Redundant replacement circuit, chip and computing device integrating storage and computing

By setting up a redundant memory unit and a redundant enable signal generation unit in the memory integrated circuit, the automatic replacement and direct transmission of data is realized, which solves the problem of data delay in the memory integrated design, and improves the performance and stability of the circuit sex.

CN113851175BActive Publication Date: 2024-11-29NANJING HOUMO TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111136019.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-09-27
Publication Date
2024-11-29
Estimated Expiration
2041-09-27

AI Technical Summary

Technical Problem

The prior art cannot effectively replace redundantly with stored data and product data in the integrated storage and computing design, resulting in a large data delay when the storage and computing unit has bad points.

Method used

A redundant replacement circuit integrating storage and computing is designed. By setting a conventional storage unit and a redundant storage unit in the storage unit array, the redundant enable signal generation unit and logic control unit are used to realize automatic data replacement. And directly transmit the input data to the redundant memory unit through the redundant path to avoid data delays caused by multi-level logic selectors.

Benefits of technology

It realizes that when a memory unit appears in the memory unit, the memory unit can be replaced with the redundant memory unit at the same time, which solves the problem of large data delay and improves the memory and computing integrated circuit. Performance and stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113851175B_ABST
    Figure CN113851175B_ABST
Patent Text Reader

Abstract

The disclosed embodiment discloses a redundant replacement circuit, chip, and computing device that integrates storage and computing, wherein the circuit includes: a storage and computing unit array, a data input unit array, a redundant enable signal generating unit, and a logic control unit, wherein the storage and computing unit array includes a conventional storage and computing unit and a redundant storage and computing unit; the logic control unit is used to replace the operation of the conventional storage and computing unit containing the bad point with the operation of the redundant storage and computing unit according to the bad point address; the redundant enable signal generating unit is used to set the redundant enable signal of the input target data input unit according to the bad point address; the data input unit includes a redundant enable signal detection module, which is used to determine whether to input the input data into the redundant storage and computing unit according to the corresponding redundant enable signal. The disclosed embodiment realizes that when a bad point occurs, the storage data and the input data are simultaneously replaced into the redundant storage and computing unit, and the input data is directly transmitted to the redundant storage and computing unit by the redundant path, which can reduce data delay.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a redundant replacement circuit, chip and computing device integrating storage and computing. Background Art

[0002] Integrated storage and computing is a storage and computing design that integrates static random access memory and computing modules into one, including storage structure and multiplication and addition structure. Because it is similar to traditional memory, it has large-scale repeated storage units, so during the mass production process, storage bad points will be caused by factors such as process deviations in the storage units. Similar to traditional storage, it is usually necessary to add redundant rows or columns for redundant design. When bad points appear, they are replaced with redundant rows or columns. The integrated storage and computing design also requires redundant design to reduce the impact of bad points in storage and computing units during mass production, but the integrated storage and computing design and the architecture of the storage and computing units are fundamentally different from traditional memories, so a special redundant design is required for the integrated storage and computing design. Summary of the invention

[0003] An embodiment of the present disclosure provides a redundant replacement circuit for storage and computing, the circuit comprising: a storage and computing unit array, a data input unit array, a redundant enable signal generating unit and a logic control unit, the storage and computing unit array comprising conventional storage and computing units and redundant storage and computing units, the redundant enable signal generating unit being used to generate redundant enable signals corresponding to the storage and computing units in the storage and computing unit array respectively; the logic control unit being used to replace the operation on the conventional storage and computing unit containing bad pixels with the operation on the redundant storage and computing unit according to the bad pixel address of the conventional storage and computing unit containing bad pixels; the redundant enable signal generating unit being used to set the redundant enable signal of the input target data input unit according to the bad pixel address, wherein the target data input unit is the data input unit corresponding to the bad pixel address; the data input unit in the data input unit array comprising a redundant enable signal detection module, the redundant enable signal detection module being used to determine whether to input the input data into the redundant storage and computing unit according to the corresponding redundant enable signal.

