Method for repairing a storage device and system for repairing a storage device

Abstract

The application provides a storage device repair method and a storage device repair system. The storage device includes a storage array and a redundant storage array. Redundant storage units in the redundant storage array are used to repair and replace failed storage units in the storage array. The repair method includes obtaining information of the failed storage units in the storage array, and determining a first failed storage unit; determining a maximum failure direction of the first failed storage unit; and using a same-direction redundant storage array matched with the maximum failure direction of the first failed storage unit to repair and replace a storage row / column in the storage array in the maximum failure direction and located at the first failed storage unit. In this way, the repair rate of the failed storage units in the storage device and the qualified rate of the storage device in mass production can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of storage, in particular to a repair method of a storage device and a repair system of the storage device. BACKGROUND

[0002] Dynamic random access memory (DRAM) chips, as a kind of storage device, have been widely used in mobile phones, computers, servers and other devices due to their simple structure, high integration and low power consumption, and have played an important role in the consumer electronics field. In the manufacturing process of DRAM chips, there may be failed storage units in the storage array of the DRAM chip. In order to deal with this situation, a part of redundant storage array is reserved in the design of the DRAM chip, and redundant storage units are arranged on the redundant storage array. The redundant storage units are used to replace the failed storage units in the storage array to ensure that the DRAM chip can be used normally. At present, due to the area limitation, the number of redundant storage arrays of most DRAM chips is limited. Therefore, it is very important to utilize the limited redundant storage array to repair as many failed storage units as possible and improve the repair rate of the failed storage units. SUMMARY

[0003] In order to solve the above problems, the present application provides a repair method of a storage device and a repair system of the storage device, which can improve the repair rate of failed storage units in the storage device and improve the qualified rate of the storage device.

[0004] One technical solution adopted by the present application is to provide a repair method of a storage device, wherein the storage device includes a storage array and a redundant storage array, redundant storage units in the redundant storage array are used to repair and replace failed storage units in the storage array, the repair method includes obtaining information of the failed storage units in the storage array and determining a first failed storage unit; determining a maximum failure direction of the first failed storage unit; and using a same-direction redundant storage array matched with the maximum failure direction of the first failed storage unit to repair and replace a storage row / column in the storage array in which the first failed storage unit is located in the maximum failure direction.

[0005] In the repair method, obtaining the information of the failed storage units in the storage array and determining the first failed storage unit include obtaining address information of the failed storage units in each row and each column of the storage units in the storage array and establishing a failed storage unit distribution map; in the failed storage unit distribution map, a first failed storage unit of a first storage row / column extending in a first direction and having a failed storage unit is determined as the first failed storage unit, wherein a plurality of storage rows / columns extending in the first direction are arranged in a second direction in sequence, and the first direction and the second direction intersect each other.

[0006] The first direction is a word line direction in the memory array, and the second direction is a bit line direction in the memory array.

[0007] The maximum failure direction of the first failed memory cell is determined by obtaining a first number of failed memory cells in a storage row / column extending in the first direction in which the first failed memory cell is located, and obtaining a second number of failed memory cells in a storage column / row extending in the second direction in which the first failed memory cell is located, wherein the first number of failed memory cells and the second number of failed memory cells each include the first failed memory cell; comparing the first number and the second number, and selecting a storage row / column extending in the first direction or a storage column / row extending in the second direction with a larger number as the maximum failure direction of the first failed memory cell.

[0008] The repair method further includes updating information of failed memory cells in the memory array, and updating the first failed memory cell; and using the updated first failed memory cell to continue the steps of determining the maximum failure direction and performing the repair replacement.

[0009] When the failed memory cell replaced by the repair is removed, a redundant storage row / column in the redundant memory array in the same storage row / column as the failed memory cell replaced by the repair in the maximum failure direction is also removed synchronously.

[0010] The repair method further includes determining whether the number of failed memory cells in the memory array is zero; in response to the number of failed memory cells in the memory array not being zero, further determining whether the redundant storage row / column extending in the first direction and / or the redundant storage column / row extending in the second direction is exhausted; and in response to neither the redundant storage row / column extending in the first direction nor the redundant storage column / row extending in the second direction being exhausted, returning to continue the steps of updating information of failed memory cells in the memory array and updating the first failed memory cell.

[0011] In response to the redundant storage row / column extending in the first direction being exhausted and the redundant storage column / row extending in the second direction not being exhausted, the repair method further includes updating information of existing failed memory cells in the memory array, and counting the number of failed memory cells on each storage column / row extending in the second direction in the memory array; sorting the storage column / rows extending in the second direction in the memory array according to the number of failed memory cells on each storage column / row extending in the second direction from more to less; and using the existing redundant storage column / row extending in the second direction to replace the failed memory cells on the storage column / rows extending in the second direction in the memory array in sequence according to the sorting result.

[0012] Another technical solution adopted by the present application is to provide a repair system of a storage device, the repair system comprising a host device and a storage device, the host device being configured to execute the repair method to generate repair replacement information; and the storage device being coupled to the host device, wherein the storage device comprises a storage die, and the storage die comprises at least one memory block, and each memory block comprises a memory array and a redundant memory array.

