Extreme cases sampling method for improved variation-aware full custom design

A system for yield improvement of an integrated circuit device implementing a circuit design having a set of performance specifications including one or more performance specifications, the system comprising:

• a memory;
• a computer system coupled to the memory, the computer system configured to:
• obtain a first circuit design;
• in a first verification, verify by simulation adherence of the first circuit design to the set of performance specifications, over a first set of test cases which includes variations in one or more of a fabrication process variable and an environmental condition, the first verification including identifying, for each test case of the first set of test cases, an extent to which the first circuit design satisfies one or more performance specifications of the set of performance specifications, the first set of test cases including a respective set of randomly generated Monte Carlo parameter values for each device in a first set of devices in the first circuit design;
• store an identification of a set of Monte Carlo parameter values for each device in the first set of devices from the first set of test cases;
• determine, in dependence upon the first verification, a second set of test cases in dependence upon the first set of test cases;
• provide a second circuit design adjusted in dependence upon the first verification;
• and
• in a second verification, re-verify by simulation adherence of the second circuit design to the set of performance specifications, over the second set of test cases,
• wherein the second set of test cases includes the stored set of Monte Carlo parameter values for each device in the first set of devices which remain in the second circuit design.

A method for yield improvement of an integrated circuit device implementing a circuit design having a set of performance specifications including one or more performance specifications, the method comprising:

• a computer system obtaining a first circuit design;
• in a first verification, a computer system verifying by simulation adherence of the first circuit design to the set of performance specifications, over a first set of test cases which includes variations in one or more of a fabrication process variable and an environmental condition, the first verification including identifying, for each test case of the first set of test cases, an extent to which the first circuit design satisfies one or more performance specifications of the set of performance specifications;
• determining, in dependence upon the first verification, a second set of test cases having test cases fewer than all of the test cases in the first set of test cases;
• providing a second circuit design adjusted in dependence upon the first verification;
• and
• in a second verification, a computer system re-verifying by simulation adherence of the second circuit design to the set of performance specifications, over only the second set of test cases.

A method for yield improvement of an integrated circuit device implementing a circuit design having a set of performance specifications including one or more performance specifications, the method comprising:

• a computer system obtaining a first circuit design;
• in a first verification, a computer system verifying by simulation adherence of the first circuit design to the set of performance specifications, over a first set of test cases which includes variations in one or more of a fabrication process variable and an environmental condition, the first verification including identifying, for each test case of the first set of test cases, an extent to which the first circuit design satisfies one or more performance specifications of the set of performance specifications, the first set of test cases including a respective set of randomly generated Monte Carlo parameter values for each device in a first set of devices in the first circuit design;
• storing an identification of a set of Monte Carlo parameter values for each device in the first set of devices from the first set of test cases;
• determining, in dependence upon the first verification, a second set of test cases in dependence upon the first set of test cases;
• providing a second circuit design adjusted in dependence upon the first verification;
• and
• in a second verification, a computer system re-verifying by simulation adherence of the second circuit design to the set of performance specifications, over the second set of test cases,
• wherein the second set of test cases includes the stored set of Monte Carlo parameter values for each device in the first set of devices which remain in the second circuit design.