#include "sux.h"
#include <assert.h>
#include <ctype.h>
#include <string.h>

#if getclk
uint64_t clk[THREADS];	/* Per Thread Clock cycles. */
uint64_t tclk;		/* Total Clock cycles. */
#endif

const uint16_t tv = 0xFF50; /* Starting address of the Thread Vectors. */

#if !IO
uint64_t inst[THREADS];
#endif

#if bench
uint64_t inss;
#endif

#if debug
uint8_t subdbg;
#endif

pthread_mutex_t mutex = PTHREAD_MUTEX_INITIALIZER;
pthread_mutex_t main_mutex = PTHREAD_MUTEX_INITIALIZER;
pthread_cond_t cond = PTHREAD_COND_INITIALIZER;
pthread_cond_t main_cond = PTHREAD_COND_INITIALIZER;

uint8_t threads_done = 0;
uint8_t step = 0;
uint8_t esc = 0;

uint8_t *addr;

uint8_t kbd_rdy;

WINDOW *scr;

struct suxthr {
	struct sux sx;
	uint8_t th;
};

#if bench
double ipc;
struct timeval str[THREADS], en[THREADS];
#endif

void *run(void *args) {
	struct suxthr *thr = (void *)args;
	struct sux *cpu = &thr->sx;
	uint8_t thread = thr->th;
	uint8_t prefix = 0;
	uint8_t opcode = 0;
	union reg address;
	union reg value;
	#if getclk
	uint64_t iclk = 0;
	#endif
	#if !IO
	uint64_t ins = 0;
	#endif
#if !bench
	uint8_t lines = (6*thread)+2;
#endif
#if debug && !bench
	if (!subdbg) {
		addr[STEP_ADDR] = 1;
		step = 1;
	}
	#if keypoll
	pthread_mutex_lock(&mutex);
	#endif
	werase(scr);
	#if keypoll
	pthread_mutex_unlock(&mutex);
	#endif
#endif
	uint64_t tmpaddr = 0;
#if bench
	gettimeofday(&str[thread], 0);
#endif
	for (;;) {
		address.u64 = 0;
		value.u64 = 0;
		prefix = addr[cpu->pc[thread]];
		if ((prefix & 0x03) != 0x03) {
			prefix = 0;
		}
		cpu->pc[thread] += ((prefix & 0x03) == 0x03);
		opcode = addr[cpu->pc[thread]];
		#if debug && !bench
		if (lines > 24*(thread+1)) {
			lines = (24*thread)+2;
		}
		#if keypoll
		pthread_mutex_lock(&mutex);
		#endif
		for (uint8_t i = (24*thread)+2; i <= 24*(thread+1); i++) {
			wmove(scr, i, 0);
			waddch(scr, (i == lines) ? '>' : ' ');
		}
		wmove(scr, lines, 1);
		wclrtoeol(scr);
		wprintw(scr, "pc: $%04"PRIX64  , cpu->pc[thread]);
		wprintw(scr, ", a: $%016"PRIX64, cpu->a[thread]);
		wprintw(scr, ", b: $%016"PRIX64, cpu->b[thread]);
		wprintw(scr, ", x: $%016"PRIX64, cpu->x[thread]);
		wprintw(scr, ", y: $%016"PRIX64, cpu->y[thread]);
		wprintw(scr, ", sp: $%04X", cpu->sp[thread]);
		wprintw(scr, ", ps: $%02X", cpu->ps.u8[thread]);
		wprintw(scr, ", inst: ");
		#if keypoll
		pthread_mutex_unlock(&mutex);
		#endif
		#endif
		address.u64 = cpu->pc[thread];
		++cpu->pc[thread];
		#if getclk
		++iclk;
		#endif
		if (optype[opcode] != IMPL) {
			address.u64 = get_addr(cpu, &tmpaddr, opcode, prefix, thread);
			/* Unroll Loop by implementing Duff's Device. */
			value.u8[0] = addr[address.u64];
			switch (1 << (prefix >> 4)) {
				case 8:
					value.u8[7] = addr[address.u64+7];
					value.u8[6] = addr[address.u64+6];
					value.u8[5] = addr[address.u64+5];
					value.u8[4] = addr[address.u64+4];
				case 4:
					value.u8[3] = addr[address.u64+3];
					value.u8[2] = addr[address.u64+2];
				case 2:
					value.u8[1] = addr[address.u64+1];
			}
			#if getclk
			++iclk;
			#endif
		}
	#if debug && !bench
		#if keypoll
		pthread_mutex_lock(&mutex);
		#endif
		uint64_t operands[3];
		operands[0] = value.u64;
		operands[1] = address.u64;
		operands[2] = tmpaddr;
		disasm(cpu, operands, lines, opcode, prefix, thread);
		lines+=1;
		#if keypoll
		pthread_mutex_unlock(&mutex);
		#endif
	#endif
		switch(opcode) {
			case CPS: /* Clear Processor Status. */
				cpu->ps.u64 = 0;
				break;
			case AAB: /* Add Accumulator with carry by B register. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case ADC:  /* ADC Immediate. */
			case ADC_AB: /* ADC Absolute. */
			case ADC_Z: /* ADC Zero Matrix. */
				adc(cpu, value.u64, thread);
				break;
			case PHP: push(cpu, value.u64, cpu->ps.u8[thread], thread); break; /* PusH Processor status to stack. */