[0013] The storage device further comprises a control die, and the control die is packaged together with the storage die, wherein the control die receives the repair replacement information.

[0014] The repair method of the storage device provided by the present application can determine the first failed storage unit by obtaining the information of the failed storage unit in the memory array, determine the maximum failure direction of the first failed storage unit based on the first failed storage unit, determine the direction with the largest number of failed storage units, and finally replace the storage row / column in the maximum failure direction of the first failed storage unit in the memory array with a redundant storage row / column in the redundant memory array in the same direction, so that the redundant storage unit in the redundant memory array replaces the failed storage unit in the memory array, thereby repairing the maximum number of failed storage units in the memory array of the storage device, and further improving the repair rate of the failed storage unit of the storage device. Further, the repair of the failed storage unit of the storage device enables the storage device to be used normally. For example, if the repair method of the storage device provided by the present application is applied to the pre-production test of the storage device, the qualified rate of the production of the storage device can be improved. If the repair method of the storage device provided by the present application is applied to the repair during the normal operation of the storage device, the time of each power-on repair can be shortened. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0016] Wherein:

[0017] Figure 1 is a structural schematic diagram of an embodiment of the storage device provided by the present application;

[0018] Figure 2 is a flowchart of an embodiment of the repair method of the storage device provided by the present application;

[0019] Figure 3 is Figure 2A flowchart of step S110 in the embodiment;

[0020] Figure 4 is a matrix distribution of failed storage units provided by the present application Figure 1 A structural diagram of the embodiment;

[0021] Figure 5 is Figure 2 A flowchart of step S120 in the embodiment;

[0022] Figure 6 is a flowchart of another embodiment of the repair method of the storage device provided by the present application;

[0023] Figure 7 is Figure 6 A flowchart of step S240 in the embodiment;

[0024] Figure 8 is a flowchart of still another embodiment of the repair method of the storage device provided by the present application;

[0025] Figure 9 is a flowchart of still another embodiment of the repair method of the storage device provided by the present application;

[0026] Figure 10 is a flowchart of an embodiment of the application of the repair method of the storage device;

[0027] Figure 11 is a schematic diagram of an embodiment of a two-difference tree;

[0028] Figure 12 is a structural diagram of an embodiment of repair replacement of the matrix distribution of failed storage units provided by the present application;

[0029] Figure 13 is a structural diagram of an embodiment of the repair system of the storage device provided by the present application;

[0030] Figure 14 is Figure 13 A structural diagram of another embodiment of the storage device in the embodiment. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. It can be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings, but not all the structures. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0032] Reference to an "example" in this text means that a particular feature, structure, or characteristic described in connection with the example can be included in at least one example of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same example, nor are they necessarily mutually exclusive or alternative examples to one another. It is expressly understood that the examples described herein can be combined with each other in their various permutations.

[0033] Reference is made to Figure 1 , Figure 1 is a structural schematic diagram of an embodiment of the storage device provided by the application. As shown in Figure 1 , the storage device includes a storage array and a redundant storage array. The storage array includes storage rows and storage columns, where the storage rows can be referred to as word lines (WL) and the storage columns can be referred to as bit lines (BL). The storage rows extend along a first direction and the storage columns extend along a second direction, and a plurality of storage rows are arranged in sequence along the second direction and a plurality of storage columns are arranged in sequence along the first direction; or the storage rows extend along the second direction and the storage columns extend along the first direction, and a plurality of storage rows are arranged in sequence along the first direction and a plurality of storage columns are arranged in sequence along the second direction. The redundant storage array includes redundant storage rows (YWL) and / or redundant storage columns (YBL), and the redundant storage rows and / or the redundant storage columns can be referred to as redundant resources. The redundant storage rows extend along the first direction and the redundant storage columns extend along the second direction, and a plurality of redundant storage rows are arranged in sequence along the second direction and a plurality of redundant storage rows are arranged in sequence along the first direction; or the redundant storage rows extend along the second direction and the redundant storage columns extend along the first direction, and a plurality of redundant storage rows are arranged in sequence along the first direction and a plurality of redundant storage columns are arranged in sequence along the second direction. The redundant storage rows extend in the same direction as the storage rows, the redundant storage columns extend in the same direction as the storage columns, and the first direction and the second direction intersect each other. The redundant storage units in the redundant storage array can be used to repair and replace failed storage units in the storage array. Understandably, when the redundant storage units in the redundant storage array repair and replace the failed storage units in the storage array, the addresses of the storage rows / columns are mapped to the addresses of the redundant storage rows / columns, thereby achieving the repair and replacement of the failed storage units in the storage array by the redundant storage units in the redundant storage array.

[0034] The application provides a repair method for a storage device, reference is made to Figure 2 , Figure 2 is a flowchart of an embodiment of the repair method for a storage device provided by the application, as shown in Figure 2 , the repair method includes the following steps:

[0035] Step S110: obtaining information of failed storage units in the storage array, and determining a first failed storage unit.

