Intel Corporation

Sebastien Hily

Per H Hammarlund

Glenn J Hinton

Ronak Singhal

Stephan Jourdan

Haitham H Akkary

Deborah T Marr

John G Holm

Steven E Raasch

Matthew C Merten

Shubhendu Sekhar Mukherjee

Raanan Sade

Bratin Saha

David A Koufaty

Stanislav Shwartsman

Michael A Fetterman

Shlomo Raikin

Beeman C Strong

Mark C Davis

Alexandre J Farcy

Adrian C Moga

Arijit Biswas

David W Burns

Christopher J Cormack

Hermann Gartler

Zhongying Zhang

Morris Marden

Aravindh Baktha

Ron Shalev

Prakash Math

Matthew C Curfman

Melih Ozgul

Apparatuses, methods and storage medium associated with embedded computing, are disclosed herein. In embodiments, an embedded computing platform includes a plurality of system-on-chips (SoCs) forming a local compute cluster; and an orchestrator disposed on one of the SoCs arranged to orchestrate fail-safe operations, in response to a reported unrecoverable failure requiring shut down or partial disabling of one of the SoCs, to consolidate execution of critical workloads on one or more of remaining fully or partially operational ones of the SoCs. Other embodiments are also described and claimed.

Method and apparatus to provide an improved fail-safe system for critical and non-critical workloads of a computer-assisted or autonomous driving vehicle

In one embodiment, the present invention includes a method for controlling redundant execution such that if an exceptional event occurs, the redundant execution is stopped, non-redundant execution is performed in one of the threads until the exceptional event has been-resolved, after which a state of the threads is synchronized, and redundant execution is continued. Other embodiments are described and claimed.

Controlling non-redundant execution in a redundant multithreading (RMT) processor

In accordance with embodiments disclosed herein, there are provided methods, systems, mechanisms, techniques, and apparatuses for cutting senior store latency using store prefetching. For example, in one embodiment, such means may include an integrated circuit or an out of order processor means that processes out of order instructions and enforces in-order requirements for a cache. Such an integrated circuit or out of order processor means further includes means for receiving a store instruction; means for performing address generation and translation for the store instruction to calculate a physical address of the memory to be accessed by the store instruction; and means for executing a pre-fetch for a cache line based on the store instruction and the calculated physical address before the store instruction retires.

Method and apparatus for cutting senior store latency using store prefetching

In one embodiment, the present invention includes a method for providing a cache block in an exclusive state to a first cache and providing the same cache block in the exclusive state to a second cache when cores accessing the two caches are executing redundant threads. Other embodiments are described and claimed.

Obtaining data for redundant multithreading (RMT) execution

A redundant multithreading processor is presented. In one embodiment, the processor performs execution of a thread and its duplicate thread in parallel and determines, when in a redundant multithreading mode, whether or not to synchronize an operation of the thread and an operation of the duplicate thread.

Redundant multithreading processor

A device is presented including a first processor and a second processor. A number of memory devices are connected to the first processor and the second processor. A register buffer is connected to the first processor and the second processor. A trace buffer is connected to the first processor and the second processor. A number of memory instruction buffers are connected to the first processor and the second processor. The first processor and the second processor perform single threaded applications using multithreading resources. A method is also presented where a first thread is executed from a first processor. The first thread is also executed from a second processor as directed by the first processor. The second processor executes instructions ahead of the first processor.

Avoiding execution of instructions in a second processor by committing results obtained from speculative execution of the instructions in a first processor

Embodiments of the present invention relate to a memory management scheme and apparatus that enables efficient memory renaming. The method includes computing a store address, writing the store address in a first storage, writing data associated with the store address to a memory, and de-allocating the store address from the first storage, allocating the store address in a second storage, predicting a load instruction to be memory renamed, computing a load store source index, computing a load address, disambiguating the memory renamed load instruction, and retiring the memory renamed load instruction, if the store instruction is still allocated in at least one of the first storage and the second storage and should have effectively provided to the load the full data. The method may also include re-executing the load instruction without memory renaming, if the store instruction is not in at the first storage or in the second storage.

