Intel Corporation

Gavri Berger

Igor Yanover

Zeev Sperber

Alexander Gendler

Sagi Meller

Amit Gradstein

Simon Rubanovich

Jose Yallouz

Lev Makovsky

Ron Schneider

Edo Hachamo

Saeed Kharouf

Elkana Korem

Haim Granot

Eyal Hadas

Hanan Shomroni

Daniela Kaufman

Robert Valentine

Efraim Rotem

Raanan Sade

Ron Gabor

Nir Rosenzweig

Stanislav Shwartsman

Ian M Steiner

Nikhil Gupta

Gabi Malka

Krishnamurthy Jambur Sathyanarayana

Shmuel Zobel

Omer Vikinski

Tomer Weiner

Sumesh Subramanian

Systems, methods, and apparatuses relating to a matrix operations accelerator are described. In one embodiment, a processor includes a matrix operations accelerator circuit that includes a two-dimensional grid of fused multiply accumulate circuits that is switchable to a scheduling mode for execution of a decoded single instruction where the matrix operations accelerator circuit loads a first buffer of the two-dimensional grid of fused multiply accumulate circuits from a first plurality of registers that represents a first input two-dimensional matrix, checks if a second buffer of the two-dimensional grid of fused multiply accumulate circuits stores an immediately prior input two-dimension matrix that is the same as a second input two-dimensional matrix from a second plurality of registers that represents the first input two-dimensional matrix, and when the second buffer of the two-dimensional grid of fused multiply accumulate circuits stores the immediately prior input two-dimension matrix, from execution of a previous instruction, that is the same as the second input two-dimensional matrix: prevents reclamation of the second buffer between execution of the previous instruction and the decoded single instruction, performs an operation on the first input two-dimensional matrix from the first buffer and the immediately prior input two-dimension matrix from the second buffer to produce a resultant, and stores the resultant in resultant storage, and when the second buffer of the two-dimensional grid of fused multiply accumulate circuits does not store the immediately prior input two-dimension matrix, from execution of the previous instruction, that is the same as the second input two-dimensional matrix: loads the second input two-dimensional matrix into the second buffer of the two-dimensional grid of fused multiply accumulate circuits, performs the operation on the first input two-dimensional matrix from the first buffer and the second input two-dimension matrix from the second buffer to produce a resultant, and stores the resultant in the resultant storage.

Apparatuses, methods, and systems for instructions of a matrix operations accelerator

An apparatus and method for intelligent power virus protection in a processor. For example, one embodiment of a processor comprises: first circuitry including an instruction fetch circuit to fetch instructions, each instruction comprising an instruction type and an associated width comprising a number of bits associated with source and/or destination operand values associated with the instruction; detection circuitry to detect one or more instructions of a particular type and/or width; evaluation circuitry to evaluate an impact of power virus protection (PVP) circuitry when executing the one or more instructions based on the detected instruction types and/or widths; and control circuitry, based on the evaluation, to configure the PVP circuitry in accordance with the evaluation performed by the evaluation circuitry.

Apparatus and method for power virus protection in a processor

In one embodiment, a processor includes a current protection controller to: receive instruction width information and instruction type information associated with one or more instructions stored in an instruction queue prior to execution of the one or more instructions by an execution circuit; determine a power license level for the core based on the corresponding instruction width information and the instruction type information; generate a request for a license for the core corresponding to the power license level; and communicate the request to a power controller when the one or more instructions are non-speculative, and defer communication of the request when at least one of the one or more instructions is speculative. Other embodiments are described and claimed.

System, apparatus and method for power license control of a processor

In an embodiment, a processor includes processing cores to execute instructions; and throttling logic. The throttling logic is to: determine an average capacitance score for execution events in a sliding window; perform frequency throttling when the average capacitance score exceeds a throttling threshold; determine a count of frequency throttling instances; and in response to a determination that the count of frequency throttling instances exceeds a maximum throttling value, increase the throttling threshold and concurrently reduce a baseline frequency. Other embodiments are described and claimed.

Adjusting a throttling threshold in a processor

The present disclosure is directed to systems and methods for performing one or more operations on a two dimensional tile register using an accelerator that includes a tiled matrix multiplication unit (TMU). The processor circuitry includes reservation station (RS) circuitry to communicatively couple the processor circuitry to the TMU. The RS circuitry coordinates the operations performed by the TMU. TMU dispatch queue (TDQ) circuitry in the TMU maintains the operations received from the RS circuitry in the order that the operations are received from the RS circuitry. Since the duration of each operation is not known prior to execution by the TMU, the RS circuitry maintains shadow dispatch queue (RS-TDQ) circuitry that mirrors the operations in the TDQ circuitry. Communication between the RS circuitryand the TMU provides the RS circuitry with notification of successfully executed operations and allows the RS circuitry to cancel operations where the operations are associated with branch mispredictions and/or non-retired speculatively executed instructions.

Accelerator systems and methods for matrix operations

A processor and method are described for performing proactive throttling of execution unit ports. For example, one embodiment of a processor core comprises: a plurality of execution unit ports within an execution stage of the processor core; a scheduler unit to schedule execution of a plurality of operations to the plurality of execution unit ports; and proactive throttling logic to limit acceleration of execution of the operations by the ports to an acceleration level which does not result in significant power supply droops.

Method and apparatus for proactive throttling for improved power transitions in a processor core

In an embodiment, a processor includes at least a first core. The first core includes execution logic to execute operations, and a first event counter to determine a first event count associated with events of a first type that have occurred since a start of a first defined interval. The first core also includes a second event counter to determine a second event count associated with events of a second type that have occurred since the start of the first defined interval, and stall logic to stall execution of operations including at least first operations associated with events of the first type, until the first defined interval is expired responsive to the first event count exceeding a first combination threshold concurrently with the second event count exceeding a second combination threshold. Other embodiments are described and claimed.

Method and apparatus to protect a processor against excessive power usage

Growing community of inventors

Haifa, Israel

Gavri Berger