piwalker/melonDS
1
0
Fork 0
mirror of https://github.com/melonDS-emu/melonDS.git synced 2025-07-23 14:19:55 -06:00
Code Issues Packages Projects Releases Wiki Activity

temp commit

Browse Source 
almost done killing ARM9/ARM7 desync, f
This commit is contained in:
StapleButter
2018-12-11 03:08:46 +01:00
parent aba4610d45
commit 78d384a754
5 changed files with 213 additions and 141 deletions
Download Patch File Download Diff File Expand all files Collapse all files

212
src/NDS.cpp
View File

@ -31,7 +31,7 @@
#include "Wifi.h"
#include "Platform.h"
u64 arm9total=0, arm7total=0, arm9timer=0, arm7timer=0, systotal=0;
namespace NDS
{
@ -736,100 +736,73 @@ u32 RunFrame()
{
// TODO: give it some margin, so it can directly do 17 cycles instead of 16 then 1
CalcIterationCycles();
s32 arm9cycles;
u64 kiki = arm9total;
if (CPUStop & 0x80000000)
{
// GXFIFO stall
// we just run the GPU and the timers.
// the rest of the hardware is driven by the event scheduler.
s32 cycles = GPU3D::CyclesToRunFor();
GPU3D::Run(cycles);
u32 timermask = TimerCheckMask[0];
if (timermask & 0x1) RunTimer(0, cycles);
if (timermask & 0x2) RunTimer(1, cycles);
if (timermask & 0x4) RunTimer(2, cycles);
if (timermask & 0x8) RunTimer(3, cycles);
// run ARM7 and system peripherals, step by step
// as to give the finer-grained ones a chance to reschedule properly
// in case we end up running a large chunk of GXFIFO commands
s32 ndscyclesran = 0;
s32 ndscyclestorun;
for (;;)
{
ndscyclestorun = std::min(CurIterationCycles, cycles);
if (CPUStop & 0x0FFF0000)
{
s32 cycles = ndscyclestorun - ARM7Offset;
s32 critcycles = cycles;
cycles = DMAs[4]->Run(cycles);
if (cycles > 0) cycles = DMAs[5]->Run(cycles);
if (cycles > 0) cycles = DMAs[6]->Run(cycles);
if (cycles > 0) cycles = DMAs[7]->Run(cycles);
ARM7Offset = -cycles;
RunTimingCriticalDevices(1, critcycles);
}
else
{
ARM7->CyclesToRun = ndscyclestorun - ARM7Offset;
CurCPU = 2; ARM7->Execute(); CurCPU = 0;
ARM7Offset = ARM7->Cycles - ARM7->CyclesToRun;
}
RunSystem(ndscyclestorun);
ndscyclesran += ndscyclestorun;
if (ndscyclesran >= cycles) break;
CalcIterationCycles();
}
arm9cycles = GPU3D::CyclesToRunFor();
arm9cycles = std::min(CurIterationCycles, arm9cycles);
RunTightTimers(0, arm9cycles); arm9total+=arm9cycles;//arm9timer += arm9cycles;
}
else if (CPUStop & 0x0FFF)
{
s32 cycles = CurIterationCycles;
cycles = DMAs[0]->Run(cycles);
if (cycles > 0) cycles = DMAs[1]->Run(cycles);
if (cycles > 0) cycles = DMAs[2]->Run(cycles);
if (cycles > 0) cycles = DMAs[3]->Run(cycles);
//printf("DMAs been running for %d cycles, %d, asked for %d\n", CurIterationCycles-cycles, (u32)(arm9total-kiki), CurIterationCycles);
arm9cycles = CurIterationCycles - cycles;
}
else
{
s32 ndscyclestorun;
if (CPUStop & 0x0FFF)
{
s32 cycles = CurIterationCycles;
cycles = DMAs[0]->Run(cycles);
if (cycles > 0) cycles = DMAs[1]->Run(cycles);
if (cycles > 0) cycles = DMAs[2]->Run(cycles);
if (cycles > 0) cycles = DMAs[3]->Run(cycles);
ndscyclestorun = CurIterationCycles - cycles;
}
else
{
ARM9->CyclesToRun = CurIterationCycles << 1;
CurCPU = 1; ARM9->Execute(); CurCPU = 0;
ndscyclestorun = ARM9->Cycles >> 1;
}
RunTimingCriticalDevices(0, ndscyclestorun);
if (CPUStop & 0x0FFF0000)
{
s32 cycles = ndscyclestorun - ARM7Offset;
s32 critcycles = cycles;
cycles = DMAs[4]->Run(cycles);
if (cycles > 0) cycles = DMAs[5]->Run(cycles);
if (cycles > 0) cycles = DMAs[6]->Run(cycles);
if (cycles > 0) cycles = DMAs[7]->Run(cycles);
ARM7Offset = -cycles;
RunTimingCriticalDevices(1, critcycles);
}
else
{
ARM7->CyclesToRun = ndscyclestorun - ARM7Offset;
CurCPU = 2; ARM7->Execute(); CurCPU = 0;
ARM7Offset = ARM7->Cycles - ARM7->CyclesToRun;
}
RunSystem(ndscyclestorun);
ARM9->CyclesToRun = CurIterationCycles << 1;
CurCPU = 1; ARM9->Execute(); CurCPU = 0;
arm9cycles = ARM9->Cycles >> 1;
RunTightTimers(0, arm9cycles); //arm9timer += arm9cycles;
}
//arm9total += arm9cycles;
RunLooseTimers(0, arm9cycles);
GPU3D::Run(arm9cycles);
s32 ndscyclestorun = arm9cycles;
s32 zarp;
// ARM7Offset > ndscyclestorun means we are too far ahead of the ARM9
if (ARM7Offset > ndscyclestorun)
{
ARM7Offset -= ndscyclestorun;
}
else
if (CPUStop & 0x0FFF0000)
{
s32 cycles = ndscyclestorun - ARM7Offset; zarp=cycles;
cycles = DMAs[4]->Run(cycles);
if (cycles > 0) cycles = DMAs[5]->Run(cycles);
if (cycles > 0) cycles = DMAs[6]->Run(cycles);
if (cycles > 0) cycles = DMAs[7]->Run(cycles);
ARM7Offset = -cycles;
printf("ARM7 DMA: cyclestorun=%d, req=%d, offset=%d\n", ndscyclestorun, zarp, ARM7Offset);
}
else
{
ARM7->CyclesToRun = ndscyclestorun - ARM7Offset; zarp=ARM7->CyclesToRun;
CurCPU = 2; ARM7->Execute(); CurCPU = 0;
ARM7Offset = ARM7->Cycles - ARM7->CyclesToRun;
RunTightTimers(1, ARM7->Cycles); //arm7timer += ndscyclestorun;
}
//arm7total += zarp + ARM7Offset;//ARM7->Cycles;//ndscyclestorun+ARM7Offset;
systotal += ndscyclestorun;
RunLooseTimers(1, ndscyclestorun);// + ARM7Offset);
RunSystem(ndscyclestorun);
}
//printf("cycles: %ld %ld, %ld %ld, %ld\n", arm9total, arm9timer, arm7total, arm7timer, systotal);
printf("drift: [%ld] %ld %ld, %ld %ld\n", systotal, arm9total-systotal, arm9timer-systotal, arm7total-systotal, arm7timer-systotal);
return GPU::TotalScanlines;
}
@ -838,12 +811,12 @@ void Reschedule()
s32 oldcycles = CurIterationCycles;
CalcIterationCycles();
if (CurIterationCycles > oldcycles)
if (CurIterationCycles >= oldcycles)
{
CurIterationCycles = oldcycles;
return;
}
//printf("Reschedule %d->%d while in %d, %08X\n", oldcycles, CurIterationCycles, CurCPU, CPUStop);
if (CurCPU == 1) ARM9->CyclesToRun = CurIterationCycles << 1;
else if (CurCPU == 2) ARM7->CyclesToRun = CurIterationCycles - ARM7Offset;
// this is all. a reschedule shouldn't happen during DMA or GXFIFO stall.
@ -859,14 +832,20 @@ void ScheduleEvent(u32 id, bool periodic, s32 delay, void (*func)(u32), u32 para
SchedEvent* evt = &SchedList[id];
if (periodic) evt->WaitCycles += delay;
else evt->WaitCycles = delay + (ARM9->Cycles >> 1);
if (periodic)
evt->WaitCycles += delay;
else
{
if (CurCPU == 1) evt->WaitCycles = delay + (ARM9->Cycles >> 1);
else if (CurCPU == 2) evt->WaitCycles = delay + ARM7->Cycles;
else evt->WaitCycles = delay;
}
evt->Func = func;
evt->Param = param;
SchedListMask |= (1<<id);
//printf("scheduling event %d for within %d cycles\n", id, delay);
Reschedule();
}
@ -1041,7 +1020,7 @@ void ResumeCPU(u32 cpu, u32 mask)
void GXFIFOStall()
{
if (CPUStop & 0x80000000) return;
printf("GXFIFO STALL\n");
CPUStop |= 0x80000000;
if (CurCPU == 1) ARM9->Halt(2);
@ -1069,11 +1048,17 @@ u32 GetPC(u32 cpu)
void HandleTimerOverflow(u32 tid)
{
Timer* timer = &Timers[tid];
//if ((timer->Cnt & 0x84) != 0x80) return;
timer->Counter += timer->Reload << 16;
if (timer->Cnt & (1<<6))
SetIRQ(tid >> 2, IRQ_Timer0 + (tid & 0x3));
//if (tid<4) printf("[%03d] timer%d IRQ\n", GPU::VCount, tid);
//u32 delay = (0x10000 - timer->Reload) << (16 - timer->CycleShift);
//delay -= (timer->Counter - timer->Reload) >> timer->CycleShift;
//printf("timer%d IRQ: resched %d, reload=%04X cnt=%08X\n", tid, delay, timer->Reload, timer->Counter);
//ScheduleEvent(Event_TimerIRQ_0 + tid, true, delay, HandleTimerOverflow, tid);
if ((tid & 0x3) == 3)
return;
@ -1111,19 +1096,24 @@ void RunTimer(u32 tid, s32 cycles)
HandleTimerOverflow(tid);
}
void RunTimingCriticalDevices(u32 cpu, s32 cycles)
void RunTightTimers(u32 cpu, s32 cycles)
{
register u32 timermask = TimerCheckMask[cpu];
if(cpu)arm7timer+=cycles;else arm9timer+=cycles;
if (timermask & 0x1) RunTimer((cpu<<2)+0, cycles);
if (timermask & 0x2) RunTimer((cpu<<2)+1, cycles);
if (timermask & 0x4) RunTimer((cpu<<2)+2, cycles);
if (timermask & 0x8) RunTimer((cpu<<2)+3, cycles);
}
if (cpu == 0)
{
GPU3D::Run(cycles);
}
void RunLooseTimers(u32 cpu, s32 cycles)
{
register u32 timermask = TimerCheckMask[cpu];
if (timermask & 0x10) RunTimer((cpu<<2)+0, cycles);
if (timermask & 0x20) RunTimer((cpu<<2)+1, cycles);
if (timermask & 0x40) RunTimer((cpu<<2)+2, cycles);
if (timermask & 0x80) RunTimer((cpu<<2)+3, cycles);
}
@ -1138,6 +1128,16 @@ bool DMAsInMode(u32 cpu, u32 mode)
return false;
}
bool DMAsRunning(u32 cpu)
{
cpu <<= 2;
if (DMAs[cpu+0]->IsRunning()) return true;
if (DMAs[cpu+1]->IsRunning()) return true;
if (DMAs[cpu+2]->IsRunning()) return true;
if (DMAs[cpu+3]->IsRunning()) return true;
return false;
}
void CheckDMAs(u32 cpu, u32 mode)
{
cpu <<= 2;
@ -1180,12 +1180,28 @@ void TimerStart(u32 id, u16 cnt)
if ((!curstart) && newstart)
{
timer->Counter = timer->Reload << 16;
/*if ((cnt & 0x84) == 0x80)
{
u32 delay = (0x10000 - timer->Reload) << TimerPrescaler[cnt & 0x03];
printf("timer%d IRQ: start %d, reload=%04X cnt=%08X\n", id, delay, timer->Reload, timer->Counter);
CancelEvent(Event_TimerIRQ_0 + id);
ScheduleEvent(Event_TimerIRQ_0 + id, false, delay, HandleTimerOverflow, id);
}*/
}
if ((cnt & 0x84) == 0x80)
TimerCheckMask[id>>2] |= (1<<(id&0x3));
{
u32 tmask;
if ((cnt & 0x03) == 0)
tmask = 0x01 << (id&0x3);
else
tmask = 0x10 << (id&0x3);
TimerCheckMask[id>>2] |= tmask;
}
else
TimerCheckMask[id>>2] &= ~(1<<(id&0x3));
TimerCheckMask[id>>2] &= ~(0x11 << (id&0x3));
}