Bus anomaly detection method and apparatus, bus anomaly processing method, apparatus and system, device, and medium

By detecting the abnormality of the response signal on the bus interface and recording information, the problem of untimely handling of bus exceptions is solved, and the reliability and stability of the system are improved.

WO2024230553A1PCT designated stage expired Publication Date: 2024-11-14BEIJING YOUZHUJU NETWORK TECH CO LTD +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2024/090572
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-10
Filing Date
2024-04-29
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

In computer systems, bus exceptions cannot be discovered and processed in time, resulting in errors in calculations of functional components and cannot be processed in time, affecting system reliability.

Method used

By entering the detection state on the bus interface, the abnormality of the response signal is detected and abnormal information is recorded, and the interrupt signal is issued to prevent the interrupt signal from being issued, ensuring that the block is released after all response signals have no abnormalities, so as to avoid the errors continuing to propagate.

Benefits of technology

It realizes timely detection and handling of bus exceptions, avoids calculation errors, and improves the reliability and stability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024090572_14112024_PF_FP_ABST
    Figure CN2024090572_14112024_PF_FP_ABST
Patent Text Reader

Abstract

A bus anomaly detection method and apparatus, a data processing apparatus, a bus anomaly processing method, apparatus and system, an electronic device, and a storage medium. The bus anomaly detection method comprises: at an interface for accessing a bus, in response to an access initiating apparatus sending access requests to an access receiving apparatus, entering a detection state to carry out, in the detection state, anomaly detection on response information fed back by the access receiving apparatus for each access request, and preventing the access initiating apparatus from sending an interrupt signal; in response to the fact that the response information corresponding to the access request is anomalous, ending the detection state, recording bus anomaly information, and sending an interrupt signal; and if the response information corresponding to each access request sent by the access initiating apparatus is received and is not anomalous, ending the detection state to stop blocking the interrupt signal sent by the access initiating apparatus. The bus anomaly detection method at least can discover an anomalous response signal in a timely manner and process the anomalous response signal.
Need to check novelty before this filing date? Find Prior Art

Description

Bus anomaly detection and processing method, device, system, equipment and medium

[0001] This application claims priority to Chinese Patent Application No. 202310524160.5 filed on May 10, 2023, the entire text of which is incorporated by reference as a part of this application. Technical Field

[0002] Embodiments of the present disclosure relate to a bus anomaly detection method and apparatus, a data processing apparatus, a bus anomaly processing method and apparatus, a processing system, an electronic device, and a storage medium. Background Art

[0003] A bus is a common communication line used to transmit information between various functional components of a computer. Various functional components transmit data via the bus. For example, a data processing component can send a read request to a storage component via the bus based on a data processing task. The storage component responds to the read request and feeds the read result back to the data processing component via the bus. After the data processing component performs the data processing task based on the read data and stores the processing result in the storage component, it can send an interrupt signal to the upper-level component to notify it of the completion of the current data processing task.

[0004] Summary of the Invention

[0005] When various functional components access and communicate via a bus, access anomalies may occur due to various reasons, resulting in functional component calculation errors that cannot be promptly detected and processed. To address the above issues, at least one embodiment of the present disclosure provides a bus anomaly detection method and device, a data processing device, a bus anomaly processing method and device, an electronic device, and a storage medium. These methods can promptly detect and process abnormal response signals, avoid or reduce functional component calculation errors, and prevent continued operation after an error has occurred, thereby improving reliability.

[0006] At least one embodiment of the present disclosure provides a bus anomaly detection method, comprising: at an interface for accessing a bus, in response to at least one access initiating device sending an access request to at least one access receiving device, entering a detection state, so that in the detection state, response information fed back by the at least one access receiving device for each access request is detected for anomalies, and the at least one access initiating device is prevented from issuing an interrupt signal for the task corresponding to the access request; in response to at least one access initiating device issuing response information corresponding to the access request having an anomaly, ending the detection state, recording bus anomaly information based on the response information having an anomaly, and issuing an interrupt signal for the response information having an anomaly; if all response information corresponding to each access request issued by the at least one access initiating device is received and there is no anomaly, ending the detection state to stop blocking the interrupt signal issued by the access initiating device.

[0007] At least one embodiment of the present disclosure also provides a bus anomaly detection device, including a detection unit, a first processing unit, and a second processing unit. The detection unit is configured to enter a detection state at an interface for accessing the bus in response to at least one access initiating device sending an access request to at least one access receiving device, so as to perform an anomaly detection on the response information fed back by the at least one access receiving device for each access request in the detection state, and prevent the at least one access initiating device from issuing an interrupt signal for the task corresponding to the access request; the first processing unit is configured to end the detection state in response to an anomaly in the response information corresponding to at least one access request issued by the at least one access initiating device, record bus anomaly information based on the response information with the anomaly, and issue an interrupt signal for the response information with the anomaly; the second processing unit is configured to end the detection state if all response information corresponding to each access request issued by the at least one access initiating device is received and there is no anomaly, so as to stop blocking the interrupt signal issued by the access initiating device.

[0008] At least one embodiment of the present disclosure further provides a data processing device, comprising at least one data processing unit and the bus anomaly detection device described in any embodiment of the present disclosure, wherein the at least one data processing unit is configured to send an access request to at least one storage device based on a received task.

[0009] At least one embodiment of the present disclosure further provides a bus exception handling method, comprising: in response to receiving an interrupt signal sent by the data processing device described in any embodiment of the present disclosure, determining whether there is bus exception information recorded for the data processing device; if there is bus exception information recorded for the data processing device, resetting the data processing device.

[0010] At least one embodiment of the present disclosure further provides a bus exception handling device, comprising an interrupt response unit and a reset unit. The interrupt response unit is configured to respond to an interrupt signal sent by the data processing device described in any embodiment of the present disclosure and determine whether there is bus exception information recorded for the data processing device; the reset unit is configured to reset the data processing device if there is bus exception information recorded for the data processing device.

[0011] At least one embodiment of the present disclosure further provides an electronic device, comprising: the bus exception processing device described in any embodiment of the present disclosure; and the data processing device described in any embodiment of the present disclosure.

