Analog Devices,inc.

Analog Devices Technology

Adapteva Incorporated

Andreas D Olofsson

Paul J Kettle

Christopher Jacobs

Baruch Yanovitch

A method for generating a digital signal pattern at M outputs involves retrieving an instruction from memory comprising a first set of bits identifying a first group of N outputs that includes fewer than all of the M outputs, and a second set of N bits each corresponding to a respective output included in the identified first group of N outputs. For each of the M outputs that is included in the identified first group of N outputs, the signal at the output is toggled if the one of the N bits corresponding to that output is in a first state and is kept in the same state if the one of the N bits corresponding to that output is in a second state. For each of the M outputs that is not included in the identified first group of N outputs, the signal at that output is kept in the same state.

Data pattern generator with selectable programmable outputs

Multiplication engines and multiplication methods are provided for a digital processor. A multiplication engine includes multipliers, each receiving a first operand and a second operand; a local operand register having locations to hold the first operands for respective multipliers; a first operand bus coupled to the local operand register to supply the first operands from a compute register file to the local operand register; a second operand bus coupled to the plurality of multipliers to supply one or more of the second operands from the compute register file to respective multipliers; and a control unit responsive to a digital processor instruction to supply the first operands from the local operand register to respective multipliers, to supply the second operands from the compute register file to respective multipliers on the second operand bus and to multiply the first operands by the respective second operands in the respective multipliers, wherein one or more of the first operands in the local operand register are reused by the multipliers in two or more multiplication operations.

Bandwidth efficient instruction-driven multiplication engine

A mesh network has a plurality of nodes that can be arranged in a two or three dimensional arrangement. The address-based mesh routing scheme sends a full address and full data on every clock cycle. The system can include broadcasting to columns, rows, planes, or all system elements using a single data transaction. A selectable routing scheme can be random, fixed, or adaptive, depending on user configuration. A registered transaction-weight scheme is provided that can stall transactions in a mesh without having to resend data.

Mesh network

Multiplication engines and multiplication methods are provided. A multiplication engine for a digital processor includes a first multiplier to generate unequally weighted partial products from input operands in a first multiplier mode; a second multiplier to generate equally weighted partial products from input operands in a second multiplier mode; a multiplexer to select the unequally weighted partial products in the first multiplier mode and to select the equally weighted partial products in the second multiplier mode; and a carry save adder array configured to combine the selected partial products in the first multiplier mode and in the second multiplier mode.

Multi-format multiplier unit

A first output count is compared with first and second stored count values for generating an output event at a first node if the first Output count corresponds to the first or second stored count values.

Software programmable timing architecture

An apparatus for generating a digital signal pattern may comprises a memory, a program sequencer, first and second circuits, and an event execution unit. The memory may have stored therein a plurality of instructions that, when executed, cause a digital signal pattern to be generated on a plurality of nodes. The program sequencer may be configured to control a sequence in which the plurality of instructions are retrieved from the memory and executed. The first circuit may sequentially step through a plurality of different output states in response to a clock signal. The second circuit may identify an output event when an output state of the first circuit corresponds to an output state identified by retrieved instructions of a particular type. The event execution unit may control states of signals on the plurality of nodes in a manner specified by the retrieved instructions of the particular type in response to the second circuit identifying an output event.

An integrated circuit includes a substrate, a storage device formed in the substrate to hold bias settings, and operational blocks formed in the substrate, each operational block including an operational circuit and a charge pump to provide well bias voltages to the operational circuit in response to one or more of the bias settings. A method for testing an integrated circuit having two or more operational blocks includes: (a) determining a maximum operating speed of each of the blocks at a minimum supply voltage; (b) selecting a block that has a slow operating speed; (c) selecting a well bias to speed up the selected block; (d) selecting a supply voltage to meet a target operating frequency at the selected well bias and measuring power; (e) repeating acts (b)-(d) while the measured power is less than a baseline power; and (f) saving the selected well bias and supply voltage settings for operation of the integrated circuit.

Growing community of inventors

Lexington, MA, United States of America

Andreas D Olofsson