Abnormal information monitoring method and system for graphic card initialization process

Abstract

The application discloses an abnormal information monitoring method and system for a graphic card initialization process, and relates to the technical field of data monitoring. The abnormal information monitoring method is executed by the graphic card, and based on a graphic card initialization abnormality monitoring strategy corresponding to a target monitoring parameter in a graphic card initialization monitoring process, whether the target monitoring parameter is abnormal is judged. The target monitoring parameter includes a hardware monitoring parameter and a software monitoring parameter. The technical effects are achieved, that is, the abnormal monitoring of the target monitoring parameter is quickly realized, the abnormal record is stored, the abnormal reason is quickly located, the graphic card detection personnel can timely repair the abnormality, the graphic card safety risk is reduced, and finally the user experience is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data monitoring, and in particular to an abnormal information monitoring method and system for a graphics card initialization process. BACKGROUND

[0002] With the rapid development and application of artificial intelligence (AI), the graphics card initialization process includes the interaction process between the server boot stage and the graphics card through the basic input output software system (BIOS) and the video basic input output software system (VBIOS), as well as the two stages of driving initialization after entering the software system. Normal initialization is a key step to ensure the normal function of the graphics card, which is conducive to ensuring the compatibility of the graphics card, thereby improving the AI performance, supporting complex AI frameworks and tools, and optimizing the power consumption stability. Therefore, the correct driving of the initialization process of the graphics card is the basis for efficient operation of AI application.

[0003] In related technologies, whether the graphics card driver is completely installed successfully is generally determined by checking the driver version or checking the basic information of the graphics card, but the initialization stage of the current graphics card is not monitored independently. Therefore, due to the fact that the current graphics card usually ignores the initialization process monitoring, it is easy to cause the graphics card to appear card loss, application program startup abnormality or performance abnormality, and the graphics card initialization abnormality cannot be quickly located, resulting in a decline in user experience and even a security risk. SUMMARY

[0004] The present application provides an abnormal information monitoring method and system for a graphics card initialization process and a computer device to at least solve the problem that the current graphics card usually ignores the initialization process monitoring in related technologies, cannot quickly locate the abnormality reason of the graphics card initialization, causes a decline in user experience, and even has a security risk.

[0005] The present application provides an abnormal information monitoring method for a graphics card initialization process, comprising: the method is applied to a graphics card, and the graphics card comprises: a plurality of hardware circuits, a storage unit, a plurality of sensors and a control unit, wherein each sensor corresponds to a hardware monitoring parameter of a hardware circuit, and a video basic input output software system is configured in the control unit, and the method comprises:

[0006] In response to the start of the graphics card, performing a monitoring action of the graphics card initialization process through the video basic input output software system;

[0007] Obtaining a target monitoring parameter in the graphics card initialization monitoring process, the target monitoring parameter comprising: a hardware monitoring parameter and a software monitoring parameter;

[0008] The GPU initialization abnormality monitoring strategy based on the target monitoring parameter is generated by the video basic input and output software system according to preset configuration parameters of the target monitoring parameter.

[0009] When the target monitoring parameter is abnormal, the abnormal information is stored in the storage unit according to the abnormal marking mode.

[0010] The application further provides an abnormal information monitoring device of a GPU initialization process, applied to a GPU, the GPU comprising a plurality of hardware circuits, a storage unit, a plurality of sensors and a control unit, wherein each sensor corresponds to a hardware monitoring parameter of a hardware circuit, and a video basic input and output software system is configured in the control unit, and the device comprises:

[0011] The monitoring action execution module is configured to execute a monitoring action of the GPU initialization process through the video basic input and output software system in response to starting of the GPU;

[0012] The monitoring data acquisition module is configured to acquire a target monitoring parameter in the GPU initialization monitoring process, and the target monitoring parameter comprises a hardware monitoring parameter and a software monitoring parameter;

[0013] The abnormal condition determination module is configured to determine whether the target monitoring parameter is abnormal based on a GPU initialization abnormality monitoring strategy of the target monitoring parameter, wherein the GPU initialization abnormality monitoring strategy of the target monitoring parameter is generated by the video basic input and output software system according to preset configuration parameters of the target monitoring parameter;

[0014] The abnormal information storage module is configured to store abnormal information in the storage unit according to an abnormal marking mode when the target monitoring parameter is abnormal.

[0015] The application further provides a computer device, comprising a memory configured to store a computer program and a processor configured to execute the computer program to implement steps of the abnormal information monitoring method of the GPU initialization process.

[0016] The application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement steps of the abnormal information monitoring method of the GPU initialization process.

[0017] The application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement steps of the abnormal information monitoring method of the GPU initialization process.

[0018] Through the application, since the graphics card executes the abnormal information monitoring method of the graphics card initialization process, the graphics card initialization abnormal monitoring strategy corresponding to the target monitoring parameter in the graphics card initialization monitoring process is used to judge whether the target monitoring parameter is abnormal, the target monitoring parameter includes a hardware monitoring parameter and a software monitoring parameter, the technical effects of quickly realizing abnormal monitoring of the target monitoring parameter and storing abnormal records, quickly positioning abnormal reasons, facilitating graphics card detection personnel to repair abnormalities in time, reducing graphics card safety risks, and finally improving user experience are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0020] Figure 1 The structural block diagram of the graphics card initialization process abnormal information monitoring system provided in the embodiments of the present application is shown in the figure.

[0021] Figure 2 The structural block diagram of the graphics card initialization process abnormal information monitoring system provided in the embodiments of the present application is shown in the figure.

[0022] Figure 3 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0023] Figure 4 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0024] Figure 5 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0025] Figure 6 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0026] Figure 7 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0027] Figure 8 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0028] Figure 9 The flow chart of the graphics card initialization process abnormal information monitoring method provided in the embodiments of the present application is shown in the figure.

[0029] Figure 10It is a schematic diagram of the hardware structure of the computer device in the embodiments of the present application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to 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 the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0031] It should be noted that, in the description of the present application, the terms “comprise”, “contain” or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. The terms “first”, “second” and the like in the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0032] In order to enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0033] In the current AI context, the importance of graphics cards is increasingly prominent. The importance of graphics processing units (GPUs) in graphics cards is increasingly prominent. GPUs were originally designed for graphics rendering, but with their powerful parallel computing capabilities, they have become an indispensable hardware support in the AI field. The large amount of matrix operations and data processing in deep learning training and inference are decomposed into thousands of parallel tasks, and the architecture of GPU enables it to efficiently execute these tasks. The high memory bandwidth and optimized hardware architecture of GPU make it more advantageous than CPU in processing AI tasks. At the same time, GPU can also respond quickly in the model inference stage, supporting real-time decision-making. The scalability and flexibility of GPU also promote the emergence of new programming models and algorithms, driving the development of AI technology.

