A memory self-repair circuit, method and chip

By designing the register unit and selector switching conduction relationship of the self-repair chain in the memory self-repair circuit, the problem of functional module failure under the embedded memory high failure rate and low power consumption design is solved, and the self-repair and function of the memory is realized Consumption savings improve chip yield and design flexibility.

CN118471304BActive Publication Date: 2024-11-08THIS CORE TECH (WUXI) CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410917324.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-08
Estimated Expiration
2044-07-10

AI Technical Summary

Technical Problem

The failure rate of embedded memory in SoC chips continues to increase, affecting the yield of the chip. In the low-power design, some functional modules fail, resulting in breakage of the self-repair chain, affecting the overall repair function.

Method used

Design a memory self-repair circuit, including register units in the self-repair chain, including repair registers and spare registers, switch the conduction relationship through selectors, and use backup register bypass when repair registers fail to ensure the continuity of the self-repair chain and low power operation.

Benefits of technology

It realizes that the memory can still be self-repaired when the repair register fails, reduces the failure rate and power consumption, and improves the mass production yield of the chip and the feasibility of low-power design.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118471304B_ABST
    Figure CN118471304B_ABST
Patent Text Reader

Abstract

The present application proposes a memory self-repair circuit, method and chip. The memory self-repair circuit includes at least one self-repair chain. The self-repair chain includes n register units. The register unit includes a repair register, a spare register and a selector. The repair register and the spare register in the same register unit are the same. When a repair register fails, the internal conduction relationship is switched by the selector. For example, the second port and the third port in the selector are turned on, and the corresponding repair register is bypassed by the spare register. When the repair register does not work, other memories on the self-repair chain can still be self-repaired, thereby saving their losses or saving power consumption.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of chip technology, and in particular to a memory self-repair circuit, method and chip. Background Art

[0002] Embedded memory is widely used in system on chip (SoC) chip design due to its advantages of high frequency, low power consumption and small area. The size of embedded memory in SoC chip can reach 1Mbyte or larger. At present, in order to meet market demand, the size of embedded memory increases four times every few years; at the same time, the development of chip technology requires a large number of embedded memories with smaller area and faster access speed, but the failure rate is also getting higher and higher. The yield of SoC chip will be affected by embedded memory. Improving the yield of embedded memory is the key to improving the yield of the entire SoC chip. Summary of the invention

[0003] The purpose of the present application is to provide a memory self-repair circuit, method and chip to at least partially improve the above-mentioned problems.

[0004] In order to achieve the above purpose, the technical solution adopted in the embodiment of the present application is as follows:

[0005] In a first aspect, an embodiment of the present application provides a memory self-repair circuit, the memory self-repair circuit comprising a first controller, a management module, and at least one self-repair chain, the self-repair chain comprising n register units, the register unit comprising a repair register, a spare register, and a selector, the repair register and the spare register in the same register unit being the same;

[0006] The output end of the repair register is connected to the first port of the selector, the output end of the standby register is connected to the second port of the selector, the third port of the selector serves as the output end of the register unit, and the input end of the repair register and the input end of the standby register serve as the input end of the register unit;

[0007] The input end of the register unit is connected to the output end of the upper unit, and the output end of the register unit is used to connect to the input end of the lower unit;

[0008] The upper level unit of the first register unit is the first controller, the lower level unit of the nth register unit is the first controller, the upper level unit of the ith register unit is the i-1th register unit, and 2≤i≤n;

[0009] The management module is respectively connected to the control port of each selector;

[0010] The repair registers are respectively connected to the embedded memories of the corresponding functional modules, and the functional modules connected to the repair registers in the same self-repair chain are distributed in the same power domain.

[0011] Optionally, the management module is used to send a path selection instruction to the selector in each register unit to switch the internal port conduction relationship of the selector to complete the self-repair chain deformation, and the path selection instruction matches the current state of the functional module connected to the register unit;

[0012] The first controller is used to load the repair information into the deformed self-repair chain to complete the memory self-repair.

