Error positioning method and system on chip

By deploying a monitoring module in the on-chip system to filter and match transaction data packets, the problem of error location in complex on-chip networks is solved, and testing efficiency and flexibility are improved.

CN119088601APending Publication Date: 2024-12-06PHYTIUM TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411052386.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-31
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, in silicon pre-silicon test and post-silicon test, it is difficult to effectively locate the wrong position for chips in complex on-chip networks, resulting in difficult time cost and low efficiency.

Method used

Deploy the monitoring module in the on-chip system, and perform preset filtering and matching processing of transaction data packets, reduce the amount of transactions, narrow the range of error positioning, and improve the efficiency of error positioning.

Benefits of technology

It effectively reduces the time of mislocalization, improves analysis efficiency, and improves the flexibility of transaction data screening.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088601A_ABST
    Figure CN119088601A_ABST
Patent Text Reader

Abstract

The invention provides an error positioning method and a system on chip, and the method comprises the steps: receiving a transaction data message from a communication node or an on-chip module through a monitoring module disposed between communication nodes or between the on-chip module and the communication node, and filtering and / or matching the transaction data message based on a preset filtering condition and a preset matching condition to obtain a to-be-selected transaction data message, and carrying out error positioning on the system on chip according to the to-be-selected transaction data message to obtain an error positioning result. Through the monitoring module, the amount of data needing to be analyzed in the system-on-chip with high complexity can be reduced, and the efficiency of subsequent error positioning is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of chip technology, and in particular to an error location method and a system on chip. Background Art

[0002] As the complexity of on-chip networks continues to increase, pre-silicon testing and post-silicon testing have become particularly important. Pre-silicon testing and post-silicon testing can locate problems in on-chip networks, thereby improving the correctness and reliability of chips.

[0003] In the prior art, pre-silicon testing generally exposes problems by constructing stimuli, and then uses specific software to track signals to locate the source of the error. However, this process is time-consuming for chips with highly complex on-chip networks. Post-silicon testing generally involves debugging and defect tracking through debugging test software. Although this can meet most debugging and testing requirements, for increasingly complex on-chip networks and imperfect error reporting mechanisms, problems exposed in post-silicon testing are more difficult to track and locate than those exposed in pre-silicon testing, and the time cost is also more difficult to estimate. Summary of the invention

[0004] The purpose of the present application is to provide an error location method and an on-chip system to address the deficiencies in the above-mentioned prior art, so as to solve the problem in the prior art that it is difficult to locate the error position when performing pre-silicon testing and post-silicon testing on chips with high on-chip network complexity.

[0005] To achieve the above objectives, the technical solutions adopted in this application are as follows:

[0006] In a first aspect, the present application provides an error location method, which is applied to a system on chip, wherein the system on chip includes: multiple on-chip modules, an on-chip network, and multiple monitoring modules, wherein the on-chip network includes multiple communication nodes, and each of the monitoring modules is deployed between the communication nodes or between the on-chip module and the communication node, and the method includes:

[0007] The monitoring module receives a transaction data message from the communication node or the on-chip module;

[0008] The monitoring module filters and / or matches the transaction data message based on a preset filtering condition and a preset matching condition to obtain a transaction data message to be selected;

[0009] The system on chip is error located according to the transaction data message to be selected to obtain an error location result.

[0010] As a possible implementation manner, the monitoring module includes: a filtering unit, a matching unit and a recording unit, and an output end of the filtering unit is connected to an input end of the matching unit and the recording unit respectively.

[0011] As a possible implementation manner, filtering and / or matching the transaction data message based on a preset filtering condition and a preset matching condition to obtain a data message to be selected includes:

[0012] The filtering unit filters the transaction data message based on the filtering condition to obtain a filtered transaction data message, and sends the filtered transaction message to the matching unit;

[0013] The matching unit matches the filtered transaction data message based on the matching condition to obtain a transaction data message to be selected, and sends the transaction data message to be selected to the recording unit;

[0014] The recording unit records the transaction data message to be selected.

[0015] As a possible implementation manner, filtering and / or matching the transaction data message based on a preset filtering condition and a preset matching condition to obtain a transaction data message to be selected includes:

[0016] The matching unit matches the transaction data message based on the matching condition to obtain a transaction data message to be selected, and sends the transaction data message to be selected to the recording unit;

[0017] The recording unit records the transaction data message to be selected.

[0018] As a possible implementation manner, the transaction data message includes: a first channel data message and a plurality of second channel data messages;

[0019] The filtering unit filters the transaction data message based on the filtering condition to obtain a filtered transaction data message, including:

[0020] The filtering unit filters the first channel data message based on the filtering condition to obtain a filtered first channel data message;

[0021] The matching unit matches the filtered transaction data message based on the matching condition to obtain the transaction data message to be selected, including:

[0022] The matching unit matches the filtered first channel data message based on the matching condition to obtain a first channel data message to be selected;

[0023] The recording unit recording the transaction data message to be selected includes:

[0024] The recording unit records the first channel data message to be selected and a plurality of second channel data messages corresponding to the first channel data message to be selected.

