb), name(n) {}
TimedCallback callback;
const char *name;
const char* name;
};
std::vector<EventType> event_types;
@ -41,262 +38,247 @@ struct BaseEvent
s64 time;
u64 userdata;
int type;
// Event *next;
};
typedef LinkedListItem<BaseEvent> Event;
Event *first;
Event *tsFirst;
Event *tsLast;
Event* first;
Event* ts_first;
Event* ts_last;
// event pools
Event *eventPool = 0;
Event *eventTsPool = 0;
int allocatedTsEvents = 0;
Event* event_pool = 0;
Event* event_ts_pool = 0;
int allocated_ts_events = 0;
// Optimization to skip MoveEvents when possible.
std::atomic<u32> hasTsEvents;
std::atomic<bool> has_ts_events(false);
// Downcount has been moved to currentMIPS, to save a couple of clocks in every ARM JIT block
// as we can already reach that structure through a register.
int slicelength;
int g_slice_length;
MEMORY_ALIGNED16(s64) globalTimer;
s64 idledCycles;
s64 global_timer;
s64 idled_cycles;
s64 last_global_time_ticks;
s64 last_global_time_us;
static std::recursive_mutex externalEventSection;
static std::recursive_mutex external_event_section;
// Warning: not included in save state.
void(*advanceCallback)(int cyclesExecuted) = nullptr;
using AdvanceCallback = void(int cycles_executed);
AdvanceCallback* advance_callback = nullptr;
std::vector<MHzChangeCallback> mhz_change_callbacks;
void SetClockFrequencyMHz(int cpuMhz)
{
g_clock_rate_arm11 = cpuMhz * 1000000;
// TODO: Rescale times of scheduled events?
void FireMhzChange() {
for (auto callback : mhz_change_callbacks)
callback();
}
int GetClockFrequencyMHz()
{
void SetClockFrequencyMHz(int cpu_mhz) {
// When the mhz changes, we keep track of what "time" it was before hand.
// This way, time always moves forward, even if mhz is changed.
last_global_time_us = GetGlobalTimeUs();
last_global_time_ticks = GetTicks();
g_clock_rate_arm11 = cpu_mhz * 1000000;
// TODO: Rescale times of scheduled events?
FireMhzChange();
}
int GetClockFrequencyMHz() {
return g_clock_rate_arm11 / 1000000;
}
u64 GetGlobalTimeUs() {
s64 ticks_since_last = GetTicks() - last_global_time_ticks;
int freq = GetClockFrequencyMHz();
s64 us_since_last = ticks_since_last / freq;
return last_global_time_us + us_since_last;
}
Event* GetNewEvent()
{
if (!eventPool)
Event* GetNewEvent() {
if (!event_pool)
return new Event;
Event* ev = eventPool;
eventPool = ev->next;
return ev;
Event* event = event_pool;
event_pool = event->next;
return event;
}
Event* GetNewTsEvent()
{
allocatedTsEvents++;
Event* GetNewTsEvent() {
allocated_ts_events++;
if (!eventTsPool)
if (!event_ts_pool)
return new Event;
Event* ev = eventTsPool;
eventTsPool = ev->next;
return ev;
Event* event = event_ts_pool;
event_ts_pool = event->next;
return event;
}
void FreeEvent(Event* ev)
{
ev->next = eventPool;
eventPool = ev;
void FreeEvent(Event* event) {
event->next = event_pool;
event_pool = event;
}
void FreeTsEvent(Event* ev)
{
ev->next = eventTsPool;
eventTsPool = ev;
allocatedTsEvents--;
void FreeTsEvent(Event* event) {
event->next = event_ts_pool;
event_ts_pool = event;
allocated_ts_events--;
}
int RegisterEvent(const char *name, TimedCallback callback)
{
int RegisterEvent(const char* name, TimedCallback callback) {
event_types.push_back(EventType(callback, name));
return (int)event_types.size() - 1;
}