			case PHA: push(cpu, value.u64, cpu->a[thread    ], thread); break; /* PusH Accumulator to stack. */
			case PHB: push(cpu, value.u64, cpu->b[thread    ], thread); break; /* PusH B register to stack. */
			case PHY: push(cpu, value.u64, cpu->y[thread    ], thread); break; /* PusH Y register to stack. */
			case PHX: push(cpu, value.u64, cpu->x[thread    ], thread); break; /* PusH X register to stack. */
			case TAY: /* Transfer Accumulator to Y. */
			case TAX: /* Transfer Accumulator to Y. */
			case TYX: /* Transfer Y to X. */
			case TYA: /* Transfer Y to Accumulator. */
			case TXA: /* Transfer X to Accumulator. */
			case TXY: /* Transfer X to Y. */
			case TAB: /* Transfer Accumulator to B. */
			case TSX: /* Transfer Stack pointer to X. */
			case TBA: /* Transfer B to Accumulator. */
			case TXS: /* Transfer X to Stack pointer. */
				transfer(cpu, value.u64, opcode, prefix, thread);
				break;
			case JMP: /* JMP Absolute. */
			case JMP_Z: /* JMP Zero Matrix. */
			case JMP_IN: /* JMP Indirect. */
				cpu->pc[thread] = address.u64;
				break;
			case SAB: /* Subtract Accumulator with carry by B register. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case SBC: /* SBC Immediate. */
			case SBC_AB: /* SBC Absolute. */
			case SBC_Z: /* SBC Zero Matrix. */
				sbc(cpu, value.u64, thread);
				break;
			case PLP: cpu->ps.u8[thread] = pull(cpu, value.u64, thread); break; /* PuLl Processor status from stack. */
			case PLA: cpu->a[thread    ] = pull(cpu, value.u64, thread); break; /* PuLl Accumulator from stack. */
			case PLB: cpu->b[thread    ] = pull(cpu, value.u64, thread); break; /* PuLl B register from stack. */
			case PLY: cpu->y[thread    ] = pull(cpu, value.u64, thread); break; /* PuLl Y register from stack. */
			case PLX: cpu->x[thread    ] = pull(cpu, value.u64, thread); break; /* PuLl X register from stack. */
				break;
			case ABA: /* bitwise And with Accumulator, and B register. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case AND: /* AND Immediate. */
			case AND_AB: /* AND Absolute. */
			case AND_Z: /* AND Zero Matrix. */
				and(cpu, value.u64, thread);
				break;
			case STT: /* STart Thread. */
				/*cpu->crt |= value;
				for (uint8_t i = 0; i < 7; i++) {
					if ((value >> i) & 1) {
						address = (uint64_t)addr[tv+(i<<3)]
								| (uint64_t)addr[tv+1+(i<<3)] << 8
								| (uint64_t)addr[tv+2+(i<<3)] << 16
								| (uint64_t)addr[tv+3+(i<<3)] << 24
								| (uint64_t)addr[tv+4+(i<<3)] << 32
								| (uint64_t)addr[tv+5+(i<<3)] << 40
								| (uint64_t)addr[tv+6+(i<<3)] << 48
								| (uint64_t)addr[tv+7+(i<<3)] << 56;
						cpu->pc[i+1] = address;
					}
				}*/
				break;
			case BPO: /* BPO Absolute. */
			case BPO_Z: /* BPO Zero Matrix. */
				if (!getflag(N)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case OAB: /* bitwise Or with Accumulator, and B register. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case ORA: /* ORA Immediate. */
			case ORA_AB: /* ORA Absolute. */
			case ORA_Z: /* ORA Zero Matrix. */
				or(cpu, value.u64, thread);
				break;
			case SEI: /* SEt Interrupt. */
				setflag(1, I);
				break;
			case BNG: /* BNG Absolute. */
			case BNG_Z: /* BNG Zero Matrix. */
				if (getflag(N)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case XAB: /* bitwise Xor with Accumulator, and B register. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case XOR: /* XOR Immediate. */
			case XOR_AB: /* XOR Absolute. */
			case XOR_Z: /* XOR Zero Matrix. */
				xor(cpu, value.u64, thread);
				break;
			case CLI: /* CLear Interrupt. */
				setflag(0, I);
				break;
			case BCS: /* BCS Absolute. */
			case BCS_Z: /* BCS Zero Matrix. */
				if (getflag(C)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case LLB: /* Logical shift Left accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case LSL: /* LSL Immediate. */
			case LSL_AB: /* LSL Absolute. */
