WO2011007383A1 - 試験装置および救済解析方法 - Google Patents
試験装置および救済解析方法 Download PDFInfo
- Publication number
- WO2011007383A1 WO2011007383A1 PCT/JP2009/003271 JP2009003271W WO2011007383A1 WO 2011007383 A1 WO2011007383 A1 WO 2011007383A1 JP 2009003271 W JP2009003271 W JP 2009003271W WO 2011007383 A1 WO2011007383 A1 WO 2011007383A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- address
- block
- fail
- memory
- group
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/56—External testing equipment for static stores, e.g. automatic test equipment [ATE]; Interfaces therefor
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/04—Detection or location of defective memory elements, e.g. cell constructio details, timing of test signals
- G11C29/08—Functional testing, e.g. testing during refresh, power-on self testing [POST] or distributed testing
- G11C29/12—Built-in arrangements for testing, e.g. built-in self testing [BIST] or interconnection details
- G11C29/44—Indication or identification of errors, e.g. for repair
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/56—External testing equipment for static stores, e.g. automatic test equipment [ATE]; Interfaces therefor
- G11C29/56008—Error analysis, representation of errors
Definitions
- the present invention relates to a test apparatus and a repair analysis method for testing a memory.
- Test equipment that tests the memory writes predetermined data to the memory under test, reads the written data, and compares it with the expected value. Further, the test apparatus writes fail data indicating that the read data does not match the expected value in the address fail memory (AFM). Then, the test apparatus performs a repair analysis of the memory under test based on the fail data written in the AFM (see Patent Document 1).
- the test apparatus when performing a repair analysis, sequentially reads fail data from the AFM, and the number of defective cells per row address (RFC) and the number of defective cells per column address (CFC) in the memory under test. Count. It is desirable that the RFC and CFC count processing be performed at high speed in order to shorten the repair analysis time between tests and improve the overall throughput. Therefore, it is desirable for the test apparatus to configure a counter circuit that counts RFC and CFC using a high-speed memory (for example, SRAM).
- a high-speed memory for example, SRAM
- the test apparatus has to increase the capacity of the memory for storing RFC and CFC in accordance with the capacity of the memory under test.
- a test apparatus for testing a memory under test, wherein an address fail indicating whether or not a defective cell is included for each address in the memory under test.
- An address fail memory for storing data; a block fail memory for storing block fail data indicating whether or not a defective cell is included for each block including a plurality of cells in the memory under test; and a block within the memory under test.
- the address read by the reading unit for each row address in a group having a plurality of blocks in a part of the memory under test and the reading unit that reads the address fail data from the address fail memory every time A row fail counter that counts defective cells indicated in the fail data; and Per column addresses in the loop, to provide a test apparatus and a column fail counter for counting the defective cell shown in the address fail data read by the reading unit. Furthermore, a method for analyzing the repair of a memory under test in such a test apparatus is provided.
- 1 shows a configuration of a test apparatus 100 according to the present embodiment, together with a memory under test 300.
- 1 shows a configuration of a failure analysis memory unit 10 according to the present embodiment.
- the configuration of the analysis unit 20 according to the present embodiment is shown together with the address fail memory 30 and the block fail memory 40.
- An example of the configuration of the row fail counter 52, the column fail counter 54, and the total fail counter 56 according to the present embodiment is shown.
- 2 shows an exemplary configuration of a reading unit 50 according to the present embodiment.
- An example of addresses given to the address fail memory 30 and the block fail memory 40 in the DUT test is shown.
- An example of addresses given to the address fail memory 30 and the block fail memory 40 in the repair analysis is shown.
- An example of setting when the number of row address bits and the number of column address bits in a block is 7 bits is shown.
- An example of setting when the number of row address bits and the number of column address bits in a block is 6 bits is shown.
- An example of setting when the number of row address bits and the number of column address bits in a block is 10 bits is shown.
- the processing flow of the analysis part 20 which concerns on this embodiment is shown.
- FIG. 1 shows a configuration of a test apparatus 100 according to this embodiment together with a memory under test 300.
- the test apparatus 100 according to the present embodiment tests the memory under test 300 and detects defective cells. Further, the test apparatus 100 performs a repair analysis for making the memory under test 300 non-defective by electrically replacing an address line in which a defective cell exists and a spare line.
- the test apparatus 100 includes a timing generator 102, a pattern generator 104, a waveform shaper 106, a logic comparator 108, a failure analysis memory unit 10, and an analysis unit 20.
- the timing generator 102 generates a reference clock and supplies it to the pattern generator 104.
- the pattern generator 104 generates an address signal, a data signal, and a control signal to be supplied to the memory under test 300 based on the reference clock, and supplies them to the waveform shaper 106.
- the pattern generator 104 generates an expected value signal to be output from the memory under test 300 and supplies it to the logic comparator 108.
- the waveform shaper 106 shapes an applied signal based on the address signal, the data signal, and the control signal, and supplies the shaped signal to the memory under test 300.
- the logic comparator 108 compares the output signal output from the memory under test 300 in response to the application signal and the expected value signal generated by the pattern generator 104.
- the logical comparator 108 outputs a fail signal indicating a failure when the output signal and the expected value signal do not match.
- the failure analysis memory unit 10 stores the fail signal generated by the logic comparator 108 in correspondence with the address indicated by the address signal generated by the pattern generator 104. As a result, the failure analysis memory unit 10 can store fail data representing the position of the defective cell in the memory area in the memory under test 300. The configuration of the failure analysis memory unit 10 will be further described with reference to FIG.
- the analysis unit 20 When a defective cell is found in the test of the memory under test 300, the analysis unit 20 reads the fail data stored in the failure analysis memory unit 10 and performs a repair analysis of the memory under test 300. That is, the analysis unit 20 replaces the row address line and the column address line including the defective cell of the memory under test 300 with the spare line for the row address and the spare line for the column address. Analyzes whether the product can be improved. The configuration of the analysis unit 20 will be further described in FIG.
