Abstract: Array based memory abstraction in a multiprocessor computing system is disclosed. A plurality of memory resources are operably connected to an interconnect fabric. In a plurality of memory blocks, each memory block represents a contiguous portion of the plurality of memory resources. A cell is operably connected to the interconnect fabric. The cell has an agent with a fabric abstraction block, and the fabric abstraction block includes a block table having an entry for each of the plurality of memory blocks. A memory controller is associated with the agent, is operably connected to the interconnect fabric, and is configured to control a portion of the plurality of memory blocks.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a high reliability triple redundant latch with integrated testability. Three settable memory elements set an identical logical value into each settable memory element. After the settable memory elements are set, three voting structures with inputs from the first, second, and third settable memory elements, determine the logical value held on each of the settable memory elements. Data may be scanned into and out of the second settable memory element. Data is propagated through the buffer into the third settable memory element. The third settable memory element may be used to scan data out of the triple redundant latch. The propagation delay through a latch is the only propagation delay of the triple redundant latch.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a high reliability triple redundant latch with integrated testability. Three settable memory elements set an identical logical value into each settable memory element. After the settable memory elements are set, a voting structure with inputs from the second settable memory element, the third settable memory element and control to the settable memory elements determine the logical value held on the first settable memory element. Data may be scanned into and out of the second settable memory element. Data is propagated through the buffer into the third settable memory element. The third settable memory element may be used to scan data out of the triple redundant latch. The propagation delay through a latch is the only propagation delay of the triple redundant latch.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a high reliability triple redundant latch. Three settable memory elements set an identical logical value into each settable memory element. After the settable memory elements are set, three voting structures with inputs from the first, second, and third settable memory elements and control to the settable memory elements determine the logical values held on the settable memory elements. The propagation delay through a latch is the only propagation delay of the triple redundant latch.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a smaller and faster triple redundant latch. Three settable memory elements set an identical logical value into each settable memory element. After the settable memory elements are set, a voting structure with inputs from the first settable memory element, the second memory element, and control to the settable memory elements determines the logical value held on the third settable memory element. The propagation delay through the third settable memory element is the only propagation delay of the triple redundant latch.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a smaller and faster triple redundant latch. Two settable memory elements, and a voting structure/settable memory element set an identical logical value into each settable memory element, and the voting structure/settable memory element. After the settable memory elements, and the voting structure/settable memory element are set, the voting structure/settable memory element with inputs from the first settable memory element, the second memory element, and control to the settable memory elements determines the logical value held on the voting structure/settable memory element. The propagation delay through the voting structure/settable memory element is the only propagation delay of the triple redundant latch.

Abstract: In a preferred embodiment, the invention provides a circuit and method for a smaller and faster triple redundant latch with storage node recovery. An input driver is connected to the input of three transfer gates. The output of each transfer gate is connected to a separate output of one of three feedback inverters. The transfer gates are controlled by two control inputs. The inputs of the three feedback inverters are connected the output of the forward inverter/majority voter. The output from each of the three feedback inverters are inputs to the forward inverter/majority voter. The output of the forward inverter/majority voter is connected to the input of the output driver. The output of the output driver is the output of the triple redundant latch.

Abstract: A method of using a fet level simulator to check for races in a digital design. The method comprises varying digital design models input into the simulator. Each model comprises a clock gater circuit producing clocks with differing overlaps and dead times. Raw data files corresponding to each model input are generated by the fet level simulator. The raw data files preferably comprise lists of node values with corresponding time stamps. Corresponding latch node values in the raw data files are compared to identify the nodes of a circuit which are affected by races. Identifying affected latch nodes allows a race's root cause to be quickly pinpointed. Vector inputs to the fet level simulator may be varied. If vector inputs are varied, comparison of the raw data files comprises comparing the files generated for differing models with common vector inputs. Apparatus for implementing the above method is also disclosed.

Abstract: A fast nullify system and method facilitate handling of nullification dependencies in a processor that executes instructions out of order. Instructions are forwarded from an instruction fetch mechanism to a reordering mechanism, where the instructions are permitted to execute out of order. After execution of an instruction by an execution unit, instructions are retired by a retire mechanism, which transforms the results of instruction execution to the architecture state. While instructions are executed in the reordering mechanism nullifying instructions and dependent instructions that can potentially be nullified by the nullifying instructions are identified. For each dependent instruction, a determination is made as to whether the dependent instruction qualifies for a fast nullify procedure in that the dependent instruction has less operands than a number that can be read by the execution unit.