void AntiCrashCallback(u64 userdata, int cyclesLate)
{
LOG_CRITICAL(Core, "Savestate broken: an unregistered event was called.");
void AntiCrashCallback(u64 userdata, int cycles_late) {
LOG_CRITICAL(Core_Timing, "Savestate broken: an unregistered event was called.");
Core::Halt("invalid timing events");
}
void RestoreRegisterEvent(int event_type, const char *name, TimedCallback callback)
{
void RestoreRegisterEvent(int event_type, const char* name, TimedCallback callback) {
if (event_type >= (int)event_types.size())
event_types.resize(event_type + 1, EventType(AntiCrashCallback, "INVALID EVENT"));
event_types[event_type] = EventType(callback, name);
}
void UnregisterAllEvents()
{
void UnregisterAllEvents() {
if (first)
PanicAlert("Cannot unregister events with events pending");
event_types.clear();
}
void Init()
{
//currentMIPS->downcount = INITIAL_SLICE_LENGTH;
//slicelength = INITIAL_SLICE_LENGTH;
globalTimer = 0;
idledCycles = 0;
hasTsEvents = 0;
void Init() {
Core::g_app_core->down_count = INITIAL_SLICE_LENGTH;
g_slice_length = INITIAL_SLICE_LENGTH;
global_timer = 0;
idled_cycles = 0;
last_global_time_ticks = 0;
last_global_time_us = 0;
has_ts_events = 0;
mhz_change_callbacks.clear();
}
void Shutdown()
{
void Shutdown() {
MoveEvents();
ClearPendingEvents();
UnregisterAllEvents();
while (eventPool)
{
Event *ev = eventPool;
eventPool = ev->next;
delete ev;
while (event_pool) {
Event* event = event_pool;
event_pool = event->next;
delete event;
}
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
while (eventTsPool)
{
Event *ev = eventTsPool;
eventTsPool = ev->next;
delete ev;
std::lock_guard<std::recursive_mutex> lock(external_event_section);
while (event_ts_pool) {
Event* event = event_ts_pool;
event_ts_pool = event->next;
delete event;
}
}
u64 GetTicks()
{
LOG_ERROR(Core, "Unimplemented function!");
return 0;
//return (u64)globalTimer + slicelength - currentMIPS->downcount;
u64 GetTicks() {
return (u64)global_timer + g_slice_length - Core::g_app_core->down_count;
}
u64 GetIdleTicks()
{
return (u64)idledCycles;
u64 GetIdleTicks() {
return (u64)idled_cycles;
}
// This is to be called when outside threads, such as the graphics thread, wants to
// schedule things to be executed on the main thread.
void ScheduleEvent_Threadsafe(s64 cyclesIntoFuture, int event_type, u64 userdata)
{
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
Event *ne = GetNewTsEvent();
ne->time = GetTicks() + cyclesIntoFuture;
ne->type = event_type;
ne->next = 0;
ne->userdata = userdata;
if (!tsFirst)
tsFirst = ne;
if (tsLast)
tsLast->next = ne;
tsLast = ne;
void ScheduleEvent_Threadsafe(s64 cycles_into_future, int event_type, u64 userdata) {
std::lock_guard<std::recursive_mutex> lock(external_event_section);
Event* new_event = GetNewTsEvent();
new_event->time = GetTicks() + cycles_into_future;
new_event->type = event_type;
new_event->next = 0;
new_event->userdata = userdata;
if (!ts_first)
ts_first = new_event;
if (ts_last)
ts_last->next = new_event;
ts_last = new_event;
hasTsEvents.store(1, std::memory_order_release);
has_ts_events = true;
}
// Same as ScheduleEvent_Threadsafe(0, ...) EXCEPT if we are already on the CPU thread
// in which case the event will get handled immediately, before returning.
