Method and computer program product for verifying memory circuit and computing device

By dividing the address space of the chip storage circuit into multiple address blocks and using random constraints for verification, the problem of incomplete verification in the existing technology is solved, automated address allocation and function verification are realized, and the efficiency of chip verification is improved. System performance.

CN119065986AActive Publication Date: 2024-12-03SHENZHEN CORERAIN TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202411572122.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-06
Publication Date
2024-12-03
Estimated Expiration
2044-11-06

AI Technical Summary

Technical Problem

The prior art is difficult to effectively verify the boundaries and special addresses of the memory circuit in chip verification, resulting in incomplete address space verification and functional verification, low efficiency, and easy to miss problems.

Method used

By dividing the address space of the chip storage circuit into multiple address blocks, the starting address and block length are generated using a random method, the weight of boundaries and special addresses is increased, and the out-of-order arrangement and sequential selection is performed, and the memory allocation combined with the general verification method is used for memory allocation Management components perform dynamic address allocation and recycling, realizing automated address allocation and functional verification simulation.

Benefits of technology

Improves the completeness and efficiency of chip verification, reduces the workload of manual management, avoids memory leaks and other errors, enhances system performance and security, and can better support concurrent or multitasking.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119065986A_ABST
    Figure CN119065986A_ABST
Patent Text Reader

Abstract

The present invention provides a method for verifying a memory circuit, a computer program product and a computing device, the method comprising: dividing an address space of a chip memory circuit into a plurality of address blocks, including: generating an initial address and a block length for the address blocks in a random manner; randomly allocating at least one address block in the plurality of address blocks; performing function verification simulation on the chip storage circuit by using the at least one allocated address block; and collecting the initial address of the address block and the coverage rate of the block length according to multiple random results, and completing the function verification simulation after the coverage rate reaches 100%. According to the technical scheme provided by the invention, the boundary and the special address can be better verified, the completeness of chip verification is increased, the verification of the address space and the function verification can be combined, the automatic address allocation is realized, and the verification efficiency is increased.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of chip technology, and in particular to a method for verifying a storage circuit, a computer program product, and a computing device. Background Art

[0002] The storage circuit is an important component of the chip. The instructions and data for the chip to run need to be stored in the storage circuit. And as the requirements for chip computing speeds become higher and higher, the location of the storage circuit is getting closer and closer to the computing circuit. Nowadays, placing storage circuits directly inside the computing core and using on-chip storage have become an important architectural solution to the problem of data movement. When the chip is in use, in addition to using register configuration, some data or configuration may also be stored in the corresponding storage circuit random access memory space (RAM space). When the chip is actually used, the software performs dynamic memory allocation and allocates and recycles the storage space.

[0003] The dynamic memory allocation algorithms in the prior art are mostly for software use, focusing on the allocation and recycling of storage space. Each allocation should be sufficient, and after use, the recycled space should not be invalidly occupied. However, for chip verification, the commands executed and the data used are usually small, so the space is sufficient. The focus should be on finding possible scenarios based on the chip structure, and constructing the allocation and use of storage space based on these scenarios. For example, specially constructing space across two actual RAM blocks, or constructing multiple spaces with misaligned addresses, etc.

[0004] To this end, a technical solution is needed that can better verify boundaries and special addresses, increase the completeness of chip verification, combine address space verification and functional verification, realize automatic address allocation, and increase verification efficiency. Summary of the invention

[0005] The present invention aims to provide a method, a computer program product and a computing device for verifying a storage circuit, which can better verify boundaries and special addresses, increase the completeness of chip verification, combine address space verification and functional verification, realize automatic address allocation, and increase verification efficiency.

[0006] According to one aspect of the present invention, there is provided a method for verifying a storage circuit, the method comprising: Dividing the address space of the chip storage circuit into a plurality of address blocks, including: generating a starting address and a block length for the address block in a random manner; randomly assigning at least one address block among the plurality of address blocks; Performing a functional verification simulation on the chip memory circuit using the at least one allocated address block; The coverage of the start address and block length of the address block is collected according to the results of multiple random operations, and the functional verification simulation is completed when the coverage reaches 100%.

