DE102023210817A1 - Method for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit - Google Patents

Abstract

The invention relates to a method for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit (10) and/or another computing unit in a memory system (12) of the computing unit (10) having at least one buffer memory (16), comprising the following steps:
- reading in event information to be stored in the buffer memory (16) relating to an event of the computing unit (10) and/or the further computing unit, wherein the read-in event information comprises a cause for the event;
- Determining a number of event information stored in the buffer memory (16) which have the same cause for the respective event as the read-in event information, wherein
◯ the oldest event information with the same cause is deleted from the buffer memory (16) if the determined number is greater than or equal to a defined threshold value, otherwise
◯ the oldest event information is deleted from the buffer memory (16); and
- Storing the read event information in the buffer memory (16).

Description

The invention relates to a method for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit and/or a further computing unit in a memory system of the computing unit having at least one buffer memory, furthermore a computing unit and a corresponding computer program and a storage medium.

State of the art

In systems consisting of a microcontroller and/or microprocessor, so-called resets (restarts) can occur during runtime, causing the system to restart. These resets can have very different causes; for example, switching on the supply voltage can trigger a "power-on reset," a software function can trigger a software reset, a hardware error in the microcontroller/microprocessor can trigger a reset, etc.

Since these, sometimes unintentional, restarts negatively impact application availability, there is an interest in detecting and logging these events. For this purpose, it is state of the art to implement a type of reset buffer from which the list of most recently occurred resets can be read.

Reading these reset events can provide very helpful information about a device's problem during error analysis in the field and also during return analysis.

However, a major problem in embedded systems is that they typically have only a very limited amount of available memory to store such events. Therefore, a trade-off between the data to be stored and the available memory is often necessary. This leads, among other things, to a reset buffer typically only being able to store a certain number of events and then overwriting the oldest event in the buffer as new events arrive.

While this method keeps memory usage under control, old reset events are simply discarded. This is acceptable if the same reset reason occurs every time. However, discarding old reset events can be disadvantageous if the occurrence of new resets deletes older events from the buffer, leaving no trace that the specific (deleted) reset ever occurred.

From the US 2009/0222700 A1 A method for detecting a device reset history of a network communication device is known, wherein the reset has a hardware or software cause.

Disclosure of the invention

According to a first aspect, the subject matter of the present invention is a method, in particular a computer-implemented method, for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit and/or the further computing unit in a memory system of the computing unit having at least one buffer memory according to claim 1.

According to a second aspect, the present invention relates to a computing unit according to claim 10.

According to a further aspect, the present invention relates to a computer program and a machine-readable storage medium.

In the context of this application, an event of the computing unit can be understood as an event that affects the software and/or hardware of the computing unit and/or the additional computing unit. At least two different causes or reasons can exist for the same event. This means that there is an interest in logging, reading, or determining not only the frequency of events, but also the underlying cause or reason for the respective event.

The event is preferably a reset event or a restart event, wherein the reset or restart may comprise a reset of a software and/or hardware module of the and/or the further computing unit. It is also conceivable that the event is an error event of the and/or the further computing unit. In this case, the error event may comprise an error of a software and/or hardware module of the computing unit. It is further conceivable that the event relates to another event of the and/or the further computing unit.

The cause or reason of the event characterizes a trigger of the event. For example, a reset event can be triggered by a software and/or a hardware module of the and/or the additional computing unit. It is conceivable that a reset event is caused or triggered by activating or interrupting a supply voltage of the computing unit.

The event information may include, represent, or indicate the event. For example, the event information may be embodied as a digital signal that includes, represents, or indicates the event.

The computing unit comprises one or more microcontrollers and/or one or more microprocessors. Furthermore, the computing unit comprises a memory system with at least one buffer memory. Additionally, the memory system may comprise one or more additional memory units, e.g., a main memory.

In the context of the present application, a buffer memory or a buffer can be understood as a preferably non-volatile memory unit of the memory system, which is designed or configured to temporarily store or buffer data, in particular event information relating to events of the and/or the further computing unit.

For this purpose, the buffer memory has a defined memory size. In particular, the buffer memory is designed to store a defined number of event information items or entries.

The computing unit comprising the buffer memory is preferably designed as a control unit or a control device. In particular, the computing unit can be designed as an embedded system, for example, as an embedded control device. The computing unit can be designed, for example, to control a vehicle, a smart home device, a handheld power tool, a household appliance, a smartphone, etc.

The method comprises a step of reading in event information relating to an event of the computing unit and/or the further computing unit, to be stored in the buffer memory. The read-in event information comprises a cause of the event. Reading in event information, in particular by means of a hardware and/or software interface of the computing unit, can comprise receiving the event information and/or reading the event information from a storage medium, in particular a temporary storage medium. The step of reading in the event information can be preceded by a step of capturing the event information in order to provide the event information, for example, to the computing unit for storage in the buffer memory.

The method further comprises a step of determining a number of event information items stored in the buffer memory which have the same cause for the respective event as the read-in event information. If, when, or precisely in the case in which the determined number is greater than or equal to a defined threshold value, the oldest event information item with the same cause is deleted from the buffer memory. If, when, when, or precisely in the case in which the determined number is less than the defined threshold value, the oldest event information item is deleted from the buffer memory. The oldest event information item is the oldest event information item stored in the buffer memory, which is deleted regardless of the cause encompassed by the event information item.

