'SEQREQZMB' uvm_error in uvm-1.1b

[IChip]

Hi experts,

I found a uvm_error in my tb based on uvm-1.1b as below:

UVM_ERROR @ 6155000: uvm_test_top.tb0.agentA.sequencer [sEQREQZMB] The task responsible for requesting a wait_for_grant on sequencer 'uvm_test_top.tb0.agentA.sequencer' for sequence 'uvm_test_top.tb0.agentA.sequencer. seqA' has been killed, to avoid a deadlock the sequence will be removed from the arbitration queues

And there is a forever block in seqA body(), which will send items to driver all the time.

how can i stop such sequence which run on agt.main_phase, when main_phase start to jump the next phase???

It does not work using the following.

function void phase_ready_to_end(uvm_phase phase);
  302         static int count = 0;
  303         count++;
  304         if(phase.get_name()=="main") begin
  305             if(count>=20) begin
  308                 [COLOR="red"]agtA.sequencer[/COLOR].stop_sequences();
  309             end
  310         end
  311     endfunction : phase_ready_to_end

Thanks in advance

[mastrick]

I suspect the code in line 308 never executes; there is no guarantee there will be more than one call to ready_to_end. Unless somebody raises an objection when ready_to_end is called, phase_ended will happen immediately. If you remove all the "count" stuff, you should avoid the error you are getting. You could implement the phase_ready_to_end in the sequencer itself and avoid the "agtA.sequencer". I'm pretty sure you could then move it to phase_ended() and the stop_sequences() would be guaranteed to be called before the sequence is killed.

[IChip]

I suspect the code in line 308 never executes; there is no guarantee there will be more than one call to ready_to_end. Unless somebody raises an objection when ready_to_end is called, phase_ended will happen immediately. If you remove all the "count" stuff, you should avoid the error you are getting. You could implement the phase_ready_to_end in the sequencer itself and avoid the "agtA.sequencer". I'm pretty sure you could then move it to phase_ended() and the stop_sequences() would be guaranteed to be called before the sequence is killed.

Thanks. Both work.