[0007] According to some embodiments, dividing the address space of the chip memory circuit into a plurality of address blocks further comprises: Using the random constraints, weights of boundaries and / or special addresses are increased, the weights including empirical values.

[0008] According to some embodiments, the random constraints are set according to boundaries of the chip memory circuit and white box test points.

[0009] According to some embodiments, randomly allocating at least one address block of the plurality of address blocks comprises: Arrange the multiple address blocks in random order; The at least one address block is sequentially selected from the plurality of address blocks arranged in random order.

[0010] According to some embodiments, a memory allocation management component using a general verification methodology is used on the at least one allocated address block to perform dynamic address allocation and automatically control address allocation and recycling.

[0011] According to some embodiments, performing a functional verification simulation on the chip memory circuit using the allocated at least one address block includes: According to the starting address and block length of the at least one address block, setting the actual chip storage space used; A use case test is performed on the actual chip storage space.

[0012] According to some embodiments, the actual chip storage space includes a combination of multiple storage units, or is directly constructed by registers, and the boundaries and chip select controls between the multiple storage units are different.

[0013] According to some embodiments, each memory cell includes a memory row, a predetermined number of memory cell storage groups, a predetermined number of storage groups constitute a memory block, and the boundaries of the chip memory circuit include boundaries between the memory rows, boundaries between the storage groups and / or boundaries between the memory blocks.

[0014] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method as described above is implemented.

[0015] According to another aspect of the present invention, there is provided a computing device, comprising: Processor; and A memory stores a computer program, and when the computer program is executed by the processor, the method described in any one of the above items is implemented.

[0016] According to an embodiment of the present invention, by dividing the address space into multiple blocks, using multiple blocks of address space for verification simulation, collecting the coverage of the starting address and space size of the address space allocated each time, completing the verification of related functions when the coverage reaches 100%, and realizing automatic address allocation. The present invention can better verify the boundaries and special addresses, increase the completeness of chip verification, and can combine the verification of address space with the verification of functions, thereby increasing the efficiency of verification.

[0017] According to some embodiments, the present invention implements dynamic memory allocation, and when an application requests memory, the operating system kernel or runtime library automatically allocates a memory block of appropriate size, automatically detects objects that are no longer in use, and releases the memory they occupy.

[0018] According to some embodiments, automated memory management reduces the workload required for manual management, while improving system performance by avoiding problems such as memory leaks. Manual memory management is prone to errors, such as forgetting to release memory, wild pointers, double releases, etc. Automated allocation can significantly reduce such errors.

[0019] According to some embodiments, automated memory management simplifies code writing, allowing developers to focus more on business logic and manage resources more flexibly. The present invention enables the system to dynamically adjust memory resources according to actual needs and better support concurrent or multi-tasking.

[0020] According to some embodiments, the present invention can resist attempts to attack using a fixed memory layout, thereby increasing the difficulty for attackers and enhancing security.

[0021] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for describing the embodiments are briefly introduced below.

[0023] Figure 1 A flow chart of a method for verifying a memory circuit according to an example embodiment is shown.

[0024] Figure 2 A flow chart of a method for verifying dynamically allocated space according to an example embodiment is shown.

[0025] Figure 3 A flow chart of a method for selecting an address block according to an example embodiment is shown.

[0026] Figure 4 A block diagram of a computing device is shown according to an exemplary embodiment. DETAILED DESCRIPTION

[0027] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted.

[0028] In addition, the described features, structures or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present invention. However, those skilled in the art will appreciate that the technical solution of the present invention can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring various aspects of the present invention.

[0029] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0030] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0031] It should be understood that although the terms first, second, third, etc. may be used herein to describe various components, these components should not be limited by these terms. These terms are used to distinguish one component from another component. Therefore, the first component discussed below can be referred to as the second component without departing from the teachings of the present inventive concept. As used herein, the term "and / or" includes any one of the associated listed items and all combinations of one or more.

