Apparatus and method for logic optimization by redundancy addition and removal

• US 5,587,919 A
• Filed: 04/22/1994
• Issued: 12/24/1996
• Est. Priority Date: 04/22/1994
• Status: Expired due to Term

• Current Assignee: Bell Semiconductor, LLC (Hilco Inc. (d/b/a Hilco Global))
• Original Assignee: Lucent Technologies, Inc. (Nokia Corporation)
• Inventors: Entrena, Luis, Cheng, Kwang-Ting
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): PHAN, THAI Q
• Application Number:
  US08/233,791
• Time in Patent Office: 977 Days
• Field of Search: 364/488, 364/489, 364/490, 364/491, 364/578, 395/500
• US Class Current: 716/106
• CPC Class Codes: G01R 31/318342 by preliminary fault modelling, e.g. analysis, simulation
• CPC Class Codes: G06F 30/327 Logic synthesis; Behaviour synthesis, e.g. mapping logic, HDL to netlist, high-level language to RTL or netlist

Abstract

A method is disclosed for optimizing a digital logic network through the iterative addition and removal of redundant connections. Redundancy testing is performed on connections absolutely dominated by the node of the network being optimized to find redundant connections and connections that can be added to the network which would create more redundant connections elsewhere in the network which may under appropriate conditions be removed to simplify the network. Application of the redundancy addition and removal network optimization technique to selected connections in a network improves the cost efficiency of network optimization because a network is optimized by the consideration of a reduced number of candidate connections for addition.

Claims

1. A method of efficiently optimizing a digital logic network through the iterative addition and removal of redundant connections previously identified as candidate connections comprising the steps of:

selecting a target node for optimization;

analyzing the digital logic network to determine if connections to the target node are absolutely dominated by said target node, by determining if each of said connections must pass through the target node before reaching any primary output of the digital logic network;

performing redundancy testing on each fault corresponding to each connection previously determined to be absolutely dominated;

removing from the digital logic network each said fault found redundant;

identifying each candidate connection for addition suggested by an assignment value implicated during redundancy testing; and

adding to the network each said candidate connection found redundant.

2. The method of claim 1, further comprising the step of removing from the network a connection made redundant by said addition.

3. The method of claim 1, further comprising the step of computing the cost of each node of the network and selecting a node of the network having the highest cost as the target node.

4. The method of claim 3, wherein each said candidate connection is identified based on an assignment value implicated at a node outside the fanout cone and not in the input tree of said target node.

5. The method of claim 3, wherein each said candidate connection is identified based on an assignment value implicated at a node outside the fanout cone of the target node.

6. The method of claim 2, further comprising the step of computing the cost of each node of the network and selecting a node of the network having the highest cost as the target node.

7. The method of claim 6, wherein each said candidate connection is identified based on an assignment value implicated at a node outside the fanout cone and not in the input tree of said target node.

8. The method of claim 7, further comprising the step of creating for said target node an optimization table comprising at least one said suggested candidate connection for addition and at least one said connection made redundant by the addition to the network of said at least one candidate connection.

9. The method of claim 8, wherein each said fault and any said candidate connection for said target node, respectively, is a row and a column of said optimization table.

10. The method of claim 9, further comprising the step of marking a row in said optimization table in any column suggested by the fault corresponding to said row.

11. The method of claim 10, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy makes the highest cost marked fault become redundant.

12. The method of claim 10, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy has the highest sum cost of marked faults.

13. The method of claim 11, further comprising the step of selecting the candidate connection having the highest sum cost of marked faults for redundancy testing where the highest cost marked fault makes more than one candidate connection redundant.

14. The method of claim 6, wherein each said candidate connection is identified based on an assignment value implicated at a node outside the fanout cone of the target node.

15. The method of claim 14, further comprising the step of creating for said target node an optimization table comprising at least one said suggested candidate connection for addition and at least one said connection made redundant by the addition to the network of said at least one candidate connection.

16. The method of claim 15, wherein each said fault and any said candidate connection for said target node, respectively, is a row and a column of said optimization table.