[0034] The graphics card initialization process includes the interaction process between the server boot stage and the VBIOS system through the system BIOS software system. Normal initialization is a key step to ensure the normal function of the graphics card, which is conducive to ensuring the compatibility of the graphics card, improving the AI performance, supporting complex AI frameworks and tools, and optimizing the power consumption stability. Therefore, the correct driving of the initialization process of the graphics card is the basis for the efficient operation of AI applications.

[0035] The current graphics card initialization stage cannot be monitored independently. The current graphics card application is to determine whether the graphics card driver is completely installed successfully by checking the driver version or checking the basic information of the graphics card after all the function modules are driven and the data is loaded. The initialization process is not judged. Therefore, due to the current graphics card usually ignoring the initialization process monitoring, it is easy to cause the graphics card to appear card loss, application program startup abnormality or performance abnormality, and the abnormal reason cannot be quickly located, causing the user experience to decrease, and even there is a security risk.

[0036] Therefore, the embodiment provides an abnormal information monitoring method of a graphics card initialization process, which monitors and records abnormal failures in the graphics card initialization process.

[0037] In combination with the specific application environment architecture or the specific hardware architecture on which the abnormal information monitoring method of the graphics card initialization process depends, the specific application environment architecture or the specific hardware architecture is described herein.

[0038] The embodiment of the present application provides an abnormal information monitoring method of a graphics card initialization process, and the method is described in detail in combination with the execution process of the abnormal information monitoring method of the graphics card initialization process. The technical terms involved in the method must be explained.

[0039] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment.

[0040] The embodiment of the present application provides an abnormal information monitoring system of a graphics card initialization process, as shown in Figure 1 The system includes a graphics card 11, a baseboard management controller 12 and a server 13.

[0041] In Figure 1In the specific implementation, the graphics card 11 comprises a plurality of hardware circuits 111, a storage unit 112, a plurality of sensors 113, and a control unit 114, wherein each sensor 113 corresponds to monitor a hardware monitoring parameter of a hardware circuit 111, the control unit 114 is configured with a video basic input and output software system, and the server 13 communicates with the baseboard management controller 12 and the graphics card 11. The graphics card is used to execute the abnormal information monitoring method of the graphics card initialization process in the embodiment, and the method flow comprises: in response to the start of the graphics card, performing a monitoring action of the graphics card initialization process by the video basic input and output software system; obtaining a target monitoring parameter in the graphics card initialization monitoring process, the target monitoring parameter comprising: a hardware monitoring parameter and a software monitoring parameter; judging whether the target monitoring parameter is abnormal based on a graphics card initialization abnormal monitoring strategy of the target monitoring parameter, wherein the graphics card initialization abnormal monitoring strategy of the target monitoring parameter is generated by the video basic input and output software system according to preset configuration parameters of the target monitoring parameter; and when the target monitoring parameter is abnormal, storing the abnormal information to the storage unit according to an abnormal marking mode.

[0042] The plurality of hardware circuits comprise but are not limited to: a power management circuit, a clock circuit, and a temperature interface circuit, and the plurality of sensors comprise but are not limited to: a voltage sensor, a current sensor, a temperature sensor, and a frequency sensor, etc.

[0043] As shown in Figure 2 , the voltage sensor corresponds to monitor a hardware monitoring parameter of the power management circuit, for example, the hardware monitoring parameter comprises: a plurality of voltage parameters of the power management circuit, and the plurality of voltage parameters comprise but are not limited to: a core voltage parameter, a display memory voltage parameter, an input voltage parameter, and an auxiliary voltage parameter of the power management circuit. That is, the voltage sensor monitors the plurality of voltage parameters to ensure that they are within the normal voltage range.

[0044] In Figure 2 , the current sensor corresponds to monitor a hardware monitoring parameter of the power management circuit, for example, the hardware monitoring parameter comprises: an input current parameter of the power management circuit. That is, the current sensor can monitor the input current parameter of the power management circuit to determine whether there is a short circuit or overload condition, and once an abnormally large current is monitored, an alarm will be sent out quickly. The power management chip in the power management circuit will immediately cut off the power supply or take current limiting measures to prevent serious failure caused by overheating or component damage.

[0045] In Figure 2 , the temperature sensor corresponds to monitor a hardware monitoring parameter of the temperature interface circuit, and the hardware monitoring parameter comprises: a current interface temperature of the temperature interface circuit. That is, the temperature sensor monitors the temperature of the graphics card core chip, the display memory, and other key components in the temperature interface circuit in real time to prevent failure caused by excessively high temperature.

[0046] InFigure 2 In some embodiments, the frequency sensor corresponds to monitoring a hardware monitoring parameter of a clock circuit, which includes but is not limited to: core frequency and display memory frequency of a graphics card, and communication frequency between the graphics card and the server. That is, the frequency sensor is used to monitor the frequency of the graphics card clock signal.

[0047] The abnormal information monitoring method of the graphics card initialization process in the embodiment can be specifically referred to the following implementation. The graphics card initialization process in the embodiment is configured with a VBIOS software system in the control unit of the graphics card, and a BIOS software system in the server. The baseboard management controller (BMC) in the embodiment interacts with the graphics card, and can obtain the abnormal information of the graphics card initialization process. At the same time, the server interacts with the graphics card. The server not only has the function of obtaining the abnormal information of the graphics card initialization process, but also has the functions of allocating data resources for the graphics card and identifying the target graphics card model parameter of the graphics card.

[0048] The abnormal information monitoring system of the graphics card initialization process in the embodiment can quickly realize the abnormal monitoring of the target monitoring parameter and store the abnormal record based on the graphics card initialization abnormal monitoring strategy corresponding to the target monitoring parameter in the graphics card initialization monitoring process by executing the abnormal information monitoring method of the graphics card initialization process of the graphics card. In turn, it is beneficial to quickly locate the abnormal reason, facilitate the graphics card detection personnel to repair the abnormality in time, thereby reducing the graphics card security risk, and ultimately improving the user experience.

[0049] The present application provides a kind of graphics card initialization process of abnormal information monitoring method, it can be used in computer equipment, such as mobile phone, tablet computer, desktop computer, portable notebook, server, Figure 3 It is the flow chart of the graphics card initialization process of abnormal information monitoring method according to the embodiment of the application, the method is applied to the abnormal monitoring system of the graphics card initialization process in the above embodiment, such as Figure 3 As shown, the flow includes the following steps:

[0050] Step S301, in response to the start of the graphics card, the monitoring action of the graphics card initialization process is executed by the video basic input and output software system.

[0051] Specifically, when the graphics card is started, a plurality of sensors are controlled by the VBIOS software system in the graphics card to perform monitoring actions on the key hardware modules of the graphics card. The key hardware modules of the graphics card include but are not limited to power management circuit, clock circuit and temperature interface circuit, etc. For example, the VBIOS software system in the graphics card is used to perform monitoring actions on the key software parameters of the graphics card, which include but are not limited to memory parameters, resource demand parameters and software system setting parameters, etc.

