Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
2002-02-19
2004-10-05
Peikari, B. James (Department: 2186)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C710S054000
Reexamination Certificate
active
06802027
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to an electric or electronic circuit arrangement and to a method of detecting and/or identifying and/or recording at least an access violation, particularly at least a memory access violation, in a microcontroller provided particularly for a chip card or smart card.
2. Description of Related Art
Such a direct detection and/or identification and/or recording of access violations, particularly memory access violations, is conventionally only possible in the case of violations occurring upon data memory access, because these violations have a directly temporal relation between the run-down of instructions and the access to the data memory unit or the EEPROM (=Electrically Erasable Programmable Read-Only Memory).
In contrast, there is generally no fixed temporal relation between the instant of collecting or requesting the instruction code and the instant of executing the instruction code in microcontrollers and particularly in microcontrollers provided for chip cards or smart cars (for example, in the form of SXA, i.e. “smart extended architecture”). In principle, these two processes run independently of each other and are only very loosely interrelated by the fetch or request queue mechanism.
Access violations, particularly memory access violations are now conventionally detected by triggering a general break exception process. As already stated hereinbefore, the memory source causing the access violation as well as the associated code address of the instruction causing the access violation cannot be detected in microcontrollers and particularly in microcontrollers provided for chip cards or smart cards.
This impossibility of diagnosing or tracing the cause(s) of a general break exception has the result that, in the case of a memory access violation, the operating system of the circuit arrangement could hitherto not detect which memory unit triggered the violation and which instruction had exactly led to the violation. Also the circuit arrangements and methods known from the documents DE 37 36 190 A1, U.S. Pat. Nos. 5,155,829, 5,491,827 and 5,920,690 cannot eliminate these problems.
BRIEF SUMMARY OF THE INVENTION
Starting from the conventional arrangements, the present invention has for its object to provide an electric or electronic circuit arrangement and a method of the type described in the opening paragraph with which, in the case of occurrence of an access violation dug the program run, the source causing this access violation (referred to as break source) as well as the code address at which the violation occurs can be detected and/or identified and/or recorded.
This object is solved by the characteristic features defined in claim
1
for an electric or electronic circuit arrangement and by the characteristic features defined in claim
9
for a corresponding method. Advantageous embodiments and further embodiments of the present invention are defined in the dependent claims.
The teaching of the present invention is based on specially defined and/or specially encoded access violation codes which indicate an access violation and, likewise as the conventional instruction codes, are passed or moved through the fetch or request queue before these access violation codes become effective, for example, in the form of an “exception” and/or a “flag” during execution by means of the processor unit.
Requesting the instruction code or the access violation code and executing the instruction code or the access violation code substantially proceed temporally independently of each other in this case, i.e. “code memory fetch timing” and “instruction execution timing” are temporally independent of each other.
Such an access violation, which usually occurs when instruction codes are fetched or requested, is only effective in accordance with an advantageous embodiment of the present invention when the codes are actually executed after running through the fetch or request queue. In other words, this means that a violation does not occur when the fetch or request queue is erased before the code leading to a violation is executed.
With reference to the present invention, those skilled in the field of electric or electronic circuit techniques will appreciate the fact that the code access violation is detected exactly at the instant of executing the instruction code or the access violation code by the processor unit, i.e. it is not detected at the instant when the instruction code or the access violation code is fetched or requested from the relevant (code) memory unit (=the code-fetch instant). In this connection, the present invention provides the particularly valuable advantage that it can be recognized within the scope of the access violation in which (code) memory unit this access violation has taken place.
For the access violation codes which are specially defined for this purpose and indicate an access violation, and which, likewise as conventional instruction codes, are passed through the fetch or request queue, unused reserved op-codes are defined or modified in accordance with a particularly inventive further embodiment, namely a special op-code for each type of code access violation.
When an access violation occurs in one of the (memory) interface units in the case of a code-fetch, i.e. in the case of fetching or requesting the instruction code, it is not the expected (memory) op-code but the relevant access violation (op-)code of the (memory) interface unit that is entered into the fetch or request queue. The corresponding fetch or request sequence content now carries the data and information about the addresses, particularly the code addresses, and/or the type and/or the location and/or the source and/or the instant of the access violation.
In a further preferred embodiment of the present invention, the circuit arrangement and the method performed with the circuit arrangement are implemented in such a way that, only when the relevant access violation (op-)code reaches the decoding unit and is to be executed, the decoding unit recognizes that a code access violation has taken place at this location in the instruction sequence. In this case, at least a corresponding exception is triggered and/or at least a corresponding flag is set.
In a suitable embodiment of the present invention, at least a source register, particularly a break point-source register comprises several of such flags which can be set in a preferable way by the decoding unit when the decoding unit reaches one or more access violation (op-)codes. In other words, this means that this source register serves for recognizing and/or identifying and/or recording the break source or the origin of the violation when this exception is realized, i.e. when an exception routine is provided.
In accordance with an advantageous further embodiment of the present invention, the access violation (op-)code itself is also cleared or erased when clearing the request unit or when erasing the fetch or request queue, which is done in the meantine, i.e. before running the access violation code. Due to this disappearance of the access violation (op-)code from the fetch or request queue, taking place between the code-fetch and the code-run, no exception is triggered and/or no flag is set. In other words, this means that the fetch or request queue mechanism only responds to certain illegal fetch or request processes which are also actually decoded and executed.
In summary, it can be concluded that the circuit arrangement and the method provide the possibility of an exact diagnosis of one or more (memory) access violations by means of the exception routine of the operating system in relation to the instruction causing the violation and in relation to the source causing the violation (=break source).
This is particularly important in systems imposing strict reliability requirements, in which the operating system performs strong control and reliability functions mostly implemented in the (memory) interface units via the application code used in the system,
Buhr Wolfgang
Hagedorn Dieter
Mueller Detlef
Koninklijke Philips Electronics , N.V.
Peikari B. James
Shah Saumil R.
Simons Kevin
LandOfFree
Circuit arrangement and method of detecting access violation... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Circuit arrangement and method of detecting access violation..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Circuit arrangement and method of detecting access violation... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3264811