Data-driven clock port and clock signal recognition
Operations to recognize clock ports within a simulation circuit component and/or recognize a clock signal within simulation waveforms are described. One or more of the operations include generating a plurality of output values at an output port of a circuit simulation component by applying, during a simulation, a plurality of input values to a first input port of the circuit simulation component. The operations also include calculating a correlation vector based on bit sequences in the input values and bit sequences in the output values. The first input port is determined to be a clock port by applying a machine learning model to the correlation vector. One or more of the operations include determining a waveform file comprising signals from a simulation, determining a subset of the signals are bit-level signals, calculating toggle metrics for the subset of the signals, identifying a signal from the subset with a toggle metric satisfying a toggle threshold, calculating, by a processor, multiple duty cycles for the signal, and determining the signal is a clock signal based on the multiple duty cycles.
1. A method, comprising:
generating a plurality of output values at an output port of a circuit simulation component by applying, during a simulation, a plurality of input values to a first input port of the circuit simulation component;
calculating a correlation vector based on bit sequences in the plurality of input values and bit sequences in the plurality of output values; and
determining, by a processor, the first input port is a clock port by applying a machine learning model to the correlation vector.
2. The method of claim 1 , further comprising:
determining a second input port of the circuit simulation component is a power port and a power value for the second input port, wherein the second input port is fixed at the power value during the simulation.
3. The method of claim 2 , wherein the determining further comprises:
inputting a random pattern into the second input port that changes the corresponding output port value to a one or a zero.
4. The method of claim 1 , further comprising:
training the machine learning model using clock ports and correlation vector corresponding to a well-known sequential module.
5. The method of claim 1 , wherein the calculating the correlation vector further comprising:
extracting features corresponding to the plurality input values, wherein the features include on or more of a number of value changes, a number of value changes to one, and a duration of zero.
6. The method of claim 1 , comprising:
determining an input port of a circuit simulation component is a power port and a power value for the input port.
7. The method of claim 1 , wherein the correlation vector comprises correlation values that range from zero to one.
8. A method, comprising:
accessing a waveform file comprising signals from a simulation;
determining a subset of the signals are bit-level signals;
calculating toggle metrics for the subset of the signals;
identifying a signal from the subset with a toggle metric satisfying a toggle threshold;
calculating, by a processor, multiple duty cycles for the signal; and
determining the signal is a clock signal based on the multiple duty cycles.
9. The method of claim 8 , wherein the bit-level signals are binary level signals corresponding to binary bit strings.
10. The method of claim 8 , wherein the calculating toggle metrics further comprises:
calculating a number of rising and falling edges of a signal of the subset of signals in a unit of time.
11. The method of claim 8 , wherein the toggle threshold is a maximum toggle value multiplied by a value between 0 and 1.
12. The method of claim 8 , wherein a duty cycle of the multiple duty cycle corresponds to a cycle with 0-1-0-1 or 1-0-1-0 signal transitions.
13. The method of claim 8 , wherein the calculating the multiple duty cycles further comprises:
calculating a ratio of on-portion and off-portion of a cycle of the signal.
14. The method of claim 8 , wherein the determining the signal is the clock signal further comprises:
determining that the signal maintains a consistent duty cycle value over a duration of time.
15. A non-transitory computer readable medium (CRM) comprising stored instructions, which when executed by a processor, cause the processor to perform operations comprising:
generating a plurality of output values at an output port of a circuit simulation component by applying, during a simulation, a plurality of input values to a first input port of the circuit simulation component;
calculating a correlation vector based on bit sequences in the plurality of input values and bit sequences in the plurality of output values; and
determining the first input port is a clock port by applying a machine learning model to the correlation vector.
16. The non-transitory CRM of claim 15 , wherein the stored instructions, when executed by the processor, cause the processor to perform operations further comprising:
determining a second input port of the circuit simulation component is a power port and a power value for the second input port, wherein the second input port is fixed at the power value during the simulation.
17. The non-transitory CRM of claim 16 , wherein the determining the second input port of the circuit simulation component is a power port further comprises:
inputting a random pattern into the second input port that changes the corresponding output port value to a one or a zero.
18. The non-transitory CRM of claim 15 , wherein the stored instructions, when executed by the processor, cause the processor to perform operations further comprising:
training the machine learning model using clock ports and correlation vector corresponding to a well-known sequential module.
19. The non-transitory CRM of claim 15 , wherein the calculating the correlation vector further comprises:
extracting features corresponding to the plurality input values, wherein the features include on or more of a number of value changes, a number of value changes to one, and a duration of zero.
20. The non-transitory CRM of claim 15 , wherein the correlation vector comprises correlation values that range from zero to one.