[0052] Step S302, obtaining the target monitoring parameter in the graphic card initialization monitoring process, the target monitoring parameter including: hardware monitoring parameter and software monitoring parameter.

[0053] Specifically, the hardware monitoring parameter and the software monitoring parameter can be multiple monitoring parameters. The multiple hardware monitoring parameters include hardware monitoring parameters of hardware circuits monitored by multiple sensors and a target graphic card model parameter of the monitored graphic card. The multiple software monitoring parameters include a target resource requirement parameter of the graphic card and a target software system setting parameter.

[0054] The hardware monitoring parameter in the embodiment mainly includes: a hardware self-checking monitoring parameter, a target graphic card model parameter, a target graphic memory parameter, and a frequency monitoring parameter. The software monitoring parameter includes: a target resource requirement parameter and a target software system setting parameter.

[0055] In a specific example, when the hardware monitoring parameter is a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter includes: a target power supply parameter, a target current parameter, a target communication frequency, and a target temperature parameter, and the preset configuration parameter includes: a preset power supply parameter, a preset current parameter, and a preset communication frequency.

[0056] For example, the target power supply parameter includes but is not limited to a core voltage parameter, a graphic memory voltage parameter, an input voltage parameter, and an auxiliary voltage parameter. The core voltage parameter is a core parameter of the graphic card operation, and the stability of the power supply voltage is required to be high. The preset voltage parameter, i.e., the normal working voltage range, is usually between 0.8V and 1.2V. The target power supply parameter is obtained through a voltage sensor.

[0057] For example, the target current parameter is an input current parameter, which is usually less than the design value of the overcurrent protection. The target current parameter is obtained through a current sensor.

[0058] For example, the target communication frequency is usually designed in the boot stage. The basic frequency is about 200Hz, and the basic frequency is the preset communication frequency. The target communication frequency is obtained through a frequency sensor.

[0059] For example, the target temperature parameter has a preset temperature parameter of 90℃, and the preset temperature parameter is usually required to be 80℃. The target temperature parameter is obtained through a temperature sensor.

[0060] In another specific example, when the hardware monitoring parameter is a target graphic card model parameter, the corresponding preset configuration parameter includes a preset graphic card model parameter.

[0061] The target graphic card model parameter includes a manufacturer identity and a graphic card identity.

[0062] In another specific example, when the software monitoring parameter is the target memory parameter, the preset configuration parameter corresponding thereto includes the preset memory parameter.

[0063] For example, the target memory parameter is 8GB or 16GB, and the preset value corresponding thereto is the preset memory parameter.

[0064] In another specific example, when the hardware monitoring parameter is the frequency monitoring parameter, the frequency monitoring parameter includes the target memory verification frequency and the target bus frequency, and the preset configuration parameter includes the preset memory verification frequency and the preset bus frequency.

[0065] In another specific example, when the software monitoring parameter is the target resource requirement parameter, the preset configuration parameter corresponding thereto includes the preset resource configuration parameter.

[0066] In another specific example, when the hardware monitoring parameter is the target software system setting parameter, the preset configuration parameter corresponding thereto is the preset software system requirement parameter.

[0067] For example, the VBIOS software system in the graphics card interacts with the BIOS software system in the server, the server sends the preset software system requirement parameter to the graphics card, and the graphics card modifies the target software setting parameter in combination with the preset software requirement parameter.

[0068] Step S303, based on the graphics card initialization abnormal monitoring strategy of the target monitoring parameter, it is judged whether the target monitoring parameter is abnormal, wherein the graphics card initialization abnormal monitoring strategy of the target monitoring parameter is generated by the video basic input and output software system according to the preset configuration parameter of the target monitoring parameter.

[0069] The graphics card initialization abnormal detection strategy is generated in advance by the VBIOS system target monitoring parameter preset configuration parameter. For the graphics card initialization abnormal detection strategy of the hardware monitoring parameter and the software monitoring parameter in the target monitoring parameter, see the following embodiments.

[0070] Step S304, when the target monitoring parameter is abnormal, the abnormal information is stored in the storage unit according to the abnormal marking mode.

[0071] In some specific embodiments, the above step S304, when the target monitoring parameter is abnormal, the abnormal information is stored in the storage unit according to the abnormal marking mode, including:

[0072] Step a1, when the target detection parameter is abnormal, according to the abnormal information, the abnormal event is recorded according to the abnormal time point;

[0073] Step a2, determine the abnormal identification corresponding to the abnormal event;

[0074] Step a3, according to the abnormal identifier corresponding to the abnormal event, the abnormal identifier and the abnormal information are stored in the abnormal data table of the storage unit.

[0075] For example, when the target monitoring parameter is abnormal, according to the abnormal information of the hardware monitoring parameter and / or the software monitoring parameter in the target monitoring parameter, the abnormal event is recorded according to the abnormal time point, and then according to the abnormal identifier corresponding to the abnormal event, the abnormal identifier corresponding to the abnormal event can be a standard error code, and the standard error code is defined, such as 0x01=power supply abnormal event, 0x03=video memory abnormal event. The abnormal identifier corresponding to the abnormal event is stored in the abnormal data table of the storage unit.

[0076] The abnormal event in the embodiment can be accessed through the BIOS software of the server or through the video card driver, and the data can be obtained and displayed in the system log, and the data can also be obtained through a separate data line.

[0077] When the target monitoring parameter is abnormal, the abnormal information is stored in the storage unit according to the abnormal marking mode, which is beneficial to complete the abnormal record and facilitate the maintenance personnel to repair the abnormality in time.

[0078] When the target monitoring parameter is normal, it is determined that the video card initialization process is normal.

[0079] In summary, the abnormal information monitoring method of the video card initialization process in the embodiment can quickly realize the abnormal monitoring of the target monitoring parameter and store the abnormal record based on the video card initialization abnormal monitoring strategy corresponding to the target monitoring parameter in the video card initialization monitoring process, which is beneficial to quickly locate the abnormal reason, facilitate the video card detection personnel to repair the abnormality in time, and reduce the video card safety risk, and ultimately improve the user experience.

[0080] The application provides an abnormal information monitoring method of a video card initialization process, which can be applied to computer equipment such as mobile phones, tablet computers, desktop computers, portable notebooks, servers and the like, Figure 4 is a flowchart of the abnormal information monitoring method of the video card initialization process according to the embodiment of the application, and the method is applied to the abnormal monitoring system of the video card initialization process in the above embodiment, such as Figure 4 As shown in the figure, the step S303 is based on the video card initialization abnormal monitoring strategy of the target monitoring parameter to determine whether the target monitoring parameter is abnormal, and the process includes the following steps:

[0081] In a specific example, the software monitoring parameter includes a target resource requirement parameter, and the preset configuration parameter includes a preset resource configuration parameter.

