Granted Patent B2

US 7,752,578 · App. 11/551,149 · Granted Jul 6, 2010

Abstract

Claims (32)

1. A method for placement of gate-level circuit components in an integrated circuit, the integrated circuit comprising a plurality of rows of circuit blocks, each row of circuit blocks comprising a plurality of sub-rows of circuit blocks, the method comprising:

identifying a first sub-row of the integrated circuit comprising a bus and a first instance of a gate-level circuit component;

identifying at least one candidate operation to apply to the first sub-row, wherein the candidate operation affects a placement of a circuit block or a component in the first sub-row;

computing a first metric to evaluate each of said at least one candidate operation when applied to the first sub-row; and

implementing one of said at least one candidate operation based on the computed first metric, wherein at least one or more of the identifying steps, the computing or the implementing steps are carried out by a computer.

2. The method of claim 1 , wherein the first sub-row further comprises a trunk for supplying a power or ground voltage.

3. The method of claim 2 , wherein the first sub-row comprises a trunk for supplying a power voltage, and a trunk for supplying a ground voltage.

4. The method of claim 1 , wherein one of the at least one candidate operation is adding a decoupling capacitor to the first sub-row.

5. The method of claim 1 , wherein one of the at least one candidate operation is moving the first instance to a second sub-row.

6. The method of claim 5 , wherein the moving of the first instance to a second sub-row further comprises moving at least one instance on the second sub-row to make space for the first instance.

7. The method of claim 6 , wherein the second sub-row is separated from the first sub-row by no more than a threshold distance.

8. The method of claim 1 , wherein the identifying a first sub-row comprises:

identifying a first intersection between the bus and a first trunk for supplying a power or ground voltage;

identifying a second intersection between the bus and a second trunk for supplying a power or ground voltage; and wherein

the first sub-row further comprises the first intersection and the second intersection.

9. The method of claim 8 , wherein the second intersection is required to be less than a predetermined distance from the first intersection.

10. The method of claim 1 , wherein the first sub-row further comprises a second instance, and one of the at least one candidate operation is moving the second instance to a second sub-row.

11. The method of claim 1 , wherein the first metric is a voltage drop change.

12. The method of claim 11 , wherein computing the voltage drop change comprises subtracting: 1) a voltage drop corresponding to the first instance before performing the candidate operation, from 2) a voltage drop corresponding to the first instance after performing the candidate operation.

13. The method of claim 12 , wherein the implementing one of said at least one candidate operation comprises:

selecting the candidate operation having the most negative voltage drop change from among the at least one candidate operation; and

implementing the selected candidate operation.

14. The method of claim 13 , wherein the first sub-row comprises a second instance, and one of the at least one candidate operation is moving the second instance to a second sub-row, said method further comprising:

for each candidate operation, computing a voltage drop change for each instance in the first and second sub-row;

identifying a second candidate set comprising candidate operations for which each instance in the first and second sub-row have a computed voltage drop change less than a voltage drop goal threshold; and

selecting the candidate operation having the most negative voltage drop change from the second candidate set.

15. The method of claim 14 , further comprising computing a second metric for each of said at least one candidate operation, wherein the implementing one of said at least one candidate operation is further based on the computed second metric.

16. The method of claim 15 , wherein the second metric is a placement cost.

17. The method of claim 14 , wherein the implementing of one of said at least one candidate operation comprises:

identifying a first candidate set comprising candidate operations having a placement cost less than a placement cost threshold;

selecting the candidate operation having the most negative voltage drop change from the first candidate set; and

implementing the selected candidate operation.

Assignments (10)