[0025] As a possible implementation, the filtering conditions include: filtering digits, filtering masks;

[0026] The filtering the transaction data message based on the filtering condition to obtain the filtered transaction data message includes:

[0027] Determining the data to be compared in the transaction data message according to the filter digit and a preset first comparison length;

[0028] The data to be compared in the transaction data message is filtered according to the filter mask and the preset filter content to obtain the filtered transaction data message.

[0029] As a possible implementation manner, filtering the compared digits of the transaction data message according to the filter mask to obtain the filtered transaction data message includes:

[0030] Based on the number of bits of the filter mask, the bits in the data to be compared that are the same as the number of bits of the filter mask are masked to obtain filtered data, and the filtered data is filtered based on the filtering content to obtain the filtered transaction data message.

[0031] As a possible implementation, the matching condition includes: matching digits, matching masks;

[0032] The step of matching the filtered transaction data message based on the matching condition to obtain the transaction data message to be selected includes:

[0033] Determining the data to be compared of the filtered transaction data message according to the matching digits and a preset second comparison length;

[0034] The data to be compared of the filtered transaction data message is matched according to the matching mask and the preset matching content to obtain the transaction data message to be selected.

[0035] As a possible implementation manner, matching the data to be compared of the filtered transaction data message according to the matching mask to obtain the transaction data message to be selected includes:

[0036] Based on the number of bits of the matching mask, the digits in the data to be compared of the filtered transaction data message that are the same as the number of bits of the matching mask are retained to obtain matched data, and the matched data are matched based on the preset matching content to obtain the transaction data message to be selected.

[0037] As a possible implementation manner, the method further includes:

[0038] The recording unit counts the transactions to be selected corresponding to the recorded transaction data messages to be selected, and records the time information, transaction completion status, address information and data of the transactions to be selected.

[0039] As a possible implementation manner, the method further includes:

[0040] If a preset monitoring stop condition is met, the monitoring of the on-chip system is terminated, and the monitoring stop condition includes at least one of the following: the count value reaches a first preset threshold, the transaction record is completed, and the memory of the buffer is less than a second preset threshold.

[0041] In a second aspect, the present application provides a system on chip, comprising multiple on-chip modules, an on-chip network and multiple monitoring modules, wherein the on-chip network includes multiple communication nodes, each of the monitoring modules is deployed between the communication nodes or between the on-chip module and the communication node, and the on-chip system is used to execute the steps of the error location method described in the first aspect above.

[0042] The beneficial effects of the present application are as follows: by deploying a monitoring module on a transmission path that needs to be monitored, the transaction data packets on the transmission path are monitored during the debugging and testing phase, and the transaction data packets are filtered and / or matched based on preset filtering conditions and matching conditions, thereby filtering out a large number of irrelevant transactions, reducing the time spent on analyzing the huge amount of transactions in a complex NOC, fuzzily locating transactions that may be erroneous, narrowing the scope of error location, and the final data contains fewer transactions, thereby effectively reducing the amount of data that needs to be analyzed and improving the efficiency of error location. At the same time, the filtering conditions and matching conditions in the present application can be set by debugging and testing personnel based on actual needs, thereby also improving the flexibility of transaction data packet screening. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A schematic diagram of the architecture of a system on a chip provided by an embodiment of the present application is shown;

[0045] Figure 2 A schematic diagram of a flow chart of an error location method provided in an embodiment of the present application is shown;

[0046] Figure 3 A schematic diagram of a structure in which a monitoring module is provided between a CPU and an RP routing node of a NOC and between an RP routing node and a DDR module, provided in an embodiment of the present application;

[0047] Figure 4 A schematic diagram of the structure of a monitoring module provided in an embodiment of the present application is shown;

[0048] Figure 5 A flowchart of filtering and matching transaction data messages provided by an embodiment of the present application is shown;

[0049] Figure 6 A flowchart of matching transaction data messages provided by an embodiment of the present application is shown;

[0050] Figure 7 A schematic diagram of a filtering process provided by an embodiment of the present application is shown;

[0051] Figure 8 A schematic diagram of a matching process provided by an embodiment of the present application is shown;

[0052] Fig. 9 A schematic diagram of the overall process of performing message monitoring by a monitoring module provided in an embodiment of the present application is shown;

[0053] Fig.10 A schematic diagram of the structure of a computer device 100 provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0054] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0055] Pre-silicon testing generally exposes problems by constructing stimuli, and then uses specific software to track signals to locate the source of the error. However, this process is time-consuming for chips with highly complex on-chip networks. Post-silicon testing generally uses debugging and defect tracking through debugging test software. Although it can meet most debugging and testing requirements, for increasingly complex on-chip networks and imperfect error reporting mechanisms, the problems exposed in post-silicon testing are more difficult to track and locate than those exposed in pre-silicon testing, and the time cost is also more difficult to estimate.