void ScheduleEvent_Threadsafe_Immediate(int event_type, u64 userdata)
{
void ScheduleEvent_Threadsafe_Immediate(int event_type, u64 userdata) {
if (false) //Core::IsCPUThread())
{
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
std::lock_guard<std::recursive_mutex> lock(external_event_section);
event_types[event_type].callback(userdata, 0);
}
else
ScheduleEvent_Threadsafe(0, event_type, userdata);
}
void ClearPendingEvents()
{
while (first)
{
Event *e = first->next;
void ClearPendingEvents() {
while (first) {
Event* event = first->next;
FreeEvent(first);
first = e;
first = event;
}
}
void AddEventToQueue(Event* ne)
{
Event* prev = nullptr;
Event** pNext = &first;
for (;;)
{
Event*& next = *pNext;
if (!next || ne->time < next->time)
{
ne->next = next;
next = ne;
void AddEventToQueue(Event* new_event) {
Event* prev_event = nullptr;
Event** next_event = &first;
for (;;) {
Event*& next = *next_event;
if (!next || new_event->time < next->time) {
new_event->next = next;
next = new_event;
break;
}
prev = next;
pNext = &prev->next;
prev_event = next;
next_event = &prev_event->next;
}
}
// This must be run ONLY from within the cpu thread
// cyclesIntoFuture may be VERY inaccurate if called from anything else
// than Advance
void ScheduleEvent(s64 cyclesIntoFuture, int event_type, u64 userdata)
{
Event *ne = GetNewEvent();
ne->userdata = userdata;
ne->type = event_type;
ne->time = GetTicks() + cyclesIntoFuture;
AddEventToQueue(ne);
void ScheduleEvent(s64 cycles_into_future, int event_type, u64 userdata) {
Event* new_event = GetNewEvent();
new_event->userdata = userdata;
new_event->type = event_type;
new_event->time = GetTicks() + cycles_into_future;
AddEventToQueue(new_event);
}
// Returns cycles left in timer.
s64 UnscheduleEvent(int event_type, u64 userdata)
{
s64 UnscheduleEvent(int event_type, u64 userdata) {
s64 result = 0;
if (!first)
return result;
while (first)
{
if (first->type == event_type && first->userdata == userdata)
{
result = first->time - globalTimer;
while (first) {
if (first->type == event_type && first->userdata == userdata) {
result = first->time - GetTicks();
Event *next = first->next;
Event* next = first->next;
FreeEvent(first);
first = next;
}
else
{
} else {
break;
}
}
if (!first)
return result;
Event *prev = first;
Event *ptr = prev->next;
while (ptr)
{
if (ptr->type == event_type && ptr->userdata == userdata)
{
result = ptr->time - globalTimer;
prev->next = ptr->next;
Event* prev_event = first;
Event* ptr = prev_event->next;
while (ptr) {
if (ptr->type == event_type && ptr->userdata == userdata) {
result = ptr->time - GetTicks();
prev_event->next = ptr->next;
FreeEvent(ptr);
ptr = prev->next;
}
else
{
prev = ptr;
ptr = prev_event->next;
} else {
prev_event = ptr;
ptr = ptr->next;
}
}
@ -304,51 +286,44 @@ s64 UnscheduleEvent(int event_type, u64 userdata)
return result;
}
s64 UnscheduleThreadsafeEvent(int event_type, u64 userdata)
{
s64 UnscheduleThreadsafeEvent(int event_type, u64 userdata) {
s64 result = 0;
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
if (!tsFirst)
std::lock_guard<std::recursive_mutex> lock(external_event_section);
if (!ts_first)
return result;
while (tsFirst)
{
if (tsFirst->type == event_type && tsFirst->userdata == userdata)
{
result = tsFirst->time - globalTimer;
Event *next = tsFirst->next;
FreeTsEvent(tsFirst);
tsFirst = next;
}
else
{
while (ts_first) {
if (ts_first->type == event_type && ts_first->userdata == userdata) {
result = ts_first->time - GetTicks();
Event* next = ts_first->next;
FreeTsEvent(ts_first);
ts_first = next;
} else {
break;
}
}
if (!tsFirst)
if (!ts_first)
{
tsLast = nullptr;
ts_last = nullptr;
return result;
}
Event *prev = tsFirst;
Event *ptr = prev->next;
while (ptr)
{
if (ptr->type == event_type && ptr->userdata == userdata)
{
result = ptr->time - globalTimer;
Event* prev_event = ts_first;
Event* next = prev_event->next;
while (next) {
if (next->type == event_type && next->userdata == userdata) {
result = next->time - GetTicks();
prev->next = ptr->next;
if (ptr == tsLast)
tsLast = prev;
FreeTsEvent(ptr);
ptr = prev->next;
}
else
{
prev = ptr;
ptr = ptr->next;
prev_event->next = next->next;
if (next == ts_last)
ts_last = prev_event;
FreeTsEvent(next);
next = prev_event->next;
} else {
prev_event = next;
next = next->next;
}
}
@ -356,271 +331,217 @@ s64 UnscheduleThreadsafeEvent(int event_type, u64 userdata)
}
// Warning: not included in save state.