In other words, the method comprises a step of deleting event information from the buffer memory as a function of the determined number and a defined threshold value, in particular deleting the oldest event information with the same cause from the buffer memory if or when or exactly in the case in which the determined number is greater than or equal to a defined threshold value, otherwise deleting the oldest event information from the buffer memory.

It is conceivable that the defined threshold value has the same value for all causes of events in the computing unit. It is also conceivable that the defined threshold value has different values for different causes of events in the computing unit. In other words, the defined threshold value can be a cause-specific threshold value. In particular, the threshold value can be specified or specifiable in a range greater than or equal to 1 and less than or equal to 20, preferably greater than or equal to 1 and less than or equal to 10.

The method further comprises a step of storing the read-in event information in the buffer memory. In other words, the read-in event information is written or stored in the buffer memory.

The proposed method is preferably executed during operation of the computing unit. For example, the method is executed repeatedly in the described sequence of process steps, preferably whenever an event of the computing unit occurs.

The method and the devices according to the invention enable even rarely occurring causes or reasons for events in the computing unit to remain stored in the buffer memory. In particular, it is ensured that a cause for an event stored at least once in the buffer memory is not overwritten by other, more frequently occurring causes. Rather, for each cause that has occurred, at least the most recent event information remains stored in the buffer memory. In other words, not every piece of event information stored in the buffer memory remains permanently available; however, it is ensured that at least a predetermined number of event information items remain available for each cause and cannot be overwritten by other causes. This can significantly improve the diagnosis or analysis of causes of events in a computing unit, in particular reasons for reset events in the computing unit.

The proposed approach eliminates the need to filter the read event information before saving it, for example, with regard to deliberately induced and unintentional events, in order to avoid saving deliberately induced reset events at all. The storage space required for the proposed solution can be optimized by adapting it to the number of occurring causes and the choice of threshold value, so that all occurring causes of events can be reliably stored and analyzed, even in embedded systems with limited storage space. The memory requirements of the buffer memory can be managed statically, meaning no dynamic memory allocation is necessary. Rather, the memory required for the buffer memory can be precisely predicted, which is particularly advantageous for embedded systems.

It is advantageous if the read-in event information is read from a further buffer memory of the storage system, wherein the read-in event information is one, in particular the oldest, item of event information stored in the further buffer memory. It is also conceivable for the read-in event information to be the second oldest or third oldest, etc., item of event information stored in the further buffer memory. According to this embodiment, the memory system of the computing unit comprises the buffer memory and the further buffer memory. The memory system can therefore be understood as an at least two-stage system of buffer memories. In this case, an item of event information to be stored or stored in the buffer memory is event information previously stored in the further buffer memory. Thus, the further buffer memory can be understood as the first buffer memory or as a first storage stage of the storage system, and the buffer memory can be understood as the second buffer memory or as a second storage stage of the storage system.

Preferably, the additional buffer memory is configured as a FIFO buffer memory or FIFO buffer (English: "First In First Out"). In other words, the FIFO buffer memory is configured to store or save the stored data, in particular the stored event information, chronologically. The FIFO buffer memory is configured to always delete the oldest event information when there is no more storage space available in the additional buffer memory for storing the event information to be stored. This configuration ensures that at least the N most recent event information items are stored chronologically correctly, regardless of the cause of the respective event, where N represents the number of event information items that can be stored in the additional buffer memory. As a result, the N most recently occurred events and their cause can be read out at any time.

• - dem Schritt des Speicherns der eingelesenen Ereignisinformation in dem Pufferspeicher ein Schritt des Löschens der, insbesondere ältesten, in dem weiteren Pufferspeicher gespeicherten Ereignisinformation und ein Schritt des Speicherns der eingelesenen weiteren Ereignisinformation in dem weiteren Pufferspeicher nachfolgen.

It is advantageous here if the step of determining the number is preceded by a step of reading in further event information to be stored in the further buffer memory relating to a further event of the computing unit and/or the further computing unit, wherein the further event information comprises a cause for the further event, and

• - the step of storing the read-in event information in the buffer memory is followed by a step of deleting the, in particular the oldest, event information stored in the further buffer memory and a step of storing the read-in further event information in the further buffer memory.

Here, the event information deleted from the further buffer memory is the event information stored in the buffer memory. In particular, the step of determining the number can be carried out in response to the reading of the further event information to be stored in the further buffer memory. In other words, the method comprises a step of moving the oldest event information from the further buffer memory into the buffer memory in order to make storage space available in the further buffer memory for the further event information to be stored. Through this embodiment, the buffer memory and the further buffer memory can be used together to store event information, wherein the event information, regardless of the cause of the respective events, is initially stored chronologically in the additional buffer memory and only moved to the buffer memory when there is no storage space available in the additional buffer memory to store further event information.

It is also advantageous if a data record is assigned to the and/or the further event information, wherein the data record comprises data present at the time of occurrence of the and/or the further event and/or additional information regarding the cause of the or the further event and is stored together with the and/or the further event information in the respective buffer memory.