			case LSL_Z: /* LSL Zero Matrix. */
				lsl(cpu, value.u64, thread);
				break;
			case SEC: /* SEt Carry flag.*/
				setflag(1, C);
				break;
			case STA: /* STA Absolute. */
			case STA_Z: /* STA Zero Matrix. */
			case STA_ZX: /* STA Zero Matrix, Indexed with X. */
			case STA_ZY: /* STA Zero Matrix, Indexed with Y. */
			case STA_IN: /* STA Indirect. */
			case STA_IX: /* STA Indexed Indirect. */
			case STA_IY: /* STA Indirect Indexed. */
				store(cpu, address.u64, cpu->a[thread], prefix, thread);
				break;
			case STY: /* STY Absolute. */
			case STY_Z: /* STY Zero Matrix. */
			case STY_ZX: /* STY Zero Matrix, Indexed with X. */
			case STY_IN: /* STY Indirect. */
				store(cpu, address.u64, cpu->y[thread], prefix, thread);
				break;
			case STX: /* STX Absolute. */
			case STX_Z: /* STX Zero Matrix. */
			case STX_ZY: /* STX Zero Matrix, Indexed with Y. */
			case STX_IN: /* STX Indirect. */
				store(cpu, address.u64, cpu->x[thread], prefix, thread);
				break;
			case STB: /* STB Absolute. */
			case STB_Z: /* STB Zero Matrix. */
			case STB_ZX: /* STB Zero Matrix, Indexed with X. */
			case STB_ZY: /* STB Zero Matrix, Indexed with Y. */
			case STB_IN: /* STB Indirect. */
			case STB_IX: /* STB Indexed Indirect. */
			case STB_IY: /* STB Indirect Indexed. */
				store(cpu, address.u64, cpu->b[thread], prefix, thread);
				break;
			case BCC: /* BCC Absolute. */
			case BCC_Z: /* BCC Zero Matrix. */
				if (!getflag(C)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case LRB: /* Logical shift Right accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case LSR: /* LSR Immediate. */
			case LSR_AB: /* LSR Absolute. */
			case LSR_Z: /* LSR Zero Matrix. */
				lsr(cpu, value.u64, thread);
				break;
			case ARB: /* Arithmetic shift Right accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case ASR: /* ASR Immediate. */
			case ASR_AB: /* ASR Absolute. */
			case ASR_Z: /* ASR Zero Matrix. */
				asr(cpu, value.u64, thread);
				break;
			case CLC: /* CLear Carry flag. */
				setflag(0, C);
				break;
			case LDB: /* LDB Immediate. */
			case LDB_AB: /* LDB Absolute. */
			case LDB_Z: /* LDB Zero Matrix. */
			case LDB_ZX: /* LDB Zero Matrix, Indexed with X. */
			case LDB_ZY: /* LDB Zero Matrix, Indexed with Y. */
			case LDB_IN: /* LDB Indirect. */
			case LDB_IX: /* LDB Indexed Indirect. */
			case LDB_IY: /* LDB Indirect Indexed. */
				cpu->b[thread] = load(cpu, address.u64, cpu->b[thread], prefix, thread);
				break;
			case LDA: /* LDA Immediate. */
			case LDA_AB: /* LDA Absolute. */
			case LDA_Z: /* LDA Zero Matrix. */
			case LDA_ZX: /* LDA Zero Matrix, Indexed with X. */
			case LDA_ZY: /* LDA Zero Matrix, Indexed with Y. */
			case LDA_IN: /* LDA Indirect. */
			case LDA_IX: /* LDA Indexed Indirect. */
			case LDA_IY: /* LDA Indirect Indexed. */
				cpu->a[thread] = load(cpu, address.u64, cpu->a[thread], prefix, thread);
				break;
			case LDY: /* LDY Immediate. */
			case LDY_AB: /* LDY Absolute. */
			case LDY_Z: /* LDY Zero Matrix. */
			case LDY_ZX: /* LDY Zero Matrix, Indexed with X. */
			case LDY_IN: /* LDY Indirect. */
				cpu->y[thread] = load(cpu, address.u64, cpu->y[thread], prefix, thread);
				break;
			case LDX: /* LDX Immediate. */
			case LDX_AB: /* LDX Absolute. */
			case LDX_Z: /* LDX Zero Matrix. */
			case LDX_ZY: /* LDX Zero Matrix, Indexed with Y. */
			case LDX_IN: /* LDX Indirect. */
				cpu->x[thread] = load(cpu, address.u64, cpu->x[thread], prefix, thread);
				break;
			case BEQ: /* BEQ Absolute. */
			case BEQ_Z: /* BEQ Zero Matrix. */
				if (getflag(Z)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case RLB: /* Rotate Left accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case ROL: /* ROL Immediate. */
			case ROL_AB: /* ROL Absolute. */
			case ROL_Z: /* ROL Zero Matrix. */
				rol(cpu, value.u64, thread);