[0013] Optionally, if the current state of the functional module connected to the register unit is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector in the register unit to be connected;

[0014] If the current state of the functional module connected to the register unit is a normal working state, the path selection instruction instructs the first port and the third port in the selector in the register unit to be connected.

[0015] Optionally, the management module is further used to determine and generate a corresponding path selection instruction based on user instruction information, and the user instruction information includes the current status of each functional module.

[0016] Optionally, the current state of the repair register is consistent with the current state of the functional module to which it is connected.

[0017] Optionally, all backup registers are in a powered-up state.

[0018] Optionally, the first controller is further configured to read the repair information from an electronic fuse.

[0019] In a second aspect, an embodiment of the present application provides a memory self-repair method, which is applied to the above-mentioned memory self-repair circuit, and the method includes:

[0020] The management module sends a path selection instruction to the selector in each register unit to switch the internal port conduction relationship of the selector to complete the self-repair chain deformation, and the path selection instruction matches the current state of the functional module connected to the register unit;

[0021] The first controller loads the repair information into the deformed self-repair chain to complete the memory self-repair.

[0022] Optionally, if the current state of the functional module connected to the register unit is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector in the register unit to be connected;

[0023] If the current state of the functional module connected to the register unit is a normal working state, the path selection instruction instructs the first port and the third port in the selector in the register unit to be connected.

[0024] In a third aspect, an embodiment of the present application provides a chip, including: the above-mentioned memory self-repair circuit.

[0025] Compared with the prior art, the embodiments of the present application provide a memory self-repair circuit, method and chip, wherein the memory self-repair circuit includes at least one self-repair chain, the self-repair chain includes n register units, the register unit includes a repair register, a spare register and a selector, and the repair register and the spare register in the same register unit are the same. When a repair register fails, the internal conduction relationship is switched by the selector, for example, the second port and the third port in the selector are turned on, and the corresponding repair register is bypassed by the spare register. When the repair register does not work, other memories on the self-repair chain can still be self-repaired, thereby saving their losses or saving power consumption.

[0026] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0028] Figure 1 One of the memory self-repair circuits provided in the embodiments of the present application;

[0029] Figure 2 The second memory self-repair circuit provided in the embodiment of the present application;

[0030] Figure 3 A schematic diagram of the flow of a memory self-repair method provided in an embodiment of the present application.

[0031] In the figure: 100 - first controller; 200 - management module; 300 - self-repair chain; 310 - register unit; 311 - repair register; 312 - spare register; 313 - selector. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0033] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0034] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0035] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0036] In the description of the present application, it should be noted that the terms "upper", "lower", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or are the orientations or positional relationships in which the product of the application is usually placed when in use. They are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0037] In the description of this application, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "disposed" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0038] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0039] In order to reduce the failure rate of embedded memory, memory repair design can be added to the hardware to improve the chip mass production yield. For details, please refer to Figure 1 , Figure 1 This is one of the memory self-repair circuits provided in the embodiments of the present application. Figure 1 As shown, the memory repair information can be stored in the repair register (Built-In Self Repair Register, referred to as BISR register), and the repair register is strung into a BISR chain and moved to the BISR controller (BISRController, also known as the first controller) for processing, and finally the memory repair information is compressed into the electronic fuse (efuse, one-time programmable memory); the functional power management module (Functional Power Management Unit) triggers the repair sequence, decompresses the memory repair information of the electronic fuse, moves it to the BISR chain, and acts on the physical memory interface to achieve memory repair. At the same time, the repair registers of different power domains will be strung into multiple BISR chains, thereby ensuring that multiple power supplies are turned on / off in time, and the repair function can also work normally; the functional power management module determines the working order of multiple BISR chains according to the system power-on sequence. The so-called system power-on sequence is the power-on sequence of the entire chip, and the repair of each power domain memory is completed according to the power-on sequence.

