Compilation method, chip, and computer readable medium
By obtaining the chip's fault processing core information and mapping the algorithm compilation results to the unfailed processing core, the chip's inability to use caused by the fault processing core is solved, and the utilization rate of chip resources is improved and the cost of chip manufacturing is reduced.
Patent Information
- Application Number
- CN202110325748.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-03-26
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2041-03-26
AI Technical Summary
In multi-chip architectures, fault processing cores lead to chip unavailability, resulting in increased chip manufacturing costs, and the entire multi-chip architecture may need to be abandoned, with low resource utilization.
By obtaining the chip's fault processing core information, the algorithm compilation results are mapped to the unfailed processing core, avoiding the fault processing core execution of the algorithm, ensuring that the chip can execute algorithms, improving resource utilization and reducing manufacturing costs.
It effectively avoids the situation where the chip cannot be used due to fault processing cores, improves the utilization rate of chip resources and reduces the cost of chip manufacturing.
Smart Images

Figure CN112882722B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of computer technology, and in particular to a compilation method for a chip, a chip, and a computer-readable medium. Background Art
[0002] A chip (such as an artificial intelligence chip) can be composed of one or more processors, and one processor usually integrates multiple complete computing engines (or processing cores). The multiple processing cores within a processor and the processing cores of different processors can collaborate with each other to complete tasks together.
[0003] In the actual production process, some processing cores of the chip often fail due to manufacturing reasons, and these failed processing cores cannot execute algorithms (or the algorithms cannot be mapped to these failed processing cores).
[0004] In scenarios using a multi-chip architecture (such as a cloud computing center), when a chip with a processing core failure exists in the multi-chip architecture, the chip may not be used, and even the entire multi-chip architecture may need to be discarded, increasing the manufacturing cost of the chip. Summary of the invention
[0005] The present disclosure provides a compilation method for a chip, a chip, and a computer-readable medium.
[0006] In a first aspect, the present disclosure provides a compilation method for a chip, wherein the chip includes multiple processing cores, and the compilation method includes: obtaining faulty processing core information of the chip; and mapping an algorithm compilation result to a non-faulty processing core of the chip according to the faulty processing core information of the chip.
[0007] In a second aspect, the present disclosure provides a chip, comprising: multiple processing cores; an information module for obtaining faulty processing core information of the chip; and a compilation module for mapping algorithm compilation results to non-faulty processing cores of the chip based on the faulty processing core information of the chip.
[0008] In a third aspect, the present disclosure provides a computer-readable medium having a computer program stored thereon, wherein the computer program implements the above-mentioned compilation method when executed by a processing core.
[0009] In the compilation method, chip, and computer-readable medium for a chip provided by the present disclosure, the chip obtains its own fault processing core information, and maps the algorithm compilation result to the processing core of the chip that is not faulty according to the acquired chip fault processing core information. Since the algorithm compilation result is mapped to the processing core of the chip that is not faulty, the processing core of the chip that has already failed will not be mapped with the algorithm compilation result, that is, the processing core of the chip that has already failed does not need to execute the algorithm compilation result, which ensures that the chip can execute the algorithm compilation result, avoids the occurrence of the situation where the chip cannot be used, improves the utilization rate of chip resources, and reduces the cost of chip manufacturing.
[0010] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. The above and other features and advantages will become more apparent to those skilled in the art by describing detailed example embodiments with reference to the accompanying drawings, in which:
[0012] Figure 1 A flowchart of a compilation method for a chip provided by an embodiment of the present disclosure;
[0013] Figure 2 A flowchart of another compilation method for a chip provided by an embodiment of the present disclosure;
[0014] Figure 3 A flowchart of some steps of another compilation method for a chip provided by an embodiment of the present disclosure;
[0015] Figure 4 A flowchart of another compilation method for a chip provided by an embodiment of the present disclosure;
[0016] Figure 5 A flowchart of some steps of another compilation method for a chip provided by an embodiment of the present disclosure;
[0017] Figure 6 A flowchart of some steps of another compilation method for a chip provided by an embodiment of the present disclosure;
[0018] Figure 7 A schematic diagram of a compilation method for a chip provided in an embodiment of the present disclosure;
[0019] Figure 8 A schematic diagram of a compilation method for a chip provided in an embodiment of the present disclosure;
[0020] Fig. 9 A block diagram of the composition of a chip provided in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0021] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.