[0056] It can be seen that in the process of SoC (System on Chip) development, how to efficiently and accurately verify the functions and performance of large SoC systems is a difficult problem that needs to be solved urgently. For such problems, we can start with the bus protocol, and obtain more chip internal status information by analyzing the SoC on-chip network protocol, improve chip observability, and then efficiently locate, analyze and handle the problem.

[0057] The CHI (Coherent Hub Interface) protocol is a widely used on-chip protocol specification, which includes a protocol layer, a network layer, and a link layer. The operation status of multiple components interconnected by the on-chip network can be understood from the operation of the CHI protocol. In the prior art, errors can be located by tracking CHI messages, but this method requires a lot of time and has the problem of low positioning efficiency for complex on-chip networks.

[0058] Based on this, the present application proposes an error location method, which adds a CHI transaction monitoring module to some important Noc network nodes in the early design stage, and filters, matches and records the transaction data messages based on the CHI transaction monitoring module. The monitored data messages can be used for analysis and comparison during the debugging and testing process to quickly locate the source of the error.

[0059] Figure 1 This is a schematic diagram of a system-on-chip architecture given in this application, refer to Figure 1 The on-chip system includes: multiple on-chip modules, an on-chip network and multiple monitoring modules. The on-chip network includes multiple communication nodes, and each monitoring module is deployed between the communication nodes or between the on-chip module and the communication node.

[0060] Continue to refer to Figure 1The on-chip network includes multiple communication nodes, such as RP (Route Point), and the on-chip module can be a module in the chip, an interface or a node connected to the on-chip network, such as CDB (Contract Asynchronous Bridge), DDR (Double Data Rate), C2C (Chip-Chip Link Interface), CHA (Cache Home Agent), IMB (IO Master Bridge), etc.

[0061] Next, the error location method of this application is described. Figure 2 This is a flowchart of an error location method provided in this application. This method can be applied to Figure 1 In the system-on-chip shown in Figure 2 As shown, the method includes:

[0062] S201. A monitoring module receives a transaction data message from a communication node or an on-chip module.

[0063] The transaction data message may be a data message of a CHI transaction. The on-chip network communicates based on the CHI protocol, and the monitoring module may be a hardware device deployed between any hardware module and a Noc network supporting the CHI protocol, specifically, it may be deployed between communication nodes of the Noc network or between communication nodes of the Noc network and an on-chip module, and is used to monitor and record the data messages of the CHI transaction on the transmission path where the monitoring module is located during the debugging process.

[0064] It is worth noting that the specific deployment location of the monitoring module can be added in the early stage of chip design according to design requirements. A monitoring module can be set between any hardware module that needs to monitor transaction data messages and the Noc network that supports the CHI protocol to obtain transaction data messages.

[0065] Exemplarily, the monitoring module may exist between the CLUSTER interface of the CPU (Central Processing Unit) and the RP routing node of the Noc. Since the CPU needs to pass through the Noc network to access each module, the monitoring module at this position can monitor and record the CHI transactions of the CPU accessing the Noc network; the monitoring module may also exist between the CHI interface of the DDR and the RP routing node of the Noc to monitor the CHI transactions of the interaction between the DDR and the Noc; if the chip is a multi-core CPU, a monitoring module can also be constructed between the RP node and the C2C node between the cores to monitor the data packets of the data exchange between the cores.

[0066] Reference Figure 3 , is a structural diagram of a monitoring module set between the CPU and the RP routing node of the NOC and between the RP routing node and the DDR module. If an error occurs when the CPU reads the DDR data, the root cause of the error can be traced based on the monitoring module between the CPU and the RP node and the monitoring module between the RP node and the DDR. Specifically, the interaction process of the CPU in the process of reading DDR data includes: the CPU initiates a transaction by sending a read request, and the transaction is transmitted to the routing node of the Noc network in the form of a request transaction packet based on the CHI protocol. After receiving the request, the Noc network will generate a request to access the DDR. After receiving the request, the DDR may return a response signal and then transmit the data to the Noc network, or it may directly transmit the data to the Noc network. After receiving the data forwarded by the Noc network, the CPU will return a confirmation signal to the NoC network. At this point, the entire data reading operation is completed. Figure 3 The monitoring module in can monitor all CHI messages exchanged between the CPU and DDR in each step of the above process.

[0067] S202: The monitoring module filters and / or performs matching processing on the transaction data message based on a preset filtering condition and a preset matching condition to obtain a transaction data message to be selected.

[0068] The monitoring module can filter and / or match the monitored transaction data messages to obtain the selected transaction data messages, so as to reduce the number of messages that need to be analyzed, narrow the analysis scope, and improve the efficiency of subsequent error location.

[0069] The monitoring module can monitor the message during the transaction process, and filter and / or match the message to obtain the transaction data message to be selected. It can also filter and / or match all the acquired transaction data messages after the transaction is completed to obtain the transaction data message to be selected, or save all the monitored transaction data messages first, and filter and / or match all the transaction data messages when it is necessary to locate the error to obtain the transaction data message to be selected. The specific execution time of filtering and matching is not limited in this application.

