Abstract: A method, computer program product, and data processing system for minimizing the number of test sequences needed to achieve a desired level of coverage of events in testing a semiconductor design is disclosed. Test patterns are randomly generated by one or more “frontend” computers. Results from applying these patterns to the design under test are transmitted to a “backend” computer for processing. A determination is made as to which test sequences trigger events not already triggered by previously-considered test sequences. An autograde data structure is generated which further reduces the number of test sequences. A preferred embodiment of the present invention may be used to reduce the number of test sequences required, but may also be used to provide test engineers a basis for devising manually-created test sequences to test related events.

Abstract: A method, computer program product, and data processing system for determining test sequences' coverage of events in testing a semiconductor design are disclosed. Test patterns are randomly generated by one or more “frontend” computers. Results from applying these patterns to the design under test are transmitted to a “backend” computer in the form of an ordered dictionary of events and bitmap and/or countmap data structures. A “bitmap” data structure encodes Boolean information regarding whether or not a given event was covered by a particular test sequence. A “countmap” data structure includes frequency information indicating how many times a given event was triggered by a particular test sequence. The backend computer combines results from each test sequence in a cumulative fashion to measure the overall coverage of the set of test sequences.