[0036] The storage array comprises storage units, wherein when data is written and read, if the written data is consistent with the read data, the storage unit is considered as a normal storage unit; if the written data is inconsistent with the read data, the storage unit is considered as a failed storage unit, i.e. a failed storage unit. The information of the failed storage units in the storage array can be the number of the failed storage units, the addresses of the failed storage units, etc. When the information of the failed storage units in the storage array is obtained, predetermined data can be written into all the storage units in the storage array, and the read data is compared with the predetermined data. The number of the storage units with inconsistent read data and predetermined data is used as the number of the failed storage units in the storage array, and the address information of each failed storage unit is obtained. In this regard, the method of obtaining the information of the failed storage units in the storage array is not specifically limited. After the information of the failed storage units is determined, the first failed storage unit is determined. The first failed storage unit can be any one of the failed storage units in the storage array.

[0037] S120: determining the maximum failure direction of the first failed storage unit in the first direction and the second direction.

[0038] Based on the determined first failed storage unit, the maximum failure direction of the first failed storage unit in the first direction and the second direction is determined. The maximum failure direction can be the direction in which the number of failed storage units in the storage row / column in which the first failed storage unit is located is the largest in the first direction and the second direction, with the first failed storage unit as the reference point.

[0039] S130: repairing and replacing the storage row / column in which the first failed storage unit is located in the maximum failure direction of the first failed storage unit in the storage array by using the same-direction redundant storage array matching the maximum failure direction of the first failed storage unit.

[0040] After the maximum failure direction of the first failed storage unit is determined, the storage row / column in the maximum failure direction of the first failed storage unit in the storage array is repaired and replaced by using the redundant storage row / column in the redundant storage array matching the maximum failure direction of the first failed storage unit, so as to realize the repair and replacement of the failed storage units in the storage array by the redundant storage units in the redundant storage array.

[0041] For example, when the maximum failure direction of the first failed storage unit in the storage array is the storage row direction, a redundant storage row in the redundant storage array that matches the same direction as the storage row direction is used to repair and replace the storage row in the storage array where the first failed storage unit is located in the maximum failure direction of the first failed storage unit, i.e., the address of the storage row where the first failed storage unit is located is mapped to the address of a redundant storage row that is in the same direction as the maximum failure direction, so that the redundant storage unit in the redundant storage row repairs and replaces the failed storage unit in the storage row.

[0042] The repair method of the storage device provided in the present application determines the first failed storage unit by obtaining the information of the failed storage units in the storage array, determines the maximum failure direction of the first failed storage unit based on the first failed storage unit, determines the direction with the largest number of failed storage units, and finally repairs and replaces the storage row / column in the storage array where the first failed storage unit is located in the maximum failure direction of the first failed storage unit by using a redundant storage row / column in the redundant storage array that matches the same direction as the maximum failure direction of the first failed storage unit, so that the redundant storage unit in the redundant storage array repairs and replaces the failed storage unit in the storage array, thereby realizing the repair of the largest number of failed storage units in the storage array of the storage device and improving the repair rate of the failed storage units of the storage device. Further, the repair of the failed storage units of the storage device enables the storage device to be used normally. For example, if the repair method of the storage device provided in the present application is applied to the pre-production test of the storage device, the yield of the production of the storage device can be improved. If the repair method of the storage device provided in the present application is applied to the repair during the normal operation of the storage device, the time for each power-on repair can be shortened.

[0043] Optionally, referring to Figure 3 , Figure 3 is Figure 2 In an embodiment, the flowchart of step S110 is as shown in Figure 3 , step S110 obtains the information of the failed storage units in the storage array and determines the first failed storage unit, including the following steps:

[0044] Step S111: Obtain the address information of the failed storage units in each row and each column of storage units in the storage array, and establish a failed storage unit distribution map.

[0045] The address information of the failed storage units in each row and each column of storage units in the storage array is obtained, and a failed storage unit distribution map is established by using the address information of the failed storage units. The failed storage unit distribution map can be a coordinate distribution map of the failed storage units, in which each row of the storage array is corresponded to a different point on the vertical coordinate and each column of the storage array is corresponded to a different point on the horizontal coordinate. Alternatively, referring to Figure 4 ,Figure 4 is a failure storage unit distribution of the present application Figure 1 a structural schematic diagram of an embodiment, as Figure 4 As shown, each word line in the storage array corresponds to a row in the matrix, each bit line in the storage array corresponds to a column in the matrix, and the intersection position represents a failure storage unit, thereby establishing a matrix distribution diagram of the failure storage units.

[0046] Step S112: In the failure storage unit distribution diagram, a first failure storage unit of a first storage row / column extending in a first direction and having a failure storage unit is determined as the first failure storage unit, wherein a plurality of storage rows / columns extending in the first direction are arranged in sequence along a second direction, and the first direction and the second direction intersect each other.