[0070] The filtering conditions and matching conditions can be pre-configured or modified during the process of locating the error. For example, in the actual application process, after obtaining the transaction data message to be selected, if the number of the transaction data message to be selected is still large, the tester can set the filtering conditions and matching conditions multiple times to continuously narrow the error range and quickly and accurately locate the error location.

[0071] In the first implementation, the monitoring module can filter the transaction data message based on the pre-configured filtering conditions to obtain the selected transaction data message. For example, the message filtering can be performed based on the transaction ID, destination ID, source ID, monitoring attributes, etc. of the transaction data message to filter out irrelevant transactions or irrelevant content in the transaction. Among them, the transaction ID is used to indicate the transaction code related to the transaction data message, the destination ID is used to indicate the target node of the transaction data message, the source ID can indicate the source node of the transaction data message, and the monitoring attribute is used to indicate whether the transaction data message can be monitored.

[0072] In the second implementation, the monitoring module can match the transaction data message based on the pre-configured matching conditions to obtain the selected transaction data message. For example, the transaction data message can be matched based on the request operation code, transaction ID, etc. to determine the selected transaction data message that meets the matching conditions. The request operation code is used to indicate the transaction type, including different types of read requests, write requests, or no data requests.

[0073] In the third implementation, the monitoring module may first filter the transaction data message based on the pre-configured filtering conditions to obtain the filtered transaction data message, and then match the filtered transaction data message based on the pre-configured matching conditions to obtain the transaction data message to be selected.

[0074] S203: Perform error location on the system on chip according to the transaction data message to be selected to obtain an error location result.

[0075] The error location result is used to indicate the location where the error occurred, the cause of the error, etc.

[0076] During the debugging and testing process, the transaction data messages can be recorded by the monitoring module. When errors occur in pre-silicon testing and post-silicon testing, the monitoring module can execute the above-mentioned S202 step to filter and / or match the recorded transaction data messages to obtain selected transaction data messages. When locating the error, the selected transaction data messages can be read from the monitoring module through the software program in the on-chip system, and analyzed based on the selected transaction data messages to obtain the location where the error occurred.

[0077] In an embodiment of the present application, a monitoring module deployed between communication nodes or between an on-chip module and a communication node receives transaction data messages from a communication node or an on-chip module, filters and / or matches the transaction data messages based on preset filtering conditions and preset matching conditions to obtain selected transaction data messages, and locates errors in the on-chip system based on the selected transaction data messages to obtain error location results.

[0078] By deploying a monitoring module on the transmission path that needs to be monitored, the transaction data messages on the transmission path are monitored during the debugging and testing phase, and the transaction data messages are filtered and / or matched based on preset filtering conditions and matching conditions, thereby filtering out a large number of irrelevant transactions, reducing the time spent on analyzing the huge amount of transactions in complex Nocs, fuzzily locating transactions that may be erroneous, narrowing the scope of error positioning, and the data finally obtained contains fewer transactions, thereby effectively reducing the amount of data that needs to be analyzed and improving the efficiency of error positioning. At the same time, the filtering conditions and matching conditions in the present application can be set by the debugging and testing personnel based on actual needs, thereby also improving the flexibility of transaction data message screening.

[0079] Next, the specific structure of the monitoring module is described. Figure 4 As shown, the monitoring module includes: a filtering unit, a matching unit and a recording unit, and the output end of the filtering unit is connected to the input end of the matching unit and the recording unit respectively.

[0080] Among them, the filtering unit is used to filter the transaction data message or the message content of the transaction data message based on the filtering condition, the matching unit is used to match the transaction data message based on the matching condition, and the recording unit is used to record the selected transaction data message obtained after filtering and / or matching.

[0081] The output end of the filtering unit is connected to the input end of the matching unit, the output end of the filtering unit is also connected to the input end of the recording unit, the output end of the matching unit is connected to the input end of the recording unit, and the input end of the filtering unit is used to receive the transaction data message and the input timestamp. It should be understood that if only the transaction data message is filtered, the filtering unit can send the selected transaction data message to the recording unit for recording after completing the filtering, and if the transaction data message is filtered and matched, the filtering unit and the matching unit can send the selected transaction data message to the recording unit for recording after completing the filtering and matching in turn.

[0082] Continue to refer to Figure 4 The monitoring module may also include at least one register, which is connected to the recording unit and used to record the configuration information of the filtering unit and the matching unit, and store the transaction data message to be selected and the transaction package information of the transaction data message to be selected recorded by the recording unit. The recording unit may also count the transactions related to the transaction data message to be selected. The transaction package information of the transaction data message to be selected includes at least: the start timestamp of the transaction, the end timestamp of the transaction, the completion status of the transaction, the address and data, etc.

[0083] As a possible implementation, the recording unit may classify and store completed transactions and uncompleted transactions. After receiving the transaction data message to be selected, the recording unit may determine the transaction completion status of the transaction data message to be selected. If the transaction completion status of the transaction data message to be selected is completed, the transaction data message to be selected may be stored in the first register; if the transaction completion status of the transaction data message to be selected is uncompleted, the transaction data message to be selected may be stored in the second register.

