Method and system for subnet defect diagnostics through fault compositing
A method and system for subnet defect diagnostics through fault compositing is disclosed. A testing apparatus generates callout data for an integrated circuit device under test. A computer received the callout data, which includes a list of faults. Each fault of the list of faults has associated with it one or more failures and/or conflicts. In order to explain the failures, two or more faults are selected and composited, yielding a composite fault having a composite conflict count. The composite fault is assigned a score based on the composite conflict count, which score determines a candidate composite that best explains the faults of the list of faults. This procedure may be repeated to explain all the failures.
1. A computer-implemented method of diagnosing subnet defects, comprising:
(i) receiving, at a computer, callout data comprising a list of faults, each fault of the list of faults comprising failure data and conflict count, the failure data containing miscomparing measures that each fault models, and a conflict indicating that each fault predicts a failure that was not measured;
(ii) selecting a first fault on a fan-out sink of a fan-out net, the first fault explaining a first failure of the plurality of failures;
(iii) selecting a second fault on a different sink of the fan-out net, the second fault explaining a second failure that the first fault does not explain;
(iv) compositing the first fault and the second fault to yield a composite fault, wherein the composite fault comprises a combination of the failure data and conflict count of the first fault and the failure data and conflict count of the second fault;
(v) generating a composite conflict count by combining the conflict count of the first fault and the conflict count of the second fault;
(vi) assigning a score to the composite fault based on the composite conflict count; and
(vii) determining a best candidate composite fault explaining the first fault and the second fault based on the score.
2. The computer-implemented method of diagnosing subnet defects of claim 1 , further comprising repeating (ii) through (vii) until the composite fault explains all of the plurality of failures.
3. The computer-implemented method of diagnosing subnet defects of claim 1 , wherein the fault data includes a first count of test-failed (TF) faults, a second count of test-failed, simulation-detected (TFSD) faults, a third count of test-passed, simulation-detected (TPSD) faults.
4. The computer-implemented method of diagnosing subnet defects of claim 1 , wherein the score is calculated by TFSD/(TF+TPSD/a), where a is a numerical weighting factor greater than zero, TF is a count of test-failed faults, TFSD is a count of test-failed, simulation-detected faults, and TPSD is a count of test-passed, simulation-detected faults.
5. The computer-implemented method of diagnosing subnet defects of claim 1 , further comprising assigning a minimum conflict count to the composite fault.
6. The computer-implemented method of diagnosing subnet defects of claim 5 , wherein the score is given a higher value when the composite conflict count is smaller than the minimum conflict count.
7. The computer-implemented method of diagnosing subnet defects of claim 1 , wherein the plurality of failures comprises a defect caused by an inverting fault.
8. The computer-implemented method of diagnosing subnet defects of claim 7 , wherein the inverting fault is caused by a shorted net.
9. The computer-implemented method of diagnosing subnet defects of claim 1 , wherein the plurality of failures comprises a timing defect.
10. The computer-implemented method of diagnosing subnet defects of claim 1 , wherein the first fault and the second fault are logically unrelated.
11. A computer-readable storage medium having stored thereon a plurality of instructions, said plurality of instructions when executed by a computer, cause said computer to perform:
(i) receiving callout data comprising a list of faults, each fault of the list of faults comprising failure data and conflict count, the failure data containing miscomparing measures that each fault models, and a conflict indicating that each fault predicts a failure that was not measured;
(ii) selecting a first fault on a fan-out sink of a fan-out net, the first fault explaining a first failure of the plurality of failures;
(iii) selecting a second fault on a different sink of the fan-out net, the second fault explaining a second failure that the first fault does not explain;
(iv) compositing the first fault and the second fault to yield a composite fault, wherein the composite fault comprises a combination of the failure data and conflict count of the first fault and the failure data and conflict count of the second fault;
(v) generating a composite conflict count by combining the conflict count of the first fault and the conflict count of the second fault;
(vi) assigning a score to the composite fault based on the composite conflict count; and
(vii) determining a best candidate composite fault explaining the first fault and the second fault based on the score.
12. The computer-readable storage medium of claim 11 , having stored thereon a plurality of instructions, said plurality of instructions when executed by a computer, cause said computer to further perform repeating (ii) through (vii) until the composite fault explains all of the plurality of failures.
13. The computer-readable storage medium of claim 11 , wherein the fault data includes a first count of test-failed (TF) faults, a second count of test-failed, simulation-detected (TFSD) faults, a third count of test-passed, simulation-detected (TPSD) faults.
14. The computer-readable storage medium of claim 11 , wherein the score is calculated by TFSD/(TF+TPSD/a), where a is a numerical weighting factor greater than zero, TF is a count of test-failed faults, TFSD is a count of test-failed, simulation-detected faults, and TPSD is a count of test-passed, simulation-detected faults.
15. The computer-readable storage medium of claim 11 , having stored thereon a plurality of instructions, said plurality of instructions when executed by a computer, cause said computer to further perform assigning a minimum conflict count to the composite fault.
16. The computer-readable storage medium of claim 15 , wherein the score is given a higher value when the composite conflict count is smaller than the minimum conflict count.
17. The computer-readable storage medium of claim 11 , wherein the plurality of failures comprises a defect caused by an inverting fault.
18. The computer-readable storage medium of claim 17 , wherein the inverting fault is caused by a shorted net.
19. The computer-readable storage medium of claim 11 , wherein the plurality of failures comprises a timing defect.
20. The computer-readable storage medium of claim 11 , wherein the first fault and the second fault are logically unrelated.
21. A diagnostic system for fault compositing, the diagnostic system comprising:
(i) an input for receiving callout data from a tester, the callout data comprising a list of faults, and each fault of the list of faults comprising failure data and conflict count, wherein the failure data contains miscomparing measures that each fault models, and a conflict indicates that each fault predicts a failure that was not measured;
(ii) means for compositing a first fault and a second fault to yield a composite fault, wherein the first fault is selected on a fan-out sink of a fan-out net, the first fault explaining a first failure of the plurality of failures, the second fault is selected on a different sink of the fan-out net, the second fault explaining a second failure that the first fault does not explain, and the composite fault comprises a combination of the failure data and conflict count of the first fault and the failure data and conflict count of the second fault;
(iii) means for generating a composite conflict count by combining the conflict count of the first fault and the conflict count of the second fault;
(iv) means for assigning a score to the composite fault based on the composite conflict count; and
(v) means for determining a best candidate composite fault explaining the first fault and the second fault based on the score.