The data available at the occurrence or time of the occurrence of the and/or further event can be data from the computing unit and/or a device comprising the computing unit, e.g., a vehicle. The data can be in the form of environmental data. For example, the data can comprise one or more elements selected from the group: date, time, mileage of a vehicle comprising the computing unit, a supply voltage value of the computing unit, an uptime of the computing unit, a power cycle counter, a general or generic event counter, and/or a specific or cause-specific event counter (a separate counter for each cause of the event, which is incremented by one when the event occurs). A power cycle counter can be understood as a counter that represents the number of power cycles of the computing unit. This can be the number of times the control unit has been switched on and run, or, for example, the number of times an ignition has been activated in a vehicle comprising the computing unit.

The additional information regarding the cause of the event(s) may include information that further specifies or details the cause. It is conceivable that additional information regarding the cause is available or can be provided only for one or more of a multitude of causes for the events.

The data and/or the supplementary information can be provided by another unit of the computing unit, e.g., a register of a microcontroller of the computing unit, and/or a unit of a device comprising the computing unit, e.g., a sensor unit comprised by the device, and/or a unit external to the computing unit and/or the device. For example, the data and/or the supplementary information can be provided via a wireless or wired communication connection, e.g., via a CAN, Ethernet, Wi-Fi, or mobile network connection.

This design allows for a more detailed analysis of the events or a diagnosis of the computing unit and for a better understanding of the causes for the occurrence of certain events.

Alternatively, it is advantageous if a data record is assigned to the and/or the further event information, wherein the data record comprises data present at the time of occurrence of the and/or the further event and/or additional information regarding the cause of the or the further event and is stored in an additional buffer memory of the memory system, wherein the and/or the further stored event information comprises a variable referring to the stored data record. The variable referring to the stored data record can be designed as a reference, e.g. as an index or a pointer. By providing an additional buffer memory for storing the data record, it is possible to store all occurring causes, including the additional information, in the memory system of the computing unit, so that even causes can be stored and read out that should never have occurred or that were not anticipated at the time of development of the computing unit, i.e. that were unknown a priori or occur contrary to expectations. This configuration allows the data and/or information comprised in the data set to be stored additionally without having to expand the buffer memory and/or the additional buffer memory by the corresponding storage space.

• - die Anzahl der in dem Pufferspeicher gespeicherten Ereignisinformationen mit derselben Ursache wie die eingelesene Ereignisinformation nur dann ermittelt wird, wenn in dem Pufferspeicher kein Speicherplatz zum Speichern der eingelesenen Ereignisinformation verfügbar ist, ansonsten
• - die eingelesene Ereignisinformation in dem verfügbaren Speicherplatz des Pufferspeichers gespeichert wird.

It is furthermore advantageous if the method comprises a step of determining the storage space available in the buffer memory for storing the read-in event information, wherein

• - the number of event information stored in the buffer memory with the same cause as the read-in event information is only determined if there is no storage space available in the buffer memory to store the read-in event information, otherwise
• - the read event information is stored in the available memory space of the buffer memory.

According to this embodiment, the read event information is thus read without determining the number of event information stored in the buffer memory with the same cause as The read-in event information is stored in the buffer memory as long as there is sufficient memory space available or free in the buffer memory to store the read-in event information. This configuration allows the step of determining the number to be omitted if memory space is available in the buffer memory, thereby accelerating the method and allowing the available memory space for storing the event information to be fully utilized.

• - die Anzahl von in dem Pufferspeicher gespeicherten Ereignisinformationen, welche dieselbe Ursache für das jeweilige Ereignis aufweisen wie die eingelesene Ereignisinformation, nur dann ermittelt wird, wenn für keine der Ursachen die Anzahl der in dem Pufferspeicher gespeicherten Ereignisinformationen größer als der definierte Schwellenwert ist, und
• - die älteste Ereignisinformation aus dem Pufferspeicher gelöscht wird, für deren Ursache die Anzahl der in dem Pufferspeicher gespeicherten Ereignisinformationen größer als der definierte Schwellenwert ist.

It is also advantageous if, in the step of determining the number, it is first determined whether, for one or more causes, the number of event information stored in the buffer memory is greater than the defined threshold value, where

• - the number of event information items stored in the buffer memory which have the same cause for the respective event as the read-in event information is only determined if the number of event information items stored in the buffer memory is not greater than the defined threshold value for any of the causes, and
• - the oldest event information is deleted from the buffer memory, for which reason the number of event information stored in the buffer memory is greater than the defined threshold.

In other words, to determine the event information to be deleted from the buffer memory, a check is first made to determine whether the number of event information items stored in the buffer memory for one or more causes is greater than the defined threshold, possibly specific to the cause. If the number of event information items stored in the buffer memory is greater than the defined threshold for one or more causes, the oldest event information item for which the number of event information items stored in the buffer memory is greater than the defined threshold is deleted from the buffer memory.

If the number of event information items stored in the buffer memory is not greater than the defined threshold for any of the causes, the number of event information items stored in the buffer memory that have the same cause for the respective event as the read-in event information is determined. The oldest event information item with the same cause is deleted from the buffer memory if the determined number is equal to the defined threshold. If the determined number is less than the defined threshold, the oldest event information item is deleted from the buffer memory.