[0084] Next, combine Figure 4 and Figure 5 The process of filtering and / or matching the transaction data message based on the preset filtering condition and the preset matching condition to obtain the transaction data message to be selected is described. The step S202 includes:

[0085] S501: A filtering unit filters a transaction data message based on a filtering condition to obtain a filtered transaction data message, and sends the filtered transaction message to a matching unit.

[0086] S502: The matching unit matches the filtered transaction data messages based on the matching condition to obtain the transaction data messages to be selected, and sends the transaction data messages to be selected to the recording unit.

[0087] S503: The recording unit records the transaction data message to be selected.

[0088] The filter unit and the matching unit may be pre-configured in the monitoring module. If the configuration information indicates that the filter unit is enabled, after receiving the transaction data message, the filter unit may filter the transaction data message based on the filter condition.

[0089] The filtering conditions may include transaction types, transaction IDs, and addresses that do not need to be recorded. After the filtering unit filters the transaction data message based on the filtering conditions, if the configuration information indicates that the matching unit is also enabled, the filtering unit may send the filtered transaction message to the matching unit, and the matching unit may match the filtered transaction message based on the matching conditions, and use the filtered transaction data message that meets the matching conditions as the transaction data message to be selected.

[0090] When the matching unit sends the data message to be selected to the recording unit, the recording unit can record the transaction data message to be selected and the transaction package information related to the transaction data message to be selected, including the transaction start and end timestamps, transaction completion status, address and data, and add one to the count value of the current transaction data message to be selected.

[0091] As another possible implementation, the monitoring module may also only match the transaction data message. In this case, in the above step S202, the transaction data message is filtered and / or matched based on the preset filtering condition and the preset matching condition to obtain the process of the transaction data message to be selected, such as Figure 6 As shown, including:

[0092] S601: A matching unit matches transaction data messages based on matching conditions to obtain transaction data messages to be selected, and sends the transaction data messages to be selected to a recording unit.

[0093] S602: The recording unit records the transaction data message to be selected.

[0094] After the monitoring module receives the transaction data message, if the configuration information indicates that the filtering unit is not enabled and the matching unit is enabled, the filtering unit can transparently transmit the transaction data message to the matching unit, and the matching unit matches the transaction data message based on the matching conditions to determine the candidate transaction data message that meets the matching conditions.

[0095] After receiving the transaction data message to be selected, the recording unit can record the transaction data message to be selected and the transaction package information related to the transaction data message to be selected, including the transaction start and end timestamps, transaction completion status, address and data, and add one to the current transaction count value.

[0096] Furthermore, the transaction data message includes: a first channel data message and a plurality of second channel data messages.

[0097] The transaction data message can be transmitted through multiple transmission channels, and each transmission channel is used to transmit the request, response and read data of the transaction data message respectively.

[0098] Among them, the first channel data message may be a transaction data message transmitted in the first transmission channel, and the second channel data message may be a transaction data message transmitted in other transmission channels except the first transmission channel. The transmission channels include TXREQ (transmit request channel), TXRSP (transmit response channel), TXDAT (transmit data channel), RXRSP (receive response channel), and RXDAT (receive data channel). Exemplarily, the first transmission channel may be TXREQ, the first channel data message may be a transaction data message transmitted in the TXREQ channel, and the second channel data message may be a transaction data message transmitted in the TXRSP, TXDAT, RXRSP, and RXDAT channels.

[0099] Continue to refer to Figure 4 The filtering unit can receive the transaction data message and timestamp of the TXREQ channel, the recording unit receives the transaction data message and timestamp of the TXRSP, TXDAT, RXRSP, and RXDAT channels, and determines whether to record based on the selected transaction data message.

[0100] In the present application, the monitoring module can perform the above steps S501-S502 or S601 on the first channel data message to obtain the selected transaction data message after filtering and / or matching the first channel data message. At this time, the above step of filtering the transaction data message based on the filtering condition by the filtering unit to obtain the filtered transaction data message includes:

[0101] The filtering unit filters the first channel data message based on the filtering condition to obtain the filtered first channel data message.

[0102] The above-mentioned step of matching the filtered transaction data message based on the matching condition by the matching unit to obtain the transaction data message to be selected includes:

[0103] The matching unit matches the filtered first channel data message based on the matching condition to obtain the first channel data message to be selected.

[0104] The above-mentioned step of recording the transaction data message to be selected by the recording unit includes:

[0105] The recording unit records the first channel data message to be selected and a plurality of second channel data messages corresponding to the first channel data message to be selected.

[0106] Optionally, the first transmission channel may be connected to a filtering unit of the monitoring module, and the remaining transmission channels except the first transmission channel may be connected to a recording unit of the monitoring module.

