Method and apparatus for accelerating buffer memory
The method accelerates fail data copying to buffer memory by using a memory control module to manage paths and merge data, enhancing redundancy analysis efficiency in semiconductor device testing.
Patent Information
- Application Number
- KR1020240192511
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2026-11-06
- Estimated Expiration
- 2044-12-20
AI Technical Summary
Existing semiconductor device test systems face challenges in copying fail data to buffer memory at high speed, which hinders efficient redundancy analysis.
A method involving a memory control module that determines the number of first paths based on an interleaved mode, associates second paths with first paths in specific ratios, acquires fail data through these paths, and writes it to the buffer memory, utilizing OR gates for merging data and addressing buffer memory blocks.
Enables rapid copying of fail data to buffer memory, reducing redundancy analysis time by allowing parallel processing and efficient data replication.
Smart Images

Figure 112024141865230-PAT00010_ABST
Abstract
Description
Technology Field
[0001] The present disclosure relates to a method and apparatus for copying fail data from a fail memory to a buffer memory at high speed. Background Technology
[0002] A semiconductor device test device is also referred to as an Automatic Test Equipment (ATE), and it is a device that verifies whether a semiconductor device is good by applying an electrical pattern signal to the semiconductor device and analyzing the response thereto.
[0003] A semiconductor device test device includes, as internal hardware components, a power supply, a measuring instrument, an Algorithmic Pattern Generator (ALPG), a Timing Generator (TG), Pin Electronics (PE) with built-in drivers and comparators, and a processor for controlling them.
[0004] After performing a test on a semiconductor device, the semiconductor device test device sequentially stores the fail data generated from the test results in a fail memory (FM) and copies the fail data stored in the fail memory (FM) to a buffer memory (BM). Subsequently, a redundancy analysis (RA) is performed on the fail data stored in the buffer memory to determine which part of the semiconductor device's cell has a defect.
[0005] For redundancy analysis to proceed quickly, fail data must be stored in buffer memory rapidly. In other words, if fail data is stored in buffer memory at high speed, the time required for redundancy analysis can be reduced.
[0006] Accordingly, there is a growing demand for technology to quickly copy failed data to buffer memory. Prior art literature
[0007] Korean Patent Publication No. 10-2023-0072624 (Published May 25, 2023) The problem to be solved
[0008] The technical problem to be solved in some embodiments of the present disclosure is to provide a method and apparatus capable of copying data to a buffer memory at high speed.
[0009] Another technical problem to be solved in some embodiments of the present disclosure is to provide a method and system that can merge fail data and copy it to buffer memory by applying various interleave modes.
[0010] The technical problems of the present disclosure are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by a person skilled in the art of the present disclosure from the description below. means of solving the problem
[0011] A method for accelerating a buffer memory, performed by a processor according to an embodiment of the present disclosure for solving the above technical problem, may include, based on an interleaved mode, a step of determining the number of first paths related to writing data to a buffer memory; a step of associating a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a ratio of n:m, where n and m are natural numbers; a step of acquiring fail data through a specific second path; and a step of writing the acquired fail data to a buffer memory through a first path associated with the specific second path.
[0012] Additionally, the step of determining the number of first paths includes the step of determining the number of first paths to be 4 when the interleave mode is 3 interleaves, and the step of associating a plurality of second paths with the determined number of first paths in an n:m manner may include the step of associating a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a 1:1 manner when the interleave mode is 3 interleaves.
[0013] Additionally, the step of determining the number of first paths includes the step of determining the number of first paths to 2 when the interleave mode is 6 interleaves, and the step of associating a plurality of second paths with the determined number of first paths in an n:m ratio may include the step of associating a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a 2:1 ratio when the interleave mode is 6 interleaves.
[0014] Additionally, the step of acquiring fail data through a specific second path may include the step of acquiring multiple fail data through a specific second path associated in a 2:1 ratio and the step of merging the acquired multiple fail data into one fail data.