[0004] In some embodiments, the redundant enable signal detection module includes a redundant enable signal input interface, a redundant detection signal input interface, and a redundant detection signal output interface; the data input units in the data input unit array are connected in series through the redundant detection signal input interface and the redundant detection signal output interface; the redundant enable signal detection module is used to set the redundant detection signal output by the redundant detection signal output interface to the first data when the redundant enable signal is the first data indicating that the corresponding data needs to be input into the redundant storage and computing unit, and to set the redundant detection signal output by the redundant detection signal output interface to be the same as the redundant detection signal input by the redundant detection signal input interface when the redundant enable signal is the second data indicating that the corresponding data does not need to be input into the redundant storage and computing unit.

[0005] In some embodiments, the redundant enable signal detection module includes a NOR gate and a NOT gate, the two input ends of the NOR gate are respectively connected to the redundant enable signal input interface and the redundant detection signal input interface, the output end of the NOR gate is connected to the input end of the NOT gate, and the output end of the NOT gate is connected to the redundant detection signal output interface.

[0006] In some embodiments, the data input unit array includes a first preset number of data input unit groups, each data input unit group includes a second preset number of data input units; the first preset number of data input unit groups share a redundant path for inputting data to the redundant storage and computing unit, and a signal selection enhancement unit is connected to the end of the redundant path of each input unit group, and the signal selection enhancement unit is used to control the signal selection enhancement unit to be connected or disconnected according to the redundant detection signal output by the last data input unit in the corresponding data input unit group.

[0007] In some embodiments, the data input unit array includes a first sub-array and a second sub-array that are symmetrically distributed, and the data input units included in the first sub-array and the second sub-array are respectively connected in series through a redundant detection signal input interface and a redundant detection signal output interface; the redundant storage and calculation unit is located between the first sub-array and the second sub-array.

[0008] In some embodiments, the data input unit in the data input unit array further includes a data transfer selector, which is used to transfer the corresponding input data to the redundant storage and computing unit or the corresponding conventional storage and computing unit according to the corresponding redundancy enable signal.

[0009] In some embodiments, the data transmission selector includes a first branch and a second branch, the first branch includes a first switch, and the second branch includes a second switch; the first switch is used to transmit the corresponding input data to the redundant storage and computing unit when the corresponding redundant enable signal is the first data; the second switch is used to transmit the corresponding input data to the corresponding conventional storage and computing unit when the corresponding redundant enable signal is the second data.

[0010] In some embodiments, the first branch further includes an inverter and a leakage protection circuit, and the leakage protection circuit is used to cut off the power supply or ground path of the inverter when the first switch is disconnected.

[0011] According to another aspect of an embodiment of the present disclosure, a chip is provided, which includes the above-mentioned storage-computing integrated redundant replacement circuit.

[0012] According to another aspect of an embodiment of the present disclosure, a computing device is provided, which includes the above chip.

[0013] The redundant replacement circuit, chip and computing device of the integrated storage and computing provided by the above-mentioned embodiments of the present disclosure are provided by setting conventional storage and computing units and redundant storage and computing units in the storage and computing unit array, the redundant enable signal generating unit generates redundant enable signals corresponding to the storage and computing units in the storage and computing unit array respectively, the logic control unit replaces the operation of the conventional storage and computing unit containing bad points with the operation of the redundant storage and computing unit according to the bad point address of the conventional storage and computing unit containing bad points, the redundant enable signal generating unit sets the redundant enable signal of the input target data input unit according to the bad point address, and the redundant enable signal detection module included in the data input unit determines whether to input the input data into the redundant storage and computing unit according to the corresponding redundant enable signal, thereby realizing that in the integrated storage and computing circuit, when a bad point appears in the storage and computing unit, the storage data and the input data can be replaced into the redundant storage and computing unit at the same time. At the same time, compared with the traditional data selection circuit design architecture, the input data needs to pass through a multi-stage logic selector, resulting in a large data delay problem in the data transmitted to the redundant storage and computing unit. The input data in the embodiment of the present disclosure is directly transmitted to the redundant storage and computing unit by a redundant path, which solves the above-mentioned large data delay problem.