- the test apparatus 100 tests the memory under test 300 and stores the test result in the failure analysis memory unit 10 as fail data. After the test, when a defective cell is found in the test of the memory under test 300, the test apparatus 100 performs a repair analysis of the memory under test 300.
- each block includes memory cells that are accessed two-dimensionally by a plurality of row address lines and a plurality of column address lines.
- the test apparatus 100 performs relief analysis for each of a plurality of groups in which all the blocks in the memory under test 300 are grouped into one group. More specifically, the test apparatus 100 reads the fail data stored in the failure analysis memory unit 10 for each group, and determines the number of defective cells (RFC) for each row address for each block in the group. The number of defective cells (CFC) per column address for each block and the total number of defective cells (TFC) for each block in the group are counted.
- RRC defective cells
- the test apparatus 100 determines the number of defective cells for each row address (RFC) for each block in the group, the number of defective cells for each column address (CFC) for each block in the group, and the number of defective cells for each block in the group. Based on the number of cells (TFC), the repair analysis of the memory under test 300 is performed for each group. Note that, depending on the repair analysis method, the test apparatus 100 may not count the number of defective cells (TFC) for each block in the group.
- FIG. 2 shows a configuration of the failure analysis memory unit 10 according to the present embodiment.
- the failure analysis memory unit 10 includes an address fail memory (AFM) 30, a block fail memory (BFM) 40, an AFM address formatter 12, a BFM address formatter 14, an AFM address selector (MUX-A) 16, and a BFM. Address selector (MUX-B18).
- AFM address fail memory
- BFM block fail memory
- MUX-A AFM address selector
- MUX-B18 Address selector
- the address fail memory 30 has a memory area having the same address configuration as the memory under test 300.
- the address fail memory 30 stores address fail data indicating the presence / absence of a defective cell for each address in the memory under test 300.
- the block fail memory 40 has a memory area having the same address configuration as that of a block obtained by dividing the memory area in the memory under test 300 into a large number.
- the block fail memory 40 stores block fail data indicating the presence / absence of a defective cell for each block in the memory under test 300.
- the AFM address formatter 12 converts the address signal generated by the pattern generator 104 into an address (AFM_Address) to be given to the address fail memory 30. That is, the AFM address formatter 12 formats and outputs the address signal from the pattern generator 104 as the address of the address fail memory 30 so as to correspond to the address of the memory under test 300. With this formatting function, the address fail memory 30 can be associated with various address configurations of the memory under test 300.
- the BFM address formatter 14 converts the address signal generated by the pattern generator 104 into an address (BFM_Address) to be given to the block fail memory 40. That is, the BFM address formatter 14 formats and outputs the address of the block fail memory 40 so that the address from the pattern generator 104 corresponds to the address for selecting a plurality of blocks in the memory under test 300. With this format function, the block fail memory 40 can be made to correspond to various block address configurations of the memory under test 300.
- the AFM address selector 16 gives the address output from the AFM address formatter 12 to the address fail memory 30 in the DUT test, and gives the address given from the analysis unit 20 to the address fail memory 30 in the defect repair analysis.
- the address signal is switched as follows.
- the BFM address selector 18 gives the address output from the BFM address formatter 14 to the block fail memory 40 in the DUT test, and gives the address given from the analysis unit 20 to the block fail memory 40 in the defect repair analysis.
- the address signal is switched as follows.
- Such a failure analysis memory unit 10 operates as follows in the DUT test. First, the address fail memory 30 and the block fail memory 40 are cleared before the DUT test. When the DUT test is started, an address signal designating an address on the memory under test 300 and write data are output from the pattern generator 104 and the data is written into the memory under test 300. When the written data is read from the memory under test 300, the address signal and expected value data are output from the pattern generator 104, and the data read from the DUT is compared with the expected value data by the logical comparator 108. The failure analysis memory unit 10 is supplied with an address signal from the pattern generator 104 and a fail signal indicating whether or not the cell at the read address is defective.
- the address fail memory 30 stores “1” at the address specified by the AFM address formatter 12 and the cell specified by the address signal. If is not a defective cell, nothing is stored in the address fail memory 30. As a result, the address fail memory 30 can store address fail data indicating the presence or absence of a defective cell for each address in the memory under test 300.
- the block fail memory 40 stores “1” at the address specified by the BFM address formatter 14 and the cell specified by the address signal. If the cell is not a defective cell, nothing is stored in the block fail memory 40. Thereby, the block fail memory 40 can store block fail data indicating the presence / absence of a defective cell for each block including a plurality of cells in the memory under test 300.
- the defect analysis memory unit 10 operates as follows in defect repair analysis.
- the failure analysis memory unit 10 is given an address (AFM_Address) for reading the address fail memory 30 and an address (BFM_Address) for reading the block fail memory 40 from the analysis unit 20.
- AFM_Address address for reading the address fail memory 30
- BFM_Address address for reading the block fail memory 40 from the analysis unit 20.
- the address fail memory 30 outputs address fail data corresponding to the address (AFM_Address) from the analysis unit 20 to the analysis unit 20.
- the block fail memory 40 outputs block fail data corresponding to the block address (BFM_Address) from the analysis unit 20 to the analysis unit 20.
- FIG. 3 shows a configuration of the analysis unit 20 according to the present embodiment, together with an address fail memory (AFM) 30 and a block fail memory (BFM) 40.
- the analysis unit 20 includes a group designation unit 42, a block address pointer (BAP) 44, a block address generation unit 46, an update unit 48, a reading unit 50, a row fail counter (RFC) 52, and a column fail counter ( CFC) 54, total fail counter (TFC) 56, calculation unit 58, and control unit 60.
- BAP block address pointer
- CFC column fail counter
- TFC total fail counter
- the group designating unit 42 has a register (Register_A) for designating a group for repair analysis, and outputs a value for designating the group.