[0012] At least one embodiment of the present disclosure further provides an electronic device, comprising a processor and a memory, wherein the memory stores at least one computer program module; wherein the at least one computer program module is configured to be executed by the processor to implement the bus anomaly detection method described in any embodiment of the present disclosure, or to implement the bus anomaly handling method described in any embodiment of the present disclosure.

[0013] At least one embodiment of the present disclosure further provides a storage medium storing non-transitory computer-readable instructions, which, when executed by a computer, implement the bus anomaly detection method described in any embodiment of the present disclosure, or implement the bus anomaly handling method described in any embodiment of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other features, advantages and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same reference numerals represent the same elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0015] FIG1 is a schematic diagram of an electronic device;

[0016] FIG2 is a flow chart of a bus anomaly detection method provided by some embodiments of the present disclosure;

[0017] FIG3 is a schematic diagram of an electronic device provided by some embodiments of the present disclosure;

[0018] FIG4 is a schematic diagram of a state change provided by some embodiments of the present disclosure;

[0019] FIG5 is a schematic diagram of another electronic device provided by some embodiments of the present disclosure;

[0020] FIG6 is a schematic block diagram of a bus anomaly detection device provided by some embodiments of the present disclosure;

[0021] FIG7 is a schematic diagram of a data storage device and a verification unit provided in some embodiments of the present disclosure;

[0022] FIG8 is a flow chart of a bus exception handling method provided in some embodiments of the present disclosure;

[0023] FIG9 is a schematic block diagram of a bus exception processing device provided by some embodiments of the present disclosure;

[0024] FIG10 is a schematic block diagram of a processing system provided by some embodiments of the present disclosure;

[0025] FIG11 is a schematic block diagram of an electronic device provided in some embodiments of the present disclosure;

[0026] FIG12 is a schematic block diagram of another electronic device provided by some embodiments of the present disclosure; and

[0027] FIG13 is a schematic diagram of a storage medium provided in some embodiments of the present disclosure. DETAILED DESCRIPTION

[0028] The following describes embodiments of the present disclosure in more detail with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0029] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0030] As used herein, the term "including" and its variations are open-ended, i.e., "including but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment," the term "another embodiment" means "at least one additional embodiment," and the term "some embodiments" means "at least some embodiments." Other terms are defined in the following description.

[0031] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0032] It should be noted that the modifications of "one" and "plurality" mentioned in this disclosure are illustrative rather than restrictive, and those skilled in the art will understand that unless the context clearly indicates otherwise, they should be understood as "one or more." "Plurality" should be understood as two or more.

[0033] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0034] FIG1 is a schematic diagram of an electronic device.

[0035] As shown in Figure 1, taking the hardware accelerator 120 as an example, the hardware accelerator 120 may include multiple acceleration units 121. For example, the neural network accelerator may include CDMA WT (Convolution Core Weight DMA), CDMA DAT (Convolution Core Data DMA), SDP (Single Data Point Processor), PDP (Planar Data Processor), CDP (Cross-channel Data Processor) and BDMA (bridge DMA) acceleration units, among which DMA (Direct Memory Access) is a direct memory access device.

[0036] For example, the processor 130 can use software to configure the registers of the hardware accelerator 120, for example, by offloading tasks (such as computing tasks) to the registers of the hardware accelerator 120 via the bus 110. The processor 130 can be, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or other devices. Each acceleration unit 121 in the hardware accelerator 120 obtains corresponding data from the memory 140 via the bus 110 according to the software configuration. The memory 140 can be, for example, a DRAM (Dynamic Random Access Memory) or an SLC (System Level Cache). After each acceleration unit 121 obtains data from the memory 140, it can perform calculations on the obtained data and then write the calculation results back to the memory 140 via the bus 110. After each acceleration unit 121 completes the assigned computing task, it can pull up an interrupt signal to use the interrupt signal to notify the processor 130 that the accelerated computing is complete and that the next calculation can be performed. For example, in some embodiments, each acceleration unit 121 in the hardware accelerator 120 may also communicate with other devices or components other than the processor and the memory through the bus 110 .

[0037] For example, in the above process, each acceleration unit 121 in the hardware accelerator 120 can be connected to the bus 110 through the data interface 122, and then exchange data with devices such as the memory 140 through the bus 110. The data interface 122 can be, for example, an ACE-Lite (lightweight ACE) bus interface, where ACE stands for AXI Coherency Extensions and AXI stands for Advanced eXtensible Interface. When the various acceleration units 121 in the hardware accelerator 120 are running, various abnormal situations may occur. For example, when communicating with external devices such as the memory 140 through the bus 110, abnormalities may occur due to various reasons, causing the hardware accelerator to make calculation errors or hang, and cannot be discovered and processed in time.

[0038] At least one embodiment of the present disclosure provides a bus anomaly detection method and device, a data processing device, a bus anomaly processing method and device, an electronic device and a storage medium. After detecting that an access request has been issued, the detection state is entered to start detecting the returned response signal. If an anomaly is detected, it is recorded and an interrupt signal is issued. The abnormal response signal can be discovered in time and the processor and other devices are notified to handle the anomaly, thereby avoiding or reducing the calculation error or hanging of the access initiating device. In addition, in the detection state, each access initiating device can be prevented from issuing an interrupt signal. After confirming that all response signals have been returned and the detection has passed, the interrupt signal issued by each access initiating device is released. This can ensure that the reading and writing operations of each access initiating device are correct before notifying the processor and other devices that the current task processing is completed, and then the next calculation is executed, thereby avoiding continuing to run downward when an error has occurred, thereby improving reliability.

[0039] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0040] At least one embodiment of the present disclosure provides a bus anomaly detection method, comprising: at an interface for accessing a bus, in response to at least one access initiating device sending an access request to at least one access receiving device, entering a detection state, so that in the detection state, response information fed back by the at least one access receiving device for each access request is detected for anomalies, and the at least one access initiating device is prevented from issuing an interrupt signal for the task corresponding to the access request; in response to anomalies in response information corresponding to at least one access request issued by the at least one access initiating device, ending the detection state, recording bus anomaly information based on the response information with the anomaly, and issuing an interrupt signal for the response information with the anomaly; if the response information corresponding to each access request issued by the at least one access initiating device is received and there is no anomaly, ending the detection state to stop blocking the interrupt signal issued by the access initiating device.