[0047] In the failure storage unit distribution diagram, a first failure storage unit of a first storage row / column extending in a first direction and having a failure storage unit is determined as the first failure storage unit, which can be understood as selecting the first direction in the failure storage unit distribution diagram and determining the first failure storage unit as the first failure storage unit appearing in the storage row / column extending in the first direction. Wherein a plurality of storage rows / columns extending in the first direction are arranged in sequence along a second direction, and the first direction and the second direction intersect each other. For example, when the failure storage unit distribution diagram is a coordinate distribution diagram, the first direction can be the longitudinal coordinate direction or the horizontal coordinate direction, and correspondingly the second direction is the horizontal coordinate direction or the longitudinal coordinate direction. When the first direction is selected as the horizontal coordinate direction in the coordinate distribution diagram, the second direction is the longitudinal coordinate direction, and when determining the first failure storage unit, the first failure storage unit appearing in the horizontal coordinate direction is determined as the first failure storage unit.

[0048] Alternatively, when the failure storage unit distribution diagram is a matrix distribution diagram, the first direction can be a row or a column in the matrix distribution diagram, and correspondingly the second direction is a column or a row in the matrix distribution diagram. When the first direction is selected as a row in the matrix distribution diagram, the second direction is a column in the matrix distribution diagram, and when determining the first failure storage unit, the first failure storage unit appearing in the row direction in the matrix distribution diagram is determined as the first failure storage unit. For example, as shown in the failure storage unit matrix distribution diagram, Figure 4 As shown, each row in R0-R6 in the failure storage unit matrix distribution diagram corresponds to a storage row, or each word line in the storage array; each column in C0-C6 corresponds to a storage column, or each bit line in the storage array. The first direction is the storage row direction, and the first failure storage unit appearing in R0-R6 is determined, i.e., the failure storage unit at the position (R0, C0) in the failure storage unit matrix distribution diagram, which is determined as the first failure storage unit.

[0049] The embodiment obtains address information of the failed storage units in each row and each column of storage units in the storage array, and establishes a failed storage unit distribution map, so as to determine a first failed storage unit of a first storage row / column extending in the first direction and having the failed storage unit as the first failed storage unit, thereby realizing selection of the first failed storage unit. Further, the method for selecting the first failed storage unit is simple and easy to implement, and can improve the repair efficiency of the failed storage unit.

[0050] Optionally, the first direction is a word line direction in the storage array, and the second direction is a bit line direction in the storage array.

[0051] Optionally, the first direction is a word line direction in the storage array, and the second direction is a bit line direction in the storage array. Figure 5 , Figure 5 is Figure 2 An embodiment of step S120 is shown in a flowchart, which includes the following steps: Figure 5

[0052] Step S121: obtaining a first quantity of failed storage units in a storage row / column extending in the first direction and having the first failed storage unit, and obtaining a second quantity of failed storage units in a storage column / row extending in the second direction and having the first failed storage unit, wherein the first quantity of failed storage units and the second quantity of failed storage units each include the first failed storage unit.

[0053] After the first failed storage unit is determined, the first quantity of failed storage units in the storage row / column extending in the first direction and having the first failed storage unit is obtained, and the second quantity of failed storage units in the storage column / row extending in the second direction and having the first failed storage unit is obtained. Since the first direction and the second direction intersect each other, the first quantity of failed storage units and the second quantity of failed storage units each include the first failed storage unit.

[0054] Step S122: comparing the first quantity and the second quantity, and selecting the storage row / column extending in the first direction or the storage column / row extending in the second direction with the larger quantity as the maximum failure direction of the first failed storage unit.

[0055] The first quantity and the second quantity obtained are compared, and the storage row / column extending in the first direction or the storage column / row extending in the second direction with the larger quantity is selected as the maximum failure direction of the first failed storage unit, that is, the direction with the largest quantity of failed storage units in the storage row / column extending in the first direction or the storage column / row extending in the second direction is selected as the maximum failure direction of the first failed storage unit.

[0056] ​For example, if the first failed storage unit is located in a storage row extending in the first direction, and the number of failed storage units in the storage row is 3, and the number of failed storage units in the storage column in which the first failed storage unit is located is 4, then the storage column in which the first failed storage unit is located is taken as the maximum failure direction.

[0057] The embodiment selects the first number of failed storage units in the storage row / column in which the first failed storage unit is located extending in the first direction, and the second number of failed storage units in the storage column / row in which the first failed storage unit is located extending in the second direction, so as to take the direction with the larger number of failed storage units in the storage row / column in which the first failed storage unit is located as the maximum failure direction of the first failed storage unit. Thus, when the redundant storage array is used to repair and replace the storage row / column in which the first failed storage unit is located extending in the maximum failure direction, more failed storage units can be repaired, and thus the repair rate of the failed storage units of the storage device is improved.

[0058] Optionally, referring to Figure 6 , Figure 6 is a flowchart of another embodiment of the repair method of the storage device provided in the present application. As shown in Figure 6 , the repair method of the embodiment comprises the following steps:

[0059] Step S210: acquiring information of failed storage units in the storage array, and determining a first failed storage unit.

[0060] Step S210 can be implemented by steps S111 and S112, which are not described herein.

[0061] Step S220: determining a maximum failure direction of the first failed storage unit in a first direction and a second direction.

[0062] The details of step S220 are described in step S120, which are not described herein. Step S220 can also be implemented by steps S121 and S122.

[0063] Step S230: using a redundant storage array matching the maximum failure direction of the first failed storage unit to repair and replace the storage row / column in which the first failed storage unit is located extending in the maximum failure direction.