This configuration ensures that those event information items whose number exceeds the defined threshold are deleted first. For example, if there is available storage space in the buffer memory, a number of event information items with the same cause that is greater than the defined threshold can initially be stored. Only when there is no more storage space available in the buffer memory to store a read-in event information item is the oldest of these event information items deleted. This ensures that, despite the interim storage of a number of event information items with the same cause that is greater than the defined threshold, no event information items with the same cause that are less than or equal to the threshold are deleted.

It is further advantageous if the method comprises a step of reading out the stored event information from at least one of the buffer memories, preferably from the buffer memory, the further buffer memory, and the additional buffer memory, of the storage system. In the reading out step, one or more data records associated with the stored event information can be read out, which records comprise data present at the time the events occurred and/or additional information regarding the cause of the events. The reading out step can be followed by a step of analyzing or diagnosing the computing unit with regard to the occurrence of the events and their causes based on the read out event information. Furthermore, the method can comprise a step of adapting a software and/or hardware module of the computing unit based on the read out event information. This embodiment allows the computing unit to be analyzed particularly reliably, and even rare and/or unexpectedly occurring events can be diagnosed in order to appropriately modify the software and/or hardware of the computing unit.

• - im Schritt des Auslesens ferner die gespeicherte(n) Zählerstandsinformation(en) aus dem weiteren Speicherbereich ausgelesen werden und
• - für die ausgelesenen Ereignisinformationen basierend auf der/den ausgelesenen Zählerstandsinformation(en) jeweils ein Zählerstand ermittelt wird.

It is further advantageous if the method comprises a step of storing counter reading information associated with the and/or the further event information relating to a total number of events of the and/or the further computing unit and/or a total number of events of the and/or the further computing unit with a defined cause for the event in a further memory area of the memory system, wherein

• - in the reading step, the stored meter reading information is read from the further memory area and
• - a meter reading is determined for each of the read event information based on the meter reading information(s) read.

The total number of events of the and/or the further computing unit with a defined cause for the event can be a total number of events of the and/or the further computing unit with a single defined cause or a total number of events of the and/or the further computing unit with a defined group of causes. It is conceivable that the total number of events of the and/or the further computing unit is a total number of deliberately caused events or causes. Alternatively, it is conceivable that the total number of events of the and/or the further computing unit is a total number of events excluding deliberately caused events or causes. The counter reading information can comprise a counter which represents a total number of events of the and/or the further computing unit and/or a total number of events of the and/or the further computing unit with a defined cause for the event.

Preferably, in the step of storing the meter reading information, meter reading information already stored or present in the further memory area of the storage system is overwritten. In the step of reading, the read meter reading information regarding the total number of events of the and/or the further computing unit can be assigned to the most recent event information. Furthermore, the read meter reading information regarding the total number of events of the and/or the further computing unit with this defined cause can be assigned to the most recent event information with a defined, specific cause.

For read-out older event information, the counter reading to be determined or assigned can be determined or assigned based on the read-out counter reading information, in particular by decrementing the corresponding counter represented by the counter reading information. The determined or assigned counter represents a total number of events of the and/or the further computing unit and/or a total number of events of the and/or the further computing unit with a defined cause at the time the event occurred. This configuration allows the storage requirements of the buffer memory to be further reduced and, in addition, the storage requirements for storing the counter reading information in the further memory area of the storage system to be minimized.

It is advantageous if the memory system of the computing unit comprises a further buffer memory in addition to the buffer memory. The further buffer memory is configured to store event information relating to events of the and/or the further computing unit, wherein the event information each includes a cause for the event. The buffer memory is preferably configured to store event information previously stored in the further buffer memory. The memory system can therefore be understood as an at least two-stage system of buffer memories. The further buffer memory is preferably designed as a FIFO buffer memory. This configuration makes it possible to provide a two-stage reset history buffer. This makes it possible to store at least the N most recent event information items, regardless of the cause of the respective event, where N represents the number of event information items that can be stored in the further buffer memory. As a result, the N most recently occurring events and their cause can be read out at any time.

It is also advantageous if the memory system of the computing unit further comprises an additional buffer memory configured to store a data set with data present at the time the event information occurred and/or additional information regarding the cause of the event information. This configuration allows the data and/or information contained in the data set to be additionally stored without having to expand the buffer memory and/or the additional buffer memory by the corresponding storage space.

Embodiments of the invention described above and below can be implemented in hardware and/or software. The implementation can be carried out using a digital storage medium, for example an (E)EPROM, a FLASH memory, a hard disk, or another magnetic or optical memory on which electronically readable control signals are stored that can interact or interact with a programmable hardware component, in particular a programmable logic component, in such a way that the respective method or individual steps thereof are carried out.

According to one embodiment, the computing unit comprises a programmable hardware component, in particular a programmable logic component, or is designed as such.

A programmable hardware component can be formed by a processor, a computer processor (CPU = Central Processing Unit), a graphics processor (GPU = Graphics Processing Unit), a computer, a computer system, an application-specific integrated circuit (ASIC = Application-Specific Integrated Circuit), an integrated circuit (IC = Integrated Circuit), a system on a chip (SOC = System on Chip), a programmable logic element, a field-programmable gate array with a microprocessor (FPGA = Field Programmable Gate Array) or a complex programmable logic device (CPLD = Complex Programmable Logic Device).