[0041] Figure 2 is a flow chart of a bus anomaly detection method provided by some embodiments of the present disclosure. As shown in Figure 2, in at least one embodiment, the method includes the following operations.

[0042] Step S210: At the interface for accessing the bus, in response to at least one access initiating device sending an access request to at least one access receiving device, a detection state is entered, so that in the detection state, an abnormality detection is performed on the response information fed back by the at least one access receiving device for each access request, and the at least one access initiating device is prevented from issuing an interrupt signal for the task corresponding to the access request.

[0043] Step S220: In response to an abnormality in the response information corresponding to at least one access request issued by the at least one access initiating device, end the detection state, record bus abnormality information based on the abnormal response information, and issue an interrupt signal for the abnormal response information.

[0044] Step S230: If the response information corresponding to each access request sent by the at least one access initiating device is received and no abnormality exists, the detection state is ended to stop blocking the interrupt signal sent by the access initiating device.

[0045] FIG3 is a schematic diagram of an electronic device provided by some embodiments of the present disclosure.

[0046] As shown in FIG3 , a bus anomaly detection device 323 may be provided at an interface 322 of the hardware accelerator 320 for accessing the bus. The bus anomaly detection method of at least one embodiment of the present disclosure may be executed by the bus anomaly detection device 323 .

[0047] For example, in step S210, the at least one access initiating device may include at least one acceleration unit of a hardware accelerator. The hardware accelerator may include, for example, a neural network accelerator. The access receiving device may include a memory, and the memory may include, for example, a dynamic random access memory or a system-level cache, etc. However, the present disclosure is not limited to this, and the bus anomaly detection method of the embodiment of the present disclosure may also be used in scenarios where other devices and components (such as processors, controllers, etc.) initiate access requests externally. In the following embodiments of the present disclosure, the access initiating device is an acceleration unit of a hardware accelerator and the access receiving device is a memory as an example for explanation.

[0048] For example, the processor 330 can send a computing task to the hardware accelerator 320 via the bus 310. Each acceleration unit 321 of the hardware accelerator 320 then sends a read request to the memory 340 via the bus 310 based on the computing task. In response to the read request, the memory 340 reads the computational data required for the computing task according to the destination address of the read request and feeds the computational data back to the acceleration unit 321 via the bus 310. After the acceleration unit 321 completes the accelerated computation based on the computational data, it can send a write request to the memory 340 via the bus 310. The write request includes the computational result data to be written to the memory 340. In response to the write request, the memory 340 writes the computational result data to the destination address of the write request and feeds back the result of the write operation to the acceleration unit 321 via the bus 310. After receiving the result of the write operation, the acceleration unit 321 can send an interrupt signal to the processor 330 to notify the processor 330 that the current computing task is complete and that the next computing task can be executed.

[0049] For example, the bus may include a first channel for transmitting a write request, a second channel for transmitting write data, a third channel for transmitting response information corresponding to the write request, a fourth channel for transmitting a read request, and a fifth channel for transmitting response information corresponding to the read request. In some of the following embodiments, the response information corresponding to the write request is also referred to as the response information of the write request, and the response information corresponding to the read request is also referred to as the response information of the read request.

[0050] For example, as shown in Figure 3, the bus may include five channels: the AW channel, the W channel, the B channel, the AR channel, and the R channel. The AW / W / B channels are used for write operations. The AW channel is used to send write request information and can serve as the first channel mentioned above. The W channel is used to send data to be written and can serve as the second channel mentioned above. The B channel is used to transmit the response information returned by the access receiving device after processing a write request (i.e., the write request response information) and can serve as the third channel mentioned above. The write request response information may include a write response status signal (bresp) indicating whether the access receiving device has correctly processed the write request. The AR channel and the R channel are used for read operations. The AR channel is used to transmit read requests and can serve as the fourth channel mentioned above. The R channel is used to transmit the response information returned by the access receiving device after processing a read request (i.e., the read request response information). The read request response information includes the read data (rdata), an error message (rpoison), and a read response status signal (rresp). The R channel can serve as the fifth channel mentioned above.

[0051] For example, the access request sent by the at least one access initiating device to the at least one access receiving device in step S210 may include a read request and / or a write request. The response information fed back by the at least one access receiving device in step S210 for each access request may include response information for the read request and / or response information for the write request.

[0052] For example, the bus anomaly detection device 323 may monitor each access request issued by the acceleration unit 321 to a device such as the memory 340 , and may receive response information corresponding to each access request.

[0053] For example, in the idle state, in response to at least one access initiating device sending an access request to at least one access receiving device, the bus anomaly detection device 323 enters the detection state. For example, the bus anomaly detection device 323 can have multiple operating states, including an idle state, a detection state, and an anomaly state. In the idle state, the bus anomaly detection device 323 can enter the detection state when it detects that the acceleration unit 321 has issued at least one access request. In the detection state, the bus anomaly detection device 323 can wait for response information corresponding to each access request issued by the acceleration unit 321, and after receiving each response information, perform an anomaly detection on the response information. For example, in the detection state, the bus anomaly detection device 323 can wait for signals returned by the third channel and the fifth channel, and perform an anomaly detection on the signals of the third channel and the fifth channel.

[0054] FIG4 is a schematic diagram of a state change provided by some embodiments of the present disclosure.