[0022] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.
[0023] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.
[0024] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "including" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded. "Connected" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.
[0025] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined as such herein.
[0026] Figure 1 A flowchart of a compilation method for a chip provided by an embodiment of the present disclosure.
[0027] Reference Figure 1 , an embodiment of the present disclosure provides a compilation method, which is used for a chip.
[0028] Here, the chip refers to a chip that integrates multiple processing cores (such as a many-core chip or a multi-core chip), that is, a chip has multiple processing cores.
[0029] The compilation method specifically includes:
[0030] S101. Obtain chip fault processing core information.
[0031] The chip has a computing unit, such as an ARM (Advanced RISC Machine) or APU (Accelerated Processing Unit) of the chip to obtain the fault processing core information of the chip.
[0032] Among them, a faulty processing core refers to a processing core in the chip that fails or is temporarily unusable due to various reasons, such as a processing core that fails due to manufacturing reasons, a processing core that fails during use, a processing core that is temporarily unusable due to overheating, errors, etc.
[0033] Since the processing core routing of the chip is not prone to failure, although the faulty processing core does not have computing capabilities (such as the ability to run the executable file corresponding to the algorithm), it still has a high probability of having routing functions and can still transmit information.
[0034] The fault processing core information refers to the information of the fault processing core of the chip, such as the number of the fault processing cores of the chip, the absolute coordinates of the fault processing core in the chip, and an ID (Identity Document).
[0035] After a chip is manufactured, the absolute coordinates and ID information of each processing core have been determined. Before the chip leaves the factory, the chip information will be written into the chip's efuse (one-time programmable memory).
[0036] The information stored in the chip efuse includes the basic information of the chip: such as the power supply voltage that the chip can use, the version number of the chip, the production date, etc. After the chip is produced, the chip will be tested. When the chip is tested, the fault processing core information of the chip (such as the number of fault processing cores, the absolute coordinates of the fault processing core in the chip, ID, etc.) can be obtained, and this information can also be written into the chip efuse.
[0037] The ARM or APU of the chip can obtain the fault processing core information of the chip by reading the efuse (one-time programmable memory) of the chip.
[0038] S102 . Map the algorithm compilation result to the processing core of the chip that is not faulty according to the faulty processing core information of the chip.
[0039] After the ARM or APU of the chip obtains the faulty processing core information of the chip, the algorithm compilation result is mapped to the processing core of the chip that is not faulty according to the faulty processing core information of the chip.
[0040] The algorithm compilation result is a compilation result obtained by compiling the algorithm to be compiled corresponding to the chip.
[0041] Among them, the algorithm to be compiled corresponding to the chip refers to the algorithm to be implemented on the chip. The algorithm to be compiled is the method and process for solving the problem, and is a piece of logic, while the chip is the physical device that actually executes the method and process for solving the problem.
[0042] The chip implements the algorithm to be compiled by running an executable file (such as a binary executable file) corresponding to the algorithm to be compiled and executable by the chip. The process of converting the algorithm to be compiled into an executable file is the process of compiling the algorithm to be compiled. The executable file executable by the chip is the algorithm compilation result. The step of compiling the algorithm to be compiled and obtaining the algorithm compilation result can be completed by a server connected to the chip. After obtaining the algorithm compilation result, the server sends the algorithm compilation result to the chip.
[0043] Since the chip integrates multiple processing cores, it is the processing core of the chip that ultimately executes the algorithm compilation result (ie, the executable file). Therefore, after obtaining the algorithm compilation result, it is necessary to map the algorithm compilation result to the processing core of the chip.