[0015] Additionally, the step of determining the number of first paths includes the step of determining the number of first paths to one when the interleave mode is 12 interleaves, and the step of associating a plurality of second paths with the determined number of first paths in an n:m ratio may include the step of associating a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a 4:1 ratio when the interleave mode is 12 interleaves.
[0016] Additionally, the step of acquiring fail data through a specific second path may include the step of acquiring multiple fail data through a specific second path associated in a 4:1 ratio and the step of merging the acquired multiple fail data into one fail data.
[0017] In addition, multiple failed data may be ORed and merged into a single failed data.
[0018] In addition, multiple lines connected to multiple memories included in a specific fail memory group are input to an OR gate, and a line formed from the output of the OR gate can form a second path.
[0019] Additionally, the step of writing fail data to buffer memory may include the step of identifying an address and the step of writing fail data to a memory block of buffer memory associated with the identified address.
[0020] Additionally, the buffer memory acceleration method further includes the step of grouping a plurality of buffer memories according to the number of determined first paths, and the step of writing fail data to the buffer memory may include the step of identifying a plurality of buffer memories of a specific group associated with a specific second path and the step of writing fail data to a memory block included in the plurality of buffer memories of the identified specific group.
[0021] A semiconductor test device according to one embodiment of the present disclosure for solving a technical problem comprises a fail memory including a plurality of memories, a buffer memory including a plurality of memories, and a memory control module that controls reading of the fail memory and writing of the buffer memory. The memory control module determines the number of first paths related to writing data to the buffer memory based on an interleave mode, associates a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a ratio of n:m, where n and m are natural numbers, acquires fail data through a specific second path, and can write the acquired fail data to the buffer memory through a first path associated with the specific second path. Brief explanation of the drawing
[0022] FIG. 1 is a block diagram illustrating the configuration of a semiconductor test device according to one embodiment of the present disclosure. FIG. 2 is an illustrative diagram for explaining a method of storing data at high speed in a buffer memory according to some embodiments of the present disclosure. FIG. 3 is a diagram illustrating fail data being written to a buffer memory in interleaved mode according to some embodiments of the present disclosure. FIGS. 4 and FIGS. 5 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated in 3-interleaved according to some embodiments of the present disclosure. FIGS. 6 and FIGS. 7 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated with 6 interleaves, according to some embodiments of the present disclosure. FIGS. 8 and 9 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated in 12 interleaves according to some embodiments of the present disclosure. FIG. 10 is a flowchart for explaining a buffer memory acceleration method according to one embodiment of the present disclosure. Specific details for implementing the invention
[0023] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the attached drawings. The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the attached drawings. However, the technical concept of the present disclosure is not limited to the following embodiments but can be implemented in various different forms. The following embodiments are provided merely to complete the technical concept of the present disclosure and to fully inform those skilled in the art of the scope of the present disclosure, and the technical concept of the present disclosure is defined only by the scope of the claims.
[0024] It should be noted that when assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the present disclosure, if it is determined that a detailed description of related known components or functions could obscure the essence of the present disclosure, such detailed description is omitted.
[0025] Unless otherwise defined, all terms used herein (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which this disclosure pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. The terms used herein are for describing the embodiments and are not intended to limit this disclosure. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text.
[0026] Additionally, terms such as first, second, A, B, (a), (b), etc., may be used to describe the components of the present disclosure. These terms are intended only to distinguish the components from other components and do not limit the nature, order, or sequence of the components. Where it is stated that a component is "connected," "coupled," or "joined" to another component, it should be understood that the component may be directly connected or joined to the other component, but that another component may also be "connected," "coupled," or "joined" between each component.
[0027] The components, steps, actions, and / or elements mentioned in "comprises" and / or "comprising" as used in the specification do not exclude the presence or addition of one or more other components, steps, actions, and / or elements.
[0028] Hereinafter, several embodiments of the present disclosure will be described in detail with reference to the attached drawings.
[0029] FIG. 1 is a block diagram illustrating the configuration of a semiconductor test device according to one embodiment of the present disclosure.