[0032] The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the present invention are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0033] Those skilled in the art will appreciate that the drawings are merely schematic diagrams of example embodiments, and the modules or processes in the drawings are not necessarily necessary for implementing the present invention, and therefore cannot be used to limit the protection scope of the present invention.

[0034] The dynamic memory allocation algorithms in the prior art are mostly for software use, focusing on the allocation and recycling of storage space. Each allocation should be sufficient, and after use, it should not be invalidly occupied. However, for chip verification, the commands executed and the data used are usually small, so the space is sufficient. The focus should be on finding possible scenarios based on the chip structure, and constructing the allocation and use of storage space based on these scenarios. For example, specially constructing space across two actual random access registers (RAM), or constructing multiple spaces with misaligned addresses, etc.

[0035] At present, the most commonly used verification method is to allocate several blocks of space in sequence according to the space size requirements, or to directly divide the storage space equally. This can only complete very simple scenario verification. Or it is necessary to manually write the address space allocation for each scenario and manually check the coverage, which is inefficient and easy to miss. If a method similar to the memory management unit (MMU) is used, it is too complicated and usually there is neither the need nor enough time to build it. In addition, the storage module verification may construct a separate scan of the storage space boundary, but on the one hand, it takes time to construct a use case separately, and on the other hand, it is not combined with the actual function, and there may be verification omissions of coupled functions.

[0036] The software looks like a continuous piece of storage space, but it is actually made up of many small storage circuits. It also contains various control circuits, which are also part of the chip design and need to be verified. Therefore, the storage module is usually verified and scanned separately.

[0037] To this end, the present invention proposes a method for chip verification space allocation, which can better verify boundaries and special addresses, increase the completeness of chip verification, combine address space verification and function verification, realize automatic address allocation, and increase verification efficiency.

[0038] Exemplary embodiments of the present invention are described below with reference to the accompanying drawings.

[0039] Figure 1 A flow chart of a method for chip verification space allocation according to example embodiments is shown.

[0040] See also Figure 1 The method of the present invention is used in the process of chip verification, and the configured address starting point and address segment length are obtained through random constraints to set the actual chip storage space used for calculation.

[0041] At S101, the address space of the chip storage circuit is divided into a plurality of address blocks, including: generating a start address and a block length for the address block in a random manner.

[0042] According to some embodiments, the address space boundary and white box test points are extracted; random constraints and functional coverage models are written based on the white box test points. White box test points are test points extracted based on current actual circuit specifications, which are actual and non-abstract.

[0043] According to some embodiments, the random constraints are used to increase the weight of the boundary and / or special address, and the weight includes an empirical value. The random constraints are set according to the boundary and white box test points of the chip storage circuit. The address starting point and address range of the address space configured are obtained through the random constraints; according to the address starting point and the address segment length, the actual chip storage space used for calculation is set.

[0044] According to the verification scenario, the address space is divided into a predetermined number of blocks; the address starting point and the address range of each address space block are randomly generated; the weights of the boundaries and special addresses are increased using the random constraints. Different random constraints and weights are written according to the boundaries of the address space and the chip select control; the problem value is determined and assigned a specific weight, i.e., a higher weight.

[0045] For example, a storage space of 1K=1024 bytes is not directly composed of a 1024-byte SRAM, but may be composed of multiple smaller SRAMs, such as 16 64-byte SRAMs, or directly constructed by registers. In each case, the boundaries and chip select controls between multiple SRAMs are different, and different random constraints and weights can be written accordingly.

[0046] The weights are set by the verification personnel based on their experience, and the values ​​that are most likely to have problems are given higher weights. Generally, the boundaries of SRAM block splicing are special addresses, and there may be other special addresses based on the actual circuit design, which need to be specifically identified.

[0047] In S103, at least one address block among the plurality of address blocks is randomly allocated.

[0048] According to some embodiments, the plurality of address blocks are arranged in random order, and the at least one address block is sequentially selected from the plurality of address blocks arranged in random order, see Figure 3 The flowchart of the method for selecting an address block according to an example embodiment is shown, and the address space disorder processing is performed in a simulation tool.