- the group means a memory area in the memory under test 300 divided into a large number of blocks, and a plurality of these blocks are collected. For example, if the memory under test 300 is divided into 256 blocks, the 16 blocks are combined into one group.
- the group may be an area specified by some bits on the upper side in a block address that specifies a block in the memory under test 300.
- the group designation unit 42 may be a register that stores a group value, and the value may be rewritten by the control unit 60. Further, the group designation unit 42 may be a counter, and the value may be incremented by 1 from an initial value (for example, 0).
- the block address pointer (BAP) 44 sequentially generates block addresses within the group.
- the block address in the group is an address that designates a block from which address fail data is read in the group.
- the block address pointer 44 is incremented by 1 from an initial value (for example, 0) every time an update instruction is given by the update unit 48.
- the block address generation unit 46 synthesizes the block address in the group in which the block address pointer (BAP) 44 is generated and the group value output by the group designating unit 42 to read a block for reading the block fail data from the block fail memory 40.
- a designated block address (BFM_Address) is generated.
- the block address generation unit 46 assigns the group value output from the group specification unit 42 to the higher-order bits, and assigns the block address in the group generated by the block address pointer (BAP) 44 to the lower-order bits.
- An address (BFM_Address) is generated.
- the block address generation unit 46 gives the generated block address (BFM_Address) to the block fail memory 40. Thereby, the analysis unit 20 can read the block fail data from the block fail memory 40. In addition, the block address generation unit 46 gives the generated block address (BFM_Address) to the block fail memory 40 and the reading unit 50.
- the update unit 48 receives the block fail data stored in the block address (BFM_Address) generated by the block address generation unit 46 from the block fail memory 40. Further, the updating unit 48 receives from the reading unit 50 an end flag indicating that address fail data has been read from all addresses in one block designated by the block address generation unit 46.
- the update unit 48 sets the block address pointer (BAP) 44 when the received block fail data does not indicate a failure (that is, when the block designated by the block address generation unit 46 does not include a defective cell). Update the block address in the group by incrementing. Further, even when the updating unit 48 receives the end flag from the reading unit 50, the updating unit 48 increments the block address pointer 44 to update the intra-group block address.
- BAP block address pointer
- the reading unit 50 reads the address fail data from the address fail memory 30 for each block in the memory under test 300. More specifically, the reading unit 50 reads the address fail data by giving the address fail memory 30 an address (AFM_Address) that sequentially specifies the cells in the specified block.
- AMF_Address an address that sequentially specifies the cells in the specified block.
- the reading unit 50 receives a block address (BFM_Address) and block fail data.
- BFM_Address block address
- the reading unit 50 uses the received block address (BFM_Address).
- An address (AFM_Address) for sequentially specifying the cells in the specified block is output. Further, the reading unit 50 outputs an end flag when the output of the addresses of all the cells in the block is completed.
- the reading unit 50 receives the intra-group block address.
- the reading unit 50 internally generates an intra-block row address and an intra-block column address.
- a row address is generated from the row address in the block and the block address in the group, and is given to the row fail counter 52.
- a column address is generated from the intra-block column address and the intra-group block address and is given to the column fail counter 54.
- the reading unit 50 gives the block address in the group including the address (AFM_Address) given to the address fail memory 30 to the total fail counter 56.
- AFM_Address address given to the address fail memory 30 to the total fail counter 56.
- the row fail counter (RFC) 52 counts the defective cells indicated in the address fail data read by the reading unit 50 for each row address of each block in the group.
- An example of the configuration of the low-fail counter 52 will be further described with reference to FIG.
- the column fail counter (CFC) 54 counts the defective cells indicated in the address fail data read by the reading unit 50 for each column address of each block in the group. An example of the configuration of the column fail counter 54 will be further described with reference to FIG.
- the total fail counter (TFC) 56 counts defective cells indicated in the address fail data read by the reading unit 50 for each block in the group. An example of the configuration of the total fail counter 56 will be further described with reference to FIG.
- the calculation unit 58 performs a repair analysis for electrically replacing row and column address lines including defective cells with spare lines for each block in each group. More specifically, the calculation unit 58 calculates the number of defective cells per row address (RFC) counted by the row fail counter 52 and the defective cells for each column address counted by the column fail counter 54. The row or column address line including the defective cell is searched for the repair analysis based on the number (CFC) and the total number of defective cells (TFC) counted by the total fail counter 56 for each block.
- RRC defective cells per row address
- the control unit 60 After the repair analysis for one group is completed by the arithmetic unit 58, the control unit 60 writes a group value indicating a group to be repaired next in a register (Register_A) in the group designating unit 42.
- the group designating unit 42 is a counter
- the control unit 60 notifies the group designating unit 42 that the repair analysis has been completed after the repair analysis is performed on one group. In this case, the group specifying unit 42 increments the group value in response to receiving the notification.
- the control unit 60 performs initial setting on the reading unit 50 and the like based on the number of row addresses and the number of column addresses in the block.
- FIG. 4 shows an example of the configuration of the row fail counter 52, the column fail counter 54, and the total fail counter 56 according to the present embodiment.
- the low fail counter 52 includes an RFC memory (RFCM) 62 and an adder 63.
- RFCM RFC memory
- the RFC memory 62 has a storage area corresponding to the number of row address bits for each block in the group.
- the RFC memory 62 receives the address fail data output from the address fail memory 30 at the write enable terminal.
- the RFC memory 62 receives a row address (RFC_Address) in the group at an address terminal.
- the adder 63 reads a value from the storage area of the RFC memory 62 specified by the row address (RFC_Address), adds 1 to the read value, and outputs it.
- the RFC memory 62 writes the value output by the adder 63 to the storage area designated by the row address (RFC_Address) when the address fail data indicates a failure (for example, 1), and the address fail data indicates that the address fail data is defective. When not shown, the value output by the adder 63 is not written.
- the row fail counter 52 having such a configuration can count defective cells indicated by the address fail data read from the address fail memory 30 by the reading unit 50 for each row address of each block in the group.
- the column fail counter 54 includes a CFC memory (CFCM) 64 and an adder 65.
- the CFC memory 64 has a storage area corresponding to the number of column address bits for each block in the group.
- the CFC memory 64 receives the address fail data output from the address fail memory 30 at the write enable terminal.
- the CFC memory 64 receives the column address (CFC_Address) in the group at the address terminal.
- the adder 65 reads a value from the storage area of the CFC memory 64 specified by the column address (CFC_Address), adds 1 to the read value, and outputs the result.
- the CFC memory 64 writes the value output by the adder 65 to the storage area specified by the column address (CFC_Address) when the address fail data indicates failure, and when the address fail data does not indicate failure, the adder 65 Does not write the value output by.
- the column fail counter 54 having such a configuration can count defective cells indicated in the address fail data read from the address fail memory 30 by the reading unit 50 for each column address of each block in the group.
- the total fail counter 56 includes a TFC memory (TFCM) 66 and an adder 67.
- the TFC memory 66 has at least a storage area corresponding to the number of blocks in the group.
- the TFC memory 66 receives the address fail data output from the address fail memory 30 at the write enable terminal. Further, the TFC memory 66 receives an intra-group block address (TFC_Address) that designates a block in the group at an address terminal.
- TFC_Address intra-group block address
- the adder 67 reads a value from the storage area of the TFC memory 66 designated by the intra-group block address (TFC_Address), adds 1 to the read value, and outputs it.
- the TFC memory 66 writes the value output from the adder 67 to the storage area specified by the intra-group block address.
- the TFC memory 66 The value output by 67 is not written.
- the total fail counter 56 having such a configuration can count the defective cells indicated in the address fail data read from the address fail memory 30 by the reading unit 50 for each block in the group.
- FIG. 5 shows an example of the configuration of the reading unit 50 according to the present embodiment.
- the reading unit 50 includes a start detection unit 72, a row address pointer (RAP) 74, a column address pointer (CAP) 76, an address control unit 78, a memory address generation unit 80, a row address generation unit 82, a column And an address generation unit 84.
- RAP row address pointer
- CAP column address pointer
- the start detection unit 72 controls the operation period of the row address pointer 74 and the column address pointer 76. More specifically, the start detection unit 72 operates the row address pointer 74 and the column address pointer 76 after receiving block fail data from the block fail memory 40 until receiving an end flag from the end detection unit 90.
- the row address pointer (RAP) 74 generates a row address in the block.
- the row address pointer 74 may be a counter that sequentially and cyclically generates each row address from the beginning to the end in the block.
- a value for example, 0
- the row address pointer 74 increments the counter value for each clock while the enable signal is given from the start detection unit 72.
- the row address pointer 74 When the counter value reaches the value indicating the last row address in the block, the row address pointer 74 generates a carry signal to the column address pointer 76, and at the next clock, the counter value is set to the first row address in the block. Return to the value indicating.
- Column address pointer (CAP) 76 generates a column address in the block.
- the column address pointer 76 may be a counter that sequentially generates each column address from the beginning to the last column address in the block.
- a value for example, 0
- the row address pointer 74 is set for each carry from the row address pointer 74 (that is, the counter value of the row address pointer 74 is set to the first row address in the block while the enable signal is given from the start detection unit 72). Increment the counter value (on every return).
- the address control unit 78 controls the operation of the row address pointer 74 and the column address pointer 76. More specifically, the address control unit 78 determines combinations of row addresses and column addresses that specify all addresses in a block according to the number of row addresses and column addresses in one block. Control is performed to sequentially output from the address pointer 76.
- the address control unit 78 includes a row carry selector 86, a column carry selector 88, and an end detection unit 90.
- the row carry selector 86 the maximum value of the row address pointer 74 (that is, the number of row addresses in one block) is set.
- the row carry selector 86 generates a carry signal when the counter value of the row address pointer 74 reaches the maximum value.
- the column address pointer 76 increments the counter value by this carry signal. Further, after the row address pointer 74 reaches the set maximum value, the value returns to the initial value.
- the maximum value of the column address pointer 76 (that is, the number of column addresses in one block) is set.
- the column carry selector 88 generates a carry signal when the counter value of the column address pointer 76 reaches the maximum value.
- the end detection unit 90 outputs an end flag at the timing when the carry signal is generated from both the row carry selector 86 and the column carry selector 88. That is, the end detection unit 90 outputs an end flag at the timing when the value of the row address pointer 74 reaches the last row address in the block and the value of the column address pointer 76 reaches the last column address in the block. To do. Further, after the column address pointer 76 reaches the set maximum value, the value returns to the initial value.
- the end flag is given to the start detection unit 72.
- the start detection unit 72 detects the row address pointer 74 and the column address pointer at the timing when the counter value of the row address pointer 74 reaches the last row address and the counter value of the column address pointer 76 reaches the last column address.
- the count operation of 76 can be stopped.
- the end flag is given to the update unit 48.
- the update unit 48 can update the value of the next block by incrementing the value of the block address pointer 44.
- the memory address generation unit 80 receives a block address (BFM_Address) from the block address generation unit 46, a row address from the row address pointer 74, and a column address from the column address pointer 76. Then, the memory address generation unit 80 combines the block address including the group value, the row address, and the column address, and generates an address (AFM_Address) for reading the address fail data from the address fail memory 30.
- BFM_Address block address
- AMF_Address address for reading the address fail data from the address fail memory 30.
- the row address generator 82 receives the intra-group block address from the block address pointer 44 and the row address from the row address pointer 74. The row address generation unit 82 combines these to generate an address (RFC_Address) to be given to the row fail counter 52.
- the column address generator 84 receives the intra-group block address from the block address pointer 44 and the column address from the column address pointer 76. Then, the column address generation unit 84 combines these to generate an address (CFC_Address) to be given to the column fail counter 54.