[0014] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments 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. In the drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a structural diagram of a redundant replacement circuit integrating storage and computing provided by an exemplary embodiment of the present disclosure.

[0017] Figure 2 It is a structural diagram of a redundant enable signal detection module provided by an exemplary embodiment of the present disclosure.

[0018] Figure 3 It is a structural diagram of a plurality of redundant enable signal detection modules connected in series according to an exemplary embodiment of the present disclosure.

[0019] Figure 4 It is a schematic diagram of the internal structure of a redundant enable signal detection module provided by an exemplary embodiment of the present disclosure.

[0020] Figure 5 It is a structural diagram of grouping a data input unit array provided by an exemplary embodiment of the present disclosure.

[0021] Figure 6 It is a schematic diagram of the internal structure of a data transmission selector provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] Below, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0023] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0024] Those skilled in the art can understand that the terms "first" and "second" in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate the necessary logical order between them.

[0025] It should also be understood that in the embodiments of the present disclosure, “plurality” may refer to two or more than two, and “at least one” may refer to one, two, or more than two.

[0026] It should also be understood that any component, data or structure mentioned in the embodiments of the present disclosure can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0027] In addition, the term "and / or" in the present disclosure is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present disclosure generally indicates that the associated objects before and after are in an "or" relationship.

[0028] It should also be understood that the description of the various embodiments in the present disclosure focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced to each other, and for the sake of brevity, they will not be described one by one.

[0029] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0030] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0031] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0032] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0033] Application Overview

[0034] The existing technology is mainly applicable to the architecture of traditional memory, and can only perform redundant replacement on the data stored in the rows and columns of the memory. However, the input data of the storage and computing unit of the storage and computing integrated design is divided into two types. One is the storage data, and the other is the input data for computing with the storage data. The redundant design needs to perform redundant replacement on these two types of data respectively. The existing traditional memory redundant design cannot meet the storage and computing integrated design's requirements for replacing storage data and product data.

[0035] Exemplary Structure

[0036] Figure 1 This is a schematic diagram of a redundant replacement circuit for storage and computing provided by an exemplary embodiment of the present disclosure. The various components included in the circuit can be integrated into one chip, or can be set in different chips or circuit boards, and data communication links are established between these chips or circuit boards.

[0037] like Figure 1 As shown, the circuit includes: a storage computing unit array 101, a data input unit array 102, a redundant enable signal generating unit 103 and a logic control unit 104. The storage computing unit array 101 includes conventional storage computing units and redundant storage computing units. The redundant enable signal generating unit 103 is used to generate redundant enable signals corresponding to the storage computing units in the storage computing unit array 101. Usually, a redundant storage computing unit is set in each storage computing unit array, such as Figure 1 As shown, the storage and computing unit array includes a row of redundant storage and computing units. When multiple redundant storage and computing units need to be set, the storage and computing unit array can be divided into multiple sub-arrays, and each sub-array is provided with a redundant storage and computing unit. Conventional storage and computing units and redundant storage and computing units are used to store data, receive input data, and perform calculations on the stored data and the input data in a preset manner (e.g., multiplication).

[0038] Typically, each data input unit in the data input unit array 102 corresponds to at least one conventional storage and calculation unit. Figure 1As shown, each row of storage and computing units corresponds to a data input unit. Each storage and computing unit may include a redundant enable signal input interface. If a bad pixel is detected in a conventional storage and computing unit, a redundant enable signal (e.g., digital 1) indicating the bad pixel may be input into the corresponding data input unit. If no bad pixel appears in the conventional storage and computing unit, a redundant enable signal (e.g., digital 0) indicating normality may be input into the corresponding data input unit.