Also advantageous is a computer program product or computer program with program code that can be stored on a machine-readable carrier or storage medium such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out, implement and/or control the steps of the method according to one of the embodiments described above, in particular when the program product or program is executed on a computer or a computing unit.

In the following, the invention will be explained in more detail with reference to the drawings.

• 1 eine schematische Darstellung einer Recheneinheit gemäß einer Ausführungsform; und
• 2 ein Ablaufdiagramm eines Verfahrens zum Speichern einer Ereignisinformation bezüglich eines Ereignisses einer Recheneinheit gemäß einer Ausführungsform.

To do this, show

• 1 a schematic representation of a computing unit according to an embodiment; and
• 2 a flowchart of a method for storing event information relating to an event of a computing unit according to one embodiment.

1 shows a computing unit 10 with a microcontroller and/or microprocessor (not shown) and a memory system 12. The memory system 12 comprises a first buffer memory 14, a second buffer memory 16, and a third buffer memory 18. The memory system 12 further comprises a main memory (not shown) with a computer program. The computer program comprises instructions which, when executed by the processor 10, cause the computing unit 10 to store event information relating to events of the computing unit 10 in one or more of the buffer memories 14, 16, 18 or buffers 14, 16, 18 according to the method described below. Alternatively or additionally, the memory system 12 of the computing unit 10 can comprise further such third buffer memories 18.

The computing unit 10 is embodied, for example, as a control unit 10 of a vehicle, in particular an embedded control unit 10. The computing unit 10 is configured to read event information, to be stored in the first buffer memory 14, relating to an event of the computing unit 10 and/or another computing unit, e.g., another control unit or a sensor unit of the vehicle. The read-in event information includes a cause for the event.

Furthermore, the computing unit 10 is configured to determine a storage space available in the first buffer memory 14 for storing the read-in event information. The computing unit 10 is configured to store the read-in event information in the available storage space of the first buffer memory 14 if sufficient storage space is available in the first buffer memory 14 to store the read-in event information.

In addition, the computing unit 10 is configured to determine a number of event information items stored in the second buffer memory 16 which have the same cause for the respective event as the oldest event information item stored in the first buffer memory 14.

In addition, the computing unit 10 is configured to delete the oldest event information with the same cause from the second buffer memory 16 if the determined number is greater than or equal to a defined threshold, and alternatively—that is, if the determined number is less than the defined threshold—to delete the oldest event information from the second buffer memory 16. The causes for the event contained in the oldest event information in the first buffer memory 14 and the event contained in the oldest event information in the second buffer memory 16 can be different from one another.

Furthermore, the computing unit 10 is configured to store the oldest event information from the first buffer memory 14 in the second buffer memory 16. Furthermore, the computing unit 10 is configured to delete the oldest event information from the first buffer memory 14. Furthermore, the computing unit 10 is configured to store the read-in event information in the first buffer memory 14.

According to this embodiment, the first buffer memory 14 is designed as a FIFO (First In First Out) buffer 14. Here, the first buffer memory 14 can be used as a first memory stage 14 of the memory system 12 and the second buffer memory 16 can be used as a second memory stage 16 of the memory storage system 12. The third buffer memory 18 can be understood as a storage stage 18 that is subordinate to the first and second storage stages 14, 16.

The event information stored in the first buffer memory 14 and the second buffer memory 16 is stored chronologically. The most recent event information stored in the second buffer memory 16 is older than the oldest event information stored in the first buffer memory 14. Each event information stored in the second buffer memory 16 is previously stored in the first buffer memory 14.

In the present case, the memory system 12 is used to store reset or restart events of the computing unit 10. In other words, according to this embodiment, a multi-stage reset history buffer is described for storing or saving reset reasons, which occur in particular on embedded systems 10, such as an automotive control unit 10. For this purpose, reset events, along with other relevant data, can be stored in the buffer memories 14, 16, 18 of the memory system 12 using the reset history buffer. The stored information can then be read out, for example, in the event of problems in the field or during a return analysis.

According to the proposed approach, when a reset event occurs, the reset event or event information representing the reset event can be added to the first buffer memory 14 (arrow 20 in 1 ). This entry in the first buffer memory 14 is assigned unique identification information, according to this embodiment an identification number and a cause for the reset event or a reset reason, and stored in the first buffer memory 14 (e.g. PORST for "Power-on Reset", SW_RST for "Software Reset", SMU_RST for "Safety Management Unit (SMU) Reset" or "Hardware Reset"). The identification number is stored in the "No." column of the first buffer memory 14. The cause of the reset event is stored in the "Info" column of the first buffer memory 14.

Furthermore, one or more data records can also be filed or saved, which contain additional information about the occurrence of the reset event. This additional information can include data present at the time of the reset event or at the time the reset event occurred, such as the date, time, mileage of a vehicle containing the computing unit 10, a value of a supply voltage of the computing unit 10, the uptime of the computing unit 10, the power cycle counter, a general or generic reset counter (this counter is incremented by 1 for each unintentional reset) and/or a specific reset counter per reset (a separate counter for each reset reason, which is incremented by 1 when the reset occurs). The data records with the additional information are stored in the "Info+" column of the first buffer memory 14 and correspondingly in the "Info+" column of the second buffer memory 16.