[0064] The details of step S230 are described in step S130, which are not described herein.

[0065] Step S240: updating the information of failed storage units in the storage array, and updating the first failed storage unit.

[0066] After the redundant storage array repairs and replaces the storage row / column where the first failed storage unit is located in the maximum failed direction of the failed storage units in the storage array, the information of the failed storage units in the storage array is updated, and the first failed storage unit is updated based on the updated information of the failed storage units in the storage array.

[0067] Step S250: Continue to perform the steps of determining the maximum failed direction and repairing and replacing with the updated first failed storage unit.

[0068] After the updated first failed storage unit is determined, the steps of determining the maximum failed direction and repairing and replacing of the first failed storage unit are continued to be performed with the updated first failed storage unit, so as to continue to repair the failed storage units of the storage device.

[0069] The repair method of the embodiment improves the utilization rate of the redundant storage array in the storage device by updating the information of the failed storage units in the storage array and updating the first failed storage unit, so as to continue to select the maximum failed direction and perform the steps of repairing and replacing based on the updated first failed storage unit.

[0070] Optionally, in order to improve the repair efficiency of the failed storage units in the storage array, refer to Figure 7 , Figure 7 is Figure 6 The flowchart of one embodiment of step S240 in the embodiment is shown in Figure 7 The step S240 updates the information of the failed storage units in the storage array and updates the first failed storage unit, which includes the following steps:

[0071] Step S241: Remove the failed storage unit repaired and replaced to update the failed storage unit distribution map of the storage array.

[0072] On the basis of the original failed storage unit distribution map, the failed storage unit repaired and replaced is removed, so that the distribution map after the failed storage unit repaired and replaced is removed is used as the updated failed storage unit distribution map of the storage array.

[0073] Step S242: Determine the updated first failed storage unit in the updated failed storage unit distribution map of the storage array.

[0074] The first failed storage unit is re-determined in the updated failed storage unit distribution map of the storage array, so as to update the first failed storage unit.

[0075] The embodiment updates the distribution map of the failed storage units of the storage array by removing the failed storage unit replaced by the repair replacement, so that the distribution map after removing the failed storage unit replaced by the repair replacement is the updated distribution map of the failed storage units of the storage array, and the first failed storage unit is determined based on the updated distribution map of the failed storage units of the storage array, so that the first failed storage unit can be quickly updated, thereby improving the repair efficiency of the failed storage units of the storage device.

[0076] Optionally, when the failed storage unit replaced by the repair replacement is removed, the redundant storage row / column in the redundant storage array which is in the same storage row / column as the failed storage unit replaced by the repair replacement in the maximum failure direction is also removed. Wherein, the removal of the redundant storage row / column in the redundant storage array which is in the same storage row / column as the failed storage unit replaced by the repair replacement in the maximum failure direction means that the redundant storage row / column used for repairing and replacing the storage row / column in which the first failed storage unit in the maximum failure direction is located in the storage array is marked, so as to avoid the marked redundant storage row / column from being repeatedly used in the subsequent repair replacement process.

[0077] Optionally, based on the above-mentioned embodiment, refer to Figure 8 , Figure 8 is a flowchart of another embodiment of the repair method of the storage device provided by the present application, as shown in Figure 8 the repair method further comprises the following steps:

[0078] Step S360: determining whether the number of failed storage units in the storage array is zero.

[0079] After the redundant storage row / column in the redundant storage array is repaired and replaced in the storage row / column in which the first failed storage unit in the maximum failure direction is located in the storage array, part of the failed storage units in the storage array are repaired, and it is determined whether there are still failed storage units in the storage array which have not been repaired and replaced, i.e. whether the number of failed storage units in the storage array is zero. Wherein, the determination of whether the number of failed storage units in the storage array is zero can further determine whether the number of failed storage units in the storage array is zero by reacquiring the information of the failed storage units in the storage array. Preferably, the number of failed storage units in the storage array can be determined by confirming whether the number of failed storage units in the storage array is zero in the updated distribution map of the failed storage units of the storage array.

[0080] Step S370: in response to the number of failed storage units in the storage array not being zero, further determining whether the redundant storage row / column extending along the first direction and / or the redundant storage column / row extending along the second direction is exhausted.

[0081] In response to the number of failed storage units in the storage array not being zero, i.e. there are still failed storage units in the storage array that have not been repaired and replaced, it is further determined whether the redundant storage rows / columns extending in the first direction and / or the redundant storage columns / rows extending in the second direction are exhausted.

[0082] Step S380: In response to the redundant storage rows / columns extending in the first direction and the redundant storage columns / rows extending in the second direction not being exhausted, the step of updating the information of the failed storage units in the storage array and updating the first failed storage unit is returned to continue.

[0083] If the redundant storage rows / columns extending in the first direction and the redundant storage columns / rows extending in the second direction are not exhausted, in response to the redundant storage rows / columns extending in the first direction and the redundant storage columns / rows extending in the second direction not being exhausted, the step of updating the information of the failed storage units in the storage array and updating the first failed storage unit is returned to continue to repair and replace the first failed storage unit in the storage array in the maximum failed direction by using the redundant storage rows / columns extending in the first direction and the redundant storage columns / rows extending in the second direction that are not exhausted.