[0040] like Figure 1 As shown, power supply 0, power supply 1, ..., and power supply n represent different power domains, each power domain corresponds to a BISR chain, and the number of repair registers in each BISR chain is not necessarily the same. Each repair register is connected to a corresponding embedded memory.

[0041] Figure 1 ① represents the efuse address, ② represents the efuse clock, ③ represents the efuse data, ④ represents the efuse read completion flag, ⑤ represents the repair enable, ⑥ represents the repair clock, and ⑦ represents the power enable [n-1:0].

[0042] The functional power management module can trigger the repair sequence through repair enable, repair clock and power enable.

[0043] There are the following possible scenarios in Soc chip design, which will affect Figure 1 The normal operation of the memory self-repair circuit shown in the figure results in the inability to complete the memory repair. When the functional modules in the same power domain are connected into a BISR chain, if the low-power design requires that only some functional modules will be enabled to save power, or some functional modules fail during mass production, the entire BISR chain will be broken, which will cause the repair register to not work properly and affect the entire BISR chain.

[0044] exist Figure 1 In the solution shown, the memory repair problem of different power domains during function use can be solved by stringing the repair registers of multiple power domains into multiple BISR chains. However, it also faces the problem that the memory repair BISR chain can also work normally when some functional modules of the same power domain (PowerDomain) are turned on and off.

[0045] If the above problems cannot be solved, some good products will not be put on the market, which will cause losses and waste. The most typical example is the multi-core (multiple CPU cores) design of the processor, where only some cores are available. Another problem will affect the implementation of low power consumption, as some functional modules in the same power domain cannot be turned off at will.

[0046] In order to overcome the above problems, the present application also provides a possible implementation method, please refer to Figure 2 , Figure 2 This is the second memory self-repair circuit provided in the embodiment of the present application.

[0047] like Figure 2As shown, the memory self-repair circuit includes a first controller 100, a management module 200 and at least one self-repair chain 300, the self-repair chain 300 includes n register units 310, the register unit 310 includes a repair register 311, a spare register 312 and a selector 313, the repair register 311 and the spare register 312 in the same register unit 310 are the same, wherein n is greater than or equal to 2.

[0048] The first controller 100 may be, but is not limited to, a BISR controller, also referred to as a BISR controller.

[0049] It should be understood that the repair register 311 and the spare register 312 have the same length. When repairing the embedded memory of the functional module, the repair register 311 needs to be used, and the spare register 312 cannot complete the repair of the embedded memory of the functional module.

[0050] The output end of the repair register 311 is connected to the first port of the selector 313, the output end of the spare register 312 is connected to the second port of the selector 313, the third port of the selector 313 serves as the output end of the register unit 310, and the input end of the repair register 311 and the input end of the spare register 312 serve as the input end of the register unit 310.

[0051] The input end of the register unit 310 is connected to the output end of the upper unit, and the output end of the register unit 310 is used to connect to the input end of the lower unit.

[0052] Among them, the upper unit of the 1st register unit 310 is the first controller 100, the lower unit of the nth register unit 310 is the first controller 100, the upper unit of the i-th register unit 310 is the i-1th register unit 310, and the lower unit of the i-1th register unit is the i-th register unit, 2≤i≤n.

[0053] It should be understood that the inputs of the repair register 311 and the spare register 312 are from the same source, and the outputs are controlled by the selector 313 .

[0054] The management module 200 is connected to the control port of each selector 313 respectively.

[0055] Optionally, the management module 200 can process the power switch signal to control the corresponding power domain to power on or off. The repair registers 311 are respectively connected to the embedded memories of the corresponding functional modules, and the functional modules connected to the repair registers 311 in the same self-repair chain 300 are distributed in the same power domain.

[0056] Optionally, the functional module may be but is not limited to a processor core (CPU core). It should be understood that the number of functional modules is the same as the number of register units 310 , and Ci in the present application represents the i-th functional module.