Method for and a trailing store buffer for use in memory renaming

Methods and apparatus for resolving false dependencies associated with speculatively executing load instructions in a processor core are described. In one embodiment, physical addresses of a load operation and a store operation are compared in response to a determination that the load operation may be potentially dependent on the store operation. Other embodiments are also described.

Resolving false dependencies of speculative load instructions

In one embodiment, the present invention includes an apparatus having a first counter to count dispatches of a senior request in a memory unit, a second counter to count cycles of a processor coupled to the memory unit, and a controller coupled to the first and second counters to execute one or more one remediation measures with respect to the senior request based on a value of at least one of the counters. Other embodiments are described and claimed.

Detecting and resolving locks in a memory unit

In one embodiment, the present invention includes a predictor to predict contention of an operation to be executed in a program. The operation may be processed based on a result of the prediction, which may be based on multiple independent predictions. In one embodiment, the operation may be optimized if no contention is predicted. Other embodiments are described and claimed.

Predicting contention in a processor

Microarchitecture policies and structures partition execution resource clusters. In disclosed microarchitecture embodiments, micro-operations representing a sequential instruction ordering are partitioned into a two sets. To one set of micro-operations execution resources are allocated from a cluster of execution resources that can perform memory access operations but not branching operations. To the other set of micro-operations execution resources are allocated from a cluster of execution resources that can perform branching operations but not memory access operations. The first and second sets of micro-operations may be executed out of sequential order but are retired to represent their sequential instruction ordering.

Method and apparatus for microarchitecture partitioning of execution clusters

A method is disclosed. The method includes scheduling a load operation at least twice the size of a maximum access supported by a memory device, dividing the load operation into a plurality of separate load operation segments having a size equivalent to the maximum access supported by the memory device, and performing each of the plurality of load operation segments. A further method is disclosed where a temporary register is used to minimize the number of memory accesses to support unaligned accesses.

Load mechanism

A memory disambiguation apparatus includes a store queue, a store forwarding buffer, and a version count buffer. The store queue includes an entry for each store instruction in the instruction window of a processor. Some store queue entries include resolved store addresses, and some do not. The store forwarding buffer is a set-associative buffer that has entries allocated for store instructions as store addresses are resolved. Each entry in the store forwarding buffer is allocated into a set determined in part by a subset of the store address. When the set in the store forwarding buffer is full, an older entry in the set is discarded in favor of the newly allocated entry. A version count buffer including an array of overflow indicators is maintained to track overflow occurrences. As load addresses are resolved for load instructions in the instruction window, the set-associative store forwarding buffer can be searched to provide memory disambiguation.

Memory disambiguation for large instruction windows

Embodiments of the present invention provide a method, apparatus and system for memory renaming. In one embodiment, a decode unit may decode a load instruction. If the load instruction is predicted to be memory renamed, the load instruction may have a predicted store identifier associated with the load instruction. The decode unit may transform the load instruction that is predicted to be memory renamed into a data move instruction and a load check instruction. The data move instruction may read data from the cache based on the predicted store identifier and load check instruction may compare an identifier associated with an identified source store with the predicted store identifier. A retirement unit may retire the load instruction if the predicted store identifier matches an identifier associated with the identified source store. In another embodiment of the present invention, the processor may re-execute the load instruction without memory renaming if the predicted store identifier does not match the identifier associated with the identified source store.

Method and system for transforming memory location references in instructions

Embodiments of the present invention relate to a memory management scheme and apparatus that enables efficient cache memory management. The method includes writing an entry to a store buffer at execute time; determining if the entry's address is in a first-level cache associated with the store buffer before retirement; and setting a status bit associated with the entry in said store buffer, if the address is in the cache in either exclusive or modified state. The method further includes immediately writing the entry to the first-level cache at or after retirement when the status bit is set; and de-allocating the entry from said store buffer at retirement. The method further may comprise resetting the status bit if the cacheline is allocated over or is evicted from the cache before the store buffer entry attempts to write to the cache.

Growing community of inventors

Hillsboro, OR, United States of America

Sebastien Hily