[0030] Referring to FIG. 1, the semiconductor test device (1) includes an ALPG (Algorithmic Pattern Generator) (11), a timing generator (12), pin electronics (13), a fail memory (FM) (14), a buffer memory (BM) (15), a memory control module (16), and a test analysis module (17).
[0031] ALPG (11) can generate logic data sequentially using data stored in memory. The logic data may include an address, data, control signal, etc. to be applied to a semiconductor device (DUT) (100). The logic data may be provided to the timing generator (12) along with a clock signal in the form of pulse data expressed as "0" and "1".
[0032] The timing generator (12) can generate a timing signal based on pulse data input from the ALPG (11) through multiple channels. The generated timing signal can be transmitted to pin electronics (13).
[0033] Fin electronics (13) can generate a test signal for electrical testing of a semiconductor device (100) using a timing signal and apply the generated test signal to the semiconductor device (100). Fin electronics (13) may include a DCL (Driver Comparator Logic, not shown) that receives a timing signal and a pattern signal, modulates the pattern signal based on the timing signal, and generates a test signal.
[0034] The fail memory (14) can store fail data related to the test results of the semiconductor device (100). The fail data may be generated based on a signal received from the semiconductor device (100) after a test signal is applied to the semiconductor device (100).
[0035] According to one embodiment, the fail memory (14) may be composed of a plurality of DRAMs to store fail data in an interleaved manner. In this case, the plurality of DRAMs may form a plurality of fail memory groups. For example, the first DRAM, the second DRAM, and the third DRAM may form a first fail memory group, and the fourth to sixth DRAMs may form a second fail memory group.
[0036] According to one embodiment, the fail memory (14) can be designed to have a structure that can be converted in the capacity direction by varying the memory usage capacity according to the change in interleave mode. For example, it can be designed to expand in the capacity direction when the number of interleaves is small, and shrink in the capacity direction when the number of interleaves is large to meet system requirements.
[0037] The buffer memory (15) can read and write the failed data stored in the failed memory (14). According to one embodiment, a plurality of failed data stored in the failed memory (14) may be merged, and the merged failed data may be written to the buffer memory (15). According to one embodiment, the buffer memory (15) may be divided into a plurality of groups to rearrange the failed data.
[0038] The memory control module (16) includes at least one processor and can control the read / write operation of the buffer memory (15) and the read / write operation of the fail memory (14). The memory control module (16) can control the failure data to be stored in an interleave manner, in which the failure data is divided and stored in multiple DRAMs, so as to store the failure data at an address of the fail memory (14) without delay time and to write the failure data at high speed.
[0039] Additionally, the memory control module (16) can acquire multiple failure data contained in the failure memory (14) and write the acquired failure data to the buffer memory (15) at high speed. In order to save multiple failure data stored in the failure memory (14) to the buffer memory (15) at high speed, the memory control module (16) determines the number of first paths related to writing data to the buffer memory based on an interleaved mode, and can associate multiple second paths formed from multiple failure memory groups with the determined number of first paths in a ratio of n:m (where n and m are natural numbers). In this state where the second paths and the first paths are associated, when the failure memory (14) acquires failure data stored in the failure memory (14) through a specific second path, it can write the acquired failure data to the buffer memory (15) using the first path associated with the specific second path.
[0040] The test analysis module (17) performs redundancy analysis (RA) to extract defects in the semiconductor device (100) using fail data stored in the buffer memory (15). Through the redundancy analysis of the test analysis module (17), it determines whether the semiconductor device (100) is good or bad, and can perform a repair operation to recover the cells of the semiconductor device (100) that are failed according to the result of the determination of whether it is good or bad.
[0041] FIG. 2 is an illustrative diagram for explaining a method of storing data at high speed in a buffer memory according to some embodiments of the present disclosure. FIG. 2 illustrates a case of 3 interleaves.
[0042] Referring to FIG. 2, a plurality of fail memories can be divided into four fail memory groups (a first fail memory group, a second fail memory group, a third fail memory group, and a fourth fail memory group). The address associated with the data stored in each fail memory group may include an LSB (Least Significant Bit) and an MSB (Most Significant Bit).
