WO1994006082A1 - Circuit de memoire avec redondance - Google Patents
Circuit de memoire avec redondance Download PDFInfo
- Publication number
- WO1994006082A1 WO1994006082A1 PCT/FR1993/000842 FR9300842W WO9406082A1 WO 1994006082 A1 WO1994006082 A1 WO 1994006082A1 FR 9300842 W FR9300842 W FR 9300842W WO 9406082 A1 WO9406082 A1 WO 9406082A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- row
- rank
- decoder
- output
- circuit
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/16—Error detection or correction of the data by redundancy in hardware
- G06F11/20—Error detection or correction of the data by redundancy in hardware using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C29/00—Checking stores for correct operation ; Subsequent repair; Testing stores during standby or offline operation
- G11C29/70—Masking faults in memories by using spares or by reconfiguring
- G11C29/78—Masking faults in memories by using spares or by reconfiguring using programmable devices
- G11C29/84—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved access time or stability
- G11C29/848—Masking faults in memories by using spares or by reconfiguring using programmable devices with improved access time or stability by adjacent switching
Definitions
- the invention relates to an integrated circuit memory, produced in the form of a matrix network of rows and columns.
- the memory test shows that a row or column of memory cells is faulty, it is replaced by a row or column of redundancy elements. Seen from the outside of the integrated circuit, the memory must then appear entirely good: the implementation of the redundancy is transparent to the user.
- the word "line" whenever we want to speak indifferently of a row or a column and we will distinguish in the usual way the columns and rows of the network: the memory cells of the network matrix are connected in rows and columns; all the cells of the same column are connected to the same column conductor (called bit line) on which one can read or write information, and all the cells of the same row are connected to the same row conductor (called a word line) which is used to address a particular row of cells.
- bit line column conductor
- a word line which is used to address a particular row of cells.
- a possible architecture is one in which at least one redundancy line (row or column) is placed next to a group of lines (rows or columns), so that any faulty line in this group can be replaced by the redundancy line.
- Each line likely to be replaced is associated with a fuse; the fuse is blown to isolate the faulty line from the rest of the circuit and to connect the redundancy line instead.
- blown fuse means either the open circuit of a fuse which was conductive in the intact state, or on the contrary the setting in the conductive state of a fuse which was not conductive in the state intact.
- redundancy items there is also the ease of full memory testing, including redundancy items, and including unused redundancy items.
- unused redundancy elements are very difficult to access and cannot always be tested.
- Another objective for a memory can sometimes be the possibility of repair not only at the time of manufacture (during the wafer test), but also during use.
- an objective must remain transparency vis-à-vis the user who must not see the difference between a memory which required the implementation of the repair of a fault and a memory which did not have need this repair.
- the additional connections added to access the repair line lengthen the access time to information for repaired addresses compared to the access time to information for normal addresses.
- the present invention is intended to provide a memory architecture that takes you consider the best possible these worries.
- a memory comprising at least one network of n + 1 successive lines of memory elements, a decoder having n outputs, the output of rank j used to designate a line determined as a function of '' an address applied to the decoder, and a redundancy circuit to replace a line of elements which would be defective by a replacement line, characterized in that the redundancy circuit comprises means for activating by the output of rank j of the decoder (j varying from 1 to n) either the row of row j or the row of row j + 1.
- the word line is to be taken in the broad sense: row or column.
- each output of rank j of the decoder selects a line of the same rank j in the succession of lines. The last line, of rank n + 1 is not used. If there is redundancy due to a faulty row of rank r, then the outputs of rank j less than r each activate a row of row j corresponding, and the outputs of rank j greater than or equal to r each activate a row of respective rank j + 1.
- the line of rank j is designated by the output of rank j of the decoder when all the fuses of rank 1 to j are in the intact state.
- the row of row j + 1 is designated by the row j output from the decoder when one of the row 1 to j fuses is blown.
- n + 2 lines are provided for two repair possibilities for a group of n lines controlled by a decoder with n outputs.
- the referral circuits are then more complex since they must be able to direct the output of rank j to a line of rank j, or j + 1 or j + 2 depending on the implementation of zero, one, or two repairs. . But the principle is the same.