void RegisterAdvanceCallback(void(*callback)(int cyclesExecuted))
{
advanceCallback = callback;
void RegisterAdvanceCallback(AdvanceCallback* callback) {
advance_callback = callback;
}
bool IsScheduled(int event_type)
{
void RegisterMHzChangeCallback(MHzChangeCallback callback) {
mhz_change_callbacks.push_back(callback);
}
bool IsScheduled(int event_type) {
if (!first)
return false;
Event *e = first;
while (e) {
if (e->type == event_type)
Event* event = first;
while (event) {
if (event->type == event_type)
return true;
e = e->next;
event = event->next;
}
return false;
}
void RemoveEvent(int event_type)
{
void RemoveEvent(int event_type) {
if (!first)
return;
while (first)
{
if (first->type == event_type)
{
while (first) {
if (first->type == event_type) {
Event *next = first->next;
FreeEvent(first);
first = next;
}
else
{
} else {
break;
}
}
if (!first)
return;
Event *prev = first;
Event *ptr = prev->next;
while (ptr)
{
if (ptr->type == event_type)
{
prev->next = ptr->next;
FreeEvent(ptr);
ptr = prev->next;
}
else
{
prev = ptr;
ptr = ptr->next;
Event* prev = first;
Event* next = prev->next;
while (next) {
if (next->type == event_type) {
prev->next = next->next;
FreeEvent(next);
next = prev->next;
} else {
prev = next;
next = next->next;
}
}
}
void RemoveThreadsafeEvent(int event_type)
{
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
if (!tsFirst)
{
void RemoveThreadsafeEvent(int event_type) {
std::lock_guard<std::recursive_mutex> lock(external_event_section);
if (!ts_first)
return;
}
while (tsFirst)
{
if (tsFirst->type == event_type)
{
Event *next = tsFirst->next;
FreeTsEvent(tsFirst);
tsFirst = next;
}
else
{
while (ts_first) {
if (ts_first->type == event_type) {
Event* next = ts_first->next;
FreeTsEvent(ts_first);
ts_first = next;
} else {
break;
}
}
if (!tsFirst)
{
tsLast = nullptr;
if (!ts_first) {
ts_last = nullptr;
return;
}
Event *prev = tsFirst;
Event *ptr = prev->next;
while (ptr)
{
if (ptr->type == event_type)
{
prev->next = ptr->next;
if (ptr == tsLast)
tsLast = prev;
FreeTsEvent(ptr);
ptr = prev->next;
}
else
{
prev = ptr;
ptr = ptr->next;
Event* prev = ts_first;
Event* next = prev->next;
while (next) {
if (next->type == event_type) {
prev->next = next->next;
if (next == ts_last)
ts_last = prev;
FreeTsEvent(next);
next = prev->next;
} else {
prev = next;
next = next->next;
}
}
}
void RemoveAllEvents(int event_type)
{
void RemoveAllEvents(int event_type) {
RemoveThreadsafeEvent(event_type);
RemoveEvent(event_type);
}
//This raise only the events required while the fifo is processing data
void ProcessFifoWaitEvents()
{
while (first)
{
if (first->time <= globalTimer)
{
//LOG(TIMER, "[Scheduler] %s (%lld, %lld) ",
// first->name ? first->name : "?", (u64)globalTimer, (u64)first->time);
// This raise only the events required while the fifo is processing data
void ProcessFifoWaitEvents() {
while (first) {
if (first->time <= (s64)GetTicks()) {
Event* evt = first;
first = first->next;
event_types[evt->type].callback(evt->userdata, (int)(globalTimer - evt->time));
event_types[evt->type].callback(evt->userdata, (int)(GetTicks() - evt->time));
FreeEvent(evt);
}
else
{
} else {
break;
}
}
}
void MoveEvents()
{
hasTsEvents.store(0, std::memory_order_release);
void MoveEvents() {
has_ts_events = false;
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
std::lock_guard<std::recursive_mutex> lock(external_event_section);
// Move events from async queue into main queue