[0055] As shown in Figure 4, the bus anomaly detection device 323 can monitor the read / write channel. In the idle state, when an access request is detected, it enters the detection state. The detection state may include multiple sub-states (CHECK1 to CHECKn, where n is a positive integer). When the first access request is detected, the CHECK1 sub-state is entered. After that, each time a new access request is received, the next sub-state is switched. For example, when the second access request is detected, the CHECK1 sub-state is switched to the CHECK2 sub-state. When the third access request is detected, the CHECK2 sub-state is switched to the CHECK3 sub-state. The nth sub-state indicates that n access requests have been detected and issued, and also indicates that there are currently n response messages that need to be received and detected. During the monitoring of access requests, the bus anomaly detection device 323 also receives the response information sent back by the bus and detects each response message received. Whenever a response signal passes the detection (i.e., there is no abnormality), it switches back to the next sub-state. For example, if a response signal is received in the CHECKn sub-state and the response signal passes the detection, it switches from the CHECKn sub-state back to the CHECK(n-1) sub-state. If a response signal is received in the CHECK(n-1) sub-state and the response signal passes the detection, it switches from the CHECK(n-1) sub-state back to the CHECK(n-2) sub-state.

[0056] For example, in step S220, if the response information corresponding to at least one access request issued by the at least one access initiating device contains an abnormality, the detection state is terminated and the abnormal state is switched to. For example, as shown in FIG4 , if the bus abnormality detection device 323 receives a response signal in the CHECK(n-2) state and detects an abnormality from the response signal, the abnormality state is switched to. In the abnormal state, an interrupt signal is sent to the processor 330 in response to the abnormal response information.

[0057] For example, in step S230, if all response information corresponding to each access request issued by the at least one access initiating device is received and no abnormalities are detected, the detection state is terminated and the idle state is restored. For example, as shown in Figure 4, if all response signals corresponding to each access request issued are detected to be passed, the bus anomaly detection device 323 can sequentially switch from the CHECKn state back to the CHECK1 state, and from the CHECK1 state to the idle state. Therefore, the idle state is the state in which all response information corresponding to the access requests issued by the at least one access initiating device is received and detected to be normal, and no new access requests are issued.

[0058] For example, after the idle state is restored from the detection state, when a new access request is detected again in the idle state, the bus anomaly detection method of the embodiment of the present disclosure can be executed again.

[0059] For example, the anomalies in the response information corresponding to the access request may include: an anomaly triggered in the response signal because the access address of the access request does not meet the requirements; and / or an anomaly triggered in the response signal because an error occurs in the data read by at least one access receiving device in response to the read request.

[0060] For example, the B channel and the R channel respectively contain a write response status signal (bresp) and a read response status signal (rresp). When SLVERR (slave error) or DECERR (decoding error) appears in these two signals, it indicates that an error exception has occurred in the response signal. This exception occurs when the access address of the access request does not meet the requirements, for example, the access address is illegal.

[0061] For example, the R channel may include error information (rpoison). When the signal of the error information is at an abnormal level (e.g., a high level), it indicates that several (e.g., 8) bits of data corresponding to the error information are erroneous. When the access receiving device (e.g., a memory) finds that the read data is erroneous, the signal of the error information will jump to an abnormal level to indicate that the read data is erroneous.

[0062] For example, in an abnormal state, the bus anomaly detection device 323 may also record bus anomaly information. For example, an anomaly memory may be provided for each access initiating device. When an abnormal response message is detected, the access initiating device corresponding to the abnormal response message may be first determined as the target access initiating device. An anomaly parameter may then be written to the anomaly memory corresponding to the target access initiating device to indicate that an anomaly has occurred in the response message corresponding to the access request issued by the target access initiating device.

[0063] For example, different identifications (IDs) can be set for different acceleration units in the hardware accelerator, and a memory (e.g., a register) can be provided for each acceleration unit to record exceptions. The exception memory can be located in the hardware accelerator or outside the hardware accelerator. When the bus exception detection device 323 detects that a response message has an exception, it can first locate the corresponding acceleration unit based on the acceleration unit identification in the response message and write the exception parameters to the exception memory corresponding to the acceleration unit. For example, the bit used to represent the exception in the exception memory is set to 1, so as to locate the acceleration unit with the exception and facilitate subsequent processing.

[0064] For example, a memory (e.g., a register) for recording exceptions may also be provided for the hardware accelerator. The memory may be located in the hardware accelerator or outside the hardware accelerator. When the bus anomaly detection device 323 detects that a response message received by the hardware accelerator is abnormal, an abnormality parameter may be written to the abnormality memory corresponding to the hardware accelerator, such as setting a bit representing the abnormality to 1. Subsequently, the memory may be queried to determine whether the hardware accelerator has an abnormality.

[0065] For example, the bus anomaly detection device 323 can not only detect whether each response message received from the bus contains an anomaly, but can also indicate whether a response message has not been returned for a long time. For example, if the bus anomaly detection device 323 remains in the detection state for a long time (i.e., does not change to the idle state or abnormal state for a long time), it means that the response message corresponding to the access request has not been returned for a long time.

[0066] For example, in other embodiments, if the bus abnormality detection device 323 detects that it is in the detection state for a long time (predetermined time), it may also record this situation as an abnormal situation or send an interrupt signal to the processor.

[0067] For example, in the detection state, the bus anomaly detection device 323 can prevent each acceleration unit from sending an interrupt signal to the processor 330 for the task corresponding to the access request. After the response signals corresponding to all the access requests that have been sent are returned and pass the detection, the bus anomaly detection device 323 can release the block on the interrupt signals sent by each acceleration unit.

[0068] For example, in an abnormal state, the bus abnormality detection device 323 may send an interrupt signal to the processor to notify the processor to process the abnormality occurring in the hardware accelerator.

[0069] FIG5 is a schematic diagram of another electronic device provided by some embodiments of the present disclosure.

[0070] As shown in Figure 5, the bus anomaly detection device 323 can be provided with two or more interfaces 322, each of which can include a bus anomaly detection device 323. The data length of a single access request processed by different interfaces can be different, and each acceleration unit 321 can send the access request to the corresponding interface 322 based on the data length of the access request. The bus anomaly detection device 323 can also include a collection unit 324, which can collect the status of the bus anomaly detection device 323 in each interface and initiate an interrupt to the processor when an abnormal state occurs.