[0049] The storage circuit and the computing circuit need to work together. In high-level verification, the storage module will also be included, but the storage circuit will be abstracted and divided at a larger granularity, and only required to cooperate with other modules to achieve functional verification. Dividing the address space at a larger granularity means dividing the address space into relatively large blocks instead of fine-grained small blocks. For example, if only the capacity is considered, an 8K byte storage circuit is abstracted as a 1024*8bit array, which is directly divided into two halves, one half for input data and the other half for output data.

[0050] For the same static random access memory (SRAM), accessing it at different starting addresses may trigger conflicts (bank conflicts) in rows or storage bodies of different SRAMs that are accessed at the same time, resulting in different timings for writing / reading data. Although it does not affect the reading and writing of SRAM in a simple sense, it may cause errors if combined with the actual circuit function. For example, the actual circuit requires one clock cycle to read out the data, but if the address is in an unaligned state, the SRAM requires 2 clock cycles to read out, which will cause functional errors. If the address allocation is allocated in a simple way of alignment, the function point may not be verified, resulting in the problem being missed. For another example, the subsequent operation is to add the read data a and the read data b in each clock cycle. If new data is read in each cycle, the added data will continue to change; if a does not change in conflict and b does change in conflict, the calculation result will not meet expectations.

[0051] The address information for accessing RAM is generated by other functional circuits, but the transfer process causes mismatches in the bit width and alignment of some addresses. For example, the data bit width is different. A 32-bit address is transferred to RAM with only 16 bits left. If access starts from 0, there is no difference. If access starts from the valid position of the 17-bit address, there is a difference. If space is allocated only according to simple alignment addresses, the transfer problem of different addresses may not be discovered.

[0052] Therefore, it is necessary to shuffle multiple address spaces and set the address space in combination with the function to more completely verify the chip. Shuffle processing can be performed in simulation tools. After shuffling, different address segments can be used. If stored in sequence, only the front address segment may be used.

[0053] At S105 , a function verification simulation is performed on the chip memory circuit using the at least one allocated address block.

[0054] According to some embodiments, a memory allocation management component of a general verification methodology is used on the at least one allocated address block to perform dynamic address allocation, and address allocation and recycling are automatically controlled. The actual chip storage space used is set according to the starting address and block length of the at least one address block; and a use case test is performed on the actual chip storage space. The actual chip storage space includes a combination of multiple storage units, or is directly constructed by registers, and the boundaries and chip select controls between the multiple storage units are different.

[0055] According to some embodiments, each memory cell includes a memory row, a predetermined number of memory cell storage groups, a predetermined number of storage groups constitute a memory block, and the boundaries of the chip memory circuit include boundaries between the memory rows, boundaries between the storage groups and / or boundaries between the memory blocks.

[0056] During the chip verification process, a block of address space with a certain capacity is usually allocated to the RAM, but the specific storage location of the data can be controlled by the environment itself. However, the same allocation process as the software is usually not used, but the allocation method is designed by itself. The most commonly used verification method is to allocate several blocks of space in sequence according to the space size requirements, or directly divide the storage space equally.

[0057] According to some embodiments, after the address space is processed out of order, any of the address spaces is randomly verified in various functional verification use cases to avoid the order causing some address ranges to fail to be verified. Dynamic address allocation is performed on any of the address spaces to automatically control address allocation and recycling.

[0058] By using one or more blocks of addresses in scrambled order for verification simulation, the random address space can be verified in various functional verification use cases.

[0059] For example, suppose the on-chip storage RAM space used by the chip is 1024 bytes, composed of SRAM. The SRAM unit is a 128-bit memory, and every 4 SRAM units form a group, with a total of 512 bits, and a total of 16 groups, divided into two blocks, each with 8 groups. Each time 4 groups of address subspaces are needed, 4 groups are selected from the 8 randomly generated subspaces each time, which can cover multiple blocks of addresses in the same group, across different groups, across SRAM blocks, and other scenarios. At this time, the address space generated will not overlap, and there will be no errors caused by data overwriting.