- the memory comprises a multiplexer between the bit lines and the input / output pads of the memory.
- the multiplexer has individual multiplexing elements each corresponding to a bit line. Each multiplexing element can route a given bit line to an input / output pad or to a read or read / write amplifier connected to this pad.
- the row j switching circuit has two inputs for this purpose: one is a designation control input coming from the row j output of the decoder, so that this switching circuit is designated by the decoder when the the address corresponding to rank j is received by the decoder; the other input of the switch circuit is a switch control input; it allows you to choose one of the two sort ies of the circuit switch designated by the decoder, the choice depending on the implementation or lack of implementation of redundancy in rows 1 to j.
- This second control circuit input is controlled by an associated redundancy logic circuit, of rank j, depending in particular on the state of the fuse of rank j.
- the redundancy logic circuit of rank j which controls the switch, preferably comprises the following elements: a monostable flip-flop connected to the fuse of rank j, this flip-flop being maintained in a first state when the fuse is intact and being released towards a second state when the fuse is blown; a gate with two inputs receiving on the one hand the output of the monostable flip-flop and on the other hand an output of the logic circuit of previous rank, this gate providing an output signal applied on the one hand to the logic circuit of next rank j + 1 and on the other hand to the row j referral circuit.
- the lengths of connections and the circuitry between the repair line and the rest of the circuit are minimized: there is not a long connection length between a distant repair line and the other lines, which would tend to lengthen the access time (capacitive effect of long lines, and passage time in doors) for the repaired line compared to the access time for other lines.
- the access times are the same for all the lines, and are relatively independent of whether redundancy is brought into service or not.
- the logic added for redundancy can be logic without current consumption.
- the architecture proposed makes it possible, if desired, to very easily control the interruption of the supply of faulty lines (that is to say, we do not just disconnect them from the input / output circuits , but we cut their power), which is important in some cases and which is not allowed by all redundancy architectures.
- FIG. 1 shows the classic architecture of memory vine with redundancy circuitry for the repair of defective columns.
- FIG. 3 represents a detailed diagram of the preferred embodiment of the invention.
- a memory comprising an array of memory cells MC arranged in rows and columns.
- the cells of the same row are connected to the same word line WL originating from a row decoder DR which makes it possible to designate one row among N as a function of a row address AR.
- the cells of the same column are connected to the same bit line BL which makes it possible to read or write information in the cell located at the intersection of this bit line and the selected word line.
- there is a DC column decoder which receives a column address ⁇ C and controls a MUX multiplexer.
- the columns are then associated into P groups of n columns.
- the groups are juxtaposed or nested within each other.
- the multiplexer makes it possible to select a bit line from n, this in each group, and to connect it to a read amplifier and a write amplifier, the latter being connected to a data pad.
- two groups of n columns with respectively two multiplexers MUX1, MUX2 (simultaneously controlled by the decoder DC), two read amplifiers ALI, AL2, two write amplifiers AE1, AE2, and two data pads PI , P2.
- This additional column is designated by CRI for the first group, CR2 for the second; there is a redundancy circuit RD1, RD2 which controls access to this redundancy column and a circuit CF with fuses (n fuses), in the column decoder DC or interposed between the column decoder and the control inputs of the multiplexer, to prevent access to the defective column by simply blowing the fuse corresponding to this column.
- An additional fuse FS activates the redundancy circuits RDI, RD2 if redundancy is used.
- FIG. 2 represents the general principle of the modification made by the invention to the architecture of FIG. 1.
- the multiplexer MUX1, MUX2 therefore has, for each group, n + 1 bit line connections and can select any one from n + 1 bit lines to connect it to the corresponding pad PI or P2.
- the DC column decoder which has only n outputs, now controls the MUX multiplexer through a ⁇ IG switching circuit.
- the AIG switching circuit has been shown in FIG. 2 as distinct from the MUX multiplexer. This is a convenient representation, but it will be understood that the switch block, as well as the fuse circuit, can be strongly nested inside the multiplexer. The important thing is the operating mode with offset of one unit between the rank of the selected bit line and the rank of the decoder output, this for all the lines following the faulty line but not for those preceding it.