[0043] FIG. 2 illustrates that failed data stored in one failed memory group is replicated into one buffer memory. According to some embodiments, multiple buffer memories may be grouped. In this case, failed data stored in multiple failed memory groups may be replicated and reordered into multiple buffer memory groups. The replication of each failed data may proceed in parallel, thereby enabling the failed data to be replicated and reordered at high speed.
[0044] FIGS. 3 to 10 are illustrative diagrams for explaining a method of replicating fail data according to an interleaved mode according to some embodiments of the present disclosure.
[0045] The memory POS (POSition) exemplified in FIG. 3 is used to distinguish multiple memory regions and can be used when expanding in the capacity direction. Using POS may result in a longer access time in the capacity direction. In this case, to reduce the access time, the POS[1:0] is not used, and the enable signal is set to High to configure simultaneous access to multiple memories. Here, the value of the enable signal may change depending on POS[1:0], which represents the capacity direction.
[0046] MEMADD[3:0] is a physical address of memory and can be converted into a row / column address in a memory controller block (not shown). MEMADD[3:0] may be 16 accessible physical addresses (0x0 to 0xF) of each of multiple memories.
[0047] In FIG. 3, the structure of the algorithm of the address generator is visually illustrated. The address table (310) generated by the address generator, each interleaved signal (322, 324, 326) of the EN_Table (Enable Table), and the address table (332, 334, 336) of each interleaved signal are illustrated.
[0048] According to one embodiment, when a memory fail scan of a semiconductor test device (1) is performed with 3 interleaves, an enable signal (322) of 3 interleaves is generated, and a physical address according to an address table (332) corresponding to 3 interleaves can be matched.
[0049] Additionally, when a memory fail scan of the test of the semiconductor test device (1) is performed with 6 interleaves, an enable signal (324) of 6 interleaves is generated, and a physical address according to the address table (334) corresponding to 6 interleaves can be matched.
[0050] According to one embodiment, when a memory fail scan of a semiconductor test device (1) is performed with 12 interleaves, an enable signal (326) of 12 interleaves is generated, and a physical address according to an address table (336) corresponding to 12 interleaves can be matched.
[0051] In the embodiments described below, a plurality of fail memories (FM_MEM1 to FM_MEM12) are divided into four fail memory groups, and each fail memory group is exemplified as containing three memories (e.g., DRAM). Additionally, in the embodiments described below, four second paths formed from the plurality of fail memory groups are exemplified. That is, the number of fail memory groups and the number of second paths may be the same. Here, the second path is a path for acquiring fail data and may be configured as at least one bus.
[0052] FIGS. 4 and FIGS. 5 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated in 3-interleaved according to some embodiments of the present disclosure.
[0053] Referring to FIGS. 4 and FIGS. 5, when the interleave mode is 3 interleave, the memory control module (16) determines that the number of first paths (510) related to writing fail data to the buffer memory is 4, and can associate the first paths and the second paths in a 1:1 ratio. The number of first paths can be varied based on the interleave mode.
[0054] As illustrated in FIG. 4, four first paths may be formed from each of the first fail memory group (MEM1 to MEM3), the second fail memory group (MEM4 to MEM6), the third fail memory group (MEM7 to MEM9), and the fourth fail memory group (MEM10 to MEM12). Here, the first path may be a path related to acquiring fail data.
[0055] Multiple failed data included in multiple memories of a memory group can be merged into a single failed data. As illustrated in FIG. 4, multiple lines connected to multiple memories included in a specific memory group can be input to an OR gate. For example, multiple lines (i.e., lines related to PSO[1:0] = 00) from multiple memories (FM_MEM1, FM_MEM2, FM_MEM3) included in a first memory group (FM_MEM1, FM_MEM2, FM_MEM3) can be input to a first OR gate (410). As another example, multiple lines (i.e., lines related to PSO[1:0] = 01) from multiple memories (FM_MEM3, FM_MEM4, FM_MEM5) included in a second memory group (FM_MEM3, FM_MEM4, FM_MEM5) can be input to a second OR gate (420).
