SE517917C2 - Integrated circuit for monitoring events in computer systems - Google Patents

Abstract

An integrated circuit (1) for a computer system, where the computer system comprises at least one processor (2) and an operating system which has at least one part (4) implemented in hardware, wherein said operating system part (4) is arranged in the integrated circuit (1). The integrated circuit (1) further comprises a unit (5) for supervising events in said computer system, wherein the supervising unit (5) has means for detecting events in the computer system and means for recording information about the events and that the supervising unit (5) is connected to said operating system part (4) in such a manner that information about internal events in the operation system part is possible to transfer to the supervising unit.

Description

: be: rnan 10 15 20 25 30 35 517 917 remove the extra instructions, new errors may occur due to the timing change.

It is especially important in a real-time operating system that the timing is not affected by the monitoring. Real-time systems, unlike most other systems, are extremely sensitive to timing interruptions.

Significant for a real-time operating system is that it is deterministic, which means that it must always be predictable. Thus, the monitoring of a real-time system must not change the timing or order of events in the system.

Another known method of monitoring a computer system is to use a monitoring device implemented in hardware. Such a monitoring device comprises probes which are connected to the computer system by buses and signals. The monitoring device passively detects what is happening on the bus and collects information about events that may be of interest. The advantage of monitoring using hardware is that it does not interfere with the system it is monitoring. A problem with using a separate monitoring device is that in some modern computer systems many functions are physically integrated on the same circuit in the same enclosure. It may therefore be impossible to access all the signals needed to obtain good monitoring. The enclosure of an integrated circuit should preferably not be too large and thus has a limited number of pins. In order to obtain all the signals necessary for the monitoring, the number of pins must be increased considerably and thus the size of the enclosure must be increased and this is not desirable.

To increase the performance of a computer system, it is known to implement all or at least part of the operating system in hardware in an external device outside the processor. High performance is especially important in real-time applications, such as process control applications. A hardware operating system has a real-time core which is arranged on an integrated circuit which is physically separate from the processor and the real-time core communicates with the processor via a bus line. Such a computer system is unsuitable for monitoring software with software, as the timing of the system can be affected. Although a monitoring device can be connected to the bus between the processor and the hardware operating system, the information available on the bus is not sufficient to be able to make a reliable monitoring of the system or to find errors in the program code.

DISCLOSURE OF THE INVENTION The object of the invention is to provide an integrated circuit which comprises a hardware operating system or at least a part of an operating system which is implemented in hardware and which enables a reliable monitoring of events in the computer system.

This object is achieved with the initially indicated integrated circuit, which is characterized in that it further comprises a unit for monitoring events in said computer system, the monitoring unit having means for detecting events in the computer system and means for recording information about the events, and that the monitoring unit is connected to said operating system part in such a way that information about internal events in the operating system part is transferable to the monitoring unit.

By arranging a monitoring unit on the same integrated circuit as the hardware operating system part and connecting these to each other, necessary information for the monitoring, for example information on internal events in the operating system part, can be transmitted to the monitoring unit. Such internal information is not available on the bus between the processor and the operating system part. the information about internal events is normally inaccessible or so sensitive to disturbances, such as capacitances in lines, that it is not measurable. The monitoring unit is passive and listens to what is happening in the operating system without affecting the system and its timing. When the monitoring unit has detected an interesting event, it records information about the event. This information can then be analyzed, for example, in order to find errors in the system. Since the monitoring unit is constructed »~: || aint. 10 15 20 25 30 35 517 917 erad in hardware, there is no obstacle to also using it for monitoring arbitrary physical signals in the system.

In a preferred embodiment of the invention, said operating system is a real-time operating system and said operating system part is a real-time core. The invention is particularly advantageous if the operating system is a real-time operating system, since the timing in the system is not affected by the monitoring according to the invention. In one embodiment of the invention, the operating system part comprises means for administering when a number of sub-jobs are to be executed by the processor and said information includes information about current conditions of the sub-jobs. This is relevant information and knowledge of this contributes to improved monitoring of the system. This information is difficult to access outside the integrated circuit, but the fact that the monitoring unit and the hardware operating system part are arranged on the same integrated circuit enables this information to be easily transmitted to the monitoring unit.