while (tsFirst)
{
Event *next = tsFirst->next;
AddEventToQueue(tsFirst);
tsFirst = next;
while (ts_first) {
Event* next = ts_first->next;
AddEventToQueue(ts_first);
ts_first = next;
}
tsLast = nullptr;
ts_last = nullptr;
// Move free events to threadsafe pool
while (allocatedTsEvents > 0 && eventPool)
{
Event *ev = eventPool;
eventPool = ev->next;
ev->next = eventTsPool;
eventTsPool = ev;
allocatedTsEvents--;
while (allocated_ts_events > 0 && event_pool) {
Event* event = event_pool;
event_pool = event->next;
event->next = event_ts_pool;
event_ts_pool = event;
allocated_ts_events--;
}
}
void Advance()
{
LOG_ERROR(Core, "Unimplemented function!");
//int cyclesExecuted = slicelength - currentMIPS->downcount;
//globalTimer += cyclesExecuted;
//currentMIPS->downcount = slicelength;
//if (Common::AtomicLoadAcquire(hasTsEvents))
// MoveEvents();
//ProcessFifoWaitEvents();
//if (!first)
//{
// // WARN_LOG(TIMER, "WARNING - no events in queue. Setting currentMIPS->downcount to 10000");
// currentMIPS->downcount += 10000;
//}
//else
//{
// slicelength = (int)(first->time - globalTimer);
// if (slicelength > MAX_SLICE_LENGTH)
// slicelength = MAX_SLICE_LENGTH;
// currentMIPS->downcount = slicelength;
//}
//if (advanceCallback)
// advanceCallback(cyclesExecuted);
void ForceCheck() {
int cycles_executed = g_slice_length - Core::g_app_core->down_count;
global_timer += cycles_executed;
// This will cause us to check for new events immediately.
Core::g_app_core->down_count = 0;
// But let's not eat a bunch more time in Advance() because of this.
g_slice_length = 0;
}
void LogPendingEvents()
{
Event *ptr = first;
while (ptr)
{
//INFO_LOG(TIMER, "PENDING: Now: %lld Pending: %lld Type: %d", globalTimer, ptr->time, ptr->type);
ptr = ptr->next;
void Advance() {
int cycles_executed = g_slice_length - Core::g_app_core->down_count;
global_timer += cycles_executed;
Core::g_app_core->down_count = g_slice_length;
if (has_ts_events)
MoveEvents();
ProcessFifoWaitEvents();
if (!first) {
if (g_slice_length < 10000) {
g_slice_length += 10000;
Core::g_app_core->down_count += g_slice_length;
}
} else {
// Note that events can eat cycles as well.
int target = (int)(first->time - global_timer);
if (target > MAX_SLICE_LENGTH)
target = MAX_SLICE_LENGTH;
const int diff = target - g_slice_length;
g_slice_length += diff;
Core::g_app_core->down_count += diff;
}
if (advance_callback)
advance_callback(cycles_executed);
}
void LogPendingEvents() {
Event* event = first;
while (event) {
//LOG_TRACE(Core_Timing, "PENDING: Now: %lld Pending: %lld Type: %d", globalTimer, next->time, next->type);
event = event->next;
}
}
void Idle(int maxIdle)
{
LOG_ERROR(Core, "Unimplemented function!");
//int cyclesDown = currentMIPS->downcount;
//if (maxIdle != 0 && cyclesDown > maxIdle)
// cyclesDown = maxIdle;
void Idle(int max_idle) {
int cycles_down = Core::g_app_core->down_count;
if (max_idle != 0 && cycles_down > max_idle)
cycles_down = max_idle;
//if (first && cyclesDown > 0)
//{
// int cyclesExecuted = slicelength - currentMIPS->downcount;
// int cyclesNextEvent = (int) (first->time - globalTimer);
if (first && cycles_down > 0) {
int cycles_executed = g_slice_length - Core::g_app_core->down_count;
int cycles_next_event = (int)(first->time - global_timer);
// if (cyclesNextEvent < cyclesExecuted + cyclesDown)
// {
// cyclesDown = cyclesNextEvent - cyclesExecuted;
// // Now, now... no time machines, please.