[0082] Step S401, sending an access request of a to-be-accessed resource to the server, so that the server marks a memory region belonging to the graphics card in a memory mapping table configured with preset resource configuration parameters according to identity information of the graphics card and resource demand parameters carried in the access request, and generates an access response of the to-be-accessed resource, the access response carrying a resource allocation result of the to-be-accessed resource and a resource loading instruction.

[0083] After the interaction between the graphics card and the server, the server obtains the identity information of the graphics card and the resource demand parameters through access to the configuration space, and the resource demand parameters include information such as a required memory address range.

[0084] The server marks a memory region allocated to the graphics card in the memory mapping table according to the access request of the to-be-accessed resource sent by the graphics card, and records information such as a start address and a size.

[0085] Step S402, when the resource loading instruction is triggered, it is determined that the target resource demand parameter is abnormal if the to-be-accessed resource fails to be loaded through the resource allocation result of the to-be-accessed resource.

[0086] Step S403, when the resource loading instruction is triggered, it is determined that the target resource demand parameter is normal if the to-be-accessed resource is successfully loaded through the resource allocation result of the to-be-accessed resource, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0087] The resource loading instruction is equivalent to starting the graphics card to read the resource allocation result. When the graphics card is triggered by the resource loading instruction and fails to load the to-be-accessed resource through the resource allocation result of the to-be-accessed resource, it is determined that the target resource demand parameter is abnormal, otherwise, the target resource demand parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order. For example, the preset order is to monitor the target resource demand parameter in the first step, monitor the hardware self-check parameter in the second step, monitor the target graphics card model parameter in the third step, monitor the target graphics memory parameter in the fourth step, and monitor the frequency parameter in the fifth step. The preset order can be flexibly set according to actual conditions.

[0088] In the embodiment, the steps S401-S403 are executed. Since the graphics card and the server interact with each other, the graphics card sends an access request of a to-be-accessed resource to the server, so that the server generates an access response carrying a resource allocation result of the to-be-accessed resource and a resource loading instruction, and then the graphics card can quickly determine the abnormal condition of the target resource demand parameter in the graphics card initialization process.

[0089] In some optional embodiments, the hardware monitoring parameter includes a hardware self-check monitoring parameter, the hardware self-check monitoring parameter includes a target power supply parameter, the preset configuration parameter includes a preset power supply parameter, and the graphics card initialization abnormality monitoring strategy based on the target monitoring parameter includes:

[0090] Step b1, when the target power supply parameter is less than the preset power supply parameter for more than a preset number of times during multiple monitoring processes within a continuous time, it is determined that the target power supply parameter is abnormal.

[0091] Specifically, the target power supply parameter includes but is not limited to: external power supply parameter, core voltage parameter, display memory voltage parameter and input voltage parameter, etc. The preset power supply parameter is a hardware power supply parameter configured in advance, which can be flexibly set in combination with the actual situation.

[0092] Step b2, when the target power supply parameter is less than the preset power supply parameter for no more than a preset number of times during multiple monitoring processes within a continuous time, it is determined that the target power supply parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0093] The preset number of times can be 3 or 5, and the preset time can be 1 min. The preset number of times and the preset time can be flexibly set in combination with the actual situation. For example, when the target power supply parameter is less than the preset power supply parameter for more than 5 times during 5 monitoring processes within a continuous time, it is determined that the target power supply parameter is abnormal. Conversely, when the target power supply parameter is less than the preset power supply parameter for no more than 5 times, it is determined that the target power supply parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order. For another example, the preset order can also be the following order: first, monitor the hardware self-check parameter, second, monitor the target graphics card model parameter, third, monitor the target display memory parameter, fourth, monitor the frequency parameter, and fifth, monitor the target resource demand parameter.

[0094] In some optional embodiments, the hardware monitoring parameter includes: a hardware self-check monitoring parameter, the hardware self-check monitoring parameter includes: a target current parameter, the preset configuration parameter includes: a preset current parameter, the graphics card initialization abnormality monitoring strategy based on the target monitoring parameter includes:

[0095] Step c1, when the target current parameter is greater than the preset current parameter within a preset time, it is determined that the target current parameter is abnormal.

[0096] Step c2, when the target current parameter is not greater than the preset current parameter within a preset time, it is determined that the target current parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0097] Specifically, the target current parameter can be an input current parameter. The preset current parameter is the design value of the overcurrent protection.

[0098] For example, the preset time can be 100us, when 100us exceeds the preset current parameter, it is determined that the target current parameter is abnormal. On the contrary, it is determined that the target current parameter is abnormal, and whether the next target monitoring parameter is abnormal is determined according to the preset order. The preset order has been described above, and will not be described here.

[0099] In some optional embodiments, the hardware monitoring parameter includes a hardware self-check monitoring parameter, the hardware self-check monitoring parameter includes a target communication frequency and a target temperature parameter, the preset configuration parameter includes a preset communication frequency and a preset temperature parameter, the GPU initialization abnormality monitoring strategy based on the target monitoring parameter includes:

[0100] Step d1, when the target communication frequency in the multiple monitoring processes, the target communication frequency of any time is inconsistent with the preset communication frequency, it is determined that the target communication frequency is abnormal.

[0101] Step d2, when the target communication frequency in the multiple monitoring processes, the target communication frequency of any time is consistent with the preset communication frequency, it is determined that the target communication frequency is normal, and whether the next target monitoring parameter is abnormal is determined according to the preset order.

[0102] The preset communication frequency represents the communication frequency when the GPU interacts with the server through the VBIOS software system and the BIOS software system, which is usually about 200Hz. For example, in multiple monitoring, the target communication frequency is matched with the preset communication frequency, when the two are inconsistent, it is determined that the target communication frequency is abnormal, on the contrary, it is determined that the target communication frequency is normal. And whether the next target monitoring parameter is abnormal is determined according to the preset order. The preset order has been described above, and will not be described here.

[0103] Step d3, when the target temperature parameter is greater than the preset temperature parameter, it is determined that the target temperature parameter is abnormal.

[0104] Step d4, when the target temperature parameter is not greater than the preset temperature parameter, it is determined that the target temperature parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to the preset order.

[0105] Specifically, the target temperature parameter includes but is not limited to core temperature and video memory temperature. When the target temperature parameter is core temperature, the corresponding preset temperature parameter can be 90℃, and when the target temperature is video memory temperature, the corresponding preset temperature parameter can be 80℃. When the target temperature is greater than the preset temperature parameter, it is determined that the target temperature parameter is normal, on the contrary, it is determined that the target temperature parameter is abnormal. The preset order has been described above, and will not be described here.