[0107] After the first channel data message in the first transmission channel arrives at the filtering unit, if the filtering unit is enabled, the filtering unit filters the first channel data message based on the filtering condition to obtain the filtered first channel data message, and sends the filtered first channel data message to the matching unit; if the filtering unit is not enabled, the filtering unit directly sends the first channel data message to the matching unit.

[0108] After the filtered first channel data message or the first channel data message reaches the matching unit, if the matching unit is enabled, the matching unit matches the filtered first channel data message or the first channel data message based on the matching condition to obtain the first channel data message to be selected.

[0109] Optionally, the second channel data message corresponding to the first channel data message to be selected may have the same transaction ID as the first channel data message to be selected, so that fast recording of transaction data messages of the same transaction flow can be achieved.

[0110] When the recording unit records the first channel data message to be selected and the corresponding second channel data message, the first channel data message to be selected and the corresponding second channel data message can be merged into a transaction package, and the transaction information of the transaction package can be recorded, including the timestamps of the start and end of the transaction, the transaction completion status, the address and data, etc., and the current count value is increased by one.

[0111] The following is a further description of the above filtering based on the filtering conditions, the filtering conditions include: filtering digits, filtering masks. Specifically, the above filtering based on the filtering conditions to obtain the filtered transaction data message process, such as Figure 7 As shown, including:

[0112] S701: Determine the data to be compared in the transaction data message according to the filtering digit and a preset first comparison length.

[0113] Among them, the digit in the transaction data message refers to the data position in the transaction data message. For example, assuming that the digits are numbered from 0, the digit 1 in the transaction data message refers to the second position in the transaction data message. Correspondingly, the filter digit refers to the starting position of a part of the data that needs to be filtered in the transaction data message. By setting the value of the filter digit, it can be determined from which bit of the transaction data message to start comparing the data when filtering, and by setting the first comparison length, it can be determined how many bits of data to compare. Exemplarily, assuming that the filter digit is 3, assuming that a total of 64 bits of data are compared, the compared data is the part of the data from the 4th to the 68th bit of the transaction data message.

[0114] S702: Filter the data to be compared in the transaction data message according to the filter mask and the preset filter content to obtain a filtered transaction data message.

[0115] The filter mask can be used to determine the specific digits to be compared. For example, if a transaction ID needs to be compared, the filter mask can be used to shield the content that does not need to be compared, and only the digital content of the transaction ID that needs to be compared is retained.

[0116] The filtering content may be a preset filtering condition, such as a transaction ID, a destination ID, etc. By comparing the filtering content with the transaction data message, the transaction data message that is the same as the filtering content may be filtered out.

[0117] By setting the filtering digits and filtering masks, the digits that need to be compared in the filtering process can be determined in a refined manner, thereby improving the efficiency of the filtering process.

[0118] Furthermore, in the above step S702, the process of filtering the to-be-compared data of the transaction data message according to the filtering mask and the filtering content to obtain the filtered transaction data message includes:

[0119] Based on the number of digits of the filter mask, the digits in the data to be compared that are the same as the number of digits of the filter mask are masked to obtain filtered data, and the filtered data is filtered based on the filtering content to obtain a filtered transaction data message.

[0120] As an example, the specific content that needs to be compared is the transaction ID, and the transaction ID is from the 0th to the 11th digits. Then a filter mask can be set for the remaining digits, and the content from the 0th to the 11th digits is retained as the filtered data and compared with the filtered content. If the content from the 0th to the 11th digits is the same as the filtered content, the transaction data message is not retained. If the content from the 0th to the 11th digits is different from the filtered content, the filtered transaction data message is retained.

[0121] The following is a further description of the above matching based on matching conditions, the matching conditions include: matching digits, matching masks. The above steps of matching the filtered transaction data message based on the matching conditions to obtain the transaction data message to be selected are as follows: Figure 8 As shown, including:

[0122] S801. Determine the data to be compared of the filtered transaction data message according to the matching digits and a preset second comparison length.

[0123] The matching digit refers to the starting position of a portion of data that needs to be matched in the filtered transaction data message. By setting the value of the matching digit, it can be determined from which bit of the filtered transaction data message the data is to be matched when matching, and by setting the second comparison length, it can be determined how many bits of data are to be compared. For example, assuming that the matching digit is 2 and a total of 64 bits of data need to be compared, the data that needs to be compared is the portion of data from the 3rd bit to the 67th bit.

[0124] S802: Match the data to be compared of the filtered transaction data message according to the matching mask and the preset matching content to obtain the transaction data message to be selected.

[0125] The match mask is used to determine the specific content to be compared when matching. For example, if you need to match the destination ID, you can use the match mask to mask the remaining digits outside the destination ID, retaining only the content of the destination ID as the data to be compared, and compare it with the matching content. If the match is successful, the filtered transaction data message is retained to obtain the selected transaction data message.

[0126] The matching content may be information such as a destination ID and a transaction ID. By setting the matching content, the filtered transaction data message that is the same as the matching content may be retained.

[0127] By setting the matching digits and the matching mask, the digits that need to be matched and the specific content that needs to be matched can be determined in detail, thereby improving the efficiency of the matching process.