[0044] The ARM or APU of the chip has obtained the fault processing core information of the chip. Therefore, in the process of mapping the algorithm compilation results to the chip, it can map the algorithm compilation results to the processing core of the chip that is not faulty based on the acquired chip fault processing core information.
[0045] Since the algorithm compilation results are mapped to the processing cores of the chip that are not faulty, the processing cores of the chip that have faults will not be mapped with the algorithm compilation results, that is, the processing cores of the chip that have faults do not need to execute the algorithm compilation results, which ensures that the chip can execute the algorithm compilation results, avoids the situation where the chip cannot be used, improves the utilization rate of chip resources, and reduces the cost of chip manufacturing.
[0046] Figure 2 A flowchart of some steps of another compilation method for a chip provided by an embodiment of the present disclosure.
[0047] Reference Figure 2 Another chip compilation method provided by the embodiment of the present disclosure specifically includes:
[0048] S201. Obtain chip fault processing core information.
[0049] The ARM or APU of the chip obtains the fault processing core information of the chip including the number of the fault processing cores and the absolute coordinates of the fault processing cores by reading the efuse of the chip.
[0050] S202: Group the algorithm compilation results according to the number of chip fault processing cores to obtain group results to be mapped.
[0051] The ARM or APU of the chip groups (or folds) the algorithm compilation results according to the number of faulty processing cores of the chip, that is, divides the algorithm compilation results into compilation results (that is, grouped results to be mapped) for execution by a certain number of non-faulty processing cores.
[0052] Among them, the algorithm compilation result can specifically be the compilation result executed by multiple numbers (such as the first number) of processing cores. In general, the first number is greater than the number of non-faulty processing cores of the chip. Therefore, the ARM or APU of the chip groups the compilation results executed by the first number of processing cores according to the number of non-faulty processing cores in the chip, and the compilation results grouped into the same group are executed by a non-faulty processing core of the chip.
[0053] If the algorithm compilation result is an executable file executed by 500 processing cores, and the chip has 100 processing cores, including 6 faulty processing cores and 94 normal processing cores, the chip's ARM or APU groups the algorithm compilation results into 94 groups, and the executable file in each group is executed by a normal processing core, that is, the group result to be mapped can be an executable file for execution by 94 normal processing cores.
[0054] S203: Map the grouping results to be mapped to the processing cores of the chip that are not faulty.
[0055] The ARM or APU of the chip maps the grouping results to be mapped to the processing core corresponding to the absolute coordinates determined by the chip according to the absolute coordinates of the fault processing core of the chip.
[0056] The grouping results to be mapped are grouped according to the number of faulty processing cores. The algorithm compilation results are only divided into compilation results that can be executed by multiple non-faulty processing cores. It has not been determined which non-faulty processing core of the chip will execute each group. Therefore, after obtaining the grouping results to be mapped, the ARM or APU of the chip maps each group in the grouping results to be mapped to a non-faulty processing core of the chip according to the absolute coordinates of the faulty processing core of the chip.
[0057] If a chip has 100 processing cores, including 6 faulty processing cores and 94 normal processing cores, then the group results to be mapped are the compilation results executed by the 94 processing cores. The process of mapping the group results to be mapped to the normal processing cores of the chip is to assign a processing core corresponding to a certain absolute coordinate to each group of the group results to be mapped. For example, the first group of the group results to be mapped is the compilation result executed by the normal processing core with absolute coordinates (1,1), and the second group of the group results to be mapped is the compilation result executed by the normal processing core with absolute coordinates (1,2), and so on.
[0058] In the process of actually executing the compilation results, the processing core with the absolute coordinates (1,1) first executes the corresponding part of the operations in the executable file, and sends the execution results to other processing cores without faults according to the routing file. For example, the processing core with the absolute coordinates (1,2) receives the result. The processing core with the absolute coordinates (1,2) executes the corresponding part of the operations in the executable file according to the result, and so on, until all processing cores without faults have executed the compilation results mapped to them and implemented the algorithm to be compiled.