- FIG. 3 represents a detailed embodiment of the invention for obtaining the operation described above.
- the inputs of the multiplexer MUX are the n + 1 pairs of complementary bit lines capable of being connected to the pad P: bit line BL (j) and complementary line NBL (j) for row j, lines BL (j + l ) and NBL (j + l) for row j + 1, etc.
- the decoder, the fuse circuit, and the routing circuit are broken down each into n elements, each element corresponding to a determined rank, that is to say to a determined address supplied to the decoder; only two elements successive, of rows j and j + 1, are shown in FIG. 3.
- the edge of row j is identified in the figure between two vertical dashed lines.
- the element of rank j of the decoder referenced DC (j) provides on its output a logic level 1 if the address received represents rank j and a level 0 otherwise. This output is connected to a designation input for the AIG switching circuit element (j) of the same rank j.
- the switching element of rank j is therefore only active if the address received by the decoder corresponds to rank j.
- the AIG element (j) has two outputs making it possible to activate either the multiplexing element of the same rank MUX (j) or the multiplexing element of rank immediately following MUX (j + 1).
- the first output is connected to a command input of the MUX element (j) and makes it active (therefore connects the bit line of rank j to the pad P) by imposing a high logic level (1) on this command input; the other output is connected to the command input of the MUX element (j + l) and makes it active (therefore connects the bit line of rank j + 1 to the pad P) by imposing a high logic level (1 ) on this entry.
- the switching element is made inactive by the DC decoder (j)
- its two outputs are at a low logic level (0) and neither the bit line BL (j) nor the next line BL (j + l) cannot be connected to terminal P.
- the routing circuit therefore has an inactive state and two complementary active states.
- control input of the rank j multiplexer element receives not only the first output of the AIG switching element (j) of the same rank, but also the second output of the switching element of rank preceding d-1.
- control input of the multiplexing element MUX (j + l) receives not only the second output of the element AIG (j) but also the first output of the element AIG (j + l).
- the switching element ⁇ IG (j) has a switching control input which receives the output of an AND gate Gl (j). Depending on the state of this output, the AIG switching element (j) provides a logic level 1 (high) either on its first output (first active state of the switching circuit) or on its second output (second active state). ), the other output being at 0.
- AIG (d-l) More precisely, if the circuit of switch of previous rank (j-1) is in its second state (its second output is active), the AND gate Gl (j) necessarily puts the circuit of switch. row j in its second state, and so on for the following rows. For this, since the door in this example is an AND gate, provision is made for a zero state on the switch control input to place the switch in the second state. If any routing circuit goes into its second state, this second state therefore affects all the following routing circuits. Of course, other very similar logical structures can lead to the same general result.
- Fuses can be physical elements that can be blown electrically or by laser, or can be non-volatile memory cells (EPROM, EEPROM, UPROM) whose state is modified by electrical programming.
- the fuse is a physical element which can be electrically blown by a current flowing through it.
- the fuse F (j) is in series with a breakdown transistor Tl (j), the assembly in series between two supply terminals Vdd and Vss.
- An AND gate G2 (j) allows the transistor to be conductive to blow the fuse; this door receives on a first input the output of the DC decoder (j) of rank j (to select a single fuse to be blown which is the fuse corresponding to the column which is being tested); the door also receives on a second input a signal F which is a breakdown command. This order is only issued if the column is found to be defective.
- the fuse F (j) has one end connected to the supply voltage Vdd and another end connected to the input of a monostable rocker MS (j).
- the scale is then maintained in a forced unstable state (output at 1) when the fuse is intact. Its output is connected to the second input of the AND gate Gl (j). The blowing of the fuse releases the rocker which goes into its stable state (output at 0).
- the circuit elements of rank 1 they are identical to the others, except for the fact that the first input of the gate AND Gl (l) of rank 1 receives in permanent operation a logic level hatit, for example the supply voltage Vdd.