It is conceivable that one or more of these additional pieces of information, e.g., the date and/or time, are not available at the time of the reset event, e.g., because they have not yet been received via an interface such as a CAN bus. Preferably, this initially missing information is subsequently provided after the reset of the computing unit 10 and inserted or stored in the first buffer memory 14. If this is not possible, a default value for the missing information can be entered after a configurable timeout period.

According to an alternative implementation, it is also conceivable for the general and/or specific reset counter to manage only the current counter readings in a separate memory area of the memory system 12 and to also read these counter readings when reading the multi-level reset history buffer. Since all entries in the reset history buffer are output in chronologically correct order, the counter readings at the occurrence of the respective reset event can then be determined based on the current counter readings of the reset counter(s). This can save storage space in the buffer memories 14, 16.

The data record(s) with the additional information regarding the occurrence of the reset event can be added directly to each entry in the first buffer 14 or second buffer 16. Alternatively, the data record(s) can be swapped out to an additional buffer (not shown) or separate buffer of the storage system 12. In this case, the first buffer 14 or second buffer 16 can only reference the entry in this additional buffer or separate buffer (for example, using an index, pointer, etc.). This results in the following when moving stored event information from the first buffer 14 to the second buffer 16 - that is, when storing event information previously stored in the first buffer 14 in the second buffer 16 and deleting the stored event information from the first buffer 14: share with you because less data needs to be moved.

According to the 1 In the embodiment shown, the first buffer memory 14 is configured to store a maximum of ten entries or event information. The maximum number of storable entries or event information can be predetermined or can be predetermined or configured.

If, when entering a new reset event (arrow 20 in 1 ) there are still free memory locations available in the first buffer memory 14, the corresponding event information or the entry is stored in the next free memory location of the first buffer memory 14. However, if all memory locations in the first buffer memory 14 are already occupied, the oldest entry or the oldest event information in the first buffer memory 14 (recognizable by the lowest value in the "No." column) is moved to the second buffer memory 16 (arrow 22 in 1 ), i.e., stored in the second buffer memory 16 and deleted from the first buffer memory 14. This frees up a storage space in the first buffer memory 14, which can then be used for the new entry.

Since the oldest entry or the oldest event information in the first buffer 14 is always moved to the second buffer 16 first, a FIFO concept is in place. Moving the oldest entry to the second buffer 16 prevents this entry from being deleted or removed from the memory system 12 of the processing unit 10.

According to the 1 In the embodiment shown, the second buffer memory 16 has the same data structure as the first buffer memory 14. Thus, the reset events that are transferred or moved from the first buffer memory 14 to the second buffer memory 16 (arrow 22 in 1 ), can be moved very easily without conversion.

If a new entry or new event information comprising the respective cause of the event is to be added from the first buffer memory 14 to the second buffer memory 16, it is first checked how many event information items comprising this cause or how often this reset reason is already present or stored in the second buffer memory 16. In 1 For example, a reset event with the reset reason “PORST” is to be added. The reset reason “PORST” already exists three times in the second buffer memory 16. This means that if the new entry is added, the second buffer memory 16 would contain four entries with the reset reason “PORST”. At this point, the rule applies that a specific reset reason or a defined number X of event information items that have the same cause for the respective event as the event information to be stored may only be present or stored in the second buffer memory a maximum of X times. The value of X is freely selectable, specifiable, or configurable. According to the present embodiment, X takes on the value three, for example. This means that a maximum of three entries or event information items in the second buffer memory 16 may have the same reset reason. Since in the present example three entries are already stored in the second buffer memory 16 with the cause “PROST” and a further entry with the cause “PROST” is to be added, this leads to the oldest entry with the cause “PORST” being deleted from the second buffer memory 16 (arrow 24 in 1 ) is deleted. This can lead to old entries in the second buffer memory 16 being deleted even though free memory locations are still available in the second buffer memory 16.

According to an alternative implementation, it is conceivable to initially use free memory locations in the second buffer memory 16. As soon as no more free memory locations are available in the second buffer memory 16, it is determined or checked for which reset reasons the rule X=3 is violated, and the oldest entry is then removed for these reset reasons. This ensures that at least X entries can be stored for each reset reason, but if free or available memory space is available, additional event information or entries with the same reset reason can also be stored or saved.

When configuring the size of the second buffer memory 16, it is important to ensure that it must provide at least X times the number of all possible types of event information in memory locations. Therefore, if the computing unit 10 is to be able to store 15 different event information items, at X=3, at least 3x15=45 memory locations must be available in the second buffer memory 16.

According to the present embodiment, reset events can occur at the computing unit 10, for which additional information is provided regarding their cause. For example, in the case of a software reset event and/or a hardware reset event, in particular an SMU alarm reset event, further reset details can be determined and/or provided by software of the computing unit 10.