[0057] Optionally, when a repair register 311 fails, the internal conduction relationship is switched by the selector 313, for example, the second port and the third port in the selector 313 are connected, and the corresponding repair register 311 is bypassed by the standby register 312. When the repair register 311 does not work, other memories on the self-repair chain 300 can still be self-repaired, thereby saving their losses or saving power consumption.

[0058] Optionally, the management module 200 is used to send a path selection instruction to the selector 313 in each register unit 310 to switch the internal port conduction relationship of the selector 313 to complete the transformation of the self-repair chain 300, and the path selection instruction matches the current state of the functional module connected to the register unit 310.

[0059] Optionally, the path selection instruction is configurable, for example, when the repair register 311 needs to be selected, the path selection instruction can be configured as 0, and when the spare register 312 needs to be selected, the path selection instruction can be configured as 1. It should be understood that the path selection instruction can be but is not limited to a 1-bit selection signal.

[0060] The first controller 100 is used to load the repair information into the deformed self-repair chain 300 to complete the memory self-repair.

[0061] Optionally, if the current state of the functional module connected to the register unit 310 is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector 313 in the register unit 310 to be connected.

[0062] Optionally, the current state of the repair register 311 is consistent with the current state of the functional module to which it is connected. Normally, when the current state of the functional module is a power-off state or a failure state, the repair register 311 in the register unit 310 to which it is connected is in a non-working state, so it is necessary to switch the selector 313 to bypass the non-working repair register 311 through the spare register 312 to avoid affecting the repair work of the entire BISR chain. By increasing a small amount of design area, huge benefits are brought.

[0063] If the current state of the functional module connected to the register unit 310 is a normal working state, the path selection instruction instructs the first port and the third port in the selector 313 in the register unit 310 to be connected.

[0064] Optionally, the management module 200 is further configured to determine and generate a corresponding path selection instruction based on user instruction information, where the user instruction information includes a current state of each functional module.

[0065] In a possible implementation, the functional power management module may obtain user instruction information and manage the functional module according to the user instruction information, for example, controlling the power-on status of the functional module to put it in a power-off state or a power-on state.

[0066] In one possible scenario, when the device is low on power or has difficulty charging, it is necessary to turn on the low-power operation mode. When the user starts the low-power operation mode, the functional power management module can shut down some functional modules, and the shut-down functional modules can be pre-configured. At the same time, the corresponding user instruction information is generated, and the user instruction information includes that the current state of the shut-down functional modules is the power-off state. At this time, the management module 200 is also used to determine the generation of the corresponding path selection instruction based on the user instruction information, and control the selector 313 to switch the internal conduction port, so as to use the spare register 312 to bypass the repair register 311 in the same register unit 310. In low-power mode, the memory repair can still be completed through the self-repair chain. That is, in the present application scheme, when the device is in low-power operation mode, any functional module under the same power domain can be shut down according to the instruction.

[0067] Optionally, all the spare registers 312 are in a powered-on state to ensure that the self-repair chain is in an available state after switching.

[0068] Optionally, the first controller 100 is further configured to read repair information from the electronic fuse.

[0069] It should be understood that Figure 2 The structure shown is only a schematic diagram of a portion of the memory self-repair circuit. The memory self-repair circuit may also include Figure 2 More or fewer components as shown, or with Figure 2 Different configurations shown. Figure 2 Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0070] The memory self-repair method provided in the embodiment of the present application can be applied to but not limited to Figure 2 The memory self-repair circuit shown in the figure is shown in the figure. For the specific process, please refer to Figure 3 , the memory self-repair method includes: S501 and S502, which are specifically described as follows.

[0071] S501, the management module sends a path selection instruction to the selector in each register unit to switch the internal port conduction relationship of the selector to complete the self-repair chain deformation.

[0072] The path selection instruction matches the current state of the functional module connected to the register unit.

[0073] S502: The first controller loads the repair information into the deformed self-repair chain to complete the memory self-repair.

[0074] Optionally, if the current state of the functional module connected to the register unit 310 is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector 313 in the register unit 310 to be connected;

[0075] If the current state of the functional module connected to the register unit 310 is a normal working state, the path selection instruction instructs the first port and the third port in the selector 313 in the register unit 310 to be connected.