- a transistor T2 (j) is preferably provided to impose at zero the command of the mult iplexer MUX (j) when the line j is defective, this not to leave this line floating.
- the transistor T2 (j) is for example an N-channel transistor connected between the control of the multiplexer MUX (j) and the ground; the transistor becomes conductive when the fuse F (j) is blown.
- the designation of rank j by the column decoder connects the pad P to the bit line of rank j for all j from 1 to r-1; and it connects the pad P to the bit line of rank j + 1 for all the j from r to n; the defective column is therefore replaced by the next one and the addressing of all the other columns is shifted by one.
- FIG. 4 shows the last two rows of the circuit, namely n and n + 1, the circuit elements of rank n being identical to the elements of rank j in FIG. 3 and the elements for rank n + 1 being slightly different since they do not include a DC decoder element, an AIG switch element, or a fuse.
- the multiplexer element MUX (n + 1) is controlled only by the second output of the switching circuit of rank n.
- a transistor T2 (n + J) makes it possible to earth this second output to isolate the column of rank n + 1 if redundancy is not used. This transistor is controlled by the output of the AND gate G 1 (n) which remains at 1 as long as the redundancy is not used.
- n + 1 does not require a fuse. There are therefore in all n fuses for a decoder having. n outputs. In the redundancy circuits of the prior art, in general n fuses are required to select one of the n lines to be repaired, plus a fuse to indicate that there is a repair.
- the only difference with respect to row j is the fact that the AND gate Gl (l) receives a logic level 1 on its first input, so that the output of this door remains at 1 until the fuse F (l) is blown.
- a direct link, without gate Gl (l), could moreover be provided between the output of the monostable rocker MS (1) and the switch ATG (l).
- a transistor mounted in resistance maintains the first input of the gate at the positive supply voltage Vdd.
- the first input of the AND gate Gl (l) is connected to a PT test pad, that is to say a pad which is not connected to an external connection wire but to which one can apply a test tip during wafer test operations.
- the stud then remains isolated in normal operation.
- the structure of the invention makes it possible to benefit from a very important advantage: the possibility of testing all the columns, including the last, even if the latter is not used.
- these columns are addressed by the decoder, the pad PT being maintained at 1 as in the configuration of normal use.
- Row j designated by the decoder selects column j until the fuses are blown. If no fuse is blown during the test of columns 1 to n, the PT pad is set to zero while the decoder selects rank n. This zero crossing simulates the blowing of a fictitious fuse of rank zero, which switches all the switches from rank 1 to n to the addressing of columns of rank 2 to n + 1.
- the column n + 1 is then selected by the decoder and can be tested.
- An additional advantage of the invention is the possibility of easily removing the voltage supply from the defective bit line. It is indeed desirable that the bit line does not receive a permanent supply or even a precharge supply if it is defective. Indeed, a frequent defect is a short-circuit of the bit line with ground, and this short-circuit would remain, with its drawbacks of unnecessary current consumption even with repair by a replacement line.
- a very simple circuit is then preferably provided, controlled by the output of the monostable MS (j) to cut the supply to the line of row j if the fuse of row j is blown. In FIG. G, this possibility is shown in a particular example where the bit line is supplied by a P channel transistor connected to a supply line Vdd.
- the transistor T3 (j) supplies the bit line BL (j), and the transistor T'3 (j) supplies the complementary line NBL (j).
- this transistor would have its control gate in principle connected to ground during the instants when the bit line must be supplied.
- the gate of the transistors is controlled by means of a circuit which prohibits the conduction of the transistors T3 (j) and T'3 (j) when the fuse of rank j is blown.