In an embodiment of the invention, the operating system part communicates with one or more processors via a bus and the monitoring unit comprises means for receiving information from said bus (es) from the bus (es). Because the monitoring unit is connected to the bus, a processor can address the monitoring unit and transmit information about what is happening in the processor to the monitoring unit. The software is provided with program instructions that transmit information via the bus to the monitoring unit. The monitoring of the software can thus be done partly on events in the operating system part and partly on arbitrary information that the software itself generates. In this way, a combination of hardware and software monitoring is obtained.

This means that monitoring is further improved, as the information about internal events in the operating system is supplemented with information about events in the software. Since most of the monitoring takes place in the hardware, only a few extra instructions are needed in the software code and thus the load on the system is not as great as when all monitoring takes place in soft. - the product. Due to the small load on the system, the extra instructions do not need to be removed when the monitoring is complete, thus avoiding problems with timing.

All information about events in the processor is not available on the bus. In one embodiment of the invention, the integrated circuit comprises at least one processor, which is connected to the monitoring unit in such a way that information about internal events in the processor is transferable to the monitoring unit.

By integrating one or more processors in the computer system on the same circuit as the operating system part and the monitoring unit, it is possible to transfer information about internal events in the processor to the monitoring unit. In this way, monitoring can be further improved. In one embodiment of the invention, the integrated circuit comprises means for transmitting the stored information to an external unit for further analysis of the information. The external device can be, for example, a computer. The recorded and stored events are sent via a communication link to the external device, where they are stored in a database. The collected information can then be used, for example, for event-based debugging, analysis of the system's behavior and statistics.

DESCRIPTION OF THE DRAWINGS The present invention will now be explained by means of various embodiments described by way of example and with reference to the accompanying drawings.

Figure 1 shows a block diagram of a computer system with an integrated circuit according to an embodiment of the invention.

Figure 2 shows a block diagram of a monitoring unit arranged on the integrated circuit in Figure 1.; || --1 10 15 20 25 30 35 517 917 6 Figure 3 shows the format of an event package.

Figure 4 shows another embodiment of an integrated circuit according to the invention.

DESCRIPTION OF EMBODIMENTS A computer program is structured in different sub-jobs. A part-time job is a sequential program and performs a specific task. In a real-time operating system, the sub-jobs are prioritized among themselves and if two sub-jobs are to be executed simultaneously by the processor, the sub-job with the highest priority is performed. The body that administers when a certain part-time job is to be executed by the processor is called a scheduler and forms the core of a real-time operating system. Figure 1 shows a computer system comprising an integrated circuit 1 according to the invention and three processors 2 arranged in parallel with each other. The processors 2 and the integrated circuit 1 are connected to each other via a system bus 3. The computer system has a real-time operating system which comprises a real-time core 4 which is implemented in hardware (RTK = Real Time Kernel) on the integrated circuit 1. The real-time core 4 is arranged to essentially perform the same functions as traditional real-time operating systems that are implemented in software, such as handling, scheduling and prioritizing sub-jobs, communication and synchronization between sub-jobs and interrupt management. More about how a real-time operating system is structured can be read in the book "Utilization of hardware Parallelism in Realizing Real-Time Kernels" by Lennart Lind, ISBN 0280-4506.

The integrated circuit further comprises a monitoring unit 5 which is connected to the real-time core 4. The monitoring unit has the task of detecting and recording events in the computer system.

Events in the operating system may, for example, involve: - system calls to the operating system, such as creating sub-jobs, deleting sub-jobs, sending messages, receiving messages, etc. - ip-u 10 15 20 25 30 35 517 917 - changing the state of a sub-job such as a partial job being placed in a blocked state or unlocked from a block, - interruption request from external units.