				break;
			case BNE: /* BNE Absolute. */
			case BNE_Z: /* BNE Zero Matrix. */
				if (!getflag(Z)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case RRB: /* Rotate Right accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case ROR: /* ROR Immediate. */
			case ROR_AB: /* ROR Absolute. */
			case ROR_Z: /* ROR Zero Matrix. */
				ror(cpu, value.u64, thread);
				break;
			case BVS: /* BVS Absolute. */
			case BVS_Z: /* BVS Zero Matrix. */
				if (getflag(V)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case MAB: /* Multiply Accumulator by B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case MUL: /* MUL Immediate. */
			case MUL_AB: /* MUL Absolute. */
			case MUL_Z: /* MUL Zero Matrix. */
				mul(cpu, value.u64, thread);
				break;
			case BVC: /* BVC Absolute. */
			case BVC_Z: /* BVC Zero Matrix. */
				if (!getflag(V)) {
					cpu->pc[thread] = address.u64;
				}
				break;
			case DIV: /* DIV Immediate. */
			case DAB: /* Divide Accumulator by B. */
			case DIV_AB: /* DIV Absolute. */
			case DIV_Z: /* DIV Zero Matrix. */
				divd(cpu, value.u64, opcode, thread);
				break;
			case CLV: /* CLear oVerflow flag. */
				setflag(0, V);
				break;
			case CPB: /* CPB Immediate. */
			case CPB_AB: /* CPB Absolute. */
			case CPB_Z: /* CPB Zero Matrix. */
			case CPB_IN: /* CPB Indirect. */
			case CPB_IX: /* CPB Indexed Indirect. */
			case CPB_IY: /* CPB Indirect Indexed. */
				cmp(cpu, value.u64, cpu->b[thread], thread);
				break;
			case CAB: /* Compare Accumulator, and B. */
				value.u64 = cpu->b[thread]; /* Falls Through. */
			case CMP: /* CMP Immediate. */
			case CMP_AB: /* CMP Absolute. */
			case CMP_Z: /* CMP Zero Matrix. */
			case CMP_IN: /* CMP Indirect. */
			case CMP_IX: /* CMP Indexed Indirect. */
			case CMP_IY: /* CMP Indirect Indexed. */
				cmp(cpu, value.u64, cpu->a[thread], thread);
				break;
			case CPY: /* CPY Immediate. */
			case CPY_AB: /* CPY Absolute. */
			case CPY_Z: /* CPY Zero Matrix. */
			case CPY_IN: /* CPY Indirect. */
				cmp(cpu, value.u64, cpu->y[thread], thread);
				break;
			case CPX: /* CPX Immediate. */
			case CPX_AB: /* CPX Absolute. */
			case CPX_Z: /* CPX Zero Matrix. */
			case CPX_IN: /* CPX Indirect. */
				cmp(cpu, value.u64, cpu->x[thread], thread);
				break;
			case ENT: /* ENd Thread. */
			/*	cpu->crt &= ~value;
				for (uint8_t i = 0; i < 7; i++)
					if ((value >> i) & 1)
						cpu->pc[i+1] = cpu->pc[0]+(i+1);*/
				break;
			case INC: cpu->a[thread] = idr(cpu, cpu->a[thread], 1, thread); break;
			case INB: cpu->b[thread] = idr(cpu, cpu->b[thread], 1, thread); break;
			case INY: cpu->y[thread] = idr(cpu, cpu->y[thread], 1, thread); break;
			case INX: cpu->x[thread] = idr(cpu, cpu->x[thread], 1, thread); break;
			case DEC: cpu->a[thread] = idr(cpu, cpu->a[thread], 0, thread); break;
			case DEB: cpu->b[thread] = idr(cpu, cpu->b[thread], 0, thread); break;
			case DEY: cpu->y[thread] = idr(cpu, cpu->y[thread], 0, thread); break;
			case DEX: cpu->x[thread] = idr(cpu, cpu->x[thread], 0, thread); break;
			case JSR_IN: /* JSR Indirect. */
			case JSR: /* Jump to SubRoutine. */
			case JSR_Z: /* JSR Zero Matrix. */
				value.u64 = cpu->pc[thread];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-0] = value.u8[7];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-1] = value.u8[6];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-2] = value.u8[5];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-3] = value.u8[4];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-4] = value.u8[3];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-5] = value.u8[2];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-6] = value.u8[1];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-7] = value.u8[0];
				cpu->sp[thread] -= 8;
				cpu->pc[thread] = address.u64;
				break;
			case INC_AB: /* INC Absolute. */
			case INC_Z: /* INC Zero Matrix. */
				idm(cpu, address.u64, prefix, 1, thread);
				break;
			case NOP: /* No OPeration. */
				break;
			case RTI: /* ReTurn from Interrupt routine. */
				cpu->sp[thread]   += 1;
				cpu->ps.u8[thread] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread])]; /* Falls through. */
			case RTS: /* ReTurn from Subroutine. */
				cpu->sp[thread] += 8;
				value.u8[0] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-7)];
				value.u8[1] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-6)];
				value.u8[2] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-5)];
				value.u8[3] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-4)];
				value.u8[4] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-3)];
				value.u8[5] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-2)];
				value.u8[6] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]-1)];
				value.u8[7] = addr[(cpu->stk_st[thread] << 16)+(cpu->sp[thread]  )];
				cpu->pc[thread] = value.u64;
				break;
			case DEC_AB: /* DEC Absolute. */
			case DEC_Z: /* DEC Zero Matrix. */
				idm(cpu, address.u64, prefix, 0, thread);
				break;
			case BRK: /* BReaK. */
			case WAI: /* WAit for Interrupt. */
				if (opcode == WAI) {
					pthread_mutex_lock(&main_mutex);
					pthread_cond_signal(&main_cond);
					pthread_mutex_unlock(&main_mutex);
					pthread_mutex_lock(&mutex);
					pthread_cond_wait(&cond, &mutex);
					pthread_mutex_unlock(&mutex);
				}
				value.u64 = cpu->pc[thread];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-0] = value.u8[7];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-1] = value.u8[6];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-2] = value.u8[5];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-3] = value.u8[4];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-4] = value.u8[3];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-5] = value.u8[2];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-6] = value.u8[1];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-7] = value.u8[0];
				addr[(cpu->stk_st[thread] << 16)+cpu->sp[thread]-8] = cpu->ps.u8[thread];
				cpu->sp[thread] -= 9;
				setflag(1, I);
				if (opcode == BRK) {
					value.u8[0] = addr[0xFFE0];
					value.u8[1] = addr[0xFFE1];
					value.u8[2] = addr[0xFFE2];
					value.u8[3] = addr[0xFFE3];
					value.u8[4] = addr[0xFFE4];
					value.u8[5] = addr[0xFFE5];
					value.u8[6] = addr[0xFFE6];
					value.u8[7] = addr[0xFFE7];
				} else {
					value.u8[0] = addr[0xFFA0];
					value.u8[1] = addr[0xFFA1];
					value.u8[2] = addr[0xFFA2];
					value.u8[3] = addr[0xFFA3];
					value.u8[4] = addr[0xFFA4];
					value.u8[5] = addr[0xFFA5];
					value.u8[6] = addr[0xFFA6];
					value.u8[7] = addr[0xFFA7];
					kbd_rdy &= (uint8_t)~(1 << thread);
				}
				cpu->pc[thread] = value.u64;
			default:
				break;
		}
		#if !IO
		ins++;
		#endif
		#if !bench
		if (step) {
			pthread_mutex_lock(&main_mutex);
			pthread_cond_signal(&main_cond);
			pthread_mutex_unlock(&main_mutex);
			pthread_mutex_lock(&mutex);
			pthread_cond_wait(&cond, &mutex);
			pthread_mutex_unlock(&mutex);
			#if debug
			wrefresh(scr);
			#endif
		}
		#endif
		#if debug && !bench
		#if keypoll
		pthread_mutex_lock(&mutex);
		#endif
		wmove(scr, (6*thread)+1, 0);
		wprintw(scr, "Instructions executed: %"PRIu64, ins);
		#if getclk
		wprintw(scr, ", Clock cycles: %"PRIu64, iclk);
		#endif
		if (!step && !subdbg) {
			wrefresh(scr);
		}
		#if keypoll
		pthread_mutex_unlock(&mutex);
		#endif
		#elif bench
		if (ins >= BENCH_INST) {
			pthread_mutex_lock(&main_mutex);
			threads_done++;
			inst[thread] = ins;
			#if getclk
			clk[thread] = iclk;
			#endif
			pthread_cond_signal(&main_cond);
			pthread_mutex_unlock(&main_mutex);
			gettimeofday(&en[thread], 0);
			break;
		}
		#endif
	}
	return NULL;
}