[0039] It should be noted that Figure 1 The circuit structure shown is only schematic, and any number of storage and computing unit arrays, input and output unit arrays, redundant enable signal generating units, and logic control units can be set as needed. For example, multiple storage and computing unit arrays can correspond one to one with the storage and computing units in each row, and each row includes multiple storage and computing units that store different bits of multi-bit data respectively (for example, the same row includes two storage and computing units, which store the upper 4 bits and lower 4 bits of 8-bit data respectively).

[0040] In this embodiment, the logic control unit 104 is used to replace the operation on the regular storage and computing unit containing the bad pixel with the operation on the redundant storage and computing unit according to the bad pixel address of the regular storage and computing unit containing the bad pixel.

[0041] Among them, bad pixels are usually determined when the storage and computing units are tested. For example, when the circuit provided by the present embodiment is tested after the wafer is returned, the storage and computing unit array is written and read row by row, and it is found that the read data is different from the written data in one or several bits, which is regarded as a possible bad pixel. Then the wafer process analysis and detection is performed on the possible bad pixels (storage and computing units). If it is found that the bad pixels are caused by process reasons, it is defined as the occurrence of bad pixels. After determining the conventional storage and computing unit where the bad pixels appear, the address of the conventional storage and computing unit can be determined as the bad pixel address.

[0042] The above-mentioned operations on the redundant storage and computing unit may include any operations, such as read operations, write operations, etc.

[0043] Usually, the address of the conventional storage unit where the bad point is located is first input into the logic control unit as the bad point address. After the address is input for read / write operation, the logic control unit will compare it with the above bad point address. If the comparison is inconsistent, the conventional storage unit corresponding to the input address will continue to perform regular read / write. If the comparison is consistent, a replacement control signal will be generated to lower the word line potential of the conventional storage unit to be replaced, so that the conventional storage unit to be replaced will no longer perform read / write operations, and at the same time, the word line potential of the redundant storage unit will be pulled up, so that the redundant storage unit replaces the conventional storage unit where the bad point is located to perform read / write operations. In this way, the replacement of the storage unit is completed. This part of the replacement is mainly the replacement of the data stored in the storage unit array.

[0044] In this embodiment, the redundant enable signal generating unit is further used to set the redundant enable signal input to the target data input unit according to the bad pixel address. The target data input unit is the data input unit corresponding to the bad pixel address. Figure 1 As shown, the redundancy enable signal input to the target data input unit 1021 is digital 1, indicating that a bad pixel has appeared in the conventional redundancy storage and calculation unit corresponding to the target data input unit 1021.

[0045] In this embodiment, if Figure 2 As shown, the data input unit 201 in the data input unit array includes a redundant enable signal detection module 2011, and the redundant enable signal detection module 2011 is used to determine whether to input the input data into the redundant storage and calculation unit according to the corresponding redundant enable signal. As an example, the redundant enable signal detection module 2011 may include a conventional analog single-pole double-throw switch device, and if the input redundant enable signal is a digital 1, the data input unit is connected to the redundant storage and calculation unit 202, and if the input redundant enable signal is a digital 0, the data input unit is connected to the corresponding conventional storage and calculation unit 203.

[0046] The circuit provided by the above-mentioned embodiment of the present disclosure is provided by setting a conventional storage and computing unit and a redundant storage and computing unit in the storage and computing unit array, the redundant enable signal generating unit generates redundant enable signals corresponding to the storage and computing units in the storage and computing unit array respectively, the logic control unit replaces the operation of the conventional storage and computing unit containing the bad point with the operation of the redundant storage and computing unit according to the bad point address of the conventional storage and computing unit containing the bad point, the redundant enable signal generating unit sets the redundant enable signal of the input target data input unit according to the bad point address, and the redundant enable signal detection module included in the data input unit determines whether to input the input data into the redundant storage and computing unit according to the corresponding redundant enable signal, thereby realizing that in the storage and computing integrated circuit, when a bad point appears in the storage and computing unit, the storage data and the input data can be replaced into the redundant storage and computing unit at the same time. At the same time, compared with the traditional data selection circuit design architecture, the input data needs to pass through a multi-level logic selector, resulting in a large data delay when being transmitted to the redundant storage and computing unit. The input data in the embodiment of the present disclosure is directly transmitted to the redundant storage and computing unit by a redundant path, which solves the above-mentioned large data delay problem.