- the bit lines are permanently supplied by the transistors, and consequently the grid of the transistors of rank j is connected by means of an inverter to the output of the monostable flip-flop likewise MS rank (j). If the bit lines are not permanently supplied, the grid is connected to a logic circuit which receives the output of the monostable flip-flop and which takes account of this output to prevent the conduction of the transistors of rank j.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- For Increasing The Reliability Of Semiconductor Memories (AREA)
- Design And Manufacture Of Integrated Circuits (AREA)
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP93919413A EP0659291A1 (fr) | 1992-09-08 | 1993-09-03 | Circuit de memoire avec redondance |
KR1019950700895A KR950703176A (ko) | 1992-09-08 | 1993-09-03 | 리던던시 구조를 갖는 메모리 회로(Memory circuit with redundancy architecture) |
JP6506930A JPH08501178A (ja) | 1992-09-08 | 1993-09-03 | 冗長性を有するメモリ回路 |
US08/393,004 US5506807A (en) | 1992-09-08 | 1995-03-03 | Memory circuit with redundancy |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR92/10695 | 1992-09-08 | ||
FR9210695A FR2695493B1 (fr) | 1992-09-08 | 1992-09-08 | Circuit de mémoire avec redondance. |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1994006082A1 true WO1994006082A1 (fr) | 1994-03-17 |
Family
ID=9433295
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR1993/000842 WO1994006082A1 (fr) | 1992-09-08 | 1993-09-03 | Circuit de memoire avec redondance |
Country Status (6)
Country | Link |
---|---|
US (1) | US5506807A (fr) |
EP (1) | EP0659291A1 (fr) |
JP (1) | JPH08501178A (fr) |
KR (1) | KR950703176A (fr) |
FR (1) | FR2695493B1 (fr) |
WO (1) | WO1994006082A1 (fr) |
Families Citing this family (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3325456B2 (ja) * | 1996-05-22 | 2002-09-17 | 株式会社アドバンテスト | メモリリペア方法ならびにそのメモリリペア方法が適用される電子ビームメモリリペア装置およびメモリ冗長回路 |
GB9417269D0 (en) * | 1994-08-26 | 1994-10-19 | Inmos Ltd | Memory and test method therefor |
JPH09147595A (ja) * | 1995-11-24 | 1997-06-06 | Nec Corp | 半導体記憶装置 |
US5870341A (en) * | 1997-06-19 | 1999-02-09 | Sun Microsystems, Inc. | Memory column redundancy circuit |
KR100331542B1 (ko) * | 1998-10-09 | 2002-06-20 | 윤종용 | 불량메모리셀어레이블락들을스킵할수있는어드레스디코더를구비하는반도체메모리장치및이를사용하는복합반도체장치 |
JP4439683B2 (ja) * | 1999-06-03 | 2010-03-24 | 三星電子株式会社 | リダンダンシ選択回路を備えたフラッシュメモリ装置及びテスト方法 |
FR2817982B1 (fr) * | 2000-12-08 | 2003-10-24 | St Microelectronics Sa | Circuit memoire a redondance partagee |
US6862230B2 (en) * | 2002-03-19 | 2005-03-01 | Broadcom Corporation | Efficient column redundancy techniques |
DE10256487B4 (de) * | 2002-12-03 | 2008-12-24 | Infineon Technologies Ag | Integrierter Speicher und Verfahren zum Testen eines integrierten Speichers |
US6928377B2 (en) * | 2003-09-09 | 2005-08-09 | International Business Machines Corporation | Self-test architecture to implement data column redundancy in a RAM |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0477809A2 (fr) * | 1990-09-28 | 1992-04-01 | Intergraph Corporation | Lignes et colonnes redondantes à haute vitesse pour mémoires à semi-conducteurs |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH01224999A (ja) * | 1988-03-04 | 1989-09-07 | Mitsubishi Electric Corp | 半導体記憶装置 |
JP2837433B2 (ja) * | 1989-06-05 | 1998-12-16 | 三菱電機株式会社 | 半導体記憶装置における不良ビット救済回路 |
US5301153A (en) * | 1992-06-03 | 1994-04-05 | Mips Computer Systems, Inc. | Redundant element substitution apparatus |
KR960002777B1 (ko) * | 1992-07-13 | 1996-02-26 | 삼성전자주식회사 | 반도체 메모리 장치의 로우 리던던시 장치 |
-
1992
- 1992-09-08 FR FR9210695A patent/FR2695493B1/fr not_active Expired - Fee Related
-
1993
- 1993-09-03 JP JP6506930A patent/JPH08501178A/ja active Pending