[0071] According to at least one embodiment of the present disclosure, after detecting that an access request has been issued, a detection state is entered to start detecting the returned response signal. If an abnormality is detected, it is recorded and an interrupt signal is issued. The abnormal response signal can be discovered in time and the processor and other devices are notified to handle the abnormality, thereby avoiding or reducing calculation errors or hangs of the access initiating device.

[0072] According to at least one embodiment of the present disclosure, in the detection state, each access initiating device can be prevented from sending an interrupt signal. After confirming that all response signals are returned and the detection is passed, the blocking of the interrupt signal sent by each access initiating device is released. This ensures that the reading and writing operations of each access initiating device are correct before notifying the processor and other devices that the current task processing is completed, and then executing the next step of calculation, avoiding continuing to run downward when an error has occurred, thereby improving reliability.

[0073] According to at least one embodiment of the present disclosure, by setting the detection status, an abnormal situation in which a response signal does not return for a long time can be reflected.

[0074] According to at least one embodiment of the present disclosure, the access initiating device may be a hardware accelerator, which can detect abnormalities in a response signal received by the hardware accelerator.

[0075] According to at least one embodiment of the present disclosure, by recording abnormal parameters in an abnormal memory corresponding to a target access initiating device according to an abnormal response signal, the abnormal access initiating device can be located, facilitating subsequent processing.

[0076] According to at least one embodiment of the present disclosure, by detecting the signals transmitted by the two third channels and the fifth channel, abnormal signals can be detected accurately and quickly.

[0077] It should be noted that in the embodiments of the present disclosure, the order in which the various steps of the bus anomaly detection method are executed is not limited. Although the execution process of the various steps is described above in a specific order, this does not constitute a limitation on the embodiments of the present disclosure. The various steps in the bus anomaly detection method can be executed serially or in parallel, which can be determined according to actual needs. The bus anomaly detection method can also include more or fewer steps.

[0078] At least one embodiment of the present disclosure further provides a bus anomaly detection device. FIG6 is a schematic block diagram of a bus anomaly detection device provided in some embodiments of the present disclosure. As shown in FIG6 , the bus anomaly detection device 400 includes a detection unit 410 , a first processing unit 420 , and a second processing unit 430 .

[0079] Detection unit 410 is configured to enter a detection state at an interface for accessing a bus, in response to at least one access initiating device sending an access request to at least one access receiving device. In the detection state, detection unit 410 detects anomalies in response information fed back by at least one access receiving device for each access request, and prevents at least one access initiating device from issuing an interrupt signal for the task corresponding to the access request. For example, detection unit 410 may execute step S210 of the bus anomaly detection method shown in FIG2 .

[0080] The first processing unit 420 is configured to, in response to a detection of an abnormality in response information corresponding to at least one access request issued by at least one access initiating device, terminate the detection state, record bus abnormality information based on the abnormal response information, and issue an interrupt signal for the abnormal response information. For example, the first processing unit 420 may execute step S220 of the bus abnormality detection method shown in FIG2 .

[0081] The second processing unit 430 is configured to terminate the detection state if all response information corresponding to each access request issued by at least one access initiating device is received and no abnormality exists, thereby stopping the blocking of the interrupt signal issued by the access initiating device. For example, the second processing unit 430 may execute step S230 of the bus anomaly detection method shown in Figure 2.

[0082] For example, the detection unit 410, the first processing unit 420, and the second processing unit 430 may be hardware, software, firmware, or any feasible combination thereof. For example, the detection unit 410, the first processing unit 420, and the second processing unit 430 may be dedicated or general-purpose circuits, chips, or devices, or may be a combination of a processor and memory. The embodiments of the present disclosure do not limit the specific implementation of the detection unit 410, the first processing unit 420, and the second processing unit 430.

[0083] It should be noted that in the embodiments of the present disclosure, the various units of bus anomaly detection device 400 correspond to the various steps of the aforementioned bus anomaly detection method. For the specific functions of bus anomaly detection device 400, reference can be made to the relevant description of the bus anomaly detection method above and will not be repeated here. The components and structure of bus anomaly detection device 400 shown in FIG6 are merely exemplary and non-restrictive. Bus anomaly detection device 400 may also include other components and structures as needed.

[0084] At least one embodiment of the present disclosure further provides a data processing device, which includes at least one data processing unit and a bus anomaly detection device. The at least one data processing unit is configured to send an access request to at least one storage device based on a received task.

[0085] For example, the data processing apparatus may include a hardware accelerator, and the data processing unit may include an acceleration unit.

[0086] For example, regarding the data processing device, you can refer to the relevant description of the hardware accelerator 320 in any of the above embodiments. Regarding the bus abnormality detection device, you can refer to the relevant description of the bus abnormality detection device 323 in any of the above embodiments. Regarding the data processing unit, you can refer to the relevant description of the access initiation device and the acceleration unit 321 in any of the above embodiments. No further details will be given here.

[0087] For example, the data processing device may further include a data storage device and a verification unit, wherein the data storage device is configured to store calculation data of at least one data processing unit, and the verification unit is configured to verify and correct data output by the data storage device.

[0088] FIG7 is a schematic diagram of a data storage device and a verification unit provided in some embodiments of the present disclosure.

[0089] As shown in FIG7 , the data memory may be a convolution buffer (CBUF) 510, which may be used to cache convolution data (e.g., feature maps) and weights, etc., thereby reducing the excessive reading of frequently used data from external memory. The convolution buffer may be located at the second level of the convolution core. The convolution buffer may be implemented as a static random-access memory (SRAM), for example, a 512KB SRAM organized into 16 banks (memory arrays), each bank being 32KB. The convolution buffer may have two input ports and three output ports.