[0106] The embodiment can quickly determine the abnormal condition of the hardware self-checking monitoring parameter in the initialization process of the graphics card by the graphics card initialization abnormality monitoring strategy of the hardware self-checking monitoring parameter.

[0107] The application provides an abnormal information monitoring method of a graphics card initialization process, which can be applied to a computer device, such as a mobile phone, a tablet computer, a desktop computer, a portable notebook computer, a server, and the like. Figure 5 is a flowchart of the abnormal information monitoring method of the graphics card initialization process according to the embodiment of the application, and the method is applied to the abnormal monitoring system of the graphics card initialization process in the above embodiment, such as Figure 5 As shown in the above step S303, the graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, and the process includes the following steps:

[0108] The software monitoring parameter includes a target graphics card model parameter, and the preset configuration parameter includes a preset graphics card model parameter.

[0109] Step S501 receives a request instruction for identifying the target graphics card model parameter sent by the server.

[0110] For example, the server sends the request instruction for identifying the target graphics card model parameter to the graphics card through a peripheral component interconnect express (PCIe) bus.

[0111] Step S502 sends the target graphics card model parameter to the server according to the request instruction, so that the server matches the target graphics card model parameter and returns the matching result to the graphics card.

[0112] Specifically, the server sends a reading instruction of the target graphics card model parameter to the graphics card through a BIOS software system and a PCIe bus, accesses a configuration register, obtains the target graphics card model parameter such as an identity document (ID) and a hardware device ID of the graphics card, matches the target graphics card model parameter with a preset graphics card model parameter built in the BIOS software system, and returns the matching result to the graphics card.

[0113] Step S503 determines that the target graphics card model parameter is abnormal when the matching result indicates that the target graphics card model parameter fails to match the preset graphics card model parameter and the current identification time is greater than a preset identification time.

[0114] Step S504 determines that the target graphics card model parameter is normal when the matching result indicates that the target graphics card model parameter successfully matches the preset graphics card model parameter and the current identification time is not greater than the preset identification time, and determines whether the next target monitoring parameter is abnormal according to a preset order.

[0115] For example, when the target graphics card model parameter fails to match the preset graphics card model parameter, i.e., the target graphics card model parameter does not match the preset graphics card model parameter, and the current identification time is greater than the preset identification time, which can be 500 ms, it is determined that the target graphics card model parameter is abnormal, otherwise, it is determined that the target graphics card model parameter is normal, and the next target monitoring parameter is determined whether abnormal according to the preset order. The preset order has been described above and will not be repeated here.

[0116] By executing the above steps S501-S504, the server combines the target graphics card model parameter sent by the graphics card with the preset graphics card model parameter for matching, and sends the matching result to the graphics card, which is beneficial to the graphics card to quickly determine the abnormality of the target hardware model parameter in the graphics card initialization process.

[0117] The application provides an abnormal information monitoring method for a graphics card initialization process, which can be applied to computer equipment such as mobile phones, tablet computers, desktop computers, portable notebooks, servers, etc. Figure 6 is a flowchart of the abnormal information monitoring method for the graphics card initialization process according to an embodiment of the application, which is applied to the abnormal monitoring system for the graphics card initialization process in the above embodiment, such as Figure 6 As shown in the above step S303, the target monitoring parameter is determined whether abnormal based on the graphics card initialization abnormal monitoring strategy of the target monitoring parameter, and the flowchart includes the following steps:

[0118] In a specific example, the software monitoring parameter includes a target graphics memory parameter, and the preset configuration parameter includes a preset graphics memory parameter.

[0119] Step S601: An initialization instruction is sent to a graphics memory unit, and the initialization instruction is used to instruct a control unit to establish a communication link with the graphics memory unit.

[0120] Specifically, the graphics memory unit can be a graphics memory particle, and the control unit is a control center of the graphics card.

[0121] Step S602: In response to that the control unit and the graphics memory unit successfully establish the communication link, a physical address space of the graphics memory unit is scanned.

[0122] Step S603: A target graphics memory parameter is obtained from the physical address space of the graphics memory unit.

[0123] For example, the target graphics memory parameter is 8 GB, 16 GB, etc.

[0124] Step S604: When a current error between the target graphics memory parameter and the preset graphics memory parameter is not less than a preset error, it is determined that the target graphics memory parameter is abnormal.

[0125] Step S605, when the current error between the target memory parameter and the preset memory parameter is less than the preset error, it is determined that the target memory parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to the preset order.

[0126] For example, if the current error between the target memory parameter and the preset memory parameter is not greater than 0.5%, it is determined that the target memory unit is abnormal, otherwise, it is determined that the target memory unit is normal, and whether the next target monitoring parameter is abnormal is determined according to the preset order.

[0127] The embodiment can quickly determine the abnormality of the target memory parameter in the initialization process of the graphics card by comparing the target memory parameter obtained from the physical address space of the memory unit with the preset memory parameter through the communication link between the memory unit and the control unit.

[0128] The present application provides an abnormal information monitoring method for the initialization process of a graphics card, which can be applied to computer equipment such as mobile phones, tablet computers, desktop computers, portable notebooks, servers, etc. Figure 7 is a flowchart of the abnormal information monitoring method for the initialization process of a graphics card according to an embodiment of the present application. The method is applied to the abnormal monitoring system for the initialization process of a graphics card in the above embodiment, such as Figure 7 As shown in the above step S303, the target monitoring parameter is determined to be abnormal or not based on the graphics card initialization abnormal monitoring strategy of the target monitoring parameter. The flowchart includes the following steps:

[0129] The hardware monitoring parameters include frequency monitoring parameters, and the frequency monitoring parameters include target memory verification frequency and target bus frequency. The preset configuration parameters include preset communication frequency, preset memory verification frequency and preset bus frequency.

[0130] Step S701, according to the preset memory verification frequency, the core clock signal of the graphics card is controlled to be synchronized with the reference clock signal of the mainboard.

[0131] The preset memory verification frequency is the core frequency of the graphics card. The graphics card initializes the core frequency controller of the graphics card through the VBIOS software system, configures the preset memory verification frequency, and the preset memory verification frequency is usually the base frequency. Through the preset memory verification frequency, the core clock signal of the graphics card is controlled to be synchronized with the reference clock signal of the mainboard.

[0132] Step S702, the target memory verification frequency is matched with the preset memory verification frequency, and the target bus frequency is matched with the preset bus frequency.

[0133] The target bus frequency can be a PCIe bus.

[0134] Step S703, when the matching result between the target memory check frequency and the preset memory check frequency, and the matching result between the target bus frequency and the preset bus frequency do not satisfy the preset condition, it is determined that the target memory check frequency and the target bus frequency are abnormal.