// if (cyclesDown < 0)
// cyclesDown = 0;
// }
//}
if (cycles_next_event < cycles_executed + cycles_down) {
cycles_down = cycles_next_event - cycles_executed;
// Now, now... no time machines, please.
if (cycles_down < 0)
cycles_down = 0;
}
}
//INFO_LOG(TIME, "Idle for %i cycles! (%f ms)", cyclesDown, cyclesDown / (float)(g_clock_rate_arm11 * 0.001f));
LOG_TRACE(Core_Timing, "Idle for %i cycles! (%f ms)", cycles_down, cycles_down / (float)(g_clock_rate_arm11 * 0.001f));
//idledCycles += cyclesDown;
//currentMIPS->downcount -= cyclesDown;
//if (currentMIPS->downcount == 0)
// currentMIPS->downcount = -1;
idled_cycles += cycles_down;
Core::g_app_core->down_count -= cycles_down;
if (Core::g_app_core->down_count == 0)
Core::g_app_core->down_count = -1;
}
std::string GetScheduledEventsSummary()
{
Event *ptr = first;
std::string GetScheduledEventsSummary() {
Event* event = first;
std::string text = "Scheduled events\n";
text.reserve(1000);
while (ptr)
{
unsigned int t = ptr->type;
while (event) {
unsigned int t = event->type;
if (t >= event_types.size())
PanicAlert("Invalid event type"); // %i", t);
const char *name = event_types[ptr->type].name;
const char* name = event_types[event->type].name;
if (!name)
name = "[unknown]";
text += Common::StringFromFormat("%s : %i %08x%08x\n", name, (int)ptr->time,
(u32)(ptr->userdata >> 32), (u32)(ptr->userdata));
ptr = ptr->next;
text += Common::StringFromFormat("%s : %i %08x%08x\n", name, (int)event->time,
(u32)(event->userdata >> 32), (u32)(event->userdata));
event = event->next;
}
return text;
}
void Event_DoState(PointerWrap &p, BaseEvent *ev)
{
p.Do(*ev);
}
void DoState(PointerWrap &p)
{
std::lock_guard<std::recursive_mutex> lk(externalEventSection);
auto s = p.Section("CoreTiming", 1);
if (!s)
return;
int n = (int)event_types.size();
p.Do(n);
// These (should) be filled in later by the modules.
event_types.resize(n, EventType(AntiCrashCallback, "INVALID EVENT"));
p.DoLinkedList<BaseEvent, GetNewEvent, FreeEvent, Event_DoState>(first, (Event **)nullptr);
p.DoLinkedList<BaseEvent, GetNewTsEvent, FreeTsEvent, Event_DoState>(tsFirst, &tsLast);
p.Do(g_clock_rate_arm11);
p.Do(slicelength);
p.Do(globalTimer);
p.Do(idledCycles);
}
} // namespace

View File

@ -1,9 +1,11 @@
// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
// Copyright (c) 2012- PPSSPP Project / Dolphin Project.
// Licensed under GPLv2 or any later version
// Refer to the license.txt file included.
#pragma once
#include <string>
// This is a system to schedule events into the emulated machine's future. Time is measured
// in main CPU clock cycles.
@ -12,15 +14,15 @@
// See HW/SystemTimers.cpp for the main part of Dolphin's usage of this scheduler.
// The int cyclesLate that the callbacks get is how many cycles late it was.
// The int cycles_late that the callbacks get is how many cycles late it was.