In the case of a software reset, the additional information or details about a triggered reset can be stored or saved by the software. In the case of a hardware reset, in particular the SMU alarm reset, internal registers of the microcontroller included in the computing unit 10 can be read to obtain the additional information about the reason for the reset. This mechanism is a major advantage over other implementations because it offers the possibility of identifying reset reasons during a later error analysis that were not yet considered or were unknown at the time of development of the computing unit 10 and that nevertheless occurred due to an error in the software and/or hardware of the computing unit 10.

If a reset event occurs for which no additional information regarding its cause or no extended reset details are available or provided, the corresponding field in the "Index" column in the first buffer 14 and the second buffer 16 remains empty or is filled with a placeholder value. However, if a reset event occurs for which further reset details can be read, this information is entered or stored in the "Index" column. For a software reset, for example, the value to be entered can simply be a numeric value that states what exactly led to the reset event in the software. 1 In the example shown, two software-based or software-related reset events are stored in the first buffer memory 14. For example, a value of 5 specified in the "Index" column represents a reset requested by the diagnostics, and a value of 6 specified in the "Index" column represents a reset that was required to switch from an application to a bootloader of the computing unit 10.

In the case of more complex reset events, such as the aforementioned SMU alarm reset event, it may also be necessary to store a large amount of reset details or a multitude of additional information regarding the cause of the reset event. In order to avoid having to store large amounts of data in the first buffer 14 or the second buffer 16, and thus to reserve this storage space for the reset details for all event information or entries in the reset buffer, a reference to another memory area of the memory system 12 can be used. In this case, the second buffer 16 contains three entries with the reset reason "SMU_RST." These contain the values 21, 22, and 23 in their "Index" column. These values correspond to the indices in the third buffer 18 of the memory system 12, i.e., a reference to a specific entry in this buffer 18 (arrows 26 in 1 ). Using this mechanism, these large and "unwieldy" reset details can be outsourced to a separate buffer without inflating the entries in the first buffer 14 and the second buffer 16. By completely storing the additional information regarding the cause of the reset events or the reset details in the "Info++" column in the third buffer 18, reset reasons that occur due to a misconfiguration or an error in the microcontroller of the computing unit 10 will later also be visible. This configuration eliminates the need to provide a filter to filter out unexpected events. This reduces the development time of the computing unit 10 because the identification and evaluation of all fundamentally possible events required for implementing the filter can be dispensed with.

According to an alternative embodiment, the additional information regarding the cause of the reset events or the reset details can be pre-filtered before being stored in the storage system 12 in order to store less data in the buffer memories 14, 16, 18 and to save further storage space.

To determine whether the value in the "Index" column of the second buffer 16 is a directly usable value or an index to one or more additional buffers of the storage system 12, the cause or reset reason is used. In the example, a software reset has a directly usable value, while an SMU reset always includes an index to the third buffer 18. In an alternative embodiment, the associated buffer can be derived not from the reset reason, but from the value range of the entry in the "Index" column. For example, values 0-99 could refer to directly usable details for a software reset, values 100-199 could represent an index to buffer A, values 200-299 an index to buffer B, etc.

According to this embodiment, the entries in the third buffer memory 18 are deleted as soon as the associated or corresponding entry referencing this entry in the first buffer memory 14 or in the second buffer memory 16 is deleted.

When configuring the size of the third buffer memory 18, care must be taken to ensure that it has a size which corresponds at least to a sum of a maximum number of events that can be stored in the second buffer memory 16. information with a reference, in particular in the “Index” column, to the third buffer memory 18 and the number of event information stored in the first buffer memory 14.

The event information stored in the memory system 12, in particular the reset reasons or the causes of the reset events, can be very easily read out in chronologically correct order from the first buffer memory 14, the second buffer memory 16, and the third buffer memory 18. This can be done, for example, by a diagnostic service in the field or via a JTAG (Joint Test Action Group) debugger in the factory in the case of a return analysis. Furthermore, those skilled in the art will be aware of other options for reading out this data via various communication channels of the computing unit 10.

2 shows a flowchart of a method for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit in a memory system of the computing unit comprising at least one buffer memory. The computing unit can, for example, be the computing unit 10 according to 1 The method as a whole is designated by reference numeral 100.

The method 100 is preferably executed during operation of the computing unit 10. In particular, the method 100 is repeatedly executed in the described sequence of method steps, preferably always when an event of the computing unit 10 occurs or occurs.

In step 110, event information relating to an event of the computing unit 10 and/or another computing unit is read in to the first buffer memory 14. The read-in event information includes a cause for the event.

In step 120, a storage space available in the first buffer memory 14 for storing the read-in event information is determined.

If sufficient storage space is available in the first buffer memory 14 to store the read-in event information, the read-in event information is stored in the available storage space of the first buffer memory 14 in a step 190. t.

If there is insufficient storage space available in the first buffer memory 14 to store the read-in event information, the oldest event information stored in the first buffer memory 14 is read in in step 130. In step 140, a number of event information items stored in the second buffer memory 16 are determined which have the same cause for the respective event as the oldest event information stored in the first buffer memory 14.

In step 150, the oldest event information with the same cause is deleted from the second buffer memory 16 if or when or exactly in the case in which the determined number is greater than or equal to a defined threshold value.