[0084] The embodiment determines whether there are still failed storage units in the storage array that need to be repaired by determining whether the number of failed storage units in the storage array is zero, and when there are still failed storage units in the storage array that need to be repaired, it is further determined whether the redundant storage rows / columns extending in the first direction and / or the redundant storage columns / rows extending in the second direction are exhausted, so as to continue repairing the failed storage units in the storage array by using the redundant storage resources that are not exhausted, thereby improving the repair efficiency of the failed storage units in the storage array and further improving the repair rate of the storage device.

[0085] Optionally, on the basis of the above embodiment, with reference to Figure 9 , Figure 9 is a flowchart of another embodiment of the repair method of the storage device provided by the present application, as shown in Figure 9 In response to the redundant storage rows / columns extending in the first direction being exhausted and the redundant storage columns / rows extending in the second direction not being exhausted, the repair method further comprises the following steps:

[0086] Step S491: updating the information of the existing failed storage units in the storage array and counting the number of failed storage units in each storage column / row extending in the second direction in the storage array.

[0087] When the redundant storage rows / columns extending in the first direction are exhausted, but the redundant storage columns / rows extending in the second direction are not exhausted, the redundant storage columns / rows extending in the second direction can also continue to repair the failed storage cells in the storage array, at this time, the information of the existing failed storage cells in the storage array is updated, and the number of failed storage cells on each storage column / row extending in the second direction in the storage array is counted.

[0088] Step S492: The number of failed storage cells on each storage column / row extending in the second direction in the storage array is sorted from more to less.

[0089] The number of failed storage cells on each storage column / row extending in the second direction in the storage array is sorted from more to less.

[0090] Step S493: The existing redundant storage columns / rows extending in the second direction are used to repair and replace the failed storage cells on the storage columns / rows extending in the second direction in the storage array in order according to the sorting result.

[0091] The existing redundant storage columns / rows extending in the second direction are used to repair and replace the failed storage cells on the storage columns / rows extending in the second direction in the storage array in order according to the sorting result of the number of failed storage cells on each storage column / row extending in the second direction in the storage array, so that the existing redundant storage columns / rows extending in the second direction repair as many failed storage cells on the storage columns / rows extending in the second direction as possible. Until the number of failed storage cells in the storage array is zero, or the existing redundant storage columns / rows extending in the second direction are exhausted, the repair of the failed storage cells in the storage array is stopped.

[0092] For example, when the redundant storage rows extending in the first direction are exhausted, but the redundant storage columns extending in the second direction are not exhausted, the number of failed storage cells on each storage column extending in the second direction is counted, and the storage columns are sorted according to the number of failed storage cells from more to less, and finally the redundant storage columns are used to repair and replace the storage columns in the sorting result to repair and replace the failed storage cells on the storage columns.

[0093] The embodiment updates the information of the existing failed memory cells in the storage array when the redundant storage rows / columns extending along the first direction are consumed but the redundant storage rows / columns extending along the second direction are not consumed, and counts the number of the failed memory cells on each storage row / column extending along the second direction in the storage array, so as to sort the number of the failed memory cells on each storage row / column in descending order, and finally uses the existing redundant storage rows / columns extending along the second direction to repair and replace according to the descending order of the number of the failed memory cells on each storage row / column extending along the second direction, so as to repair more failed memory cells in the storage array by using the limited redundant storage rows / columns, thereby improving the repair rate of the failed memory cells in the storage device.

[0094] Referring to Figure 10 , Figure 10 is a flowchart of an embodiment of the repair method of the storage device. As shown in Figure 10 , in the repair method of the embodiment, the storage rows in the storage array extend along the first direction, or the first direction is the word line direction in the storage array; the storage columns in the storage array extend along the second direction, or the second direction is the bit line direction in the storage array, and the repair method of the embodiment can be implemented by binary tree traversal. The binary tree traversal refers to visiting all nodes in the binary tree along a certain search route once and only once. Referring to Figure 11 , Figure 11 is a schematic diagram of an embodiment of a binary tree, as shown in Figure 11 , the traversal order of the nodes can be divided into three ways: pre-order, in-order and post-order. The pre-order traversal order is root node -> left subtree -> right subtree, which corresponds to the traversal order in Figure 11 : A -> B -> D -> H -> E -> I -> J -> C -> F -> K -> G -> L -> M; the in-order traversal order is left subtree -> root node -> right subtree, which corresponds to the traversal order in Figure 11 : H -> D -> B -> I -> E -> J -> A -> F -> K -> C -> L -> G -> M; and the post-order traversal order is left subtree -> right subtree -> root node, which corresponds to the traversal order in Figure 11 ​The traversal order in the tree is: H->D->I->J->E->B->K->F->L->M->G->C->A. In this embodiment, the traversal order used by the repair method is similar to the pre-order traversal, but not exactly the same. In the pre-order traversal, the root node is visited first, and then the left sub-tree; since the left sub-tree is also the root node of the level, the left sub-tree of the next level is continuously visited, and the traversal is performed in this order until the left and right sub-trees of the last level; then the right sub-tree of the previous level is visited in turn; when returning to the first level, the right sub-tree is also visited in the order of root node->left sub-tree->right sub-tree. This traversal order can be implemented by a recursive algorithm.

