Granted Patent B2

US 8,670,970 · App. 13/549,599 · Granted Mar 11, 2014

Abstract

Claims (67)

1. A method for characterizing performance, over a simulation duration, of an electronic circuit having one or more cells, the method comprising:

(a) executing a first computer-based simulation of the performance of the electronic circuit over the simulation duration to characterize values of a cell parameter for at least one cell;

(b) executing multiple computer-based simulations of the performance of at least one cell, each of the multiple simulations simulating the performance of the at least one cell at a specific point in time within the simulation duration, based on at least one of the simulated values of the cell parameter corresponding to the specific point in time obtained during the first computer-based simulation; and

(c) generating one or more metrics characterizing the performance of the electronic circuit based on results from the multiple simulations, wherein:

the electronic circuit comprises a path p, defined as a sequence of one or more intermediate cells m connecting a first cell to a second cell, the total number of intermediate cells denoted m p ;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in the path over the simulation duration;

for the at least one intermediate cell in the path, each of the multiple simulations comprises simulating the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the at least one intermediate cell utilizing the cell parameter value corresponding to the specific point in time; and

step (c) comprises generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell.

2. The method of claim 1 , wherein the cell parameter is the at least one cell's effective power supply voltage V eff,m , wherein V eff,m is equal to the high power supply voltage V DD,m for the at least one cell minus the low power supply voltage V SS,m for the at least one cell.

3. The method of claim 2 , wherein the first simulation is performed by a computer-based system-level power-grid simulation tool.

4. The method of claim 1 , wherein:

the path is a clock path;

the first cell is a clock source cell; and

the second cell is a clock leaf cell.

5. The method of claim 4 , wherein:

the electronic circuit comprises a plurality of clock paths;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in at least two of the clock paths in the electronic circuit over the simulation duration;

for the at least one intermediate cell in the at least two clock paths, each of the multiple simulations comprises simulating the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the value of the cell parameter corresponding to the specific point in time; and

step (c) comprises generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell in the at least two clock paths.

6. The method of claim 4 , wherein the results from the multiple simulations comprise transition times of at least one of rising-edge transitions and falling-edge transitions of a clock signal propagating along the clock path p.

7. The method of claim 6 , wherein the one or more metrics comprise maximum pulse width, wherein the maximum pulse width is the largest difference between transition times corresponding to two consecutive transitions in the results from the multiple simulations.

8. The method of claim 6 , wherein the one or more metrics comprise minimum pulse width, wherein the minimum pulse width is the smallest difference between transition times corresponding to two consecutive transitions in the results from the multiple simulations.

9. The method of claim 6 , wherein the one or more metrics comprise maximum cycle-to-cycle jitter, wherein the maximum cycle-to-cycle jitter is the largest difference between (1) the difference between (i) the transition time that a first transition is received at the input of the clock leaf cell and (ii) the transition time that the first transition is received at the input of the first intermediate cell in clock path p and (2) the difference between (i) the transition time that a second transition, corresponding to a transition of a similar kind immediately following the first transition, is received at the input of the clock leaf cell and (ii) the transition time that the second transition is received at the input of the first intermediate cell, in the results from the multiple simulations.

10. The method of claim 6 , wherein the one or more metrics comprise maximum periodic jitter, wherein the maximum periodic jitter is the difference between (1) the largest difference between (i) the transition time that a first transition is received at the input of the clock leaf cell and (ii) the transition time that the first transition is received at the input of the first intermediate cell in clock path p and (2) the smallest difference between (i) the transition time that a second transition, of the same type as the first transition, is received at the input of the clock leaf cell and (ii) the transition time that the second transition is received at the input of the first intermediate cell, in the results from the multiple simulations.

11. The method of claim 1 , wherein:

the cell parameter is the cell's effective power supply voltage V eff,m ; and

for the at least one intermediate cell, each of the multiple simulations comprises:

(i) calculating a weighted effective power supply voltage V eff * (p,i,m) of the at least one intermediate cell for a signal event i; and

(ii) implementing a computer-based simulation tool to simulate the at least one intermediate cell's cell delay Dc (p,i,m) of the signal event i at the at least one intermediate cell's weighted effective power supply voltage V eff * (p,i,m) .

12. The method of claim 11 , wherein V eff * (p,i,m) is a weighted average of (i) the cell's effective power supply voltage V eff,m at a first time at which the signal event i arrives at an input of intermediate cell m and (ii) the cell's effective power supply voltage V eff,m at a second time corresponding to the first time plus a nominal delay value for intermediate cell m, wherein the nominal delay value is the intermediate cell's cell delay Dc (p,i,m) of the signal event i at an ideal effective power supply voltage.