[0076] It should be noted that the memory self-repair method provided in this embodiment can perform the functions shown in the above circuit flow embodiment to achieve the corresponding technical effects. For the sake of brief description, for parts not mentioned in this embodiment, please refer to the corresponding contents in the above embodiments.

[0077] The present application also provides a chip, which includes Figure 2 The memory self-repair circuit shown.

[0078] In summary, the present embodiment provides a memory self-repair circuit, method and chip. The memory self-repair circuit includes at least one self-repair chain. The self-repair chain includes n register units. The register unit includes a repair register, a spare register and a selector. The repair register and the spare register in the same register unit are the same. When a repair register fails, the internal conduction relationship is switched by the selector. For example, the second port and the third port in the selector are turned on, and the corresponding repair register is bypassed by the spare register. When the repair register does not work, other memories on the self-repair chain can still be self-repaired, thereby saving their losses or saving power consumption.

[0079] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0080] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A memory self-repair circuit, characterized in that: The memory self-repair circuit includes a first controller, a management module and at least one self-repair chain, the self-repair chain includes n register units, the register unit includes a repair register, a spare register and a selector, and the repair register and the spare register in the same register unit are the same; The output end of the repair register is connected to the first port of the selector, the output end of the standby register is connected to the second port of the selector, the third port of the selector serves as the output end of the register unit, and the input end of the repair register and the input end of the standby register serve as the input end of the register unit; The input end of the register unit is connected to the output end of the upper unit, and the output end of the register unit is used to connect to the input end of the lower unit; The upper level unit of the first register unit is the first controller, the lower level unit of the nth register unit is the first controller, the upper level unit of the ith register unit is the i-1th register unit, and 2≤i≤n; The management module is respectively connected to the control port of each selector; The repair registers are respectively connected to the embedded memories of the corresponding functional modules, and the functional modules connected to the repair registers in the same self-repair chain are distributed in the same power domain; When the user starts the low power operation mode, the functional power management module shuts down some functional modules, which are pre-configured, and generates corresponding user instruction information, which includes that the current state of the functional modules of the shut down part is the power off state; at this time, the management module is also used to determine and generate corresponding path selection instructions based on the user instruction information, control the selector to switch the internal conduction port, so as to use the standby register to bypass the repair register in the same register unit; The functional module is a processor core, the current state of the repair register is consistent with the current state of the functional module connected thereto, the management module is used to send a path selection instruction to the selector in each register unit to switch the internal port conduction relationship of the selector to complete the self-repair chain deformation, and the path selection instruction matches the current state of the functional module connected to the register unit; The first controller is used to load the repair information into the deformed self-repair chain to complete the memory self-repair; If the current state of the functional module connected to the register unit is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector in the register unit to be connected; If the current state of the functional module connected to the register unit is a normal working state, the path selection instruction instructs the first port and the third port in the selector in the register unit to be connected.

2. The memory self-repair circuit according to claim 1, characterized in that: All backup registers are in a powered-up state.

3. The memory self-repair circuit according to claim 1, characterized in that: The first controller is further configured to read the repair information from the electronic fuse.

4. A memory self-repair method, characterized in that: Applied to the memory self-repair circuit of claim 1, the method comprising: The management module sends a path selection instruction to the selector in each register unit to switch the internal port conduction relationship of the selector to complete the self-repair chain deformation, and the path selection instruction matches the current state of the functional module connected to the register unit; The first controller loads the repair information into the deformed self-repair chain to complete the memory self-repair; If the current state of the functional module connected to the register unit is a power-off state or a failure state, the path selection instruction instructs the second port and the third port in the selector in the register unit to be connected; If the current state of the functional module connected to the register unit is a normal working state, the path selection instruction instructs the first port and the third port in the selector in the register unit to be connected.

5. A chip, characterized in that: include: The memory self-repair circuit according to any one of claims 1 to 3.