[0059] As reference Figure 7 , chip (i.e. Figure 7 The largest box in the Figure 7 The chip groups the algorithm compilation results according to the number of faulty processing cores, obtains the group results to be mapped, and maps the group results to be mapped to each non-faulty processing core of the chip.
[0060] Specifically, refer to Figure 8 The ARM or APU of the chip obtains the algorithm compilation result, groups the algorithm compilation result according to the number of faulty processing cores, obtains the group result to be mapped, and maps the current compilation result to each of its own non-faulty processing cores. The algorithm compilation result and the group result to be mapped can be saved in the storage space of the chip, such as DDR (double data rate synchronous dynamic random access memory).
[0061] Reference Figure 3 , mapping the to-be-mapped group result to a processing core of a chip that is not faulty (S203) includes:
[0062] S301 . Map the grouping results to the processing cores of the chip that are not faulty according to the computing capabilities of the processing cores of the chip that are not faulty.
[0063] The ARM or APU of the chip determines the absolute coordinates of the non-faulty processing cores of the chip based on the absolute coordinates of the faulty processing cores of the chip, obtains the computing power of the non-faulty processing cores of the chip, and maps the grouping results to the non-faulty processing cores of the chip based on the computing power and absolute coordinates of the non-faulty processing cores.
[0064] Since multiple processing cores of the chip work together, the process of mapping the grouping results to the processing cores of the chip that are not faulty needs to consider both the rationality of the computing power of each processing core and the efficiency of the collaboration between the processing cores, such as data transmission and information transfer efficiency.
[0065] The rationality of the computing amount of the non-faulty processing cores can be obtained based on the computing power of the non-faulty processing cores. The computing amount allocated to each non-faulty processing core should match the standard computing amount that the non-faulty processing core can provide (that is, the maximum computing amount that the non-faulty processing core can provide in one calculation).
[0066] For example, the standard computing capacity that a non-faulty processing core can provide is A (MB). According to actual usage experience, the actual computing capacity of the non-faulty processing core is preferably 70%-80% of the standard computing capacity. If the actual computing capacity is too small, it will waste the chip performance, and if the actual computing capacity is too large, it will cause excessive computing load on the chip. Therefore, the standard computing capacity of the non-faulty processing core can be referred to, and the group results to be mapped can be reasonably mapped to the non-faulty processing core of the chip.
[0067] The data transmission efficiency between non-faulty processing cores can be obtained based on the absolute coordinates of the non-faulty processing cores. The greater the distance between two non-faulty processing cores, the longer it takes to transmit data and the lower the transmission efficiency. Therefore, in the process of reasonably mapping the group results to be mapped to the non-faulty processing cores of the chip, the group results that need to be transmitted should be mapped to the non-faulty processing cores that are closer as much as possible.
[0068] Figure 4 A flowchart of another chip compilation method provided by an embodiment of the present disclosure.
[0069] Reference Figure 4 , the compilation method for a chip in an embodiment of the present disclosure specifically includes:
[0070] S401. Obtain chip fault processing core information.
[0071] The ARM or APU of the chip obtains the fault processing core information of the chip including the number of the fault processing cores and the absolute coordinates of the fault processing cores by reading the efuse of the chip.
[0072] S402: Group the algorithm compilation results according to the pre-grouping results and the number of chip fault processing cores to obtain the grouping results to be mapped.
[0073] The ARM or APU of the chip groups (or folds) the algorithm compilation results according to the pre-grouping results and the number of chip fault processing cores, and divides the algorithm compilation results into compilation results (i.e., group results to be mapped) for execution by a certain number of non-faulty processing cores.
[0074] The pre-grouping result is a grouping result obtained by grouping the algorithm compilation results when the number of non-faulty processing cores in the chip is a preset number.
[0075] The pre-grouping result can be obtained by the chip's ARM or APU before step S402, by grouping the algorithm compilation results according to the preset number of non-faulty processing cores of the chip, or by other devices (such as servers) grouping the algorithm compilation results to obtain the pre-grouping result and then sending it to the chip.