17. The method of claim 16, further comprising the step of marking a row in said optimization table in any column suggested by the fault corresponding to said row.

18. The method of claim 17, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy makes the highest cost marked fault become redundant.

19. The method of claim 17, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy has the highest sum cost of marked faults.

20. The method of claim 18, further comprising the step of selecting the candidate connection having the highest sum cost of marked faults for redundancy testing where the highest cost marked fault makes more than one candidate connection redundant.

21. A method of efficiently optimizing a digital logic network through the iterative addition and removal of redundant connections previously identified as candidate connections comprising the steps of:

selecting in cost descending order a node of the network not previously selected as a target node for optimization, and in succession for each said target node selected,

analyzing the digital logic network to determine if connections to each said target node selected are absolutely dominated by each said target node, by determining if each of said connections must pass through each said target node before reaching any primary output of the digital logic network;

performing redundancy testing on each fault corresponding to each connection previously determined to be absolutely dominated by assigning values to the nodes of the network;

removing each said fault found redundant from the network and identifying each candidate connection for addition suggested by an assignment value implicated during redundancy testing; and

adding to the network each said candidate connection found redundant.

22. The method of claim 21, further comprising the step of removing from the network a connection made redundant by said addition.

23. The method of claim 22, wherein each said candidate connection is identified based on an assignment value implicated at a node outside the fanout cone and not in the input tree of said target node.

24. The method of claim 21, wherein optimization of the network continues until the network cannot be further optimized.

25. The method of claim 24, wherein optimization of the network ceases before a determination is made that the network cannot be further optimized.

26. The method of claim 23, further comprising the step of creating for said target node an optimization table comprising at least one said suggested candidate connection for addition and at least one said connection made redundant by the addition to the network of said at least one candidate connection.

27. The method of claim 26, wherein each said fault and any said candidate connection for said target node, respectively, is a row and a column of said optimization table.

28. The method of claim 27, further comprising the step of marking a row in said optimization table in any column suggested by the fault corresponding to said row.

29. The method of claim 28, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy makes the highest cost marked fault become redundant.

30. The method of claim 28, further comprising the step of selecting each candidate connection in descending order for redundancy testing according to which candidate connection not previously tested for redundancy has the highest sum cost of marked faults.

31. The method of claim 29, further comprising the step of selecting the candidate connection having the highest sum cost of marked faults for redundancy testing where the highest cost marked fault makes more than one candidate connection redundant.

32. The method of claim 31, wherein, where any connection made redundant by the addition of a candidate connection to a particular target node has been removed from the network, redundancy testing of said particular target node continues from the step of identifying each fault corresponding to each connection absolutely dominated by said particular target node.

33. The method of claims 32, wherein a candidate connection found redundant is not added to the network and the only one connection made redundant by addition of said redundant candidate connection is not removed from the network where the number of fanouts of the node corresponding to said one redundant connection is greater than the number of fanouts of the node corresponding to said redundant candidate connection.

34. The method of claim 33, wherein optimization of the network continues until the network cannot be further optimized.

35. The method of claim 34, wherein optimization of the network ceases before a determination is made that the network cannot be further optimized.

36. An apparatus for efficiently optimizing a digital logic network through the iterative addition and removal of redundant connections previously identified as candidate connections comprising;

a means for selecting a target node for optimization;

a means for analyzing the digital logic network to determine if connections to said target node are absolutely dominated by said target node, by determining if each of said connections must pass through said target node before reaching and primary output of the digital logic network;

a means for performing redundancy testing on each fault corresponding to each connection previously determined to be absolutely dominated by assigning mandatory values to the nodes of the network;

a means for removing each said fault found redundant from the network;

a means for identifying each candidate connection for addition suggested by a mandatory value assigned during redundancy testing; and

a means for adding any said candidate connection found redundant to the network.

37. The apparatus of claim 36, further comprising a means for removing from the network any connection made redundant by said addition.

38. The apparatus of claim 37, further comprising a means for determining whether to continue optimization of the network.