[0135] Step S704, when the matching result between the target memory check frequency and the preset memory check frequency, and the matching result between the target bus frequency and the preset bus frequency satisfy the preset condition, it is determined that the target memory check frequency and the target bus frequency are normal, and the next target monitoring parameter is judged whether abnormal according to the preset order.

[0136] For example, the ratio between the target memory check frequency and the preset memory check frequency meets the design specification, and the target bus frequency satisfies the stability requirement, the error between the target memory check frequency and the preset memory check frequency is less than the preset error (±0.5%), and the synchronization time between the target memory check frequency and the preset memory check frequency is less than or equal to the time threshold (600ms). That is, when the matching result between the target memory check frequency and the preset memory check frequency, and the matching result between the target bus frequency and the preset bus frequency satisfy the preset condition, it is determined that the target memory check frequency and the target bus frequency are normal, and the next target monitoring parameter is judged whether abnormal according to the preset order. Otherwise, it is determined that the target memory check frequency and the target bus frequency are abnormal. The preset order has been described above and will not be repeated here.

[0137] The embodiment can quickly determine the abnormality of the frequency monitoring parameter in the initialization process of the graphics card by executing the above steps S701-S704 through the graphics card initialization abnormality monitoring strategy of the frequency monitoring parameter.

[0138] The application provides an abnormal information monitoring method for the initialization process of a graphics card, which can be applied to computer equipment such as mobile phones, tablet computers, desktop computers, portable notebooks, servers, etc. Figure 8 is a flowchart of the abnormal information monitoring method for the initialization process of a graphics card according to an embodiment of the application, which is applied to the abnormal monitoring system for the initialization process of a graphics card in the above embodiment, such as Figure 8 As shown in the above step S303, the graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to judge whether the target monitoring parameter is abnormal, and the flowchart includes the following steps:

[0139] In a specific example, the software monitoring parameter includes a target software system setting parameter, and the preset configuration parameter includes a preset software system requirement parameter.

[0140] Step S801, receiving a preset software system requirement parameter loading request sent by a server.

[0141] The preset software system requirement parameters include, but are not limited to, a target rate, a power management state setting parameter, a maximum value of a payload, a program loading parameter, and the like.

[0142] In step S802, according to the preset software system requirement parameters, when the target software setting parameter of the configuration register fails to be modified, it is determined that the target software setting parameter is abnormal.

[0143] In step S803, according to the preset software system requirement parameters, when the target software setting parameter of the configuration register is successfully modified, it is determined that the target software setting parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0144] In the process of interaction between the server and the graphics card, the server transmits the requirement to the graphics card in the form of an instruction according to the preset software system requirement parameters, the graphics card modifies the target software setting parameter in the configuration register, when the target software setting parameter of the configuration register fails to be modified, it is determined that the target software setting parameter is abnormal, otherwise, it is determined that the target resource requirement parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0145] By executing the above steps S801-S803, the graphics card initialization abnormal monitoring strategy of the target resource requirement parameter can quickly determine the abnormal condition of the target resource requirement parameter in the graphics card initialization process.

[0146] The embodiment of the application also provides an abnormal information monitoring device of a graphics card initialization process, as shown in Figure 9 The device is applied to a graphics card, and the graphics card includes a plurality of hardware circuits, a storage unit, a plurality of sensors, and a control unit, wherein each sensor corresponds to a hardware monitoring parameter of a hardware circuit, and a video basic input and output software system is configured in the control unit.

[0147] The monitoring action execution module 901 is configured to execute a monitoring action of a graphics card initialization process through a video basic input and output software system in response to starting of the graphics card.

[0148] The monitoring data acquisition module 902 is configured to acquire a target monitoring parameter in a graphics card initialization monitoring process, and the target monitoring parameter includes a hardware monitoring parameter and a software monitoring parameter.

[0149] The abnormal condition determination module 903 is configured to determine whether the target monitoring parameter is abnormal based on a graphics card initialization abnormal monitoring strategy of the target monitoring parameter, and the graphics card initialization abnormal monitoring strategy of the target monitoring parameter is generated by the video basic input and output software system according to preset configuration parameters of the target monitoring parameter.

[0150] The abnormal information storage module 904 is configured to store the abnormal information to the storage unit according to the abnormal marking mode when the target monitoring parameter is abnormal.

[0151] The features of the embodiments of the abnormal information monitoring device of the graphic card initialization process can be referred to the related descriptions of the embodiments of the abnormal information monitoring method of the graphic card initialization process, which will not be repeated here.

[0152] In some optional embodiments, the software monitoring parameter comprises an abnormal condition determination module 903, which is specifically configured to:

[0153] sending an access request of the to-be-accessed resource to the server, so that the server marks the memory area belonging to the graphic card in the memory mapping table configured with the preset resource configuration parameter according to the identity information of the graphic card and the resource demand parameter carried in the access request, and generates an access response of the to-be-accessed resource, the access response carrying a resource allocation result and a resource loading instruction of the to-be-accessed resource;

[0154] determining that the target resource demand parameter is abnormal when the resource loading instruction is triggered and the to-be-accessed resource fails to be loaded through the resource allocation result of the to-be-accessed resource;

[0155] determining that the target resource demand parameter is normal when the resource loading instruction is triggered and the to-be-accessed resource is successfully loaded through the resource allocation result of the to-be-accessed resource, and judging whether the next target monitoring parameter is abnormal according to a preset order.

[0156] In some optional embodiments, the hardware monitoring parameter comprises a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter comprises a target power supply parameter, the preset configuration parameter comprises a preset power supply parameter, and the abnormal condition determination module 903 is further specifically configured to:

[0157] determining that the target power supply parameter is abnormal when the target power supply parameter is less than the preset power supply parameter for more than a preset number of times in a plurality of monitoring processes in a continuous time;

[0158] determining that the target power supply parameter is normal when the target power supply parameter is less than the preset power supply parameter for not more than a preset number of times in a plurality of monitoring processes in a continuous time, and judging whether the next target monitoring parameter is abnormal according to a preset order.

[0159] In some optional embodiments, the hardware monitoring parameter comprises a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter comprises a target current parameter, the preset configuration parameter comprises a preset current parameter, and the abnormal condition determination module 903 is further specifically configured to:

[0160] determining that the target current parameter is abnormal when the target current parameter is greater than the preset current parameter within a preset time;

[0161] When the target current parameter is not greater than the preset current parameter within the preset time, it is determined that the target current parameter is normal, and it is determined whether the next target monitoring parameter is abnormal according to the preset order.

[0162] In some optional embodiments, the hardware monitoring parameter includes a hardware self-check monitoring parameter, the hardware self-check monitoring parameter includes a target communication frequency and a target temperature parameter, the preset configuration parameter includes a preset communication frequency and a preset temperature parameter, and the abnormal condition determination module 903 is further specifically used for:

[0163] When the target communication frequency is inconsistent with the preset communication frequency at any time in the multiple monitoring processes, it is determined that the target communication frequency is abnormal.