[0056] OR gates (410, 420, 430, 440) are formed as many times as the number of fail memory groups, and second paths (450) can be formed as many times as the number of OR gates (410, 420, 430, 440). In FIGS. 4, 6, and 8, four second paths (450) are shown as being formed.
[0057] The memory control module (16) can read the failed data stored in each of the first failed memory group (FM_MEM1 to FM_MEM3), the second failed memory group (FM_MEM4 to FM_MEM6), the third failed memory group (FM_MEM7 to FM_MEM9), and the fourth failed memory group (FM_MEM10 to FM_MEM12). After that, the memory control module (16) can write the read failed data to each of the plurality of buffer memories (BM_MEM1, BM_MEM2, BM_MEM3, BM_MEM4). At this time, the memory control module (16) can write the failed data read from a specific failed memory group to a specific buffer memory based on a mutually related path.
[0058] As illustrated in FIGS. 4 and 5, a second path formed from a first fail memory group (FM_MEM1 to FM_MEM3) may be associated with a first path formed from a first buffer memory (BM_MEM1), and a second path formed from a second fail memory group (FM_MEM4 to FM_MEM6) may be associated with a first path formed from a second buffer memory (BM_MEM2). Additionally, a second path formed from a third fail memory group (FM_MEM7 to FM_MEM9) may be associated with a first path formed from a third buffer memory (BM_MEM3), and a second path formed from a fourth fail memory group (FM_MEM10 to FM_MEM12) may be associated with a first path formed from a fourth buffer memory (BM_MEM4).
[0059] In this state where multiple first paths and multiple second paths are associated in a 1:1 ratio, fail data acquired from a specific second path can be written to a buffer memory through a specific first path associated with the specific second path.
[0060] Specifically, in the case where the first path and the second path are 1:1 related according to 3 interleaves, the memory control module (16) can record fail data acquired from the first path of the first fail memory group (FM_MEM1 to FM_MEM3) in the first buffer memory (BM_MEM1) and fail data acquired from the first path of the second fail memory group (FM_MEM4 to FM_MEM6) in the second buffer memory (BM_MEM2). Additionally, the memory control module (16) can record fail data acquired from the first path of the third fail memory group (FM_MEM7 to FM_MEM9) in the third buffer memory (BM_MEM3) and fail data acquired from the first path of the fourth fail memory group (FM_MEM10 to FM_MEM12) in the fourth buffer memory (BM_MEM4).
[0061] According to one embodiment, sequential addresses are generated through an address generator, and a memory control module (16) identifies an address associated with this storage cycle and can write fail data to a memory block of the buffer memory associated with the identified address. FIG. 5 illustrates that fail data is written to a memory block of the buffer memory in a numbered order.
[0062] According to Figures 4 and 5, fail data can be written to buffer memory during 1T in 3 interleaves.
[0063] FIGS. 6 and FIGS. 7 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated with 6 interleaves, according to some embodiments of the present disclosure.
[0064] Referring to FIGS. 6 and FIGS. 7, when the interleaved mode is 6 interleaved, the memory control module (16) determines that the number of first paths (710) related to writing data to the buffer memory is 2, and can associate the first paths (710) and the second paths (650) in a 1:2 ratio.
[0065] As illustrated in FIG. 6, four first paths may be formed from each of the first fail memory group (MEM1 to MEM3), the second fail memory group (MEM4 to MEM6), the third fail memory group (MEM7 to MEM9), and the fourth fail memory group (MEM10 to MEM12).
[0066] According to one embodiment, the memory control module (16) can determine a number of buffer memory groups equal to the number of second paths. In FIG. 7, the first buffer memory (BM_MEM1) and the third buffer memory (BM_MEM3) are included in the first buffer memory group, and the second buffer memory (BM_MEM2) and the fourth buffer memory (BM_MEM4) are included in the second buffer memory group.
[0067] Multiple fail data obtained from two second paths can be merged and written to a buffer memory through a single first path. According to one embodiment, multiple fail data are ORed, and the fail data of the calculated result can be written to a buffer memory through a single first path.