[0060] Assume that the chip has a requirement that all four groups of address space must be accessed once in a maximum of two clock cycles, and the SRAM used has a limitation, that is, each group supports at most one access source in each clock cycle. When the four subspaces are randomly assigned to different groups, the simulation can pass, but when more than two subspaces are randomly assigned to the same group, for example, three different access sources access the same group, which requires three clock cycles to complete. This does not meet the chip requirements and the simulation fails, thereby discovering a register transfer level (RTL) problem.

[0061] In S107, the coverage of the start address and block length of the address block is collected according to the results of multiple random operations, and the functional verification simulation is completed when the coverage reaches 100%.

[0062] According to some embodiments, the coverage is collected by writing a functional coverage model and performing simulation.

[0063] Preferably, on a randomly taken block of addresses, UVM (Universal Verification Methodology) functional component UVM_MEM_MAM is further used to perform dynamic address allocation, which can automatically control the allocation and recycling of addresses.

[0064] UVM_MEM_MAM is a component for memory management. It is mainly used to create and manage an address space in the Universal Verification Methodology (UVM) environment. This component can help users avoid some common errors when initializing the address space and provide a unified way to manage the life cycle of the address space. UVM_MEM_MAM is a component for address space management in UVM, which simplifies the initialization process of the address space and provides a unified interface to manage the address space. When using UVM for verification, the reasonable use of these components can help verification engineers build and maintain the verification environment more efficiently.

[0065] The UVM_MEM_MAM tool can be used to manage the storage space. The tool can quickly convert the randomly obtained space start address and space size constraint information into array control information, thereby adapting to the UVM methodology and quickly completing the address space management.

[0066] By setting the address space start address (start_offset) and address space length (n_bytes) properties of UVM_MEM_MAM, the random constraint result can be realized. Then use the reserved address space region function (request_region) to obtain the address block that meets the constraints. For example, the 8 subspaces mentioned above can be used: uvm_mem_mam mam; uvm_mem_region region; mam.n_bytes = 128; mam.start_offset = 16; mam.end_offset = 1024; region = mam.request_region(8); An exemplary embodiment of the technical solution according to the present invention is described below.

[0067] Assume that the on-chip RAM space used by the chip is 1024 bytes, which is composed of SRAM. The SRAM unit is a 128-bit memory, and every 4 SRAM units form a group, with a total of 512 bits, and a total of 16 groups, divided into two blocks, each with 8 groups. The main boundaries that can be identified are the 128-bit line boundary (each integer multiple of 128 bits can be regarded as a boundary point), 16 512-bit group boundaries, and 2 block boundaries of 8 groups.

[0068] Assuming that the current access granularity is 8bit=1byte, set the starting address constraint of address allocation: address_start dist {0:=1, 15:=1, 16:=1, 31:=1, 32:=1, 47:=1, 48:=1, 63:=1,64:=1, 8*64:=1, [1:1023]: / 1}, that is, the weight of each boundary point = 1, and the overall weight of the [1:1023] range = 1, so the probability of the boundary point appearing is higher. Set the starting address to 11 different random values, and the probability of occurrence is 1 / 11. Then the random weight of each boundary point is larger, and the random weight of the middle address in the [1:1023] range is smaller, which makes it easier to randomly reach the key boundary point.

[0069] Set the allocated address space length constraint: mem_size dist {[1:15]: / 1, 16: / 1, 32: / 1,33: / 1, 48: / 1, 49: / 1, 64: / 1, 65: / 1, 128: / 1, 129: / 1, 192: / 1, 193: / 1, 256: / 1,257: / 1, 512: / 1, 513: / 1, 1024: / 1, [1:1023]: / 1}. For example, if each line is 128 bits, then 128 bits will just fill up one line, while 129 bits will just overflow by 1 bit and need to occupy 2 lines. These values ​​are set based on the boundaries judged by the verifier. The probability of random occurrence of a specific length is higher, and it is easier to cover the boundary conditions.