- 1993-09-03 KR KR1019950700895A patent/KR950703176A/ko not_active Application Discontinuation
- 1993-09-03 EP EP93919413A patent/EP0659291A1/fr not_active Ceased
- 1993-09-03 WO PCT/FR1993/000842 patent/WO1994006082A1/fr not_active Application Discontinuation
-
1995
- 1995-03-03 US US08/393,004 patent/US5506807A/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0477809A2 (fr) * | 1990-09-28 | 1992-04-01 | Intergraph Corporation | Lignes et colonnes redondantes à haute vitesse pour mémoires à semi-conducteurs |
Also Published As
Publication number | Publication date |
---|---|
JPH08501178A (ja) | 1996-02-06 |
US5506807A (en) | 1996-04-09 |
EP0659291A1 (fr) | 1995-06-28 |
FR2695493A1 (fr) | 1994-03-11 |
KR950703176A (ko) | 1995-08-23 |
FR2695493B1 (fr) | 1994-10-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP0317014B1 (fr) | Unité de mémoire vive à plusieurs modes de test et ordinateur muni de telles unités | |
EP0674264B1 (fr) | Circuit de sélection d'éléments de mémoire redondants et mémoire "FLASH EEPROM" comportant ledit circuit | |
EP0858032B1 (fr) | Circuit pour la réparation d'un bit défectueux dans un dispositif de mémoire à semi-conducteurs et méthode de réparation | |
KR100237583B1 (ko) | 용장 기억 소자를 포함하는 메모리를 가진 집적회로 및 메모리의 동작 방법 | |
EP0666572A1 (fr) | Bascule bistable non volatile programmable, à état initial prédéfini, notamment pour circuit de redondance de mémoire | |
FR2688328A1 (fr) | Circuit a redondance de rangees pour dispositif a memoire a semi-conducteurs pour reparer ou remplacer une cellule defectueuse d'un reseau de cellules a memoire. | |
EP0544568B1 (fr) | Circuit de lecture de fusible de redondance pour mémoire intégrée | |
EP0568439A1 (fr) | Procédé et circuit de détection de fuites de courant dans une ligne de bit | |
EP0645714B1 (fr) | Circuit de redondance dynamique pour mémoire en circuit intégré | |
WO1994006082A1 (fr) | Circuit de memoire avec redondance | |
EP0275752B1 (fr) | Circuit intégré comportant des éléments d'aiguillage vers des éléments de redondance dans une mémoire | |
US5995419A (en) | Repairable memory cell for a memory cell array | |
FR2736175A1 (fr) | Circuit de reparation de cellules de memoire en panne dans une memoire a semi-conducteurs | |
FR2641391A1 (fr) | Composant logique programmable effacable rapide | |
FR2611301A1 (fr) | Memoire integree avec redondance de colonnes de donnees | |
EP0665559B1 (fr) | Bascule bistable non volatile programmable, à reduction de parasites en mode de lecture, notamment pour circuit de redondance de mémoire | |
EP0470897B1 (fr) | Circuit intégré de mémoire avec redondance et adressage amélioré en mode de test | |
FR2576133A1 (fr) | Memoire en circuit integre a haute fiabilite | |
EP0675441B1 (fr) | Dispositif matriciel de fusibles de redondance pour mémoire intégrée et procédé de mise en oeuvre | |
KR100697441B1 (ko) | 메모리 셀 및 기준 셀을 포함하는 집적 메모리 | |
EP1168179B1 (fr) | Circuit de mémoire dynamique comportant des cellules de secours | |
JP2001210092A (ja) | 半導体記憶装置 | |
FR2576132A1 (fr) | Memoire en circuit integre | |
FR2888660A1 (fr) | Systeme redondance colonne pour une memoire en circuit integre | |
US5896329A (en) | Repairable memory cell for a memory cell array |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): JP KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1993919413 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 08393004 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 1993919413 Country of ref document: EP |
|
WWR | Wipo information: refused in national office |
Ref document number: 1993919413 Country of ref document: EP |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 1993919413 Country of ref document: EP |