To monitor events in the operating system, the monitoring unit 5 is connected directly in the hardware to a number of selected signals in the real-time core 4. The connection is realized as one or more physical lines 9 from internal signals in the real-time core 4 to the monitoring unit 5. In this way, it is possible to gain access to valuable information about what is happening in the operating system. These internal signals contain, among other things, information about the state of the sub-jobs in the system, the priority of the sub-jobs, internal communication activities and internal and external interruptions.

In some cases, there is a need to be able to generate events from the software, for example to be able to see if you have passed certain control points or report the contents of a certain memory.

Such events can be produced by inserting software instructions into the program code. The monitoring unit can also be used to record general information generated directly from the software, i.e. information that comes directly from the execution of the software on one or more of the processors 2 in the system. In order to be able to monitor events in the software and receive information from the software, the monitoring unit 5 is connected to the system bus 3. The software generates the information by, via the system bus, writing data to a special register in the monitoring unit 5.

It is also possible to connect the monitoring unit to an arbitrary signal or hardware unit on the integrated circuit for monitoring the signal or hardware unit. Examples of such hardware units are buses and internal memories. This means that the monitoring unit can also function as a standard logic analyzer. Thus, the monitoring unit 5 can also be used to find functional faults in the hardware. This is especially valuable in situations where conventional troubleshooting methods are problematic to use, for example due to high system speeds or that the signals are difficult to access due to uaii: | »» || 10 15 20 25 30 35 517 917 in that the integrated circuit has a limited number of pins. A hardware unit 6 on the integrated circuit 1 is connected via physical lines 10 to the monitoring unit 5. The monitoring unit 5 is further connected via a parallel communication link 7 to an external computer 8. The monitoring unit 5 listens passively to events, logic and / or system level, in real-time kernel 4 and interesting events are recorded. The recorded events are then transferred to a database in the external computer where further processing and analysis of the information takes place.

The integrated circuit can either be implemented in ASlC (Application Specific Integrated Circut) or in a programmable hardware such as FPGA (Field Programmable Gate Array). All components are integrated on the same plate, which is preferably made of silicon.

Figure 2 shows a block diagram of the internal structure of the monitoring unit 1. The monitoring unit comprises an event detector 11, a timer 12, an event recorder 13, an event buffer 14 and an interface 15 to the external computer. Inputs to the monitoring unit are hard-wired signals 9 from the real-time core, signals on the system bus 3 and signals 10 from the hardware unit 6. The inputs are received by the event detector 11 which detects if any event to be recorded has occurred. The event detector 11 is a comparator which compares the input signals with internally predetermined conditions. There are a number of conditions that define which events are to be monitored and these conditions are hard-coded in the event detector. The event detector detects if an event has occurred by comparing the input signals with the predetermined conditions.

The event detector 11 comprises three different detectors 16-18 with different predetermined conditions. The first detector 16 receives the internal signals 9 from the real-time core and detects events in the operating system. The second detector 17 comprises a register which receives information from the software via the bus 3 and the second nal-u 10 15 20 25 30 35 517 917 detects when information is written to the register. The third detector 18 receives the signals 10 from the other hardware units and detects events in the hardware units.

When any of the conditions are met, i.e. an event has been detected, the event is reported to the event recorder 13.

The task of the event recorder is to make a packet with information about the event which can then be transmitted via the interface 15 to the external computer. The event recorder 11 is activated by one or more start signals from the event detector and at the same time receives an identification number from the event detector. The identification number provides information to the event recorder about which signals are to be packaged together with the detected event. When the event recorder is activated, it creates an information packet including the event identification number, signals associated with the event and the current time when the event was detected. The event package must have a certain predetermined format. The time is read from the timer 12 connected to the event recorder 13.

