摘要:
A method is provided for storing data in a memory device having memory cells arranged in memory cell rows and memory cell columns. The method can include a step for providing redundant memory cells in the memory device. The method can also include a step for localizing defective cells. Further, the method can include a step of accessing the redundant memory cells by means of a predeterminable access mode. The method can also include a step of bypassing defective memory cells of the memory device in a manner dependent on the predeterminable access mode during operation of the memory device for accessing redundant memory cells and for replacement by redundant memory cells. Further, the method can include a step for providing redundant memory cells for storing additional information describing a defect correction.
摘要:
The present invention relates to an interconnect structure for an integrated circuit (1) having a first interconnect (B1; B1′; B1″), which is composed of a plurality of interconnect sections (A11-A16; A11′-A16′; A11″-A14″) lying in a first and a second interconnect plane (M0, M1); and a second interconnect (B2; B2′; B2″), which runs adjacent to the first interconnect (B1; B1′; B1″) and which is composed of a plurality of interconnect sections (A21-A25; A21′-A25′; A21″-A23″) lying in the first and second interconnect planes (M0, M1); the first and second interconnects (B1; B1′; B1″; B2; B2′; B2″) being offset with respect to one another in the longitudinal direction in such a way that the interconnect sections (A12, A14, A16; A12′, A14′, A16′; A12″, A14″) of the first interconnect (B1; B1′; B1″) which lie in the first interconnect plane (M0) run at least in sections beside the interconnect sections (A22, A24; A22′; A24′; A21″, A23″) of the second interconnect (B2; B2′; B2″) which lie in the second interconnection plane (M1), and that the interconnect sections (A11, A13, A15; A11′, A13′, A15′; A11″, A13″) of the first interconnect (B1; B1′; B1″) which lie in the second interconnect plane (M1) run at least in sections beside the interconnect sections (A21, A23, A25; A21′, A23′, A25′; A22″) of the second interconnect (B2; B2′; B2″) which lie in the first interconnect plane (M0). The invention also provides a corresponding fabrication method.
摘要:
According to one embodiment, a dynamic memory is provided. The dynamic memory can include a memory matrix having a plurality of memory cells arranged in rows and columns. The memory cells in a row can be connected by in each case one of a plurality of word lines. The memory cells in a column can be connected by in each case one of a plurality of bit lines. The dynamic memory can also include a sense amplifier for reading data from the memory cells via the plurality of bit lines. Further, the dynamic memory can include a row address decoder and a column address decoder for generating memory-internal address in a manner dependent on a memory-external address signal. The dynamic memory can also include a sequence control device for cyclically generating refresh addresses for carrying for carrying out a refresh operation of the memory cells.
摘要:
A method for replacing defective memory cells of a random access memory device of a data processing apparatus, in which, during the operation of the data processing apparatus, a defective memory cell is replaced by a replacement memory cell in the random access memory device by using a control instruction.
摘要:
A method for integrating imperfect semiconductor memory devices having functional and defective memory cells into a data processing apparatus. The defective memory cells are assigned defect addresses or defect address ranges. Before a memory access of the data processing apparatus is carried out, the address of the memory access is compared with the defect addresses or defect address ranges and is recoded in the event of correspondence.
摘要:
A memory system for network broadcasting applications, such as video/audio applications, has at least one memory which is divided into a plurality of addressable memory units, which have a respective dedicated output for interchanging data. The inputs of a matrix switch are connected to a respective output of a different memory unit. The matrix switch is operated such that a plurality of the memory units are connected to its output in a sequential order. A first sequence of memory units and a second sequence of memory units are connected to its output independently. This results in a memory system, which can handle a number of requests to the same memory at staggered times. The interaction of the individual memory units with the matrix switch allows a high data throughput and a short access time.
摘要:
A memory system for network broadcasting applications, such as video/audio applications, has at least one memory which is divided into a plurality of addressable memory units, which have a respective dedicated output for interchanging data. The inputs of a matrix switch are connected to a respective output of a different memory unit. The matrix switch is operated such that a plurality of the memory units are connected to its output in a sequential order. A first sequence of memory units and a second sequence of memory units are connected to its output independently. This results in a memory system, which can handle a number of requests to the same memory at staggered times. The interaction of the individual memory units with the matrix switch allows a high data throughput and a short access time.
摘要:
To be able to test a plurality of identical semiconductor circuit devices in a particularly rapid yet reliable manner, a test method includes carrying out the tests in parallel and substantially simultaneously on the plurality of semiconductor circuit devices and driver lines—used in the process—of a test device to the semiconductor circuit devices simultaneously and jointly for all the semiconductor circuit devices. In such a case, test results are read from a plurality of input/output channels in compressed form. Furthermore, as an alternative or in addition thereto, the semiconductor circuit devices to be tested are disposed and connected up in at least one stack.
摘要:
A method for classifying semiconductor components by their performance characteristics includes reading an identifier associated with a component and storing, in conjunction with the identifier, first and second performances values for that component. These performance values represent the component's achieved operating speed at each of two different temperatures. The component is then allocated to a speed category on the basis of the first and second performance values.
摘要:
Disclosed is a test method for testing a data store having an integrated test data compression circuit where the data store has a memory cell array with a multiplicity of addressable memory cells, read/write amplifiers for reading and writing data to the memory cell via an internal data bus in the data store and a test data compression circuit which compresses test data sequences, which are each read serially from the memory cell array, with stored reference test data sequences in order to produce a respective indicator data item which indicates whether at least one data error has occurred in the test data sequence which has been read.