- the reading unit 50 gives the received intra-group block address to the total fail counter 56 (TFC_Address).
- the reading unit 50 can read address fail data from the address fail memory 30 for each block. Further, the reading unit 50 can output an end flag when the output of the addresses of all the cells in the block is completed. Further, the reading unit 50 can generate addresses to be given to the row fail counter 52, the column fail counter 54, and the total fail counter 56.
- FIG. 6 shows an example of addresses given to the address fail memory 30 and the block fail memory 40 in the DUT test.
- the address fail memory 30 and the block fail memory 40 are given addresses from the pattern generator 104 in the DUT test.
- the address fail memory 30 is given an address of 12 bits for X address (X0 to X11) and 12 bits for Y address (Y0 to Y11).
- the block fail memory 40 is given the same address as the upper bits of the X address and Y address of the address fail memory 30.
- the block fail memory 40 is given an address of X address 5 bits (X7 to X11) and Y address 5 bits (Y7 to Y11).
- FIG. 7 shows an example of addresses given to the address fail memory 30 and the block fail memory 40 in the repair analysis.
- the address fail memory 30 and the block fail memory 40 are given addresses from the analysis unit 20 in the repair analysis.
- the configuration of the address given from the analysis unit 20 is the same as the configuration of the address given from the pattern generator 104 in the DUT test.
- the address fail memory 30 is provided with addresses corresponding to the X address 12 bits and the Y address 12 bits during the test from the analysis unit 20.
- R0 to R6 are bits 0 to 6 of the row address pointer (RAP) 74
- C0 to C6 are bits 0 to 6 of the column address pointer (CAP) 76
- B0 to B6 are bits of the block address pointer (BAP) 44.
- 0 to 6 and D0 to D2 indicate bits 0 to 2 of data indicating the group value.
- the block fail memory 40 is given an address of 5 bits for X address and 5 bits for Y address from the analysis unit 20.
- the address structure of the memory under test 300 is indicated by 16M words in which the total number of address bits is 24 bits (0 to 23) and the total number of blocks is 10 bits (B0 to B6 + D0 to D2). 1024 blocks, the number of groups is 8 groups indicated by address 3 bits (D0 to D2), the number of blocks in 1 group is 128 blocks indicated by address 7 bits (B0 to B6), and the number of row addresses in 1 block is This indicates that the number of column addresses in one block is 128M words indicated by address 7 bits (R0 to R6), and the number of column addresses in one block is 128M words indicated by address 7 bits (C0 to C6).
- X0 of X address bit 0 at the time of testing corresponds to R0 of row address bit 0 at the time of analysis.
- FIG. 8 shows an example of setting when the number of row address bits and the number of column address bits in a block are 7 bits each.
- the control unit 60 determines the effective bit width of the row address pointer (RAP) 74 and the column address pointer according to the number of row address bits and the number of column address bits in one block of the memory under test 300.
- the effective bit width of (CAP) 76 is set.
- the effective row address generated from the row address pointer 74 is 7 bits (R0 to R6) and the effective address address generated from the column address pointer 76
- the column address is set to 7 bits (C0 to C6).
- bits indicated by shading indicate bits that are not used, and bits that are not shaded indicate valid bits.
- control unit 60 may set 1 to the lower 7 bits of the row carry selector 86 and the column carry selector 88 and set 0 to the remaining bits.
- the row carry selector 86 and the column carry selector 88 generate respective carry signals by determining whether the set value matches the row address value and the column address value.
- the control unit 60 sets a block address pointer (BAP) that generates an intra-group block address according to the bit width of the addresses of the RFC memory 62 in the row fail counter 52 and the CFC memory 64 in the column fail counter 54.
- BAP block address pointer
- Set the effective bit width Specifically, a value obtained by subtracting the number of row address bits in the block from the bit width of the address (RFC_Address) of the RFC memory 62, or the number of column address bits in the block from the bit width of the address (CFC_Address) of the CFC memory 64 The smaller value of the value obtained by subtracting is used as the effective bit width of the block address pointer (BAP).
- the bit width of the address (RFC_Address) of the RFC memory 62 and the address (CFC_Address) of the CFC memory 64 is 14 bits, and the number of row address bits and column address bits in the block is 7 bits.
- the effective bit width of the address pointer (BAP) is set to 7 bits (B0 to 6).
- control unit 60 designates the group from the bit width of the block address (BFM_Address) for designating the block in the memory under test 300 and the effective bit width of the block address pointer (BAP) for generating the intra-group block address.
- the effective bit width of the register (Register_A) to be set is set. Specifically, the number of bits obtained by subtracting the number of bits of the block address pointer from the number of bits of the block address is the effective bit width of the group value that can be set in the register (Register_A).
- bit width of the block address (BFM_Address) is 10 bits. Therefore, in this example, the effective bit width of the register (Register_A) for designating the group is 3 bits (D0 to D2).
- the analysis unit 20 reads the address (AFM_Address) from which the address fail data is read from the address fail memory 30 and the address from which the block fail data is read from the block fail memory 40. (BFM_Address) can be generated. Furthermore, the analysis unit 20 can generate an address (RFC_Address) given to the RFC memory 62 of the row fail counter 52 and an address (CFC_Address) given to the CFC memory 64 of the column fail counter 54.
- FIG. 9 shows an example of setting when the number of row address bits and the number of column address bits in a block are 6 bits each.
- the control unit 60 sets the effective bit width of the row address pointer (RAP) 74 to 6 bits (R0 to R5), and the column address.
- the effective bit width of the pointer (CAP) 76 is set to 6 bits (C0 to C5).
- the effective bit width of the block address pointer (BAP) 44 is set to 8 bits (B0 to B7). Furthermore, the effective bit width of the register (Register_A) is set to 2 bits (D0 to D1).
- FIG. 10 shows an example of setting when the number of row address bits and the number of column address bits in a block are 10 bits each.