int main(int argc, char **argv) {
	struct suxthr thr[THREADS];
	char *tmp = malloc(2048);
	addr = malloc(0x04000000);
	#if bench
	inss = 0;
	#endif
	int v = 0;

	if (argc != 2) {
		if (asmmon("stdin") == 2) {
			return 0;
		}
	} else {
		#if debug
		subdbg = !strcmp(argv[1], "programs/subeditor.s");
		#endif
		if (asmmon(argv[1]) == 2) {
			return 0;
		}
	}
	sprintf(tmp, "\033[2J\033[H");
	fwrite(tmp, sizeof(char), strlen(tmp), stdout);
	fflush(stdout);
	if(!scr) {
		scr = initscr();
	}
	nodelay(stdscr, 0);
	crmode();
	noecho();
	nl();
	curs_set(1);
	werase(scr);
	scrollok(scr, 1);
	start_color();
	use_default_colors();
	init_pair(1, COLOR_WHITE, -1);
	attron(COLOR_PAIR(1) | A_BOLD);
	wmove(scr, 0, 0);
	wrefresh(scr);
	pthread_t therads[THREADS];
	int result;
	uint16_t vec = 0xFFC0;
	uint8_t offset;
	for (int i = 0; i < THREADS; i++) {
		thr[i].sx.sp[i] = 0xFFFF;
		thr[i].sx.stk_st[i] = i+1;
		offset = (i) ? ((i-1) << 3) : 0;
		vec = (i) ? 0xFF50 : 0xFFC0;
		thr[i].sx.a[i] = 0;
		thr[i].sx.b[i] = 0;
		thr[i].sx.x[i] = 0;
		thr[i].sx.y[i] = 0;
		thr[i].sx.pc[i] = (uint64_t)addr[vec+0+offset]
				| (uint64_t)addr[vec+1+offset] << 8
				| (uint64_t)addr[vec+2+offset] << 16
				| (uint64_t)addr[vec+3+offset] << 24
				| (uint64_t)addr[vec+4+offset] << 32
				| (uint64_t)addr[vec+5+offset] << 40
				| (uint64_t)addr[vec+6+offset] << 48
				| (uint64_t)addr[vec+7+offset] << 56;

