Simulation problem of greatest common divisor in Cascade simulator

[james_eali]

My question is how to implement a GCD chip in Cascade. I've read the guide but I could not get it done. Its at the behavioral level and I'm using three states, namely, idle, compare and finish. Here is my code:

#include <cascade/Cascade.hpp>
class GCD_core : public Component
{
    DECLARE_COMPONENT(GCD_core);
public:
    GCD_core(COMPONENT_CTOR) { GCD.setType(PORT_LATCH); }
    const char *Component_name;
    Input <u32> A, B;
    Input <bit> start;
    Output <bit> done;
    Output <u32> GCD;
 
    Clock clk;

    enum states{idle, compare, finish};
 
    Signal(states, P_state);
    Signal(states, N_state);

    void reset()
    {
        done << 0;
        GCD << 0;
        P_state = idle;
        N_state = idle;
    }

    void update()
    {
        switch (P_state)
        {
        case idle:
            if (start == 1)
            {
                N_state = compare;
            }
            else
                N_state = P_state;
            break;
        case compare:
            if (A == B)
            {
                N_state = finish;
                done = 1;
                GCD << A;
            }
            else if (A > B)
            {
                A = A - B;
                N_state = P_state;
            }
            else
            {
                B = B - A;
                N_state = P_state;
            }
            break;
        case finish:
            done << 1;
            GCD << A;
            start << 0;
        default:
            N_state = idle;
            break;
        }
    }
    void archive(Archive &) {}
};
class GCD_core_TB : public Component
{
    DECLARE_COMPONENT(GCD_core_TB);
public:
    GCD_core_TB(const char *name,COMPONENT_CTOR):Component_name(name)
    {
        DUT.clk << clk;
        DUT.start << start;
        DUT.A << A;
        DUT.B << B;
        DUT.GCD << GCD;
        DUT.done << done;
    }
 
    Input(u32, A);
    Input(u32, B);
    Input(bit, start);
    Output(u32, GCD);
    Output(bit, done);

 
    Clock clk;
    void reset(){}
 
    void update()
    {
        clk.generateClock(500);
        Sim::init();
        while (1)
        {
            char buff[64];
            printf("A = ");  int a = (int)(fgets(buff, 64, stdin));
            printf("\nB = ");  int b = (int)(fgets(buff, 64, stdin));
            if (*buff == 'q')
                break;
            if (*buff == 'l')
            {
                SimArchive::loadSimulation("GCD.vcd");
                printf("Simulation restored from GCD.vcd\n");
            }
            else if (*buff == 's')
            {
                SimArchive::saveSimulation("GCD.vcd");
                printf("Simulation saved to GCD.vcd\n");
                continue;
            }
            this->A << a; this->B << b;
            Sim::run();
            printf("\nGCD is: %d", this->GCD);
            Sim::reset();
        }
    }
    void archive(Archive &) {}
    const char *Component_name;
protected:
    GCD_core DUT;
};

int main(int csz, char ** rgsz)
{
    descore::arseTraces(csz, rgsz);
    Parameter::arseCommandLine(csz, rgsz);
    GCD_core_TB chip("GCD_TEST");
    Sim::dumpSignals(chip.Component_name);
    return 0;
}

I could not find any related forum to state my problem. So I would be grateful if anyone with Cascade experience could guide me on this.