[0095] The repair method of this embodiment is implemented based on the pre-order traversal of the binary tree, and the specific steps are as follows: obtaining the information of the failed storage units in the storage array, establishing the matrix distribution diagram of the failed storage units of the storage array, referring to Figure 12 , Figure 12 is a structural schematic diagram of the repair and replacement of the failed storage unit matrix distribution diagram of this application, as shown in Figure 12As shown, each row in R0-R6 in the failed storage cell matrix distribution diagram corresponds to each word line in the storage array, or in other words, corresponds to a storage row; each column in C0-C6 corresponds to each bit line in the storage array, or in other words, corresponds to a storage column. Wherein, when determining the first failed storage cell, the first direction is the word line direction in the storage array, and the second direction is the bit line direction in the storage array, that is, the first failed storage cell appearing in the rows R0-R6 in the failed storage cell matrix distribution diagram is determined as the first failed storage cell, at this time, the first failed storage cell is the failed storage cell labeled (R0, C0) in the failed storage cell matrix distribution diagram. Based on the first failed storage cell, the number of failed storage cells in the R0 row direction and the C0 column direction is counted respectively, then the number of failed storage cells in the R0 row direction is 3, and the number of failed storage cells in the C0 column direction is 2, and it is confirmed that the maximum failure direction is the R0 row direction; a redundant storage row in the redundant storage array matched with the R0 row direction is used to replace and repair the R0 row, so as to repair and replace a word line in the storage array on which the first failed storage cell in the maximum failure direction R0 row is located. After repairing and replacing the failed storage cells in the storage array, the failed storage cell matrix distribution diagram is updated, that is, the failed storage cells on the R0 row are removed, and a redundant storage row is also removed. Based on the updated failed storage cell matrix distribution diagram, it is determined whether the number of failed storage cells is zero, if the number of failed storage cells is zero, the repair and replacement is ended; in response to the number of failed storage cells not being zero, it is further determined whether the redundant storage row and the redundant storage column are exhausted. In response to neither the redundant storage row nor the redundant storage column being exhausted, the first failed storage cell and the maximum failure direction are updated, and the steps of repair and replacement are continued to be executed. In response to the redundant storage row being exhausted and the redundant storage column not being exhausted, the number of failed storage cells in each bit line direction is counted and sorted in descending order according to the number, and the existing redundant storage column in the bit line direction is sequentially repaired and replaced to the failed storage cells on each bit line in the storage array according to the sorting result. Similarly, in response to the redundant storage row in the word line direction not being exhausted and the redundant storage column in the bit line direction being exhausted, the number of failed storage cells in each word line direction is counted and sorted in descending order according to the number, and the existing redundant storage row in the word line direction is sequentially repaired and replaced to the failed storage cells in each word line direction in the storage array according to the sorting result.

[0096] The application also provides a repair system of a storage device, referring to Figure 13 , Figure 13 is a structural schematic diagram of an embodiment of the repair system of the storage device provided by the application, as Figure 13As shown, the repair system 10 includes a host device 110 and a storage device 120. The host device 110 is configured to execute the repair method of any of the above embodiments to generate repair replacement information. The storage device 120 is coupled to the host device 110 and receives the repair replacement information. The storage device 120 includes a storage die 121 including at least one memory block (not labeled in the figure), and each memory block includes a memory array (not labeled in the figure) and a redundant memory array (not labeled in the figure). It can be understood that the host device 110 is configured to execute the repair method of any of the above embodiments to generate repair replacement information, and the storage device 120 receives the repair replacement information to repair and replace the failed memory cells in the memory array using the redundant resources of the redundant memory array.

[0097] The repair system 10 of the storage device provided in the present application includes the host device 110 and the storage device 120. The host device 110 is configured to execute the repair method of any of the above embodiments to generate repair replacement information, and the storage device 120 repairs the failed memory cells in the memory array based on the repair replacement information. That is, the host device 110 determines the first failed memory cell by obtaining the information of the failed memory cells in the memory array, determines the maximum failure direction of the first failed memory cell based on the first failed memory cell, determines the direction with the largest number of failed memory cells, and finally repairs and replaces the memory row / column in the maximum failure direction of the first failed memory cell in the memory array with a redundant memory row / column in the same direction of the redundant memory array to repair and replace the failed memory cells in the memory array with the redundant memory cells in the redundant memory array. Thus, the maximum number of failed memory cells in the memory array of the storage device 120 is repaired, and the repair rate of the failed memory cells of the storage device 120 is improved. Further, the repair of the failed memory cells of the storage device 120 enables the storage device 120 to be used normally. For example, if the repair method of the storage device provided in the present application is applied to the pre-production test of the storage device 120, the yield of the storage device 120 can be improved. If the repair method of the storage device provided in the present application is applied to the repair of the storage device 120 during normal operation, the time for each power-on repair can be shortened.