[0047] In some optional implementations, such as Figure 2 As shown, the redundant enable signal detection module 2011 includes a redundant enable signal input interface 20111 , a redundant detection signal input interface 20112 , and a redundant detection signal output interface 20113 .

[0048] The data input units in the data input unit array are connected in series via a redundant detection signal input interface and a redundant detection signal output interface. Figure 3As shown, it shows a schematic diagram of redundant enable signal detection modules included in multiple data input units being connected in series in sequence. Figure 3 The redundant detection signal output interface (indicated by RP_L in the figure) of the previous level redundant enable signal detection module is connected to the redundant detection signal input interface (indicated by RP_R in the figure) of the next level redundant enable signal detection module.

[0049] The redundant enable signal detection module is used to set the redundant detection signal output by the redundant detection signal output interface to the first data when the redundant enable signal is the first data indicating that the corresponding data needs to be input into the redundant storage and computing unit, and to set the redundant detection signal output by the redundant detection signal output interface to be the same as the redundant detection signal input by the redundant detection signal input interface when the redundant enable signal is the second data indicating that the corresponding data does not need to be input into the redundant storage and computing unit.

[0050] As an example, the first data may be digital 1, and the second data may be digital 0. When the redundant enable signal is digital 1, the redundant detection signal output by the redundant detection signal output interface is 1; when the redundant enable signal is digital 0, the redundant detection signal output by the redundant detection signal output interface is equal to the redundant detection signal input by the redundant detection signal input interface. The redundant enable signal detection module of the above function may be implemented by an OR gate or by other methods.

[0051] By connecting each data input unit in series through a redundant detection signal input interface and a redundant detection signal output interface, this implementation method can make it possible that when a redundant enable signal received by one of the data input units indicates the existence of a bad pixel, the redundant detection signals output by the data input unit and the data input unit located after the data input unit are both the first data, thereby helping to more accurately determine the position of the target data input unit and providing a basis for subsequently determining the path for transmitting the input data to the redundant storage and computing unit.

[0052] Optional, such as Figure 4 As shown, the redundant enable signal detection module may include a NOR gate 401 and a NOT gate 402, wherein the two input ends of the NOR gate 401 are respectively connected to a redundant enable signal input interface (port represented by EN in the figure) and a redundant detection signal input interface (port represented by RP_R in the figure), the output end of the NOR gate 401 is connected to the input end of the NOT gate 402, and the output end of the NOT gate 402 is connected to a redundant detection signal output interface (port represented by RP_L in the figure). The function of the redundant enable signal detection module is realized by the NOR gate and the NOT gate, and compared with directly using the OR gate, the circuit performance is more stable and the anti-interference ability is stronger, which can effectively improve the overall performance of the storage-computing integrated circuit.

[0053] In some optional implementations, the data input unit array includes a first preset number of data input unit groups, each of which includes a second preset number of data input units. As an example, if the data input unit array includes 256 data input units, the 256 data input units can be divided into 4 data input unit groups, each of which includes 64 data input units.

[0054] A first preset number of data input unit groups share a redundant path for inputting data to a redundant storage and computing unit, and a signal selection reinforcement unit is connected to the end of the redundant path of each input unit group. The signal selection reinforcement unit is used to control the signal selection reinforcement unit to be connected or disconnected according to a redundant detection signal output by the last data input unit in the corresponding data input unit group.