[0068] For example, in FIG. 6, among the four second paths (650), two fail data outputs from the first OR gate (410) and the second OR gate (420) can be merged into one first fail data. Additionally, among the four second paths (650), two fail data outputs from the third OR gate (430) and the fourth OR gate (440) can be merged into one second fail data.
[0069] The memory control module (16) can read the fail data stored in each of the first fail memory group (FM_MEM1 to FM_MEM3), the second fail memory group (FM_MEM4 to FM_MEM6), the third fail memory group (FM_MEM7 to FM_MEM9), and the fourth fail memory group (FM_MEM10 to FM_MEM12). Afterward, the memory control module (16) can merge the fail data of the first fail memory group (FM_MEM1 to FM_MEM3) and the fail data of the second fail memory group (FM_MEM4 to FM_MEM6) into a single first fail data among the acquired fail data. Additionally, the memory control module (16) can identify a first path associated with two second paths that acquired the merged first fail data, and write the merged first fail data to the buffer memory through the identified first path. Here, the identified first path may be a path that can access the buffer memory included in the first buffer memory group.
[0070] In FIG. 7, it is illustrated that fail data is first written to the first buffer memory (BM_MEM1) included in the first buffer memory group, and then fail data is written to the third buffer memory (BM_MEM3).
[0071] Additionally, the memory control module (16) can merge the failure data of the third failure memory group (FM_MEM7 to FM_MEM9) and the failure data of the fourth failure memory group (FM_MEM10 to FM_MEM12) among the acquired failure data into a single second failure data. Additionally, the memory control module (16) can identify a first path associated with two second paths that acquired the merged second failure data, and write the merged first failure data to a buffer memory through the identified first path. Here, the identified first path may be a path that can access the buffer memory included in the second buffer memory group.
[0072] In FIG. 7, it is illustrated that fail data is first written to the second buffer memory (BM_MEM2) included in the second buffer memory group, and then fail data is written to the fourth buffer memory (BM_MEM4).
[0073] Figure 7 illustrates that fail data is written in parallel to memory blocks of the buffer memory in numbered order.
[0074] As exemplified in Figures 6 and 7, fail data can be written to the buffer memory within a time of 1T or less in 6 interleaves.
[0075] FIGS. 8 and 9 are illustrative diagrams for explaining a method for replicating fail data at high speed when operated in 12 interleaves according to some embodiments of the present disclosure.
[0076] Referring to FIGS. 8 and FIGS. 9, when the interleaved mode is 12 interleaved, the memory control module (16) determines the number of first paths (910) related to writing data to the buffer memory to be 1, and can associate the first path (910) and the second path (850) in a 1:4 ratio.
[0077] According to one embodiment, the memory control module (16) can determine as many buffer memory groups as there are second paths. In FIG. 9, the first buffer memory (BM_MEM1) to the fourth buffer memory (BM_MEM4) are illustrated as being included in one buffer memory group.
[0078] Multiple fail data obtained from four second paths can be merged into one and recorded in a buffer memory through a single first path. According to one embodiment, multiple fail data are ORed, and the failed data of the calculated result can be recorded in a buffer memory through a single first path.
[0079] For example, in Fig. 8, among the four second paths (850), four fail data output from the first OR gate (410) to the fourth OR gate (440) can be merged into one fail data.
[0080] The memory control module (16) can read the fail data stored in each of the first fail memory group (FM_MEM1 to FM_MEM3), the second fail memory group (FM_MEM4 to FM_MEM6), the third fail memory group (FM_MEM7 to FM_MEM9), and the fourth fail memory group (FM_MEM10 to FM_MEM12). Among the failed data, the memory control module (16) can merge the fail data of the first fail memory group (FM_MEM1 to FM_MEM3), the fail data of the second fail memory group (FM_MEM4 to FM_MEM6), the fail data of the third fail memory group (FM_MEM7 to FM_MEM9), and the fail data of the fourth fail memory group (FM_MEM10 to FM_MEM12) into a single fail data. Additionally, the memory control module (16) can identify one first path associated with four second paths that have acquired merged fail data, and write the merged first fail data to the buffer memory through the identified first path. Here, the identified first path may be a path that can access the buffer memory included in the buffer memory group.