[0090] For example, at least one embodiment of the present disclosure adds an ECC (Error Correcting Code, error checking and correction) protection function to the convolution buffer 510. An ECC check code is generated at the input port of the convolution buffer 510 by a generation module 520 (e.g., ECC GEN), and the generated check code is stored in the convolution buffer 510 together with the data. The WMB in Figure 7 represents the weight mask bit (Weight Mask Bit). ECC check is performed at the output port of the convolution buffer 510 by a check unit 530 (ECC CHECK). ECC check can detect single-bit errors (CE ERROR, Correctable ERROR) and correct them, and can also detect multi-bit errors (UC ERROR, Uncorrectable ERROR). For UC ERROR, its address in the convolution buffer 510 can be recorded and an interrupt can be sent to the processor. For CE ERROR, the bit that causes the error can be corrected. By performing ECC detection and correction on the convolution cache 510, the data in the convolution cache 510 can be protected, data errors can be reduced or avoided, and calculation errors caused by data errors can be avoided, thereby improving reliability.

[0091] At least one embodiment of the present disclosure further provides a bus exception handling method. FIG8 is a flow chart of a bus exception handling method provided by some embodiments of the present disclosure. As shown in FIG8, in at least one embodiment, the method includes the following operations.

[0092] Step S610: In response to receiving an interrupt signal sent by a data processing device, determining whether there is bus abnormality information recorded for the data processing device.

[0093] Step S620: If there is bus abnormality information recorded for the data processing device, the data processing device is reset.

[0094] For example, the bus exception handling method can be executed by the processor 330 shown in Figures 3 and 5, and the data processing device is, for example, the hardware accelerator 320 shown in Figures 3 and 5. When the hardware accelerator detects that the response signal is abnormal, it sends an interrupt signal to the processor 330. After receiving the interrupt signal, the processor 330 can first check whether the abnormal parameters are recorded in the abnormal memory corresponding to the hardware accelerator 320. If so, it means that the hardware accelerator 320 has an abnormality, and the hardware accelerator 320 can be reset. The processor 330 can also check the abnormal memory corresponding to each acceleration unit to determine which acceleration unit has an abnormality.

[0095] According to at least one embodiment of the present disclosure, after receiving an interrupt signal sent by a data processing device, the data processing device is checked to see if an abnormality occurs, and reset processing is performed when an abnormality occurs. This allows timely processing of abnormalities occurring in the data processing device, avoiding or reducing calculation errors or hangs of the data processing device, and ensuring the correctness of subsequent calculations.

[0096] For example, resetting the data processing device may include: waiting for the data processing unit of the data processing device to complete the current task, and then resetting the data processing device; after the reset process, reloading the task corresponding to the abnormal response signal to the data processing device.

[0097] For example, if an error is detected in the response signal corresponding to the first acceleration unit, the system can first wait for other acceleration units to complete processing the currently ongoing tasks, and then reset the hardware accelerator 320. For example, the reset signal of the hardware accelerator can be pulled high, maintained for a period of time, and then released. After the reset, the abnormal task can be reloaded into the register of the hardware accelerator 320 so that the first acceleration unit can reprocess the task. Since the error in the response signal may be caused by reasons such as the software configuring an incorrect address, or by an error caused by an occasional system failure, performing a reset operation and re-executing the task can solve the problem.

[0098] For example, the bus anomaly handling method may further include: in response to not receiving an interrupt signal sent by the data processing device within a predetermined time period, checking whether the current state of the bus anomaly detection device of the data processing device is a detection state; if the current state of the bus anomaly detection device of the data processing device is a detection state, resetting the processing system in which the data processing device resides. The predetermined time period can be set as needed and is not limited in the present embodiment.

[0099] For example, if the processor 330 does not receive an interrupt signal from the hardware accelerator 320 for a long time, it can check whether the hardware accelerator 320 is in a detection state. If so, it means that the hardware accelerator 320 has not returned a response signal for a long time. In this case, the processor 330 can also reset the entire processing system where the hardware accelerator 320 is located to resolve the abnormal situation.

[0100] For example, if it is detected that the response signal corresponding to the first acceleration unit has not returned for a long time, the system can first wait for other acceleration units to complete processing of the currently ongoing tasks, and then reset the entire processing system where the hardware accelerator 320 is located. The entire processing system where the hardware accelerator 320 is located may include hardware accelerators, buses, processors, memory, and other devices. After the reset, the task corresponding to the abnormal response signal can be reloaded to the hardware accelerator 320.

[0101] It should be noted that in the embodiments of the present disclosure, the order in which the various steps of the bus anomaly handling method are executed is not limited. Although the execution process of the various steps is described above in a specific order, this does not constitute a limitation on the embodiments of the present disclosure. The various steps in the bus anomaly handling method can be executed serially or in parallel, which can be determined according to actual needs. The bus anomaly detection method can also include more or fewer steps.

[0102] At least one embodiment of the present disclosure further provides a bus exception handling device. FIG9 is a schematic block diagram of a bus exception handling device provided by some embodiments of the present disclosure. As shown in FIG9 , the bus exception handling device 700 may include an interrupt response unit 710 and a reset unit 720 .

[0103] The interrupt response unit 710 is configured to, in response to receiving an interrupt signal sent by the data processing device according to any embodiment of the present disclosure, determine whether bus exception information is recorded for the data processing device. For example, the interrupt response unit 710 may execute step S610 of the bus exception handling method shown in FIG8 .

[0104] The reset unit 720 is configured to reset the data processing device if bus abnormality information recorded for the data processing device exists. For example, the reset unit 720 may execute step S620 of the bus abnormality processing method shown in FIG8 .

[0105] For example, the interrupt response unit 710 and the reset unit 720 can be hardware, software, firmware, or any feasible combination thereof. For example, the interrupt response unit 710 and the reset unit 720 can be a dedicated or general circuit, chip, or device, or can be a combination of a processor and a memory. The embodiments of the present disclosure do not limit the specific implementation of the interrupt response unit 710 and the reset unit 720.

[0106] It should be noted that in the embodiments of the present disclosure, the various units of bus exception handling device 700 correspond to the various steps of the aforementioned bus exception handling method. For the specific functions of bus exception handling device 700, reference can be made to the relevant description of the bus exception handling method above and will not be repeated here. The components and structure of bus exception handling device 700 shown in FIG9 are merely exemplary and non-restrictive. Bus exception handling device 700 may also include other components and structures as needed.

