Method for performing failure mode and effects analysis of an integrated circuit and computer program product therefor

• US 7,937,679 B2
• Filed: 04/11/2008
• Issued: 05/03/2011
• Est. Priority Date: 04/13/2007
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: YOGITECH S.p.A (Intel Corporation)
• Inventors: Mariani, Riccardo
• Primary Examiner(s): ROSSOSHEK, YELENA
• Application Number:
  US12/101,585
• Publication Number:
  US 20080276206A1
• Time in Patent Office: 1,117 Days
• Field of Search: 716/1, 716/2, 716/4, 716/5, 716/6, 716/7, 716/11, 716/18, 714/26, 714/37, 714/47, 714/57, 703/1, 703/2, 703/14, 703/22
• US Class Current: 716/104
• CPC Class Codes: G01R 31/2846 using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
• CPC Class Codes: G06F 11/2252 using fault dictionaries
• CPC Class Codes: G06F 2111/08 Probabilistic or stochastic CAD
• CPC Class Codes: G06F 2117/02 Fault tolerance, e.g. for transient fault suppression
• CPC Class Codes: G06F 30/33 Design verification, e.g. functional simulation or model checking

Abstract

A method for performing failure mode and effects analysis (FMEA) on integrated circuits including preparing a FMEA database of an integrated circuit under design and computing FMEA results from the FMEA database. Information is automatically extracted from an integrated circuit description. The extraction of information includes reading integrated circuit information, partitioning the circuit in invariant and elementary sensitive zones (SZ), using the information in the preparation step of a FMEA database. Optionally a FMEA validation stage may be performed with which FMEA computed results are compared with FMEA measured results to obtain FMEA validated results.

Claims

1. A method for performing failure mode and effects analysis (FMEA) on integrated circuits comprising:

preparing by a processor a FMEA database of a integrated digital circuit under design and computing FMEA results from said FMEA database, said FMEA database comprising records relative to zones (SZ) of said circuit;

automatically extracting information from a integrated circuit description, at least at the gate level of abstraction, said extraction of information comprising;

reading integrated circuit information;

partitioning said circuit in elementary sensitive zones (SZ);

said partitioning including selecting as zones registers from a netlist at gate level; and

extracting logic cone information associated to said sensitive zones, said extracting comprising extracting input cone information and extracting output cone information;

composing a sensitive zone database including said sensitive zone partition and the related logic cone information;

using said information in the preparing the FMEA database, importing said sensitive zone database in said FMEA database;

automatically computing by the processor on the basis of the information in said sensitive zone database fault model statistics for each record including failure rates for each sensitive zone; and

using said fault model statistics for each record including failure rates for each sensitive zone for automatically computing in said FMEA database failure rates for each FMEA record and on the basis of said computer failure rates computing FMEA results.

2. The method of claim 1 further comprising extracting the circuit usage profile from a circuit workload and of using said profile in said preparation step of a FMEA database.

3. The method of claim 1 further comprising entering information related to the circuit safety specifications and based on safety norms guidelines in said preparation step of a FMEA database.

4. The method of claim 1 further comprising a step of validation of the FMEA database using the circuit workload and a subsequent step of extraction of measured FMEA results from said measured FMEA database or a repetition of said preparation step of a FMEA database.

5. The method of claim 4 wherein the step of validation of the FMEA database includes a workload acceptance phase, that includes a functional coverage measurement of said circuit workload and a fault coverage measurement that makes use of the fault list database.

6. The method of claim 5, further comprising testing the results of the functional coverage measurement and of the fault coverage measurement.

7. The method of claim 6, wherein the testing yields a positive result, includes performing a fault injection operation for local faults and a fault injection operation for global faults in the sensitive zones (SZ) indicated in the sensitive zone database.

8. The method of claim 6, further comprising performing a fault coverage analysis taking all diagnostic outputs as observation points (OP) for the fault coverage measurement.

9. The method of claim 8, wherein said fault coverage analysis includes three sequential runs, a first run taking as observation points (OP) all the primary input/outputs of said integrated circuit under design coming from the circuit description and the circuit workload, a second run taking all the diagnostic outputs of said microelectronic circuits as observation points (OP), a third run measuring the diagnostic coverage due to SW diagnostic test for each sensitive zone (SZ).

10. The method of claim 8, further comprising a post-processing for elaborating the results of fault injection and fault coverage steps and a collection of the FMEA validation results.

11. The method of claim 10, wherein said post-processing and said collection includes a post processing flow for the norm IEC 61508 suitable for collecting indices required by said IEC 61508 norm.