[0128] Furthermore, the process of matching the data to be compared of the filtered transaction data message according to the matching mask to obtain the transaction data message to be selected includes:

[0129] Based on the number of bits of the matching mask, the bits in the data to be compared of the filtered transaction data message that are the same as the number of bits of the matching mask are retained to obtain matched data, and the matched data are matched based on the preset matching content to obtain the transaction data message to be selected.

[0130] The data to be compared is matched based on the matching content, and it can be determined whether the content of each digit in the data to be compared is the same as the matching content. If they are the same, the filtered transaction data message is used as a transaction data message to be selected.

[0131] It is worth noting that if the transaction data message is not filtered by the filtering unit, but is directly passed by the filtering unit to the matching unit for matching, the matching unit can match the transaction data message based on the matching digit and the matching mask, and determine the transaction data message with the same matching content as the selected transaction data message.

[0132] After obtaining the transaction data message to be selected, the method of the present application further includes:

[0133] The recording unit counts the pending transactions corresponding to the recorded pending transaction data messages, and records the time information, transaction completion status, address information and data of the pending transactions.

[0134] The recording unit may include a counter for counting the number of transaction data messages to be selected, and packaging the transaction data messages to be selected belonging to the same transaction into a transaction package to obtain the transaction to be selected, and recording the transaction information of the transaction to be selected.

[0135] Optionally, the time information of the selected transaction may include a start timestamp and an end timestamp of the selected transaction. The transaction completion status includes completed and uncompleted. The address information may be a source node ID and a destination node ID of the selected transaction, and the data may be data related to the selected transaction, such as data read by the transaction or data written by the transaction.

[0136] It should be noted that the transaction information recorded by the recording unit includes but is not limited to the above-mentioned time information, transaction completion status, address information and data. For example, the transaction information may also include a transaction type, which is used to indicate whether the transaction is a read transaction or a write transaction.

[0137] The recording unit may also determine whether to stop monitoring based on the recorded transactions, including:

[0138] If the preset monitoring stop condition is met, the monitoring of the system on chip is terminated, and the monitoring stop condition includes at least one of the following: the count value reaches a first preset threshold, the transaction record is completed, and the memory of the buffer is less than a second preset threshold.

[0139] When at least one of the following conditions is met, such as the count value of the counter reaches a first preset threshold, the completion status of the monitored transaction is converted to completed, the memory of the buffer is less than a second preset threshold, etc., the monitoring module can stop monitoring. It should be understood that the above monitoring stop conditions are only possible examples given in this application, and other monitoring stop conditions can also be set, such as the monitoring duration reaches a preset duration, etc., and this application does not limit this.

[0140] Among them, the first preset threshold can be set by the tester based on actual needs.

[0141] In a first implementation manner, the counter may be used to count the number of transactions to be selected. For example, if the first preset threshold is n, the monitoring module may stop monitoring when the number of transactions to be selected is n.

[0142] In the second implementation manner, the counter may be used to count the number of transaction data messages to be selected. For example, if the first preset threshold is m, the monitoring module may stop monitoring when the number of transaction data messages to be selected is m.

[0143] Next, combine Fig. 9 , the overall process of the monitoring module in this application for message monitoring is described. Fig. 9 After the monitoring starts, it can be determined whether the filtering unit is enabled based on the configuration information stored in the register, that is, whether message filtering is required. If the filtering unit is enabled, the corresponding message is filtered out based on the filtering conditions, and the filtered message is sent to the matching unit. If the filtering unit is not enabled, the message is sent directly to the matching unit.

[0144] Continue to determine whether the matching unit is enabled based on the configuration information in the register, that is, whether message matching is required. If the matching unit is enabled, message matching is performed according to the matching conditions, the transaction is counted and the transaction information is recorded by the recording unit. The recording unit also determines whether the monitoring stop condition is triggered. If so, the message monitoring is terminated.

[0145] The present application also provides a system on chip, referring to Figure 1 The system on chip includes multiple on-chip modules, an on-chip network and multiple monitoring modules. The on-chip network includes multiple communication nodes. Each monitoring module is deployed between communication nodes or between an on-chip module and a communication node. The system on chip is used to execute the steps of the above-mentioned error location method.

[0146] The embodiment of the present application also provides a computer device 100, which can be connected to the monitoring module for communication, and is used to analyze and locate errors in the system on chip based on the selected transaction data message obtained by the monitoring module. Fig.10 As shown, it is a schematic diagram of the structure of the computer device 100 provided in the embodiment of the present application, including: a processor 1001, a memory 1002, and optionally, a bus 1003. The memory 1002 stores machine-readable instructions executable by the processor 1001. When the computer device 100 is running, the processor 1001 communicates with the memory 1002 through the bus 1003. When the machine-readable instructions are executed by the processor 1001, the steps of the error location method in the above method embodiment are performed.

[0147] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. The computer program is executed when a processor is running, and the processor executes the steps in the above-mentioned error location method.