[0055] like Figure 5 As shown, four signal selection and reinforcement units 502 are arranged on the redundant path 501, and the redundant detection signal output interface (i.e., RP_L in the figure) of the last data input unit in each data input unit group is connected to the corresponding signal selection and reinforcement unit. The signal selection and reinforcement unit simultaneously inputs RP_L and RP_LB (i.e., the negation of RP_L). When RP_L is 1 and RP_LB is 0, the signal selection and reinforcement unit is turned on.

[0056] Combination Figure 4 The principle of the circuit shown in Figure 5 In the example, if there is no bad pixel in the storage and computing unit array, RP_R and the redundant enable signal input interface (EN) both input 0. When the redundant enable signal of the redundant enable signal detection module in a data input unit is 1, the RP_L interface of the redundant enable signal detection module outputs a digital 1, and the RP_L of the redundant enable signal detection modules in other data input units after the data input unit all output a digital 1, so that the signal selection reinforcement unit corresponding to the group where the data input unit is located and the signal selection reinforcement units corresponding to other groups after the data input unit are all turned on, and the input data in the data input unit is transmitted to the redundant storage and computing unit through the redundant path 501.

[0057] This implementation method is to group the data input unit array and set a signal selection enhancement unit on the redundant path of each group of data input units. Compared with no grouping and directly transmitting input data to the redundant storage and computing unit, this implementation method can add respective driving logic to each data input data unit group, so that the load capacitance to be driven by each group of driving logic is evenly divided, thereby reducing the load capacitance and solving the problem of large transmission delay caused by large load capacitance.

[0058] In some optional implementations, such as Figure 5 As shown, the data input unit array includes a symmetrically distributed first sub-array 503 and a second sub-array 504, and the data input units respectively included in the first sub-array 503 and the second sub-array 504 are connected in series through a redundant detection signal input interface and a redundant detection signal output interface. For the serial connection method, please refer to the above Figure 3 The redundant storage and calculation unit is located between the first sub-array 503 and the second sub-array 504 .

[0059] This implementation method divides the data input unit array into a first sub-array and a second sub-array, and sets the redundant storage and calculation unit between the first sub-array and the second sub-array. When the data input unit array is detected, each input data can be detected from both ends of the data input unit array one by one to see whether a redundant replacement operation is performed, thereby improving the detection efficiency. In addition, the transmission path of the input data to the redundant storage and calculation unit can be made as short as possible, reducing the delay in the input data transmission process and improving the performance of the storage and calculation integrated circuit.

[0060] In some optional implementations, such as Figure 6 As shown, the data input unit in the data input unit array further includes a data transfer selector 601, and the data transfer selector 601 is used to transfer the corresponding input data to the redundant storage and computing unit or the corresponding regular storage and computing unit according to the corresponding redundancy enable signal.

[0061] As an example, the data transmission selector 601 can be a device simulating a single-pole double-throw switch. If the redundancy enable signal (ie, EN in the figure) is 1, the input data is transmitted to the redundant storage unit 602, and if EN is 0, the input data is transmitted to the corresponding conventional storage unit 603.

[0062] This implementation method can effectively transmit input data to the corresponding storage and computing unit according to the redundant enable signal by setting a data transmission selector in the data input unit, thereby realizing that when a bad point occurs in the storage and computing unit array, the storage data and the input data are simultaneously transmitted to the redundant storage and computing unit, without using a multi-level logic selector, thereby avoiding the problem of large data delay in data transmission to the redundant storage and computing unit.

[0063] In some optional implementations, such as Figure 6 As shown, the data transmission selector 601 includes a first branch 6011 and a second branch 6012 , the first branch 6011 includes a first switch 60111 , and the second branch 6012 includes a second switch 60121 .

[0064] The first switch 60111 is used to transmit the corresponding input data to the redundant storage and calculation unit 602 when the corresponding redundancy enable signal is the first data.

[0065] The second switch 60121 is used to transmit the corresponding input data to the corresponding regular storage and calculation unit 603 when the corresponding redundancy enable signal is the second data.