12. The method of claim 10, wherein said post-processing and said collection includes a post processing flow for the norm ISO26262 suitable for collecting indices required by said ISO26262 norm.

13. The method of claim 11, wherein said post processing flow for the norm IEC 61508 is performed in three sequential specific runs, the runs comprising:

a first specific run for calculating as index the product of Dangerous Failure to Failure value (D*F) for each sensitive zone (SZ) taken from the sensitive zone database by a specific postprocessing;

a second specific run for calculating as index a total dangerous detected failure rate for permanent ON faults (DFF_ON) value for each sensitive zone (SZ) taken from the sensitive zone database by said specific postprocessing;

a third specific run for calculating as index a total dangerous detected failure rate for both permanent OFF and permanent ST faults (DDF_OFFST) value for each sensitive zone (SZ) taken from the database (214) by said postprocessing; and

wherein said specific postprocessing comprising parsing fault simulation reports and fault list database, dividing faults for the different sensitive zones (SZ) defined in the sensitive zone database, computing initial index values, applying a manipulation of said index values.

14. The method of claim 13, further comprising comparing between computed FMEA results and measured FMEA results and if a result of the comparing is below an acceptable threshold said repetition of said preparation step of a FMEA database is performed, if the result of said comparison yields a positive results, validated FMEA results are produced.

15. The method of claim 1, further comprising generating a fault-list (LF) by parsing said input (IL) and output (OL) logic cones and generating a fault-list database.

16. The method of claim 1, further comprising extracting information on shared elements by evaluating the correlation between each sensitive zone (SZ) in terms of shared elements, gates and nets, and creating a sharing database.

17. The method of claim 1, wherein said FMEA database includes presenting records, each of the records corresponding to a possible failure mode (FM) of a given sensitive zone (SZ) and each record includes a sensitive zone field, a failure mode field, a FMEA statistics field and a diagnostic coverage field.

18. The method of claim 17, further comprising presenting one or more field of said records for receiving for each sensitive zone (SZ) a failure mode (FM) representing the effect of faults (G) in the sensitive zone (SZ) leading to an error, on the basis of the circuit description, the safety specifications and the safety norm guidelines.

19. The method of claim 17, further comprising presenting said failure mode field of said records for receiving in the each sensitive zone (SZ) an expected main effect (ME) representing a certain failure mode (FM) of the corresponding sensitive zone (SZ) with respect to an observation point (OP).

20. The method of claim 17, wherein FMEA database automatically computes detected/undetected failure rates for each FMEA record and on the basis of said computed detected/undetected failure rates computes FMEA results with diagnostic.

21. The method of claim 19, further comprising presenting said FMEA statistics field of said records for receiving for each sensitive zone (SZ) expected usage frequency values.

22. The method of claim 21, further comprising presenting said FMEA statistics field of said records for receiving for each sensitive zone (SZ) safe and dangerous values.

23. The method of claim 22, wherein said receiving for each sensitive zone (SZ) safe and dangerous values includes receiving an estimated Safe fraction and Dangerous fraction of the possible failures for the given failure mode (FM) in the given sensitive zone (SZ), divided in transient and permanent faults.

24. The method of claim 22, wherein said FMEA database automatically computes failure rates for each FMEA record and on the basis of said computed failure rates computes FMEA results without diagnostic.

25. The method of claim 24, further comprising receiving for each sensitive zone (SZ) the diagnostic coverage.

26. The method of claim 25, wherein said receiving for each sensitive zone (SZ) the diagnostic coverage includes performing an auto-coverage computation, taking in account for DDF estimations (DDF AUTO (TR), DDF AUTO (PE)) the logic sharing between the sensitive zones (SZ) on the basis of said sharing database.

27. The method of claim 25, wherein said receiving for each sensitive zone (SZ) the diagnostic coverage includes performing an auto-coverage computation includes using for each sensitive zone (Z) a shape function to weight the sharing degree with other sensitive zones.

28. The method of claim 1, wherein said fault models are built distinguishing faults (G) originating failures (F) occurring in a sensitive zone (SZ) between permanent and transient/intermittent faults.

29. The method of claim 28, further comprising distinguishing said permanent faults in “

permanent ON”

, “

permanent OFF” and

“

permanent ST”

depending on when they are occurring in system operation mode.

30. A processing apparatus configured for performing the steps of the method of claim 1.

31. A computer program product directly loadable in the memory of a computer and comprising software code portions for performing the method of claim 1 when the product is run on a computer.