[0076] Reference Figure 5 , this step may specifically include:
[0077] S501 . When the number of non-faulty processing cores of the chip is the same as the preset number, use the pre-grouping result as the grouping result to be mapped.
[0078] S502: When the number of non-faulty processing cores of the chip is different from the preset number, group the algorithm compilation results according to the number of faulty processing cores of the chip to obtain group results to be mapped.
[0079] If the number of fault processing cores of the chip is consistent with the preset number of fault processing cores, the algorithm compilation result may not be further processed, and the pre-grouping result is directly used as the grouping result to be mapped of the chip.
[0080] If the number of fault processing cores of the chip is inconsistent with the preset number of fault processing cores, the algorithm compilation results are grouped according to the number of fault processing cores of the chip to obtain the grouping results to be mapped.
[0081] Since the preset number of fault processing cores is determined according to the characteristics of the chip, that is, among multiple chips manufactured in batches, the number of fault processing cores of most chips may be the preset number, the pre-grouping result can be used as the grouping result to be mapped for most chips, which greatly reduces the compilation workload of the chip and saves the computing resources of the chip.
[0082] Reference Figure 6 , this step may also include:
[0083] S601: When the number of non-faulty processing cores of the chip is not less than a preset number, use the pre-grouping result as the grouping result to be mapped.
[0084] S602: When the number of non-faulty processing cores of the chip is less than a preset number, group the algorithm compilation results according to the number of faulty processing cores of the chip to obtain group results to be mapped.
[0085] If the number of fault processing cores of the chip is greater than or equal to the preset number of fault processing cores, the algorithm compilation result may not be further processed, and the pre-grouping result is directly used as the grouping result to be mapped of the chip.
[0086] If the number of fault processing cores of the chip is greater than the preset number of fault processing cores, the algorithm compilation results are grouped according to the number of fault processing cores of the chip to obtain the grouping results to be mapped.
[0087] Compared with grouping the algorithm compilation results when the number of fault processing cores of the chip is different from the preset number, when the number of fault processing cores of the chip is greater than the preset number of fault processing cores, grouping the algorithm compilation results according to the number of fault processing cores of the chip can further reduce the compilation workload of the chip and save the computing resources of the chip.
[0088] S403: Map the grouping results to be mapped to the processing cores of the chip that are not faulty.
[0089] The ARM or APU of the chip maps the grouping results to be mapped to the processing core corresponding to the absolute coordinates determined by the chip according to the absolute coordinates of the fault processing core of the chip.
[0090] Since the algorithm compilation results are mapped to the processing cores of the chip that are not faulty, the processing cores of the chip that have faults will not be mapped with the algorithm compilation results, that is, the processing cores of the chip that have faults do not need to execute the algorithm compilation results, which ensures that the chip can execute the algorithm compilation results, avoids the situation where the chip cannot be used, improves the utilization rate of chip resources, and reduces the cost of chip manufacturing.
[0091] Fig. 9 A block diagram of the composition of a chip provided in an embodiment of the present disclosure.
[0092] Reference Fig. 9 , the embodiment of the present disclosure provides a chip 100, the chip 100 comprising:
[0093] A plurality of processing cores 101;
[0094] Information module 102, used to obtain chip fault processing core information;
[0095] The compiling module 103 is used to map the algorithm compilation result to the processing core of the chip that is not faulty according to the fault processing core information of the chip, and is also used to read the one-time programmable memory of the chip to obtain the fault processing core information of the chip.
[0096] In addition, an embodiment of the present disclosure further provides a computer-readable medium on which a computer program is stored, wherein the computer program implements the above-mentioned compilation method when executed by a processing core.
[0097] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.
[0098] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly noted, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.