[0148] In the embodiment of the present application, the computer program can also execute other machine-readable instructions when run by the processor to execute other methods described in the embodiment. For the specific execution method steps and principles, please refer to the description of the embodiment, which will not be repeated here.

[0149] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0150] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the technical solution that contributes to the prior art or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), disk or optical disk and other media that can store program code.

[0151] The above are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. An error location method, It is characterized in that Applied to a system on chip, the system on chip includes: a plurality of on-chip modules, an on-chip network and a plurality of monitoring modules, the on-chip network includes a plurality of communication nodes, each of the monitoring modules is deployed between the communication nodes or between the on-chip module and the communication node, the method includes: The monitoring module receives a transaction data message from the communication node or the on-chip module; The monitoring module filters and / or matches the transaction data message based on a preset filtering condition and a preset matching condition to obtain a transaction data message to be selected; The system on chip is error located according to the transaction data message to be selected to obtain an error location result.

2. The method according to claim 1, It is characterized in that The monitoring module includes: a filtering unit, a matching unit and a recording unit, and the output end of the filtering unit is connected to the input end of the matching unit and the recording unit respectively.

3. The method according to claim 2, It is characterized in that The filtering and / or matching processing of the transaction data message based on the preset filtering condition and the preset matching condition to obtain the data message to be selected includes: The filtering unit filters the transaction data message based on the filtering condition to obtain a filtered transaction data message, and sends the filtered transaction message to the matching unit; The matching unit matches the filtered transaction data message based on the matching condition to obtain a transaction data message to be selected, and sends the transaction data message to be selected to the recording unit; The recording unit records the transaction data message to be selected.

4. The method according to claim 2, It is characterized in that The filtering and / or matching processing of the transaction data message based on the preset filtering condition and the preset matching condition to obtain the transaction data message to be selected includes: The matching unit matches the transaction data message based on the matching condition to obtain a transaction data message to be selected, and sends the transaction data message to be selected to the recording unit; The recording unit records the transaction data message to be selected.

5. The method according to claim 3, It is characterized in that The transaction data message includes: a first channel data message and a plurality of second channel data messages; The filtering unit filters the transaction data message based on the filtering condition to obtain a filtered transaction data message, including: The filtering unit filters the first channel data message based on the filtering condition to obtain a filtered first channel data message; The matching unit matches the filtered transaction data message based on the matching condition to obtain the transaction data message to be selected, including: The matching unit matches the filtered first channel data message based on the matching condition to obtain a first channel data message to be selected; The recording unit recording the transaction data message to be selected includes: The recording unit records the first channel data message to be selected and a plurality of second channel data messages corresponding to the first channel data message to be selected.

6. The method according to claim 3, It is characterized in that The filtering conditions include: filtering digits and filtering masks; The filtering the transaction data message based on the filtering condition to obtain the filtered transaction data message includes: Determining the data to be compared in the transaction data message according to the filter digit and a preset first comparison length; The data to be compared in the transaction data message is filtered according to the filter mask and the preset filter content to obtain the filtered transaction data message.

7. The method according to claim 6, It is characterized in that The filtering the data to be compared in the transaction data message according to the filtering mask and the preset filtering content to obtain the filtered transaction data message includes: Based on the number of bits of the filter mask, the bits in the data to be compared that are the same as the number of bits of the filter mask are masked to obtain filtered data, and the filtered data is filtered based on the filtering content to obtain the filtered transaction data message.

8. The method according to claim 3, It is characterized in that The matching conditions include: matching digits, matching masks; The step of matching the filtered transaction data message based on the matching condition to obtain the transaction data message to be selected includes: Determining the data to be compared of the filtered transaction data message according to the matching digits and a preset second comparison length; The data to be compared of the filtered transaction data message is matched according to the matching mask and the preset matching content to obtain the transaction data message to be selected.

9. The method according to claim 8, It is characterized in that The step of matching the data to be compared of the filtered transaction data message according to the matching mask and the preset matching content to obtain the transaction data message to be selected includes: Based on the number of bits of the matching mask, the digits in the data to be compared of the filtered transaction data message that are the same as the number of bits of the matching mask are retained to obtain matched data, and the matched data are matched based on the preset matching content to obtain the transaction data message to be selected.

10. The method according to claim 3 or 4, It is characterized in that The method further comprises: The recording unit counts the transactions to be selected corresponding to the recorded transaction data messages to be selected, and records the time information, transaction completion status, address information and data of the transactions to be selected.

11. The method according to claim 10, It is characterized in that The method further comprises: If a preset monitoring stop condition is met, the monitoring of the on-chip system is terminated, and the monitoring stop condition includes at least one of the following: the count value reaches a first preset threshold, the transaction record is completed, and the memory of the buffer is less than a second preset threshold.

12. A system on a chip, It is characterized in that It includes multiple on-chip modules, an on-chip network and multiple monitoring modules, the on-chip network includes multiple communication nodes, each monitoring module is deployed between the communication nodes or between the on-chip module and the communication node, and the on-chip system is used to execute the steps of the error location method as described in any one of claims 1-11.