TRACE/FAILURE OBSERVATION SYSTEM, TRACE/FAILURE OBSERVATION METHOD, AND TRACE/FAILURE OBSERVATION PROGRAM

• US 20100299564A1
• Filed: 02/03/2009
• Published: 11/25/2010
• Est. Priority Date: 02/04/2008
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Sakai, Junji, Suzuki, Noriaki
• Granted Patent: US 8,799,753 B2
• Time in Patent Office: Days
• US Class Current: 714/45
• CPC Class Codes: G06F 11/3636 by tracing the execution of the program
• CPC Class Codes: G06F 11/3648 using additional hardware

Abstract

There is provided a trace/failure observation system which is capable of comprehensive collection of information that is needed for checking a desired operation in a system or the like where the amount of information to be observed is large, and which allows easy analysis of the desired operation. The system includes, in a system LSI to be subjected to trace/failure observation: an event detecting means for observing behavior of a portion to be observed; a first data reducing means for performing observation data reduction processing so that observation data from the event detecting means has an amount of information processable to a second data reducing means; and the second data reducing means for performing one or more steps of observation data reduction processing.

Claims

1-21. -21. (canceled)

22. A trace/failure observation system arranged in a system LSI to be subjected to trace/failure observation, the system comprising:

event detecting unit for observing behavior of the system LSI to be observed;

first data reducing unit for performing first reduction processing on observation data resulting from the event detecting unit; and

second data reducing unit for performing second reduction processing different from the first reduction processing on a result of reduction by the first data reducing unit,

the second data reducing unit controlling a reduction condition for the first reduction processing of the first data reducing unit according to a result of reference to the observation data resulting from the first data reducing unit.

23. The trace/failure observation system according to claim 22, wherein the second data reducing unit includes a control processor, a storage device, and a data reduction program.

24. The trace/failure observation system according to claim 22, wherein the second data reducing unit has a function of performing parallel processing on two or more data series.

25. The trace/failure observation system according to claim 22, wherein the second data reducing unit is constituted by a state machine function.

26. The trace/failure observation system according to claim 22, further comprising an observation data buffer that temporarily stores observation data output from the first data reducing unit.

27. The trace/failure observation system according to claim 22, further comprising:

an interrupt condition filter that makes a matching notification when detecting an event that matches a set interrupt condition; and

an interrupt control unit that performs control such as setting whether or not to make an interrupt notification to the second data reducing unit when receiving the matching notification from the interrupt condition filter.

28. The trace/failure observation system according to claim 22, further comprising:

a next condition holding unit that holds a next condition to be set in the first data reducing unit and the interrupt condition filter, and sets the next condition in the first data reducing unit and the interrupt condition filter held when receiving a condition change instruction.

31. The trace/failure observation system according to claim 22, wherein

the first data reducing unit performs the reduction processing within a cycle time from arrival of a piece of observation data to that of a next piece of observation data.

32. The trace/failure observation system according to claim 22, wherein

the first data reducing unit performs the reduction processing within a cycle time by defining a data series and determining data to pass depending on whether or not the data coincides with the defined data series.

33. The trace/failure observation system according to claim 22, wherein:

the first data reducing unit reduces data by selecting part of pieces of data; and

the second data reducing unit reduces data by selecting part of data from a group of data.

34. The trace/failure observation system according to claim 22, wherein:

the trace/failure observation system needs a period longer than or equal to a cycle time from arrival of a piece of observation data to that of a next piece of observation data to determine whether or not to change control, or to control the first data reducing unit; and

the second data reducing unit switches, depending on a result of detection by the event detecting unit, between a control of setting the first data reducing unit so as to pass observation data both pertaining to a first series of observation data and a second series of observation data and selecting data in the first series of observation data for data reduction and a control of setting the first data reducing unit so as to pass observation data pertaining to the second series of observation data.

35. The trace/failure observation system according to claim 22, wherein:

the trace/failure observation system needs a period longer than or equal to a cycle time from arrival of a piece of observation data to that of a next piece of observation data to determine whether or not to change control, or to control the first data reducing unit; and

the second data reducing unit switches, depending on a result of detection by the event detecting unit, between a control of setting the first data reducing unit so as to pass observation data both pertaining to a first series of observation data and a second series of observation data and selecting data in the first series of observation data for data reduction and a control of setting the first data reducing unit so as to pass observation data pertaining to the second series of observation data and selecting data pertaining the second series of observation data for data reduction.

29. A trace/failure observation method to be performed in a system LSI to be subjected to trace/failure observation, the method comprising:

an event detecting step of observing behavior of the system LSI to be observed;

a first data reducing step of performing first reduction processing on observation data resulting from the event detecting step; and

a second data reducing step of performing second reduction processing different from the first reduction processing on a result of reduction by the first data reducing step,

the second data reducing step including controlling a reduction condition for the first reduction processing of the first data reducing step according to a result of reference to the observation data resulting from the first data reducing step.

30. A non-transitory computer readable recording medium that records a trace/failure observation program for making a computer function as a trace/failure observation system that is arranged in a system LSI to be subjected to trace/failure observation, the system including:

event detecting means for observing behavior of the system LSI to be observed;

first data reducing means for performing first reduction processing on observation data resulting from the event detecting means; and

second data reducing means for performing second reduction processing different from the first reduction processing on a result of reduction by the first data reducing means,

the second data reducing means controlling a reduction condition for the first reduction processing of the first data reducing means according to a result of reference to the observation data resulting from the first data reducing means.