Claims
1. A compilation method, the method being applied to a chip, the chip comprising a plurality of processing cores, the method comprising: Obtaining fault processing core information of the chip; According to the faulty processing core information of the chip, mapping the algorithm compilation result to the processing core of the chip that is not faulty; The fault processing core information includes the number of fault processing cores; and mapping the algorithm compilation result to the processing core of the chip that is not faulty according to the fault processing core information of the chip includes: According to the number of the chip fault processing cores, the algorithm compilation results are grouped to obtain group results to be mapped; Mapping the to-be-mapped grouping results to a processing core of the chip that is not faulty; The step of grouping the algorithm compilation results according to the number of the chip fault processing cores to obtain the group results to be mapped includes: According to the pre-grouping result and the number of the chip fault processing cores, the algorithm compilation results are grouped to obtain the grouping result to be mapped, wherein the pre-grouping result is the grouping result obtained by grouping the algorithm compilation results when the number of the chip non-fault processing cores is a preset number; Wherein, when the number of non-faulty processing cores of the chip is not less than a preset number, the pre-grouping result is used as the grouping result to be mapped.
2. The compiling method according to claim 1, wherein: The obtaining of the fault processing core information of the chip includes: The one-time programmable memory of the chip is read to obtain the fault processing core information of the chip.
3. The compiling method according to claim 1, wherein: The fault processing core information also includes the absolute coordinates of the fault processing core; The step of mapping the algorithm compilation result to the processing core of the chip that is not faulty according to the faulty processing core information of the chip comprises: According to the number of the chip fault processing cores, the algorithm compilation results are grouped to obtain group results to be mapped; The to-be-mapped grouping results are mapped to the processing cores of the chip that are not faulty.
4. The compiling method according to claim 3, wherein: Mapping the grouping result to a processing core of the chip that is not faulty includes: The grouping results are mapped to the processing cores of the chip that are not faulty according to the computing capabilities of the processing cores of the chip that are not faulty.
5. The compiling method according to claim 1, wherein: The grouping of the algorithm compilation results according to the pre-grouping results and the number of the chip fault processing cores to obtain the grouping results to be mapped includes: In a case where the number of non-faulty processing cores of the chip is the same as the preset number, taking the pre-grouping result as the grouping result to be mapped; In the case that the number of non-faulty processing cores of the chip is different from the preset number, the algorithm compilation results are grouped according to the number of faulty processing cores of the chip to obtain the grouping results to be mapped.
6. The compiling method according to claim 1, wherein: The grouping of the algorithm compilation results according to the pre-grouping results and the number of the chip fault processing cores to obtain the grouping results to be mapped includes: When the number of non-faulty processing cores of the chip is less than a preset number, the algorithm compilation results are grouped according to the number of faulty processing cores of the chip to obtain grouping results to be mapped.
7. A chip, comprising: Multiple processing cores; An information module, used to obtain fault processing core information of the chip; A compiling module, used for mapping the algorithm compilation result to the processing core of the chip that is not faulty according to the faulty processing core information of the chip; The fault processing core information includes the number of fault processing cores; and mapping the algorithm compilation result to the processing core of the chip that is not faulty according to the fault processing core information of the chip includes: According to the number of the chip fault processing cores, the algorithm compilation results are grouped to obtain group results to be mapped; Mapping the to-be-mapped grouping results to a processing core of the chip that is not faulty; The step of grouping the algorithm compilation results according to the number of the chip fault processing cores to obtain the group results to be mapped includes: According to the pre-grouping result and the number of the chip fault processing cores, the algorithm compilation results are grouped to obtain the grouping result to be mapped, wherein the pre-grouping result is the grouping result obtained by grouping the algorithm compilation results when the number of the chip non-fault processing cores is a preset number; Wherein, when the number of non-faulty processing cores of the chip is not less than a preset number, the pre-grouping result is used as the grouping result to be mapped.
8. The chip according to claim 7, wherein: The information module is also used to read the one-time programmable memory of the chip to obtain the fault processing core information of the chip.
9. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by the processing core, the computer program implements the compilation method according to any one of claims 1 to 6.