[0066] Specifically, Figure 6 As shown, the first switch 60111 and the second switch 60121 can be controlled by EN. If EN is 1, the first switch 60111 is turned on, the second switch 60121 is turned off, and the input data is transmitted to the redundant storage unit 602 through the first branch 6011; if EN is 0, the first switch 60111 is turned off, the second switch 60121 is turned on, and the input data is transmitted to the regular storage unit 603 through the second branch 6012.

[0067] Optionally, an inverted signal ENB of the redundant enable signal EN can be obtained by setting an inverter 6013 in the data transmission selector 601. The first switch 60111 and the second switch 60121 are two input devices, and the first switch 60111 and the second switch 60121 are controlled by EN and ENB at the same time. If EN is 1 and ENB is 0, the first switch 60111 is turned on and the second switch 60121 is turned off, and the input data is transmitted to the redundant storage and calculation unit 601 through the first branch 6011; if EN is 0 and ENB is 1, the first switch 60111 is turned off and the second switch 60121 is turned on, and the input data is transmitted to the regular storage and calculation unit 602 through the second branch 6012. By controlling the states of the two switches simultaneously by EN and ENB, the interference of signal fluctuations on the line on the switches can be further reduced, and the performance of the storage and calculation integrated circuit can be improved.

[0068] This implementation method can further improve the anti-interference capability of the data transmission selector by setting two branches and two switches for the data transmission selector, so that the accuracy of switching the transmission path of the input data according to the redundant enable signal is higher.

[0069] In some optional implementations, such as Figure 6 As shown, the first branch 6011 further includes a leakage protection circuit and an inverter 60112 . The leakage protection circuit is used to cut off the power supply or ground path of the inverter 60112 when the first switch 60111 is disconnected.

[0070] Specifically, Figure 6 As shown, the leakage protection circuit on the first branch 6011 may include a PMOS tube 60113 and an NMOS tube 60114 as shown in the figure. When EN is 0 and ENB is 1, the first branch 6011 is cut off, and the input of the inverter 60112 is in an uncertain state. To prevent the inverter 60112 from leaking in this case, the power supply or ground path of the inverter 60112 can be cut off by the PMOS tube 60113 and the NMOS tube 60114 on both sides of the inverter.

[0071] The leakage protection circuit provided by this implementation can effectively prevent the inverter leakage problem caused by the first branch being turned off, and further improve the stability of the storage and computing integrated circuit.

[0072] The embodiment of the present disclosure further provides a chip, on which a redundant replacement circuit for storage and computing is integrated. The technical details of the redundant replacement circuit for storage and computing are as follows: Figure 1-Figure 6 As shown in the relevant description, no further description is given here.

[0073] The embodiments of the present disclosure also provide a computing device, which includes the chip described in the above embodiments. In addition, the computing device may also include an input device, an output device, and necessary memory, etc. Among them, the input device may include such as a mouse, a keyboard, a touch screen, a communication network connector, etc., for inputting storage data in the storage and calculation unit array or input data in the data input unit array. The output device may include such as a display, a printer, and a communication network and its connected remote output device, etc., for outputting calculated data, etc. The memory is used to store the data input by the above input device, and the data generated during the operation of the redundant replacement circuit of the storage and calculation. The memory may include a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0074] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0075] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referenced to each other.

[0076] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including," "comprising," "having," and the like are open words, referring to "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0077] The circuits of the present disclosure may be implemented in many ways. For example, the circuits of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps of the method used in the circuit is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as programs recorded in a recording medium, which include machine-readable instructions for implementing the functions of the circuits of the present disclosure. Thus, the present disclosure also covers recording media storing programs for executing the functions of the circuits of the present disclosure.

[0078] It should also be noted that in the circuit of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as an equivalent solution of the present disclosure.