[0070] The above combinations can cover different starting positions, and cover each time occupying one or more SRAM spaces, so that the functions under different SRAM combinations can be verified. For example, in a random, address_start=16 (byte), mem_size = 128 (byte), then the ram space is divided into 8 subspaces. And each subcontrol starts from address 16+i*128, where i is the number of the subspace, and each subspace will span 3 groups. During simulation, the order of the 8 subspaces is disrupted, and one of them is randomly taken out, such as the address space of 16-144, which can cover the boundary of the group. When 400-528 is used, the boundary of the SRAM block can also be covered.

[0071] At the same time, functional coverage modeling is performed on the starting address and the address space length. Through multiple random tests of use cases, all boundary points and access combinations of the actual SRAM structure can be covered, thereby achieving simultaneous verification of the actual function and the SRAM structure characteristics. For example, there is a defect (bug) in the RTL. When the size is 128 bytes, it is considered to occupy the space of two groups, so only two chip selects are generated. Therefore, if the first two sequentially aligned spaces are always used, the error cannot be found; in this embodiment, since it spans three address spaces, the error can be found. For example, the read-write conflict control of multiple SRAM blocks may also be missed if there are no multiple allocated accesses.

[0072] Figure 2 A flow chart of a method for verifying dynamically allocated space according to an example embodiment is shown.

[0073] In S201, the address space boundary and white box test points are extracted.

[0074] Address space boundaries refer to the starting and ending addresses of a storage space region in a storage space management system. In white box testing, these boundaries are identified to ensure the correct allocation, access, and release of storage space regions.

[0075] When extracting white-box test points, ensure that all code paths are tested, including loops, branches, and exception handling paths. For each function, test boundary conditions, such as maximum, minimum, and zero values ​​of parameters. Test storage space operation functions to ensure that they handle boundary conditions correctly and test whether storage space access is out of bounds.

[0076] In S203, random constraints and functional coverage models are written according to the white box test points.

[0077] Random constraints are used to guide the process of randomly generating test data to ensure that the generated test data can cover specific test scenarios.

[0078] Writing random constraints and functional coverage models based on the white box test points is an important step to ensure the adequacy and effectiveness of the test. Random constraints and functional coverage models are written for the storage space management module to ensure that functions such as storage space allocation, access and release are fully tested.

[0079] In S205, the address starting point and address range of the configured address space are obtained through the random constraint.

[0080] By obtaining the address starting point and address range of the configured address space through random constraints, it can be ensured that during the test process, the generated test data can cover different address ranges and conform to the expected address space configuration. In order to work properly, some constants need to be defined to specify the boundaries and size range of the address space.

[0081] In S207, the actual chip storage space used for calculation is set according to the address starting point and the address segment length.

[0082] The actual chip storage space used for calculation is set according to the address starting point and address segment length, which can be achieved by defining corresponding data structures and configuration parameters. First, a data structure is defined to represent the information of the storage space block, including the address starting point and length. Next, a storage space management module is defined to manage multiple storage space blocks and set the storage space according to the given address starting point and length. According to the above definition, a storage space management instance is created and the actual chip storage space is set. Through random constraints and coverage models, the correctness and robustness of the storage space management module under various boundary conditions can be ensured.

[0083] Figure 4 A block diagram of a computing device is shown according to an exemplary embodiment.

[0084] like Figure 4 As shown, computing device 30 includes processor 12 and memory 14. Computing device 30 may also include bus 22, network interface 16, and I / O interface 18. Processor 12, memory 14, network interface 16, and I / O interface 18 may communicate with each other via bus 22.

[0085] The processor 12 may include one or more general-purpose CPUs (Central Processing Units, processors), microprocessors, or application-specific integrated circuits, etc., for executing relevant program instructions. According to some embodiments, the computing device 30 may also include a high-performance graphics card (GPU) 20 for accelerating the processor 12.

[0086] The memory 14 may include a machine system readable medium in the form of a volatile memory, such as a random access memory (RAM), a read-only memory (ROM) and / or a cache memory. The memory 14 is used to store one or more programs including instructions and data. The processor 12 may read the instructions stored in the memory 14 to execute the above-mentioned method according to the embodiment of the present invention.