[0107] At least one embodiment of the present disclosure further provides a processing system. Figure 10 is a schematic block diagram of a processing system provided by some embodiments of the present disclosure. As shown in Figure 10, the electronic device 800 may include a bus exception handling device 810, a data processing device 820, and a bus 830.

[0108] For example, the processing system may be a system on chip (SoC), or an electronic device such as a computer or a server.

[0109] For example, the bus exception handling device 810 may refer to the bus exception handling device provided in any of the above embodiments, and may be, for example, the processor 330 shown in Figures 3 and 5. The data processing device 820 may refer to the data processing device provided in any of the above embodiments, and may be, for example, the hardware accelerator 320 shown in Figures 3 and 5.

[0110] At least one embodiment of the present disclosure further provides an electronic device, and Figure 11 is a schematic block diagram of an electronic device provided by some embodiments of the present disclosure. As shown in Figure 11, the electronic device 900 includes a processor 910 and a memory 920. The memory 920 is used to store non-transitory computer-readable instructions (such as one or more computer program modules). The processor 910 is used to run non-transitory computer-readable instructions, and when the non-transitory computer-readable instructions are run by the processor 910, one or more steps in the bus anomaly detection method and / or bus anomaly handling method described above can be executed. The memory 920 and the processor 910 can be interconnected through a bus system and / or other forms of connection mechanisms (not shown).

[0111] For example, the processor 910 may be a central processing unit (CPU), a digital signal processor (DSP), or other processing units with data processing capabilities and / or program execution capabilities, such as a field programmable gate array (FPGA). For example, the central processing unit (CPU) may be an X86 or ARM architecture. The processor 910 may be a general-purpose processor or a dedicated processor, and may control other components in the electronic device 900 to perform desired functions.

[0112] For example, the memory 920 may include any combination of one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, an erasable programmable read-only memory (EPROM), a portable compact disk read-only memory (CD-ROM), a USB memory, a flash memory, etc. At least one computer program module may be stored on the computer-readable storage medium, and the processor 910 may execute the at least one computer program module to implement various functions of the electronic device 900. The computer-readable storage medium may also store various applications and various data, as well as various data used and / or generated by the applications.

[0113] It should be noted that, in the embodiment of the present disclosure, the specific functions and technical effects of the electronic device 900 can refer to the above description of the bus anomaly detection method and / or bus anomaly processing method, which will not be repeated here.

[0114] Figure 12 is a schematic block diagram of another electronic device provided in some embodiments of the present disclosure. This electronic device 1000 is, for example, suitable for implementing the bus anomaly detection method and / or bus anomaly handling method provided in embodiments of the present disclosure. The electronic device 1000 may be a user terminal, etc. It should be noted that the electronic device 1000 shown in Figure 12 is merely an example and does not impose any limitations on the functionality and scope of use of the embodiments of the present disclosure.

[0115] As shown in FIG12 , the electronic device 1000 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 1010, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1020 or a program loaded from a storage device 1080 into a random access memory (RAM) 1030. Various programs and data required for the operation of the electronic device 1000 are also stored in the RAM 1030. The processing device 1010, the ROM 1020, and the RAM 1030 are connected to each other via a bus 1040. An input / output (I / O) interface 1050 is also connected to the bus 1040.

[0116] Typically, the following devices may be connected to the I / O interface 1050: an input device 1060 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1070 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1080 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1090. The communication device 1090 may allow the electronic device 1000 to communicate with other electronic devices wirelessly or by wire to exchange data. Although FIG12 shows the electronic device 1000 with various devices, it should be understood that it is not required to implement or have all of the devices shown, and the electronic device 1000 may alternatively implement or have more or fewer devices.

[0117] For example, according to an embodiment of the present disclosure, the bus anomaly detection method and / or the bus anomaly handling method can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the above-mentioned bus anomaly detection method and / or bus anomaly handling method. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 1090, or installed from the storage device 1080, or installed from the ROM 1020. When the computer program is executed by the processing device 1010, the functions defined in the bus anomaly detection method and / or bus anomaly handling method provided in the embodiment of the present disclosure can be executed.

[0118] At least one embodiment of the present disclosure further provides a storage medium for storing non-transitory computer-readable instructions, which, when executed by a computer, can implement the bus anomaly detection method and / or bus anomaly handling method described in any embodiment of the present disclosure.

[0119] Figure 13 is a schematic diagram of a storage medium provided in some embodiments of the present disclosure. As shown in Figure 13, storage medium 1100 is used to store non-transitory computer-readable instructions 1110. For example, when non-transitory computer-readable instructions 1110 are executed by a computer, one or more steps of the bus anomaly detection method and / or bus anomaly handling method described above may be performed.

[0120] For example, the storage medium 1100 can be applied to the electronic device 900 described above. For example, the storage medium 1100 can be the memory 920 in the electronic device 900 shown in FIG11 . For example, the relevant description of the storage medium 1100 can refer to the corresponding description of the memory 920 in the electronic device 900 shown in FIG11 , and will not be repeated here.

[0121] It should be noted that the storage medium (computer-readable medium) mentioned above in the present disclosure may be a computer-readable signal medium or a non-transitory computer-readable storage medium or any combination of the two. Non-transitory computer-readable storage media may be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or components, or any combination thereof. More specific examples of non-transitory computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a non-transitory computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a non-transitory computer-readable storage medium that can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0122] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as the Hyper Text Transfer Protocol (HTTP), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), internetworks (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

[0123] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0124] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including the at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0125] Alternatively, the computer-readable medium carries one or more programs, which, when executed by the electronic device, causes the electronic device to: receive a node evaluation request including at least two Internet Protocol addresses; select an Internet Protocol address from the at least two Internet Protocol addresses; and return the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0126] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, such as a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0127] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code includes at least one executable instruction for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0128] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0129] The functions described above herein may be performed, at least in part, by at least one hardware logic component. For example, and without limitation, exemplary types of hardware logic components that may be used include: a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), an application specific standard product (ASSP), a system on a chip (SOC), a complex programmable logic device (CPLD), and the like.