[0098] In other embodiments, referring to Figure 14 , Figure 14 is Figure 13 the structure of another embodiment of the storage device in the embodiment, as Figure 14As shown, the storage device 120 further comprises a control die 122 connected with the storage die 121, and the control die 122 is coupled with the host device 110, wherein the control die 122 is packaged together with the storage die 121, and the control die 122 is configured to receive the repair replacement information. That is, the control die 122 receives the repair replacement information generated by the host device 110, and the control die 122 repairs and replaces the failed storage unit in the storage array based on the repair replacement information using the redundant resource of the redundant storage array.

[0099] The present embodiment enables the storage device 120 to repair the failed storage unit in the storage array according to the change of the failed storage unit in the storage array by arranging the control die 122 in the storage device 120.

[0100] In other embodiments, the repair method and the repair system of the present application can be applied to the test stage of the storage device 120, such as the CP (Chip Probe) test stage. In such embodiments, the host device 110 can be implemented by the host computer of the test machine, which executes any one of the repair methods in the foregoing repair method embodiments to generate the repair replacement information for use in repairing the failed storage unit when the storage device 120 is normally operated after being shipped.

[0101] The above description is merely an embodiment of the present application, and does not limit the patent scope of the application. Any equivalent structure or equivalent process conversion according to the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A repair method of a storage device, characterized by, The storage device includes a storage array and a redundant storage array, a redundant storage unit in the redundant storage array is used for repair replacement of a failed storage unit in the storage array, and the repair method includes: acquiring information of a failed storage unit in the storage array, and determining a first failed storage unit; Determine the maximum failure direction of the first failed storage unit in the first direction and the second direction; Repair and replace the storage row / column in which the first failed storage unit is located in the maximum failure direction of the first failed storage unit in the storage array using the storage row / column of the same direction redundant storage array matched with the maximum failure direction of the first failed storage unit; Wherein, the maximum failure direction of the first failed storage unit in the first direction and the second direction includes: Acquire the first number of failed storage units in the storage row / column in which the first failed storage unit is located along the first direction, and acquire the second number of failed storage units in the storage column / row in which the first failed storage unit is located along the second direction, wherein the first number of failed storage units and the second number of failed storage units respectively include the first failed storage unit; Compare the first number and the second number, and select the storage row / column extending along the first direction or the storage column / row extending along the second direction with a larger number as the maximum failure direction of the first failed storage unit.

2. The repair method of claim 1, wherein The acquisition of the information of the failed storage unit in the storage array and the determination of the first failed storage unit includes: acquiring the address information of the failed storage unit in each row and each column of the storage unit in the storage array, and establishing a failed storage unit distribution map; In the failed storage unit distribution map, determine a first failed storage unit of a first storage row / column extending along the first direction with a failed storage unit as the first failed storage unit, wherein a plurality of storage rows / columns extending along the first direction are arranged in sequence along the second direction, and the first direction and the second direction intersect each other.

3. The repair method of claim 2, wherein, The first direction is the word line direction in the storage array, and the second direction is the bit line direction in the storage array.

4. The repair method according to claim 2, characterized in that, Further comprising: Update the information of the failed storage unit in the storage array, and update the first failed storage unit; With the updated first failed storage unit, continue to perform the steps of determining the maximum failure direction and the repair replacement.

5. The repair method of claim 4, wherein The update of the information of the failed storage unit in the storage array and the update of the first failed storage unit include: removing the repaired and replaced failed storage unit to update the failed storage unit distribution map of the storage array; In the updated failed storage unit distribution map of the storage array, update the first failed storage unit.

6. The repair method of claim 5, wherein When removing the repaired and replaced failed storage unit, synchronously remove the redundant storage row / column in the redundant storage array which is in the same storage row / column as the repaired and replaced failed storage unit in the maximum failure direction.

7. The repair method according to claim 6, characterized in that, Further comprising: Determine whether the number of failed storage units in the storage array is zero; in response to the number of failed memory cells in the memory array not being zero, further determining whether a redundant memory row / column extending in the first direction and / or a redundant memory column / row extending in the second direction is exhausted; in response to neither the redundant memory row / column extending in the first direction nor the redundant memory column / row extending in the second direction being exhausted, returning to continue performing the step of updating the information of the failed memory cells in the memory array and updating the first failed memory cell.

8. The repair method of claim 7, wherein in response to the redundant memory row / column extending in the first direction being exhausted and the redundant memory column / row extending in the second direction not being exhausted, the repair method further comprises: updating the information of the existing failed memory cells in the memory array and counting the number of failed memory cells on each memory row / column extending in the second direction in the memory array; ranking the number of failed memory cells on the memory row / column extending in the second direction in the memory array from more to less; repairing and replacing the failed memory cells on the memory row / column extending in the second direction in the memory array in order with the existing redundant memory column / row extending in the second direction.

9. A repair system for a storage device, characterized by comprising: a host device performing the repair method as claimed in any one of claims 1-8 to generate repair and replacement information; a memory device coupled to the host device, wherein the memory device comprises a memory die including at least one memory block, and each memory block includes a memory array and a redundant memory array.

10. The repair system of claim 9, wherein, the memory device further comprises a control die packaged together with the memory die, wherein the control die receives the repair and replacement information.

Images