- the control unit 60 sets the effective bit width of the row address pointer (RAP) to 10 bits (R0 to R9) and sets the column address pointer ( The effective bit width of (CAP) 76 is set to 10 bits (C0 to C9).
- the effective bit width of the block address pointer (BAP) 44 is set to 4 bits (B0 to B3). Furthermore, the effective bit width of the register (Register_A) is set to 6 bits (D0 to D5).
- the setting can be changed adaptively. Therefore, according to the failure analysis memory unit 10 according to the present embodiment, even when the capacity of the memory under test 300 is increased, failure repair analysis is performed without increasing the capacity of the memory that stores the number of defective cells. Can do. Furthermore, according to the failure analysis memory unit 10, it is not necessary to increase the capacity of the memory that stores the number of defective cells in accordance with the increase in the capacity of the memory under test 300. Therefore, the capacity of the memory that stores the number of defective cells. The cost can be reduced by reducing itself.
- FIG. 11 shows a processing flow of the analysis unit 20.
- the analysis unit 20 executes the following processing from step S31 to step S41 in the defect repair analysis processing.
- the analysis unit 20 executes the processing from step S32 to step S40 for each group (a part of a plurality of blocks in the memory under test 300) (S31, S41).
- the analysis unit 20 counts the number of defective cells for each row address (RFC) for each block in the group and the number of defective cells for each column address (CFC) for each block in the group.
- the number of defective cells (TFC) for each block in the group is initialized to 0 (S32).
- the analysis unit 20 executes the processing from step S34 to step S38 for each block of the group (S33, S39).
- the analysis unit 20 reads the block fail data of the block from the block fail memory 40, and determines whether there is a defective cell of the block (S34). If there is no defective cell in the block (No in S34), the analysis unit 20 advances the process to step S39. When there is a defective cell in the block (Yes in S34), the analysis unit 20 advances the process to Step S35.
- the analysis unit 20 increments the row address pointer from 0 to the maximum value of the effective bit width for each block, generates a carry signal when the maximum value is reached, and 0 in the next cycle. Return to, and perform increment operation. At the same time, the analysis unit 20 increments the column address pointer from 0 to the maximum effective bit width every time a carry signal is generated from the row address pointer (S35, S38).
- the analysis unit 20 reads the address fail data from the address fail memory at the address specified by the row address and the column address from the address fail memory 30, and determines whether there is a defective cell at the address (S36). . If there is no defective cell at the address (No in S36), the process proceeds to step S38. If there is a defective cell at the address (Yes in S36), the analysis unit 20 advances the process to step S37.
- step S37 the analysis unit 20 increments the number of defective cells (RFC) corresponding to the row address in the block by one.
- the analysis unit 20 increments the number of defective cells (CFC) corresponding to the column address in the block by one. Further, the analysis unit 20 increments the number of defective cells (TFC) in the block by one.
- step S38 the analysis unit 20 determines in step S36 that the row address is not the last row address in the block and the process column address is not the last column address in the block. return.
- step S39 if the block is not the last block in the group, the analysis unit 20 returns the process to step S34. Moreover, the analysis part 20 advances a process to step S40, when the said block is the last block in the said group.
- step S40 the analysis unit 20 determines the number of defective cells for each row address (RFC) for each block in the group, the number of defective cells for each column address (CFC) for each block in the group, Based on the number of defective cells per block (TFC), a repair solution for repairing the defective cells included in the group is calculated.
- step S41 if the group is not the last group in the memory under test 300, the analysis unit 20 returns the process to step S32, and the group is the last group in the memory under test 300. If it is a group, the defect repair analysis process is terminated.
- the memory under test 300 stores a plurality of bits of data (for example, 16 bits and 32 bits of data) for one address.
- the analysis unit 20 includes a plurality of row fail counters 52, a plurality of column fail counters 54, and a plurality of total fail counters corresponding to a plurality of bits constituting one data stored in the memory under test 300. 56 may be included.
- the reading unit 50 reads a plurality of bits of data from the address fail memory 30 and corresponds to each of the plurality of row fail counters 52, the plurality of column fail counters 54, and the plurality of total fail counters 56. The number of defective cells is counted for each bit to be performed. According to such an analysis unit 20, since the reading unit 50 can be provided in common for a plurality of bits, the circuit scale can be reduced.