// So to schedule a new event on a regular basis:
// inside callback:
// ScheduleEvent(periodInCycles - cyclesLate, callback, "whatever")
// ScheduleEvent(periodInCycles - cycles_late, callback, "whatever")
#include <functional>
#include "common/common.h"
class PointerWrap;
extern int g_clock_rate_arm11;
inline s64 msToCycles(int ms) {
@ -55,55 +57,84 @@ inline s64 cyclesToUs(s64 cycles) {
return cycles / (g_clock_rate_arm11 / 1000000);
}
namespace CoreTiming {
inline u64 cyclesToMs(s64 cycles) {
return cycles / (g_clock_rate_arm11 / 1000);
}
namespace CoreTiming
{
void Init();
void Shutdown();
typedef void(*TimedCallback)(u64 userdata, int cyclesLate);
typedef void(*MHzChangeCallback)();
typedef std::function<void(u64 userdata, int cycles_late)> TimedCallback;
u64 GetTicks();
u64 GetIdleTicks();
u64 GetGlobalTimeUs();
// Returns the event_type identifier.
int RegisterEvent(const char *name, TimedCallback callback);
// For save states.
/**
* Registers an event type with the specified name and callback
* @param name Name of the event type
* @param callback Function that will execute when this event fires
* @returns An identifier for the event type that was registered
*/
int RegisterEvent(const char* name, TimedCallback callback);
/// For save states.
void RestoreRegisterEvent(int event_type, const char *name, TimedCallback callback);
void UnregisterAllEvents();
// userdata MAY NOT CONTAIN POINTERS. userdata might get written and reloaded from disk,
// when we implement state saves.
void ScheduleEvent(s64 cyclesIntoFuture, int event_type, u64 userdata = 0);
void ScheduleEvent_Threadsafe(s64 cyclesIntoFuture, int event_type, u64 userdata = 0);
/// userdata MAY NOT CONTAIN POINTERS. userdata might get written and reloaded from disk,
/// when we implement state saves.
/**
* Schedules an event to run after the specified number of cycles,
* with an optional parameter to be passed to the callback handler.
* This must be run ONLY from within the cpu thread.
* @param cycles_into_future The number of cycles after which this event will be fired
* @param event_type The event type to fire, as returned from RegisterEvent
* @param userdata Optional parameter to pass to the callback when fired
*/
void ScheduleEvent(s64 cycles_into_future, int event_type, u64 userdata = 0);
void ScheduleEvent_Threadsafe(s64 cycles_into_future, int event_type, u64 userdata = 0);
void ScheduleEvent_Threadsafe_Immediate(int event_type, u64 userdata = 0);
/**
* Unschedules an event with the specified type and userdata
* @param event_type The type of event to unschedule, as returned from RegisterEvent
* @param userdata The userdata that identifies this event, as passed to ScheduleEvent
* @returns The remaining ticks until the next invocation of the event callback
*/
s64 UnscheduleEvent(int event_type, u64 userdata);
s64 UnscheduleThreadsafeEvent(int event_type, u64 userdata);
void RemoveEvent(int event_type);
void RemoveThreadsafeEvent(int event_type);
void RemoveAllEvents(int event_type);
bool IsScheduled(int event_type);
/// Runs any pending events and updates downcount for the next slice of cycles
void Advance();
void MoveEvents();
void ProcessFifoWaitEvents();
void ForceCheck();
// Pretend that the main CPU has executed enough cycles to reach the next event.
/// Pretend that the main CPU has executed enough cycles to reach the next event.
void Idle(int maxIdle = 0);
// Clear all pending events. This should ONLY be done on exit or state load.
/// Clear all pending events. This should ONLY be done on exit or state load.
void ClearPendingEvents();
void LogPendingEvents();
// Warning: not included in save states.
void RegisterAdvanceCallback(void(*callback)(int cyclesExecuted));
/// Warning: not included in save states.
void RegisterAdvanceCallback(void(*callback)(int cycles_executed));
void RegisterMHzChangeCallback(MHzChangeCallback callback);
std::string GetScheduledEventsSummary();
void DoState(PointerWrap &p);
void SetClockFrequencyMHz(int cpuMhz);
void SetClockFrequencyMHz(int cpu_mhz);
int GetClockFrequencyMHz();
extern int slicelength;
extern int g_slice_length;
} // namespace