[0130] In the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0131] The above description is only a partial embodiment of the present disclosure and an illustration of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, the above-mentioned features can be replaced with (but not limited to) technical features with similar functions disclosed in this disclosure.

[0132] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details have been included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0133] Although the subject matter has been described in language specific to structural features and / or methodological logical acts, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are merely example forms of implementing the claims.

Claims

1. A bus anomaly detection method, comprising: At an interface for accessing a bus, in response to at least one access initiating device sending an access request to at least one access receiving device, entering a detection state, so as to, in the detection state, detect abnormalities in response information fed back by the at least one access receiving device in response to each access request, and prevent the at least one access initiating device from issuing an interrupt signal for the task corresponding to the access request; In response to a finding of an abnormality in response information corresponding to at least one access request issued by the at least one access initiating device, ending the detection state, recording bus abnormality information based on the abnormal response information, and issuing an interrupt signal for the abnormal response information; If the response information corresponding to each access request sent by the at least one access initiating device is received and no abnormality occurs, the detection state is ended to stop blocking the interrupt signal sent by the access initiating device.

2. The method according to claim 1, wherein The at least one access initiating device includes at least one acceleration unit of a hardware accelerator.

3. The method according to claim 1 or 2, wherein The bus includes: a first channel for transmitting write requests, a second channel for transmitting data to be written, a third channel for transmitting response information corresponding to the write requests, a fourth channel for transmitting read requests, and a fifth channel for transmitting response information corresponding to the read requests; Performing abnormality detection on response information fed back by the at least one access receiving device in response to each access request includes: performing abnormality detection on signals of the third channel and the fifth channel.

4. The method according to claim 3, wherein: The response information corresponding to the access request contains anomalies, including: An exception is triggered in the response signal because the access address of the access request does not meet the requirements; and / or An exception is triggered in the response signal due to an error in data read by the at least one access receiving device in response to the read request.

5. The method according to any one of claims 1 to 4, wherein: Each of the access initiating devices corresponds to an abnormal storage; Record bus exception information based on the response information of the exception, including: Determine the access initiating device corresponding to the abnormal response information as the target access initiating device; An abnormality parameter is written into the abnormality memory corresponding to the target access initiating device to indicate that an abnormality occurs in the response information corresponding to the access request issued by the target access initiating device.

6. The method according to any one of claims 1 to 5, wherein: Entering a detection state in response to at least one access initiating device sending an access request to at least one access receiving device comprises: in an idle state, entering a detection state in response to at least one access initiating device sending an access request to at least one access receiving device; If the response information corresponding to each access request issued by the at least one access initiating device is received and there is no abnormality, then the detection state is ended, including: if the response information corresponding to each access request issued by the at least one access initiating device is received and there is no abnormality, then the detection state is ended and restored to the idle state.

7. A bus anomaly detection device, comprising: a detection unit configured to, at an interface for accessing the bus, enter a detection state in response to at least one access initiating device sending an access request to at least one access receiving device, and, in the detection state, perform abnormality detection on response information fed back by the at least one access receiving device in response to each access request, and prevent the at least one access initiating device from issuing an interrupt signal for the task corresponding to the access request; a first processing unit configured to, in response to a presence of an abnormality in response information corresponding to at least one access request issued by the at least one access initiating device, end the detection state, record bus abnormality information based on the abnormal response information, and issue an interrupt signal for the abnormal response information; The second processing unit is configured to end the detection state if the response information corresponding to each access request issued by the at least one access initiating device is received and there is no abnormality, so as to stop blocking the interrupt signal issued by the access initiating device.

8. A data processing device comprising: at least one data processing unit configured to send an access request to at least one storage device based on the received task; The bus anomaly detection device according to claim 7.

9. The device according to claim 8, wherein The data processing device includes a hardware accelerator, and the data processing unit includes an acceleration unit.

10. The apparatus according to claim 9, further comprising: a data storage device configured to store calculation data of at least one of the data processing units; The verification unit is configured to verify and correct the data output by the data storage.

11. A bus exception handling method, comprising: In response to receiving an interrupt signal sent by the data processing device according to any one of claims 8 to 10, determining whether there is bus abnormality information recorded for the data processing device; If there is bus abnormality information recorded for the data processing device, the data processing device is reset.

12. The method according to claim 11, further comprising: In response to not receiving an interrupt signal sent by the data processing device within a predetermined time period, checking whether a current state of a bus abnormality detection device of the data processing device is a detection state; If the current state of the bus abnormality detection device of the data processing device is the detection state, a reset process is performed on the processing system where the data processing device is located.

13. The method according to claim 11 or 12, wherein resetting the data processing device comprises: Waiting for the data processing unit of the data processing device to complete the current task, and then resetting the data processing device; After the reset process, the task corresponding to the abnormal response signal is reloaded into the data processing device.

14. A bus exception handling device, comprising: an interrupt response unit configured to, in response to receiving an interrupt signal sent by the data processing device according to any one of claims 8 to 10, determine whether there is bus abnormality information recorded for the data processing device; The reset unit is configured to reset the data processing device if there is bus abnormality information recorded for the data processing device.

15. A processing system comprising: bus; The bus abnormality processing device according to claim 14; as well as A data processing device according to any one of claims 8 to 10.

16. An electronic device comprising: processor; a memory storing at least one computer program module; The at least one computer program module is configured to be executed by the processor to implement the bus anomaly detection method according to any one of claims 1 to 6, or to implement the bus anomaly handling method according to any one of claims 11 to 13.

17. A computer-readable storage medium storing non-transitory computer-readable instructions, wherein when the non-transitory computer-readable instructions are executed by a computer, the bus anomaly detection method according to any one of claims 1 to 6 or the bus anomaly handling method according to any one of claims 11 to 13 is implemented.

Citation Information

Patent Citations

  • Hardware reset control method and apparatus

    CN101311910A

  • A service board card abnormity recovery method and system

    CN109918230A

  • Exception handling method and device, hardware protection equipment and storage medium

    CN110716818A

  • Flash memory controller, data processing system with flash memory controller and method of operating a flash memory controller

    US20170004063A1