摘要:
A local on-chip programmable pulsewidth and delay generating circuit includes a clock generation circuit configured to receive a global clock signal and output a local clock signal. The clock generation circuit includes a pulse shaping portion which adjusts a pulse width of the global clock signal in accordance with at least one of a trailing edge delay and a leading edge delay. The leading edge delay is generated by a leading edge delay circuit, and the trailing edge delay is generated by a trailing edge delay circuit configured to apply a delay to a trailing edge of a pulse. The trailing edge delay circuit includes a delay chain having programmable stages of delay elements, each stage being independently controlled using control bits decoded from address latches.
摘要:
A D-Cache SRAM cell having a modified design in schematic and layout that exhibits increased symmetry from the circuit schematic and the physical cell layout perspectives. That is, the SRAM cell includes two read ports and minimizes asymmetry by provisioning one read port on a true side and one on the complement side. Asymmetry is additionally minimized in layout as cross coupling on both the true and complement sides rises up one level by providing from the local interconnect level a via connection to a M1 or metallization level. Moreover, the distance between the local interconnect (MC) and the gate conductor structure (PC) has been enlarged and equalized for each of the pFETs in the cross-latched SRAM cell. As a result, the SRAM cell has been rendered insensitive to overlay (local interconnect processing too close) by maximizing this MC-PC distance.
摘要:
The present invention provides a memory architecture that allows memory checking and replacement of defective words by spare elements already provided on the chip that do not increase the chip size. The method of the invention uses a separate redundant array architecture to provide address translation, so that the redundant entries are represented as the correct entry index that they are replacing.
摘要:
A high-reliability content & addressable memory using a shadow content-addressable memory (CAM) array in parallel with a primary CAM array to increase the reliability of CAM searches. The reliability of CAM searches has been less than desired because of random environmental influences that corrupt data. The shadow CAM is written, read, and searched in parallel with the primary CAM. The search results from the parallel searches are compared and, if identical, are declared valid. If the search results are not equal, corrective action is initiated. The high-reliability content-addressable memory may be used with or without priority encoders.
摘要:
A dynamic latch circuit design minimizes set and restore power without sacrificing speed. The dynamic latch circuit provides two significant power saving advantages over traditional dynamic latch designs. The first regulates dynamic restore power with the state of the latch. If the dynamic internal node of the latch has not been discharged, then the restore signal applied to the input of the latch is not transferred to the restore device attached to the node. By isolating the restore device under these conditions, additional power is not wasted boot-strapping up the already precharged node. Second, by design, the restore path and set path are separate. The input signals used to set the latch are different and isolated from those performing the restore. Therefore, there is no conducting path between the voltage source and circuit ground as the restore device turns on.
摘要:
A high-reliability content-addressable memory using a shadow content-addressable memory (CAM) array in parallel with a primary CAM array to increase the reliability of CAM searches. The reliability of CAM searches has been less than desired because of random environmental influences that corrupt data. The shadow CAM is written, read, and searched in parallel with the primary CAM. The search results from the parallel searches are compared and, if identical, are declared valid. If the search results are not equal, corrective action is initiated. The high-reliability content-addressable memory may be used with or without priority encoders.
摘要:
A Dynamic Content Addressable Memory (DCAM) cell topology that contains fewer that can perform a “hidden” refresh of stored data that does not delay nor interrupt the CAM search cycle, thereby providing SCAM-like performance. A non-destructive read operation, is performed such that the stored-data does not have to be written back because of a refresh-read operation. A reliable CAM search can be performed after a read operation and before or even while the refresh-data is being written back. Soft-error detection processes can be performed on each CAM entry during the pendency of the refresh cycle. The DCAM cell can be used in a digital system such as a digital computer and a Network Router.
摘要:
A multi-port memory is provided that includes means for receiving synchronous memory requests, means for receiving asynchronous memory requests, and means for processing the received synchronous and asynchronous memory requests simultaneously. Systems and methods that employ the multi-port memory are also provided.
摘要:
The present invention provides circuitry which facilitates user selection of alternative memory accessing techniques. The present invention provides a design approach or technique to transform the time associated with waiting for a valid "way-select" signal into cycle reduction time, thus providing a beneficial increase in the overall performance of multi-way associative cache and memory designs.