Alternatively, that is, if or when or precisely in the case in which the determined number is smaller than the defined threshold value, the oldest event information is deleted from the second buffer memory 16 in step 160. In this case, the causes for the event in the first buffer memory 14 comprised by the oldest event information and for the event in the second buffer memory 16 comprised by the oldest event information may be different from one another.

In step 170, the oldest event information from the first buffer memory 14 is stored in the second buffer memory 18.

In step 180, the oldest event information is deleted from the first buffer memory 14.

In step 190, the event information read in step 110 is stored in the first buffer memory 14.

QUOTES CONTAINED IN THE DESCRIPTION

This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2009/0222700 A1 [0007]

Claims (14)

A method (100) for storing event information relating to an event, in particular a hardware and/or software error and/or a reset event, of a computing unit (10) and/or another computing unit in a memory system (12) of the computing unit (10) having at least one buffer memory (16), comprising the following steps: - reading (130) event information relating to an event of the computing unit (10) and/or the other computing unit to be stored in the buffer memory (16), wherein the read-in event information comprises a cause for the event; - determining (140) a number of event information items stored in the buffer memory (16) that have the same cause for the respective event as the read-in event information, wherein: ◯ the oldest event information with the same cause is deleted from the buffer memory (16) if the determined number is greater than or equal to a defined threshold; otherwise: ◯ the oldest event information is deleted from the buffer memory (16); and - storing (170) the read-in event information in the buffer memory (16). Procedure (100) according to Claim 1 , characterized in that the read-in event information is read from a further buffer memory (14) of the storage system (12), wherein the read-in event information is one, in particular the oldest, item of event information stored in the further buffer memory (14). Procedure (100) according to Claim 2 , characterized in that - the step of determining (140) the number is preceded by a step of reading in (110) further event information to be stored in the further buffer memory (14) relating to a further event of the computing unit (10) and/or the further computing unit, wherein the further event information comprises a cause for the further event, and - the step of storing (170) the read-in event information in the buffer memory (16) is followed by a step of deleting (180) the, in particular oldest, event information stored in the further buffer memory (14) and a step of storing (190) the read-in further event information in the further buffer memory (14). Method (100) according to one of the preceding claims, characterized in that a data record is assigned to the and/or the further event information, wherein the data record comprises data present at the occurrence of the and/or the further event and/or supplementary information regarding the cause of the or the further event and is stored together with the and/or the further event information in the respective buffer memory (14, 16). Method (100) according to one of the Claims 1 until 3 , characterized in that a data record is assigned to the and/or the further event information, wherein the data record comprises data present at the occurrence of the and/or the further event and/or additional information regarding the cause of the or the further event and is stored in an additional buffer memory (18) of the storage system (12), wherein the and/or the further stored event information comprises a variable referring to the stored data record. Method (100) according to one of the preceding claims, characterized by a step of determining the storage space available in the buffer memory (16) for storing the read-in event information, wherein - the number of event information items stored in the buffer memory (16) with the same cause as the read-in event information is only determined if no storage space is available in the buffer memory (16) for storing the read-in event information, otherwise - the read-in event information is stored in the available storage space of the buffer memory (16). Method (100) according to one of the preceding claims, characterized in that in the step of determining (140) the number it is first determined whether for one or more causes the number of event information items stored in the buffer memory (16) is greater than the defined threshold value, wherein - the number of event information items stored in the buffer memory (16) which have the same cause for the respective event as the read-in event information is only determined if for none of the causes the number of event information items stored in the buffer memory (16) is greater than the defined threshold value, and - the oldest event information item is deleted from the buffer memory (16) for the cause of which the number of event information items stored in the buffer memory (16) is greater than the defined threshold value. Method (100) according to one of the preceding claims, characterized by a step of reading out the stored event information tions from at least one of the buffer memories (14, 16, 18) of the storage system (12). Procedure (100) according to Claim 8 , characterized by a step of storing counter reading information associated with the and/or the further event information with regard to a total number of events of the computing unit (10) and/or the further computing unit and/or a total number of events of the computing unit (10) and/or the further computing unit with a defined cause for the event in a further memory area of the memory system, wherein - in the reading step, the stored counter reading information(s) are further read out from the further memory area and - a counter reading is determined for each of the read out event information based on the read out counter reading information(s). Computing unit (10), in particular a control unit, with a memory system (12) comprising a buffer memory (16), wherein - the buffer memory (16) is configured to store event information relating to events of the computing unit (10) and/or another computing unit, wherein the event information each comprises a cause for the event, and - the computing unit (10) is configured to carry out the method (100) according to one of the preceding claims. Computing unit (10) according to Claim 10 , characterized by a further buffer memory (14) which is arranged to store event information relating to events of the computing unit (10) and/or the further computing unit, wherein the event information each comprises a cause for the event. Computing unit (10) according to Claim 10 or 11 , characterized by an additional buffer memory (18) which is arranged to store a data set with data present at the time of occurrence of the event information and/or additional information regarding the cause of the event information. Computer program comprising instructions which, when executed by a computing unit (10) according to one of the Claims 10 until 12 cause them to carry out the procedure (100) according to one of the Claims 1 until 9 to execute and/or control. Machine-readable storage medium on which the computer program according to Claim 13 is stored.

Images