[0164] When the target communication frequency is consistent with the preset communication frequency at any time in the multiple monitoring processes, it is determined that the target communication frequency is normal, and it is determined whether the next target monitoring parameter is abnormal according to the preset order.

[0165] When the target temperature parameter is greater than the preset temperature parameter, it is determined that the target temperature parameter is abnormal.

[0166] When the target temperature parameter is not greater than the preset temperature parameter, it is determined that the target temperature parameter is normal, and it is determined whether the next target monitoring parameter is abnormal according to the preset order.

[0167] In some optional embodiments, the software monitoring parameter includes a target graphics card model parameter, the preset configuration parameter includes a preset graphics card model parameter, and the abnormal condition determination module 903 is further specifically used for:

[0168] receiving a request instruction sent by the server for identifying the target graphics card model parameter;

[0169] According to the request instruction, the target graphics card model parameter is sent to the server, so that the server matches the preset graphics card model parameter with the target graphics card model parameter, and returns the matching result to the graphics card.

[0170] When the matching result prompts that the target graphics card model parameter fails to match the preset graphics card model parameter, and the current identification time is greater than the preset identification time, it is determined that the target graphics card model parameter is abnormal.

[0171] When the matching result prompts that the target graphics card model parameter matches the preset graphics card model parameter, and the current identification time is not greater than the preset identification time, it is determined that the target graphics card model parameter is normal, and it is determined whether the next target monitoring parameter is abnormal according to the preset order.

[0172] In some optional embodiments, the software monitoring parameter includes a target graphics memory parameter, the preset configuration parameter includes a preset graphics memory parameter, and the abnormal condition determination module 903 is further specifically used for:

[0173] sending an initialization instruction to the display memory unit, the initialization instruction being used to instruct the control unit to establish a communication link with the display memory unit;

[0174] in response to the control unit successfully establishing the communication link with the display memory unit, scanning a physical address space of the display memory unit;

[0175] obtaining a target display memory parameter from the physical address space of the display memory unit;

[0176] when a current error between the target display memory parameter and the preset display memory parameter is not less than a preset error, determining that the target display memory parameter is abnormal;

[0177] when the current error between the target display memory parameter and the preset display memory parameter is less than the preset error, determining that the target display memory parameter is normal, and judging whether a next target monitoring parameter is abnormal according to a preset order.

[0178] In some optional embodiments, the hardware monitoring parameter includes a frequency monitoring parameter, the frequency monitoring parameter including a target display memory verification frequency and a target bus frequency, the preset configuration parameter including a preset display memory verification frequency and a preset bus frequency, and the abnormal condition determination module 903 is further specifically used for:

[0179] controlling a core clock signal of the graphics card to be synchronized with a reference clock signal of the motherboard according to the preset display memory verification frequency;

[0180] matching the target display memory verification frequency with the preset display memory verification frequency, and matching the target bus frequency with the preset bus frequency;

[0181] when matching results between the target display memory verification frequency and the preset display memory verification frequency and between the target bus frequency and the preset bus frequency do not satisfy a preset condition, determining that the target display memory verification frequency and the target bus frequency are abnormal;

[0182] when the matching results between the target display memory verification frequency and the preset display memory verification frequency and between the target bus frequency and the preset bus frequency satisfy the preset condition, determining that the target display memory verification frequency and the target bus frequency are normal, and judging whether a next target monitoring parameter is abnormal according to a preset order.

[0183] In some optional embodiments, the software monitoring parameter includes a target software system setting parameter, the preset configuration parameter includes a preset software system requirement parameter, and the abnormal condition determination module 903 is further specifically used for:

[0184] receiving a preset software system requirement parameter loading request sent by the server;

[0185] According to the preset software system requirement parameters, when the target software setting parameter of the configuration register fails to be modified, it is determined that the target software setting parameter is abnormal;

[0186] According to the preset software system requirement parameters, when the target software setting parameter of the configuration register is successfully modified, it is determined that the target software setting parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

[0187] The features of the embodiments of the abnormal information monitoring device of the graphics card initialization process can be referred to the related descriptions of the embodiments of the abnormal information monitoring method of the graphics card initialization process, which will not be repeated here.

[0188] The embodiments of the present application also provide a computer device, as shown in the accompanying drawings, which comprises a memory 10 and a processor 20, the memory 10 stores a computer program, and the processor 20 is configured to run the computer program to execute the steps in any of the above-mentioned embodiments of the abnormal information monitoring method of the graphics card initialization process. Figure 10

[0189] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the steps in any of the above-mentioned embodiments of the abnormal information monitoring method of the graphics card initialization process when running.

[0190] In an exemplary embodiment, the above-mentioned computer readable storage medium can include but is not limited to a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0191] The embodiments of the present application also provide a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps in any of the above-mentioned embodiments of the abnormal information monitoring method of the graphics card initialization process.

[0192] The embodiments of the present application also provide another computer program product, which comprises a non-volatile computer readable storage medium, the non-volatile computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps in any of the above-mentioned embodiments of the abnormal information monitoring method of the graphics card initialization process.

[0193] ​Those skilled in the art will further realize that the mere concepts, teachings, and embodiments described herein are merely meant to provide an enabling description of the claimed application. Accordingly, modifications and / or additions, other than those explicitly described herein, can be obvious to those skilled in the art in the light of this disclosure. The claimed application is intended to embrace all such modifications and / or additions.

[0194] The above has carried out the detailed introduction to the abnormal information monitoring of the graphic card initialization process provided by the application. The principle and implementation mode of the application are described by applying specific examples in the text. The above example description is only for helping to understand the method and its core idea of the application. It should be pointed out that, for the ordinary skilled in the art, some improvements and modifications can be made to the application without departing from the principle of the application. These improvements and modifications also fall within the protection scope of the claims of the application.

Claims

1. A method for monitoring exception information of a graphic card initialization process, characterized in that, The method is applied to a graphics card, the graphics card comprising: a plurality of hardware circuits, a storage unit, a plurality of sensors and a control unit, wherein each sensor corresponds to monitor a hardware monitoring parameter of a hardware circuit, a video basic input and output software system is configured in the control unit, the method comprising: In response to the start of the graphics card, performing a monitoring action of a graphics card initialization process by the video basic input and output software system; Obtaining a target monitoring parameter in the graphics card initialization monitoring process, the target monitoring parameter comprising: a hardware monitoring parameter and a software monitoring parameter; Judging whether the target monitoring parameter is abnormal based on a graphics card initialization abnormal monitoring strategy of the target monitoring parameter, wherein the graphics card initialization abnormal monitoring strategy of the target monitoring parameter is generated by the video basic input and output software system according to preset configuration parameters of the target monitoring parameter; When the target monitoring parameter is abnormal, storing abnormal information to the storage unit according to an abnormal marking mode.