13. The method of claim 11 , wherein the computer-based simulation tool is Static Timing Analysis (STA) software.

14. The method of claim 1 , wherein step (c) is implemented using a computer-based module.

15. A computer-implemented method for characterizing performance, over a simulation duration, of an electronic circuit having one or more cells, the method comprising:

(a) receiving results corresponding to a first computer-based simulation of the performance of the electronic circuit over the simulation duration to characterize values of a cell parameter for the at least one cell;

(b) receiving results corresponding to multiple computer-based simulations of the performance of at least one cell, each of the multiple simulations simulating the performance of the at least one cell at a specific point in time within the simulation duration, based on at least one of the simulated values of the cell parameter corresponding to the specific point in time obtained during the first computer-based simulation; and

(c) generating one or more metrics characterizing the performance of the electronic circuit based on the results from the multiple simulations, wherein:

the electronic circuit comprises a path p, defined as a sequence of one or more intermediate cells m connecting a first cell to a second cell, the total number of intermediate cells denoted m p ;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in the path over the simulation duration;

for the at least one intermediate cell in the path, each of the multiple simulations comprises simulating the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the at least one intermediate cell utilizing the cell parameter value corresponding to the specific point in time; and

step (c) comprises generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell.

16. An apparatus for characterizing performance, over a simulation duration, of an electronic circuit having one or more cells, the apparatus comprising:

(a) means for receiving results corresponding to a first computer-based simulation of the performance of the electronic circuit over the simulation duration to characterize values of a cell parameter for the one or more cells;

(b) means for receiving results corresponding to multiple computer-based simulations of the performance of at least one cell, each of the multiple simulations simulating the performance of the at least one cell at a specific point in time within the simulation duration, based on at least one of the simulated values of the cell parameter corresponding to the specific point in time obtained during the first computer-based simulation; and

(c) means for generating one or more metrics characterizing the performance of the electronic circuit based on the results from the multiple simulations, wherein:

the electronic circuit comprises a path p, defined as a sequence of one or more intermediate cells m connecting a first cell to a second cell, the total number of intermediate cells denoted m p ;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in the path over the simulation duration;

for the at least one intermediate cell in the path, each of the multiple simulations comprises a simulation of the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the at least one intermediate cell utilizing the cell parameter value corresponding to the specific point in time; and

the means for generating the one or more metrics comprises means for generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell.

17. The apparatus of claim 16 , wherein the apparatus further comprises means for executing the multiple simulations.

18. The apparatus of claim 17 , wherein the apparatus further comprises means for executing the first simulation.

19. A non-transitory computer-readable medium, having encoded thereon program code, wherein, when the program code is executed by a computer, the computer implements a method for characterizing performance, over a simulation duration, of an electronic circuit having one or more cells, the method comprising:

(a) receiving results corresponding to a first computer-based simulation of the performance of the electronic circuit over the simulation duration to characterize values of a cell parameter for the one or more cells;

(b) receiving results corresponding to multiple computer-based simulations of the performance of at least one cell, each of the multiple simulations simulating the performance of the at least one cell at a specific point in time within the simulation duration, based on at least one of the simulated values of the cell parameter corresponding to the specific point in time obtained during the first computer-based simulation; and

(c) generating one or more metrics characterizing the performance of the electronic circuit based on the results from the multiple simulations, wherein:

the electronic circuit comprises a path p, defined as a sequence of one or more intermediate cells m connecting a first cell to a second cell, the total number of intermediate cells denoted m p ;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in the path over the simulation duration;

for the at least one intermediate cell in the path, each of the multiple simulations comprises simulating the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the at least one intermediate cell utilizing the cell parameter value corresponding to the specific point in time; and

step (c) comprises generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell.

20. The non-transitory computer-readable medium of claim 19 , further comprising the program code for implementing the computer-based simulation tool to simulate the operation of the electronic circuit.

21. A method for characterizing performance, over a simulation duration, of an electronic circuit having one or more cells, the method comprising:

(a) executing a first computer-based simulation of the performance of the electronic circuit over the simulation duration to characterize values of a cell parameter for at least one cell;

(b) executing multiple computer-based simulations of the performance of at least one cell, each of the multiple simulations simulating the performance of the at least one cell at a specific point in time within the simulation duration, based on at least one of the simulated values of the cell parameter corresponding to the specific point in time obtained during the first computer-based simulation; and

(c) generating one or more metrics characterizing the performance of the electronic circuit based on results from the multiple simulations, wherein:

the electronic circuit comprises a clock path p, defined as a sequence of one or more intermediate cells m connecting a clock source cell to a clock leaf cell, the total number of intermediate cells denoted m p ;

the first simulation characterizes values of the cell parameter for at least one intermediate cell in the clock path over the simulation duration;

for the at least one intermediate cell in the clock path, each of the multiple simulations comprises simulating the performance of the at least one intermediate cell at a specific point in time within the simulation duration, based on the at least one intermediate cell utilizing the cell parameter value corresponding to the specific point in time; and

step (c) comprises generating the one or more metrics based on the results of the multiple simulations of the at least one intermediate cell.

Assignments (10)