- test equipment 100 test equipment, 102 timing generator, 104 pattern generator, 106 waveform shaper, 108 logic comparator, 10 defect analysis memory unit, 12 AFM address formatter, 14 BFM address formatter, 16 AFM address selector, 18 BFM address Selector, 20 analysis unit, 30 address fail memory, 40 block fail memory, 42 group designation unit, 44 block address pointer, 46 block address generation unit, 48 update unit, 50 read unit, 52 row fail counter, 54 column fail counter, 56 Total fail counter, 58 arithmetic unit, 60 control unit, 62 RFC memory, 63 adder, 64 CFC memory, 65 adder, 66 TFC memory, 67 adder 72 start detection unit, 74 row address pointer, 76 column address pointer, 78 address control unit, 80 memory address generation unit, 82 row address generation unit, 84 column address generation unit, 86 row carry selector, 88 column carry selector, 90 End detection unit, 300 memory under test
Abstract
Description
Claims (11)
- 被試験メモリを試験する試験装置であって、
前記被試験メモリにおけるアドレス毎に、不良セルを含むか否かを示すアドレスフェイルデータを記憶するアドレスフェイルメモリと、
前記被試験メモリにおける複数のセルを含むブロック毎に、不良セルを含むか否かを示すブロックフェイルデータを記憶するブロックフェイルメモリと、
前記被試験メモリ内のブロック毎に、前記アドレスフェイルメモリから前記アドレスフェイルデータを読み出す読出部と、
前記被試験メモリ内の一部の複数のブロックを有するグループ内におけるロウアドレス毎に、前記読出部により読み出された前記アドレスフェイルデータに示された不良セルをカウントするロウフェイルカウンタと、
前記グループ内におけるカラムアドレス毎に、前記読出部により読み出された前記アドレスフェイルデータに示された不良セルをカウントするカラムフェイルカウンタと、
を備える試験装置。 - 前記グループ内におけるブロック毎に、前記読出部により読み出された前記アドレスフェイルデータに示された不良セルをカウントするトータルフェイルカウンタと、
を更に備える請求項1に記載の試験装置。 - グループ毎に、前記ロウフェイルカウンタによりカウントされた不良セルの数および前記カラムフェイルカウンタによりカウントされた不良セルの数に基づき、前記被試験メモリの救済解析をする演算部
を更に備える請求項1から2の何れかに記載の試験装置。 - 救済解析をするグループを指定するグループ値を出力するグループ指定部と、
前記グループ内における前記ブロックフェイルデータを読み出すブロックを指定するグループ内ブロックアドレスを順次に発生するグループ内ブロックアドレス発生部と、
前記グループ値および前記グループ内ブロックアドレスを合成して、前記ブロックフェイルメモリから前記ブロックフェイルデータを読み出すブロックを指定するブロックアドレスを生成するブロックアドレス生成部と、
を更に備える請求項1から3の何れかに記載の試験装置。 - 前記読出部は、
ブロック内におけるロウアドレスを指定するロウアドレス発生部と、
ブロック内におけるカラムアドレスを指定するカラムアドレス発生カラムアドレス発生部と、
一つのブロック内に含まれるロウアドレス数およびカラムアドレス数に応じて、ブロック内の全てのアドレスを指定するロウアドレスおよびカラムアドレスの組合せを、前記ロウアドレス発生部および前記カラムアドレス発生部から順次に出力させるアドレス制御部と、
前記ブロックアドレス、前記ロウアドレスおよび前記カラムアドレスを合成して、前記アドレスフェイルメモリから前記アドレスフェイルデータを読み出すアドレスを生成するメモリアドレス生成部と、
を有する請求項4に記載の試験装置。 - 前記読出部は、
前記グループ内ブロックアドレスおよび前記ロウアドレスを合成して、前記ロウフェイルカウンタに与えるアドレスを生成するロウアドレス生成部と、
前記グループ内ブロックアドレスおよび前記カラムアドレスを合成して、前記カラムフェイルカウンタに与えるアドレスを生成するカラムアドレス生成部と、
を更に有する請求項5に記載の試験装置。 - 一つのブロックに含まれるロウアドレス数およびカラムアドレス数を前記アドレス制御部に設定する制御部を更に備える請求項5から6の何れかに記載の試験装置。
- 前記グループ指定部は、前記グループ値を記憶するレジスタであり、
当該試験装置は、一つのグループについて救済解析がされた後に、次に救済解析をすべきグループを示すグル-プ値を前記グループ指定部に書き込む制御部を更に備える請求項4に記載の試験装置。 - 前記グループ指定部は、前記グループ値を記憶するカウンタであり、一つのグループについて救済解析がされる毎に、前記グループ値をインクリメントする請求項4に記載の試験装置。
- 当該試験装置は、前記被試験メモリに記憶される一つのデータを構成する複数のビットに対応した、複数のロウフェイルカウンタおよび複数のカラムフェイルカウンタを備え、
前記読出部は、前記アドレスフェイルメモリに対してアドレスを共通に指定して、前記複数のロウフェイルカウンタおよび前記複数のカラムフェイルカウンタのそれぞれに、対応するビットの不良セルの数をカウントさせる
請求項1から9の何れかに記載の試験装置。 - 被試験メモリを試験する試験装置における前記被試験メモリの救済解析方法であって、
前記試験装置は、
前記被試験メモリにおけるアドレス毎に、不良セルを含むか否かを示すアドレスフェイルデータを記憶するアドレスフェイルメモリと、
前記被試験メモリにおける複数のセルを含むブロック毎に、不良セルを含むか否かを示すブロックフェイルデータを記憶するブロックフェイルメモリと、
を備え、
前記被試験メモリ内のブロック毎に、前記アドレスフェイルメモリから前記アドレスフェイルデータを読み出し、
前記被試験メモリ内の一部の複数のブロックを有するグループ内におけるロウアドレス毎に、読み出された前記アドレスフェイルデータに示された不良セルをカウントし、
前記グループ内におけるカラムアドレス毎に、読み出された前記アドレスフェイルデータに示された不良セルをカウントし、
不良セルのカウント結果に基づいてグループ毎に前記被試験メモリの救済解析を行う
救済解析方法。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2011522619A JP5087704B2 (ja) | 2009-07-13 | 2009-07-13 | 試験装置および救済解析方法 |
KR1020117027961A KR101288113B1 (ko) | 2009-07-13 | 2009-07-13 | 시험 장치 및 구제 해석 방법 |
PCT/JP2009/003271 WO2011007383A1 (ja) | 2009-07-13 | 2009-07-13 | 試験装置および救済解析方法 |
TW099120873A TWI433159B (zh) | 2009-07-13 | 2010-06-25 | Test equipment and relief analysis method |
US13/298,207 US8325547B2 (en) | 2009-07-13 | 2011-11-16 | Test apparatus and repair analysis method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2009/003271 WO2011007383A1 (ja) | 2009-07-13 | 2009-07-13 | 試験装置および救済解析方法 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US13/298,207 Continuation US8325547B2 (en) | 2009-07-13 | 2011-11-16 | Test apparatus and repair analysis method |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2011007383A1 true WO2011007383A1 (ja) | 2011-01-20 |
Family
ID=43449007