2. The method of claim 1, wherein, The software monitoring parameter comprises: a target resource requirement parameter, the preset configuration parameter comprises: a preset resource configuration parameter, judging whether the target monitoring parameter is abnormal based on the graphics card initialization abnormal monitoring strategy of the target monitoring parameter, comprising: Sending an access request of a to-be-accessed resource to a server, so that the server marks a memory area belonging to the graphics card in a memory mapping table configured with the preset resource configuration parameter and generates an access response of the to-be-accessed resource according to identity information of the graphics card and the resource requirement parameter carried in the access request, the access response carrying a resource allocation result and a resource loading instruction of the to-be-accessed resource; When the resource loading instruction is triggered, it is determined that the target resource requirement parameter is abnormal if the to-be-accessed resource fails to be loaded through the resource allocation result of the to-be-accessed resource; When the resource loading instruction is triggered, it is determined that the target resource requirement parameter is normal if the to-be-accessed resource is successfully loaded through the resource allocation result of the to-be-accessed resource, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

3. The method of claim 1, wherein, The hardware monitoring parameter comprises: a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter comprising: a target power supply parameter, the preset configuration parameter comprising: a preset power supply parameter, judging whether the target monitoring parameter is abnormal based on the graphics card initialization abnormal monitoring strategy of the target monitoring parameter, comprising: When the target power supply parameter is less than the preset power supply parameter for more than a preset number of times in a plurality of monitoring processes within a continuous time, it is determined that the target power supply parameter is abnormal; When the target power supply parameter is less than the preset power supply parameter for not more than a preset number of times in a plurality of monitoring processes within a continuous time, it is determined that the target power supply parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

4. The method of claim 1, wherein, The hardware monitoring parameter includes a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter includes a target current parameter, the preset configuration parameter includes a preset current parameter, a graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, including: When the target current parameter is greater than the preset current parameter within a preset time, it is determined that the target current parameter is abnormal; When the target current parameter is not greater than the preset current parameter within a preset time, it is determined that the target current parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

5. The method of claim 1, wherein, The hardware monitoring parameter includes a hardware self-checking monitoring parameter, the hardware self-checking monitoring parameter includes a target communication frequency and a target temperature parameter, the preset configuration parameter includes a preset communication frequency and a preset temperature parameter, a graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, including: When the target communication frequency is inconsistent with the preset communication frequency in any one of the multiple monitoring processes, it is determined that the target communication frequency is abnormal; When the target communication frequency is consistent with the preset communication frequency in any one of the multiple monitoring processes, it is determined that the target communication frequency is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order; When the target temperature parameter is greater than the preset temperature parameter, it is determined that the target temperature parameter is abnormal; When the target temperature parameter is not greater than the preset temperature parameter, it is determined that the target temperature parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

6. The method of claim 1, wherein, The software monitoring parameter includes a target graphics card model parameter, the preset configuration parameter includes a preset graphics card model parameter, a graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, including: Receiving a request instruction sent by a server for identifying the target graphics card model parameter; According to the request instruction, the target graphics card model parameter is sent to the server, so that the server matches the preset graphics card model parameter with the target graphics card model parameter, and returns the matching result to the graphics card; When the matching result prompts that the target graphics card model parameter fails to match the preset graphics card model parameter, and the current identification time is greater than a preset identification time, it is determined that the target graphics card model parameter is abnormal; When the matching result prompts that the target graphics card model parameter matches the preset graphics card model parameter successfully, and the current identification time is not greater than the preset identification time, it is determined that the target graphics card model parameter is normal, and whether the next target monitoring parameter is abnormal is determined according to a preset order.

7. The method of claim 1, wherein, The software monitoring parameter includes a target graphics memory parameter, the preset configuration parameter includes a preset graphics memory parameter, a graphics card initialization abnormality monitoring strategy based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, including: Sending an initialization instruction to a graphics memory unit, the initialization instruction is used to instruct a control unit to establish a communication link with the graphics memory unit; In response to the control unit and the display memory unit successfully establishing a communication link, scanning the physical address space of the display memory unit; Obtaining the target display memory parameter from the physical address space of the display memory unit; When the current error between the target display memory parameter and the preset display memory parameter is not less than the preset error, it is determined that the target display memory parameter is abnormal; When the current error between the target display memory parameter and the preset display memory parameter is less than the preset error, it is determined that the target display memory parameter is normal, and whether the next target monitoring parameter is abnormal is determined in a preset order.

8. The method of claim 1, wherein, The hardware monitoring parameter includes a frequency monitoring parameter, and the frequency monitoring parameter includes a target display memory verification frequency and a target bus frequency. The preset configuration parameter includes a preset display memory verification frequency and a preset bus frequency. The abnormal monitoring strategy of the graphics card initialization based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, which includes: According to the preset display memory verification frequency, the core clock signal of the graphics card is controlled to be synchronized with the reference clock signal of the motherboard; The target display memory verification frequency is matched with the preset display memory verification frequency, and the target bus frequency is matched with the preset bus frequency; When the matching results between the target display memory verification frequency and the preset display memory verification frequency, and the matching results between the target bus frequency and the preset bus frequency do not satisfy the preset condition, it is determined that the target display memory verification frequency and the target bus frequency are abnormal; When the matching results between the target display memory verification frequency and the preset display memory verification frequency, and the matching results between the target bus frequency and the preset bus frequency satisfy the preset condition, it is determined that the target display memory verification frequency and the target bus frequency are normal, and whether the next target monitoring parameter is abnormal is determined in a preset order.

9. The method of claim 1, wherein, The software monitoring parameter includes a target software system setting parameter, and the preset configuration parameter includes a preset software system requirement parameter. The abnormal monitoring strategy of the graphics card initialization based on the target monitoring parameter is used to determine whether the target monitoring parameter is abnormal, which includes: Receiving a preset software system requirement parameter loading request sent by a server; According to the preset software system requirement parameter, when the target software setting parameter of the configuration register fails to be modified, it is determined that the target software setting parameter is abnormal; According to the preset software system requirement parameter, when the target software setting parameter of the configuration register is successfully modified, it is determined that the target software setting parameter is normal, and whether the next target monitoring parameter is abnormal is determined in a preset order.

10. An exception information monitoring system for a video card initialization process, comprising: The system includes a graphics card, a baseboard management controller and a server. The graphics card includes a plurality of hardware circuits, a storage unit, a plurality of sensors and a control unit. Each sensor corresponds to a hardware monitoring parameter of a hardware circuit. A video basic input / output software system is configured in the control unit. The server and the baseboard management controller communicate with the graphics card. The graphics card is used to execute the abnormal information monitoring method of the graphics card initialization process in any one of claims 1 to 9.

Images