		thr[i].th = i;
		#if !IO
		inst[i] = 0;
		#endif
		result = pthread_create(&therads[i], NULL, run, &thr[i]);
		assert(!result);
	}
	int c = 0;
	uint8_t step_key = 0;
	uint8_t end = 0;
	werase(scr);
	while (threads_done < THREADS && !end) {
		#if !bench
		int x, y;
		if ((step_key && step && !kbd_rdy) || !step || kbd_rdy) {
			if ((c != EOF && c !=-1)) {
				#if !keypoll
				pthread_mutex_lock(&mutex);
				pthread_cond_signal(&cond);
				pthread_mutex_unlock(&mutex);
				#endif
				pthread_mutex_lock(&main_mutex);
				curs_set(0);
				pthread_cond_wait(&main_cond, &main_mutex);
				pthread_mutex_unlock(&main_mutex);
				curs_set(1);
				c = 0;
				step_key = 0;
				addr[CTRL_ADDR] = 0;
				wrefresh(scr);
			}
		}
		#if keypoll
		pthread_mutex_lock(&mutex);
		#endif
		c = wgetch(scr);
		if (c == 19) {
			if (kbd_rdy) {
				c = wgetch(scr);
			}
			step = 1;
		} else if (c == 0x11) {
			end = 1;
			continue;
		}
		if (kbd_rdy) {
			switch (c) {
				case ERR:
					addr[CTRL_ADDR] = 0;
					break;
				default:
					if (kbd_rdy && c < 0x100) {
						addr[RX_ADDR] = (uint8_t)c;
						addr[CTRL_ADDR] = 1;
						#if debug && !bench
						wmove(scr, getmaxy(scr)-1, 0);
						wclrtoeol(scr);
						wprintw(scr, "c: %i", c);
						wmove(scr, y, x);
						#endif
					}
					break;
			}
		} else {
			if (step) {
				step = !(c == 18);
				step_key = (c == 19);
			}
		}
		addr[STEP_ADDR] = step;
		#if keypoll
		pthread_mutex_unlock(&mutex);
		#endif
		#else
		pthread_mutex_lock(&main_mutex);
		pthread_cond_wait(&main_cond, &main_mutex);
		pthread_mutex_unlock(&main_mutex);
		#endif
	}
	endwin();
#if bench
	if (threads_done == THREADS) {
		double tm_sec, tm_usec, tm[THREADS], ttm;
		#if getclk
		double clkspd;
		double mhz;
		#endif
		double ips[THREADS];
		double ipst;
		for (int i = 0; i < THREADS; i++) {
			tm_sec = (en[i].tv_sec - str[i].tv_sec);
			tm_usec = (en[i].tv_usec-str[i].tv_usec);
			tm[i] = (tm_sec*1000000)+(tm_usec);
			ips[i] = inst[i]/tm[i];
			if (i) {
				inss += inst[i];
				ttm += tm[i];
				ipst += ips[i];
				#if getclk
				tclk += clk[i];
				#endif
			} else {
				inss = inst[i];
				ttm = tm[i];
				ipst = ips[i];
				#if getclk
				tclk = clk[i];
				#endif
			}
			#if getclk
			clkspd = (tm[i]/1000000)*1000000/clk[i];
			mhz = 1000000.0/clkspd/1000000;
			#endif
			sprintf(tmp, "Instructions executed for thread %i: %"PRIu64", Instructions per Second for thread %i in MIPS: %f, tm: %f\n", i, inst[i], i, ips[i], tm[i]/1000000);
			fwrite(tmp, sizeof(char), strlen(tmp), stdout);
		}
		sprintf(tmp, "Total Instructions executed: %"PRIu64", Total Instructions per Second in MIPS: %f", inss, ipst);
		fwrite(tmp, sizeof(char), strlen(tmp), stdout);
		#if getclk
		clkspd = (ttm/1000000)*1000000/tclk;
		mhz = 1000000.0/clkspd/1000000;
		sprintf(tmp, ", Clock cycles: %"PRIu64", Clock Speed in MHz: %f", tclk, mhz);
		fwrite(tmp, sizeof(char), strlen(tmp), stdout);
		#endif
		sprintf(tmp, ", tm: %f\n", ttm/1000000);
		fwrite(tmp, sizeof(char), strlen(tmp), stdout);
		fflush(stdout);
		free(tmp);
	}
#endif
	free(addr);
	return 0;
}