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2009/003271 WO2011007383A1 (ja) | 2009-07-13 | 2009-07-13 | 試験装置および救済解析方法 |
Country Status (5)
Country | Link |
---|---|
US (1) | US8325547B2 (ja) |
JP (1) | JP5087704B2 (ja) |
KR (1) | KR101288113B1 (ja) |
TW (1) | TWI433159B (ja) |
WO (1) | WO2011007383A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013097853A (ja) * | 2011-11-04 | 2013-05-20 | Sk Hynix Inc | セルフリフレッシュパルス生成回路 |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR102070729B1 (ko) | 2013-11-12 | 2020-03-02 | 삼성전자주식회사 | 반도체 메모리 장치 및 그것을 포함하는 메모리 시스템 |
KR102252376B1 (ko) | 2014-12-08 | 2021-05-14 | 삼성전자주식회사 | 셀 특성 플래그를 이용하여 리프레쉬 동작을 제어하는 메모리 장치 |
KR20160071769A (ko) * | 2014-12-12 | 2016-06-22 | 삼성전자주식회사 | 반도체 메모리 장치 및 이를 포함하는 메모리 시스템 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH03269279A (ja) * | 1990-03-19 | 1991-11-29 | Advantest Corp | 半導体メモリ試験装置 |
JPH06259993A (ja) * | 1993-03-05 | 1994-09-16 | Advantest Corp | 半導体メモリ試験装置 |
JPH1092195A (ja) * | 1996-09-18 | 1998-04-10 | Advantest Corp | メモリ試験装置 |
JP2000123595A (ja) * | 1998-08-14 | 2000-04-28 | Advantest Corp | メモリ試験装置 |
WO2006001164A1 (ja) * | 2004-06-23 | 2006-01-05 | Advantest Corporation | 試験装置及び試験方法 |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001035189A (ja) * | 1999-07-16 | 2001-02-09 | Advantest Corp | メモリの不良救済解析方法及びこの解析方法を適用した不良救済解析器を搭載したメモリ試験装置 |
JP2005259266A (ja) | 2004-03-11 | 2005-09-22 | Advantest Corp | 試験装置及び試験方法 |
JP4402093B2 (ja) * | 2006-10-26 | 2010-01-20 | 株式会社アドバンテスト | 半導体試験装置および半導体メモリの試験方法 |
JP4472004B2 (ja) * | 2007-02-16 | 2010-06-02 | 株式会社アドバンテスト | 試験装置 |
-
2009
- 2009-07-13 JP JP2011522619A patent/JP5087704B2/ja not_active Expired - Fee Related
- 2009-07-13 WO PCT/JP2009/003271 patent/WO2011007383A1/ja active Application Filing
- 2009-07-13 KR KR1020117027961A patent/KR101288113B1/ko active IP Right Grant
-
2010
- 2010-06-25 TW TW099120873A patent/TWI433159B/zh active
-
2011
- 2011-11-16 US US13/298,207 patent/US8325547B2/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH03269279A (ja) * | 1990-03-19 | 1991-11-29 | Advantest Corp | 半導体メモリ試験装置 |
JPH06259993A (ja) * | 1993-03-05 | 1994-09-16 | Advantest Corp | 半導体メモリ試験装置 |
JPH1092195A (ja) * | 1996-09-18 | 1998-04-10 | Advantest Corp | メモリ試験装置 |
JP2000123595A (ja) * | 1998-08-14 | 2000-04-28 | Advantest Corp | メモリ試験装置 |
WO2006001164A1 (ja) * | 2004-06-23 | 2006-01-05 | Advantest Corporation | 試験装置及び試験方法 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2013097853A (ja) * | 2011-11-04 | 2013-05-20 | Sk Hynix Inc | セルフリフレッシュパルス生成回路 |
Also Published As
Publication number | Publication date |
---|---|
KR20120016637A (ko) | 2012-02-24 |
TW201112256A (en) | 2011-04-01 |
JPWO2011007383A1 (ja) | 2012-12-20 |
US20120120748A1 (en) | 2012-05-17 |
JP5087704B2 (ja) | 2012-12-05 |
KR101288113B1 (ko) | 2013-07-19 |
TWI433159B (zh) | 2014-04-01 |
US8325547B2 (en) | 2012-12-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JPH03269279A (ja) | 半導体メモリ試験装置 | |
WO2007086214A1 (ja) | 試験装置および選択装置 | |
JP4402093B2 (ja) | 半導体試験装置および半導体メモリの試験方法 | |
JP5202556B2 (ja) | 制御装置、試験装置および制御方法 | |
WO2004092755A1 (ja) | 試験装置 | |
JPH09318707A (ja) | 半導体メモリ試験方法および装置 | |
JP5087704B2 (ja) | 試験装置および救済解析方法 | |
US20120117432A1 (en) | Test apparatus | |
JP5001972B2 (ja) | メモリ補修分析自己検査機能を具えた半導体検査システム | |
JP4130811B2 (ja) | 試験装置及び試験方法 | |
JP5161964B2 (ja) | 試験装置および試験方法 | |
KR101096138B1 (ko) | 시험 장치 및 시험 방법 | |
JP4183854B2 (ja) | メモリ試験装置 | |
US20120216086A1 (en) | Test apparatus | |
JP3547070B2 (ja) | 半導体メモリ試験装置 | |
JPH11316259A (ja) | 半導体試験装置およびこれを用いた半導体試験方法 | |
JP5255710B1 (ja) | 不良情報記憶装置および試験システム | |
JP3079676B2 (ja) | 集積回路試験装置 | |
JP5326800B2 (ja) | 解析装置および解析方法 | |
JP2008071451A (ja) | 半導体試験装置 | |
JPH1186595A (ja) | 半導体メモリ試験装置 | |
JP2001004705A (ja) | 半導体デバイス試験装置及び試験方法 | |
JPH1186594A (ja) | 半導体メモリ試験装置 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 09847283 Country of ref document: EP Kind code of ref document: A1 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2011522619 Country of ref document: JP |
|
ENP | Entry into the national phase |
Ref document number: 20117027961 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 09847283 Country of ref document: EP Kind code of ref document: A1 |