[0081] In FIG. 9, it is illustrated that fail data is first written to the first buffer memory (BM_MEM1) included in the first buffer memory group, and then fail data is written to the second buffer memory (BM_MEM2).
[0082] In Fig. 9, fail data is exemplified as being written to memory blocks of the buffer memory in numbered order.
[0083] As exemplified in FIGS. 8 and 9, fail data can be written to buffer memory within a time of 1T or less in 12 interleaves.
[0084] FIG. 10 is a flowchart illustrating a buffer memory acceleration method according to one embodiment of the present disclosure. The flowchart illustrated in FIG. 10 is merely an exemplary embodiment for achieving the purpose of the present disclosure, and it is understood that some steps may be added or omitted as necessary.
[0085] Referring to FIG. 10, in step S100, the memory control module (16) can determine the number of first paths related to writing data to the buffer memory based on the currently set interleaved mode. According to some embodiments, the memory control module (16) can group a plurality of buffer memories according to the number of first paths.
[0086] According to one embodiment, the memory control module (16) may determine the number of first paths to be 4 when the interleave mode is 3 interleaves. According to some embodiments, the memory control module (16) may determine the number of first paths to be 2 when the interleave mode is 6 interleaves. Additionally, the memory control module (16) may determine the number of first paths to be 1 when the interleave mode is 12 interleaves.
[0087] In step S200, the memory control module (16) can associate a plurality of second paths formed from a plurality of fail memory groups with a determined number of first paths in a ratio of n:m. Here, n and m may be natural numbers. According to some embodiments, a plurality of switches and a plurality of logic elements may be included between the buffer memory and the fail memory, and the memory control module (16) can control the on / off of at least one switch to control the association of a plurality of second paths with a determined number of first paths in a ratio of n:m.
[0088] According to one embodiment, when the interleave mode is 3 interleave, the memory control module (16) can associate a plurality of second paths formed from a plurality of fail memory groups with a determined number of first paths in a 1:1 ratio.
[0089] According to some embodiments, the memory control module (16) can associate a plurality of second paths formed from a plurality of fail memory groups with a determined number of first paths in a 2:1 ratio when the interleave mode is 6 interleaves.
[0090] Additionally, when the interleave mode is 12 interleaves, the memory control module (16) can associate a plurality of second paths formed from a plurality of fail memory groups with a determined number of first paths in a 4:1 ratio. According to one embodiment, a plurality of fail data can be ORed and merged into one fail data.
[0091] In step S300, the memory control module (16) can acquire fail data through a specific second path. According to some embodiments, when the interleave mode is 6 interleave, the memory control module (16) can acquire multiple fail data through a specific second path associated in a 2:1 ratio and merge the acquired multiple fail data into one fail data. Additionally, when the interleave mode is 12 interleave, the memory control module (16) can acquire multiple fail data through a specific second path associated in a 4:1 ratio and merge the acquired multiple fail data into one fail data.
[0092] In step S400, the memory control module (16) can write the acquired or merged fail data to the buffer memory through the first path associated with a specific second path. According to one embodiment, the memory control module (16) can identify an address and write the fail data to a memory block of the buffer memory associated with the identified address.
[0093] According to some embodiments, the memory control module (16) can identify a plurality of buffer memories of a specific group associated with a specific second path and write fail data to memory blocks included in the identified plurality of buffer memories of the specific group.
[0094] According to one embodiment, a plurality of lines connected to a plurality of memories included in a specific fail memory group are input to an OR gate, and a line formed from the output of the OR gate can form a second path.
[0095] Although embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art will understand that the present invention may be practiced in other specific forms without altering the technical concept or essential features thereof. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of protection of the present invention shall be interpreted by the claims below, and all technical concepts within the equivalent scope shall be interpreted as being included within the scope of rights of the technical concepts defined by the present disclosure.