[0079] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0080] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A redundant replacement circuit integrating storage and computing, include: A storage and computing unit array, a data input unit array, a redundant enable signal generating unit and a logic control unit, wherein the storage and computing unit array includes conventional storage and computing units and redundant storage and computing units, and the redundant enable signal generating unit is used to generate redundant enable signals corresponding to the storage and computing units in the storage and computing unit array respectively; The logic control unit is used to replace the operation on the regular storage and computing unit containing the bad pixel with the operation on the redundant storage and computing unit according to the bad pixel address of the regular storage and computing unit containing the bad pixel; The redundant enable signal generating unit is further used to set a redundant enable signal input to a target data input unit according to the bad pixel address, wherein the target data input unit is a data input unit corresponding to the bad pixel address; The data input unit in the data input unit array includes a redundant enable signal detection module, and the redundant enable signal detection module is used to determine whether to input the input data into the redundant storage and calculation unit according to the corresponding redundant enable signal; The data input unit in the data input unit array further includes a data transfer selector, which is used to transfer the corresponding input data to the redundant storage and computing unit or the corresponding regular storage and computing unit according to the corresponding redundancy enable signal.

2. The circuit according to claim 1, in, The redundant enable signal detection module includes a redundant enable signal input interface, a redundant detection signal input interface, and a redundant detection signal output interface; The data input units in the data input unit array are connected in series via the redundant detection signal input interface and the redundant detection signal output interface; The redundant enable signal detection module is used to set the redundant detection signal output by the redundant detection signal output interface to the first data when the redundant enable signal is the first data indicating that the corresponding data needs to be input into the redundant storage and computing unit, and to set the redundant detection signal output by the redundant detection signal output interface to be the same as the redundant detection signal input by the redundant detection signal input interface when the redundant enable signal is the second data indicating that the corresponding data does not need to be input into the redundant storage and computing unit.

3. The circuit according to claim 2, in, The redundant enable signal detection module includes a NOR gate and a NOT gate, wherein two input ends of the NOR gate are respectively connected to the redundant enable signal input interface and the redundant detection signal input interface, the output end of the NOR gate is connected to the input end of the NOT gate, and the output end of the NOT gate is connected to the redundant detection signal output interface.

4. The circuit according to claim 2, in, The data input unit array includes a first preset number of data input unit groups, each data input unit group includes a second preset number of data input units; The first preset number of data input unit groups share a redundant path for inputting data to the redundant storage and computing unit, and a signal selection reinforcement unit is connected to the end of the redundant path of each input unit group, and the signal selection reinforcement unit is used to control the connection or disconnection of the signal selection reinforcement unit according to the redundant detection signal output by the last data input unit in the corresponding data input unit group.

5. The circuit according to claim 2, in, The data input unit array includes a first sub-array and a second sub-array that are symmetrically distributed, and the data input units respectively included in the first sub-array and the second sub-array are connected in series through a redundant detection signal input interface and a redundant detection signal output interface; The redundant storage and calculation unit is located between the first sub-array and the second sub-array.

6. The circuit according to claim 1, in, The data transmission selector comprises a first branch and a second branch, the first branch comprises a first switch, and the second branch comprises a second switch; The first switch is used to transmit the corresponding input data to the redundant storage and calculation unit when the corresponding redundancy enable signal is the first data; The second switch is used to transmit the corresponding input data to the corresponding conventional storage and computing unit when the corresponding redundancy enable signal is the second data.

7. The circuit according to claim 6, in, The first branch further includes an inverter and a leakage protection circuit, and the leakage protection circuit is used to cut off the power supply or ground path of the inverter when the first switch is disconnected.

8. A chip, It is characterized in that It comprises a redundant replacement circuit integrating storage and computing according to any one of claims 1-7.

9. A computing device, It is characterized in that Comprising the chip according to claim 8.

Citation Information

Patent Citations

  • Storage and calculation integrated chip and method for improving yield of storage and calculation integrated chip

    CN111611112A

  • Repair circuit and memory device including same

    CN113380317A