摘要:
A multi-bank DRAM having a hierarchical column select line architecture is described. The DRAM is provided with a plurality of memory cells which are organized in at least two banks. Each of the banks includes memory cells which are arranged in rows and columns. The memory cells store data provided by at least one bit line and at least one data line. The DRAM includes: a first switch for selecting one of the two banks; and a second switch connected to the first switch for selecting one of the columns, wherein the first and second switches couple one of the bit lines to one of the data lines, enabling data to be written into or read out of memory cells common to the selected bank and to the selected column. The first switch is controlled by a plurality of bank CSLs (BCSLs), wherein the BCSLs are shared by some of the blocks within the same bank, but not by any of the blocks in other banks. The second switch is controlled by a plurality of global CSLs (GCSLs), the GCSLs being shared by all remaining banks within a unit. The BCSLs and GCSLs are controlled by the bank column decoder and by the global column decoder.
摘要:
Row redundancy control circuits which effectively reduce design space are arranged parallel to word direction and are arranged at the bottom of the redundancy block. This architecture change makes it possible to effectively lay out the redundancy control block by introducing (1) split-global-bus shared with local row redundancy wires, (2) half-length-one-way row redundancy-wordline-enable-signal wires which allows space saving, and (3) distributed wordline enable decoders designed to take advantage of the saved space. An illegal normal/redundancy access problem caused by the address versus timing skew has also been solved. The timing necessary for this detection is given locally by using its adjacent redundancy match detection. This allows the circuit to operate completely as an address driven circuit, resulting in fast and reliable redundancy match detection. In addition, a sample wordline enable signal (SWLE) is generated by using row redundancy match detection. One two-input OR gate allows the time at which SWLE sets sample wordline (SWL) to be the same as the time at which wordline enable (WLE) signal sets wordline (WL). The time at which SWLE sets SWL remains consistent regardless of mode, eliminating the existing reliability concern. This two-input OR gate combined with row redundancy match detection works as an ideal sample wordline enable generator.
摘要:
A wide Input/Output (I/O) Random Access Memory (RAM) with more efficient redundancy. The RAM array may be divided into individual units. Each unit is further divided into subarray blocks (blocks of subarrays). Each subarray or segment is organized by one and includes one spare column and may include spare word lines. When a block is accessed, only half of the segments are accessed. Whenever a segment is accessed, the segment's spare column is not. The spare columns from the unaccessed half block are available for repairing defective columns in the accessed half block. Data from columns in the accessed half and spare columns in the unaccessed half are transferred to Local Data Lines (LDLs) and from LDLs to Master Data Lines (MDLs). Valid data from accessed column lines and from selected spare lines are provided on the MDLS to second sense amplifiers. Defective columns are electrically replaced with spares after the second stage amplifiers. Thus, all of the spare columns in each half of each subarray block are available to replace an equal number of failed columns at any location in any segment in the other half block.
摘要:
A device for controlling the voltage across an NMOS pull-up transistor including a source node which may be exposed to a variable voltage. The device further includes a gate node which may be exposed to a variable voltage. A control portion regulates the voltage applied to the gate node, wherein a differential in voltage between the source node and the gate node is limited to a desired level.
摘要:
A random intrinsic chip ID generation employs a retention fail signature. A 1st and 2nd ID are generated using testing settings with a 1st setting more restrictive than the 2nd, creating more fails in the 1st ID bit string that includes 2nd ID bit string. A retention pause time controls the number of retention fails, adjusted by a BIST engine, wherein the fail numbers satisfy a predetermined fail target. Verification confirms whether the 1st ID includes the 2nd ID bit string, the ID being the one used for authentication. Authentication is enabled by a 3rd ID with intermediate condition such that 1st ID includes 3rd ID bit string and 3rd ID includes 2nd ID bit string. The intermediate condition includes a guard-band to eliminate bit instability problem near the 1st and 2nd ID boundary. The intermediate condition is changed at each ID read operation, resulting in a more secure identification.
摘要:
An apparatus comprising connecting IDVMON monitors with through silicon vias (TSV) to allow the monitors to be connected to probe pads located on the backside of the wafer. Because the backside of the wafer have significantly more space than the front side, the probe pads for IDVMON can be accommodated without sacrificing the silicon area.
摘要:
A random intrinsic chip ID generation employs a retention fail signature. A 1st and 2nd ID are generated using testing settings with a 1st setting more restrictive than the 2nd, creating more fails in the 1st ID bit string that includes 2nd ID bit string. A retention pause time controls the number of retention fails, adjusted by a BIST engine, wherein the fail numbers satisfy a predetermined fail target. Verification confirms whether the 1st ID includes the 2nd ID bit string, the ID being the one used for authentication. Authentication is enabled by a 3rd ID with intermediate condition such that 1st ID includes 3rd ID bit string and 3rd ID includes 2nd ID bit string. The intermediate condition includes a guard-band to eliminate bit instability problem near the 1st and 2nd ID boundary. The intermediate condition is changed at each ID read operation, resulting in a more secure identification.
摘要:
A method and circuit for implementing precise eFuse resistance measurement, and a design structure on which the subject circuit resides are provided. An eFuse sense amplifier coupled to an eFuse array and used for current measurements includes balanced odd and even bitlines, and a plurality of programmable reference resistors connected to the balanced odd and even bitlines. First a baseline current measurement is made through one of the programmable reference resistors, and used to identify a network baseline resistance. A current measurement is made for an eFuse path including a selected eFuse and used to identify the resistance of the selected eFuse.
摘要:
A method and circuit for implementing eFuse resistance screening, and a design structure on which the subject circuit resides are provided. An eFuse is sensed using a first reference resistor. Responsive to the eFuse being sensed as blown with the first reference resistor, the eFuse is sensed using a second reference resistor having a higher resistance than the first reference resistor. Responsive to the eFuse being sensed as unblown with the second reference resistor, the eFuse is recorded as poorly blown. Reliability concerns are identified quickly and accurately without being required to measure the resistance of the eFuse.
摘要:
A concurrent refresh mode is realized by allowing a memory array to be refreshed by way of a refresh bank select signal, while concurrently enabling a memory access operation in another array. The refresh address management is greatly simplified by the insertion of row address counter integrated within each array. In the preferred embodiment, any combination of a plurality of the memory arrays is refreshed simultaneously while enabling a memory access operation. This concurrent mode also supports a multi-bank operation.