The event packet is then stored in the internal event buffer 14. This buffer is organized as a FIFO queue, i.e. the oldest packages come out first from the queue and the newest package comes out last (FlFO = First ln First Out). At the same time as the event packets are stored in the event buffer, an indication is stored that an event has been detected in a status register (not shown). The interface 15 transmits the packets to the external computer. Via the interface, the external computer can get an indication that an event has occurred and read out the event packet from the event buffer 14. indication of whether an event has been detected can be obtained either by reading the information in the status register or by programming the interface so that it automatically emits a signal when new events are available in the event buffer.

The size of the event buffer can be varied and is application dependent. 1:11 »|».> N 10 15 20 25 30 35 517 917 10 Sometimes you may not want to transfer all detected events to the external computer. For example, if the number of events is large. In the event detector 11 there is a filter 19 in the form of a programmable register. The filter 19 makes it possible to select which events you are interested in and just let them through. The filter 19 is connected to the interface 15 and it is possible to send instructions from the external computer to the filter about which events are to pass through the filter.

Figure 3 shows an example of what a package with information about an event might look like. The event package includes three different information fields, a first field 20 which contains the event identification number, i.e. information about the type of event in question, a second field 21 containing the time and a third field 22 containing more information about the event in the form of a plurality of parameters. An example of an event is when the processor switches and starts executing a new part-time job. Parameters contained in the third field should then contain information about the ID number of the new sub-job to be executed and which of the processors is to execute the sub-job. For the “send message” event, the parameters should include the ID number of the receiving sub-job and a pointer to the message.

Figure 4 shows an embodiment of the invention where an integrated circuit 29 comprises a processor 30, a real-time core 4 of an operating system, a monitoring unit 5 and various other hardware 6. The computer system in this embodiment has only one processor and this is integrated on the same circuit as the real-time core 4 and the monitoring unit 5. For a computer system containing several processors, of course, all the processors can be integrated on the same circuit. The monitoring unit 5 is hard-connected to a number of internal signals in the processor 30. In this way, the monitoring unit has access to internal information about events that are not visible on the system bus. Examples of such signals are signals to and from Cash memories in the processor. In the same way as in the previous embodiment, the monitoring unit 5 is connected 517 917 11 to internal signals in the real-time core 4 and to other hardware functions 6 on the circuit.

The invention is not limited to the embodiments shown but can be varied and modified within the scope of the appended claims. ; »~>:

Claims (7)

»Iaun iiiin 10 15 20 25 30 35 517 917 12 PATENTKRAV An integrated circuit (1) for a computer system, the computer system comprising at least one processor (2, 30) and an operating system, which has at least one part (4) implemented in hardware, said operating system part (4 ) is arranged on the integrated circuit (1), characterized in that the integrated circuit (1) further comprises a unit (5) for monitoring events in said computer system, the monitoring unit (5) having means (11) for detecting events in the computer system and means (13) for recording information about the events, and that the monitoring unit (5) is connected to said operating system part (4) in such a way that information about internal events in the operating system part is transferable to the monitoring unit. An integrated circuit according to claim 1, characterized in that said operating system (4) is a real-time operating system and that said operating system part is a real-time core. An integrated circuit according to claim 1 or 2, characterized in that the operating system part (4) comprises means for administering when a number of sub jobs are to be executed by the processor and that said information includes information on current states of the fl objects. An integrated circuit according to any one of claims 1-3, characterized in that said operating system part (4) communicates with at least one processor (2, 30) via a bus (3) and that the monitoring unit (5) comprises means (17) for receiving from the bus information about events in the processor. An integrated circuit according to any one of the preceding claims, characterized in that it comprises at least one processor (30), which is connected to the monitoring unit (5) in such a way that information about internal events in the processor is transferable to the monitoring unit. 517 917 13 An integrated circuit according to any one of the preceding claims, characterized in that it comprises means (15) for transmitting the recorded information to an external unit (8) for further analysis of the information. An integrated circuit according to any one of the preceding claims, characterized in that it constitutes a system integrated on silicon. A111: