SC_CTHREAD and activation on both edges

[Lubomir Bogdanov]

Hello,

I'm trying to make an SC_CTHREAD trigger on both clock edges. However, I see only pos and neg methods of the sc in clk class. So, I did this:

 

SC_CTOR(microprocessor){

SC_CTHREAD(fetch, clock.pos());
SC_CTHREAD(fetch, clock.neg());

SC_CTHREAD(decode, clock.pos());
SC_CTHREAD(execute, clock.pos());
SC_CTHREAD(interrupt, clock.pos());

}

and then in the thread I put this:

void microprocessor::fetch(){
	while(1){
		wait();
		
		read_write = 1;
		address_bus = 0x00;
		instr_data_bus = 0x00;

		if(clock){
			cout << "clock_1 " << endl;
		}
		else{
			cout << "clock_0 " << endl;
		}
	}
}

But when I run the simulation, I get the following warning:

Warning: (W505) object already exists: mcu.cpu.fetch. Latter declaration will be renamed to mcu.cpu.fetch_0
In file: ../../../src/sysc/kernel/sc_object_manager.cpp:153

 

Am I doing something wrong? The code works as expected, but since I'm new to SystemC, I suspect that

this is not how to work on both edges ...

 

Regards,

L. B.

[Eyck]

If you qant to trigger on both edges you just need to use value_changed().

You get the warning since you are creating a thread with the same method for 2 different sensitivity lists. Since the method name is used to create the thread object you get a warning when creating the second thread object. You may ignore it (and actually it is possible to switch this message off) but the better way is to use the value_changed event finder.

[Lubomir Bogdanov]

OK, thanks for your help!

[David Black]

Simply being sensitive to the clock port should work.

Replace:

SC_CTHREAD(fetch, clock.pos());
SC_CTHREAD(fetch, clock.neg());

with:

SC_CTHREAD(fetch, clock);

When you are sensitive to a port, SystemC infers sensitivity to a method default_event(). For the sc_signal class, default_event() maps to value_changed_event(). sc_clock provides an sc_signal<bool>.

[Lubomir Bogdanov]

Hello,

I tried this:

SC_MODULE(microprocessor){
  	sc_in_clk clock {"m_clock"};
  	sc_out<bool> read_write {"m_read_write"};
	sc_in<bool> interrupt_request {"m_interrupt_request"}; 
	sc_inout<sc_lv<16>> instr_data_bus {"m_instr_data_bus"};
	sc_out<sc_lv<20>> address_bus {"m_address_bus"};
  
SC_CTOR(microprocessor){
		SC_CTHREAD(fetch, clock.value_changed());
		SC_CTHREAD(decode, clock.pos());
		SC_CTHREAD(execute, clock.pos());
		SC_CTHREAD(interrupt, clock.pos());
	}
}

void microprocessor::fetch(){
	while(1){
		wait();

		if(clock){
			cout << "clock_1 " << endl;
		}
		else{
			cout << "clock_0 " << endl;
		}
	}
}

and I got the expected output without the warnings this time:

 

============== [0] 0 ==============
============== [0] 1 ==============
============== [1] 0 ==============
clock_0
============== [1] 1 ==============
clock_1
============== [2] 0 ==============
clock_0
============== [2] 1 ==============
clock_1
============== [3] 0 ==============
clock_0
============== [3] 1 ==============
clock_1
============== [4] 0 ==============
clock_0
============== [4] 1 ==============
clock_1
============== [5] 0 ==============
clock_0
============== [5] 1 ==============
clock_1

...

 

However, when I tried David's solution:

SC_CTHREAD(fetch, clock);

I got this, which I believe is not OK:

============= [0] 0 ==============
============== [0] 1 ==============
============== [1] 0 ==============
============== [1] 1 ==============
clock_1
============== [2] 0 ==============
============== [2] 1 ==============
clock_1
============== [3] 0 ==============
============== [3] 1 ==============
clock_1
============== [4] 0 ==============
============== [4] 1 ==============
clock_1
============== [5] 0 ==============
============== [5] 1 ==============
clock_1

 

which leads me to believe that I'm invoking the sc_start( ) method the wrong way:

 

int sc_main(int argc, char* argv[]){
	sc_signal<bool> clock;
	sc_lv<64> irq_lines;
	sc_signal<bool> irq;
	int i;
	microcontroller mcu("mcu");

	mcu.clock(clock);	

	for(i = 0; i < 20; i++){
		cout << "============== [" << i << "] 0 ==============" << endl;
		clock = 0;
		sc_start(1, SC_MS);

		cout << "============== [" << i << "] 1 ==============" << endl;
		clock = 1;
		sc_start(1, SC_MS);
	}

	return 0;
}

 

Anyway, I'll use Eyck's method for now ...

Cheers

[David Black]

You should push most of the code in sc_main into a top_module and use sc_clock. Advancing the simulator from sc_main is not a good way to run a simulation. You can control simulation via a free running thread in top_module such as monitor, which I illustrate below.

Take a look at https://edaplayground.com/x/JTMK

In the future when requesting help, please put your code example on EDA Playground. Your code in this posting had numerous errors that I corrected.

 

[Lubomir Bogdanov]

OK, sorry about the errors - I wasn't expecting that someone would actually test the code 🙂 I basically edited it by hand to make it more readable without caring about syntax.

 

OK, I understand now what you mean - I've seen this type of project structure in SystemC tutorials but didn't quite get it. I will correct my code ...

 

Regards,

L. B.