Claims
Claim 1 A method for performing buffer memory acceleration by a processor, comprising: determining the number of first paths related to writing data to the buffer memory based on an interleave mode; associating a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a ratio of n:m, wherein n and m are natural numbers; acquiring fail data through a specific second path; and writing the acquired fail data to the buffer memory through a first path associated with the specific second path. Claim 2 A buffer memory acceleration method according to claim 1, wherein the step of determining the number of first paths includes the step of determining the number of first paths to 4 when the interleave mode is 3 interleaves, and the step of associating the plurality of second paths with the determined number of first paths in an n:m ratio includes the step of associating the plurality of second paths formed from the plurality of fail memory groups with the determined number of first paths in a 1:1 ratio when the interleave mode is 3 interleaves. Claim 3 A buffer memory acceleration method according to claim 1, wherein the step of determining the number of the first paths includes the step of determining the number of the first paths to 2 when the interleave mode is 6 interleaves, and the step of associating the plurality of second paths with the determined number of first paths in an n:m ratio includes the step of associating the plurality of second paths formed from the plurality of fail memory groups with the determined number of first paths in a 2:1 ratio when the interleave mode is 6 interleaves. Claim 4 A method for accelerating a buffer memory according to claim 3, wherein the step of acquiring fail data through the specific second path comprises: acquiring a plurality of fail data through the specific second path associated in a 2:1 ratio; and merging the acquired plurality of fail data into one fail data. Claim 5 A method for accelerating buffer memory according to claim 1, wherein the step of determining the number of the first paths includes the step of determining the number of the first paths to one when the interleave mode is 12 interleaves, and the step of associating the plurality of second paths with the determined number of first paths in an n:m ratio includes the step of associating the plurality of second paths formed from the plurality of fail memory groups with the determined number of first paths in a 4:1 ratio when the interleave mode is 12 interleaves. Claim 6 A method for accelerating a buffer memory according to claim 5, wherein the step of acquiring fail data through the specific second path comprises: acquiring a plurality of fail data through the specific second path associated in a 4:1 ratio; and merging the acquired plurality of fail data into one fail data. Claim 7 A buffer memory acceleration method according to claim 6, wherein the plurality of failed data are ORed and merged into a single failed data. Claim 8 A buffer memory acceleration method according to claim 1, wherein a plurality of lines connected to a plurality of memories included in a specific fail memory group are input to an OR gate, and a line formed from the output of the OR gate forms a second path. Claim 9 A method for accelerating a buffer memory according to claim 1, wherein the step of writing the fail data to the buffer memory comprises: a step of identifying an address; and a step of writing the fail data to a memory block of the buffer memory associated with the identified address. Claim 10 A method for accelerating buffer memory according to claim 1, further comprising the step of grouping a plurality of buffer memories according to the number of first paths determined above, and the step of writing fail data to the buffer memory comprises: the step of identifying a plurality of buffer memories of a specific group associated with a specific second path; and the step of writing the fail data to a memory block included in the plurality of buffer memories of the identified specific group. Claim 11 A semiconductor test device comprising: a fail memory including a plurality of memories; a buffer memory including a plurality of memories; and a memory control module for controlling reading of the fail memory and writing of the buffer memory, wherein the memory control module determines the number of first paths related to writing data to the buffer memory based on an interleave mode, associates a plurality of second paths formed from a plurality of fail memory groups with the determined number of first paths in a ratio of n:m, wherein n and m are natural numbers, acquires fail data through a specific second path, and writes the acquired fail data to the buffer memory through a first path associated with the specific second path.
Citation Information
Patent Citations
Testing device, and testing method
KR1020110014625A
Method for configuring address of buffer memory for semiconductor test, and semiconductor test apparatus implementing the same method
KR1020240003898A
Semiconductor test apparatus using FPGA
KR1020240003899A
Method and apparatus for testing semiconductor
KR1020240016013A
Method for count memory fail and system thereof
KR1020240106090A