[0087] The computing device 30 may also communicate with one or more networks via the network interface 16. The network interface 16 may be a wireless network interface.

[0088] The bus 22 may include an address bus, a data bus, a control bus, etc. The bus 22 provides a path for exchanging information between components.

[0089] It should be noted that, in the specific implementation process, the computing device 30 may also include other components necessary for normal operation. In addition, those skilled in the art may understand that the above device may only include components necessary for implementing the embodiments of this specification, and need not include all components shown in the figure.

[0090] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when the program is executed by a processor. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), a network storage device, a cloud storage device, or any type of medium or device suitable for storing instructions and / or data.

[0091] An embodiment of the present invention further provides a computer program product, which includes a computer program. The computer program is operable to enable a computer to execute part or all of the steps of any one of the methods described in the above method embodiments.

[0092] Those skilled in the art can clearly understand that the technical solution of the present invention can be implemented with the help of software and / or hardware. "Unit" and "module" in this specification refer to software and / or hardware that can independently complete or cooperate with other components to complete specific functions, where the hardware can be, for example, a field programmable gate array, an integrated circuit, etc.

[0093] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0094] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0095] In the several embodiments provided by the present invention, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0097] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0098] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention.

[0099] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0100] The exemplary embodiments of the present invention are specifically shown and described above. It should be understood that the present invention is not limited to the detailed structures, configurations or implementations described herein; on the contrary, the present invention is intended to cover various modifications and equivalent configurations included in the spirit and scope of the appended clauses.

Claims

1. A method for verifying a storage circuit, the method comprising: include: Dividing the address space of the chip storage circuit into a plurality of address blocks, including: generating a starting address and a block length for the address block in a random manner; randomly assigning at least one address block among the plurality of address blocks; Performing a functional verification simulation on the chip memory circuit using the at least one allocated address block; The coverage of the start address and block length of the address block is collected according to the results of multiple random operations, and the functional verification simulation is completed when the coverage reaches 100%.

2. The method according to claim 1, It is characterized in that The address space of the chip storage circuit is divided into multiple address blocks, including: Using random constraints, weights of boundaries and / or special addresses are increased, the weights including empirical values.

3. The method according to claim 2, It is characterized in that The random constraints are set according to the boundaries of the chip memory circuit and white box test points.

4. The method according to claim 1, It is characterized in that Randomly allocating at least one address block among the plurality of address blocks, comprising: Arrange the multiple address blocks in random order; The at least one address block is sequentially selected from the plurality of address blocks arranged in random order.

5. The method according to claim 1, It is characterized in that Also includes: A memory allocation management component using a general verification methodology is used on the at least one allocated address block to perform dynamic address allocation and automatically control address allocation and recycling.

6. The method according to claim 1, It is characterized in that Using the allocated at least one address block to perform a functional verification simulation on the chip storage circuit includes: According to the starting address and block length of the at least one address block, setting the actual chip storage space used; A use case test is performed on the actual chip storage space.

7. The method according to claim 6, It is characterized in that The actual chip storage space includes a combination of multiple storage units, or is directly constructed by registers, and the boundaries and chip selection controls between the multiple storage units are different.

8. The method according to claim 7, It is characterized in that Each memory cell includes a memory row, a predetermined number of memory cell memory groups, and a predetermined number of memory groups constitute a memory block. The boundaries of the chip memory circuit include boundaries between the memory rows, boundaries between the memory groups, and / or boundaries between the memory blocks.

9. A computer program product, It is characterized in that The method comprises a computer program, which implements the method according to any one of claims 1 to 8 when being executed by a processor.

10. A computing device, It is characterized in that include: processor; as well as A memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 8 is implemented.

Citation Information

Patent Citations

  • Systems and methods for direct memory access

    CN113672529A

  • Chip system-level verification method and device

    CN117574813A

  • Simulation method and device for microcontroller chip and electronic equipment

    CN118468804A

  • Chip verification method and system

    CN118503104A

  • Verification method of storage circuit and computing equipment

    CN118571297A