Intel Corporation

Sebastian Winkel

Ethan Schuchman

Tyler N Sondag

Vineeth Mekkat

Jamison D Collins

Oleg Margulis

Jayesh Iyer

Girish Venkatasubramanian

Youfeng Wu

Koichi Yamada

Jason M Agron

Suresh Srinivas

Polychronis Xekalakis

Zhongying Zhang

Mark Dechene

Howard Chen

James E Smith

Gregor Stellpflug

Naveen Kumar

Naveen Neelakantam

H Peter Anvin

Yu Wang

David Pardo Keppel

Igor Yanover

Mohammad R Haghighat

Tianyou Li

Arun Raman

Rolf Kassa

Junyong Ding

Patrick P Lai

Jaroslaw Topp

John Faistl

Rainer Theuer

Rupert Brauch

Ulrich Bretthauer

Rakesh Ranjan

Sumit Ahuja

Michael Neilly

Ori Lempel

Janghaeng Lee

Hou-Jen Ko

Nirajan L Cooray

Rainer Theur

Yongjun Park

Gisle Dankel

Howard H Chen

Systems, apparatuses and methods may provide for developer stage technology that embeds binary code into an application binary file, wherein the binary code corresponds to vector functions and non-vector functions in statically typed source code, and generates intermediate representation (IR) data, wherein the intermediate representation data corresponds to the vector functions in the statically typed source code. Additionally, the developer stage technology embeds the IR data in the application binary file. Moreover, deployment stage technology may generate a first compilation output based on the application binary file and detect a capability change in an execution environment associated with the first compilation output. The deployment stage technology may also generate, in response to the detected capability change, a second compilation output based on the first compilation output.

Multi-stage automatic compilation for vector computations in applications

A processor includes a front end to receive an instruction. The processor also includes a core to execute the instruction. The core includes logic to execute a base function of the instruction to yield a result, generate a predicate value of a comparison of the result based upon a predication setting in the instruction, and set the predicate value in a register. The processor also includes a retirement unit to retire the instruction.

Instruction and logic for predication and implicit destination

Methods and apparatuses relating to assigning a logical thread to a physical thread. In one embodiment, an apparatus includes a data storage device that stores code that when executed by a hardware processor causes the hardware processor to perform the following: translating an instruction into a translated instruction, assigning a logical thread for the translated instruction, and providing a thread map hint for the translated instruction; and a hardware scheduler to assign a physical thread of the hardware processor to execute the logical thread based on the thread map hint.

Apparatuses and methods to assign a logical thread to a physical thread

A processor includes a store buffer to store store instructions to be processed to store data in main memory, a load buffer to store load instructions to be processed to load data from main memory, and a loop invariant code motion (LICM) protection structure coupled to the store buffer and the load buffer. The LPT tracks information to compare an address of a store or snoop microoperation with entries in the LICM and re-loads a load microoperation of a matching entry.

Method and apparatus for supporting speculative memory optimizations

A processor for redundant stores includes a front end including circuitry to decode instructions from an instruction stream, a data cache unit including circuitry to cache data for the processor, a binary translator, and a memory execution unit. The binary translator includes circuitry to identify a first region of the instruction stream including a redundant store, mark a first starting instruction of the first region with a protection designator, mark a first ending instruction of the first region with a clear designator, and store an amended instruction stream with the markings. The memory execution unit includes circuitry to track the first redundant store based on the protection designator and the clear designator to eliminate the first redundant store.

Eliminating redundant stores using a protection designator and a clear designator

In one embodiment, a processor includes logic, responsive to a first instruction, to perform an operation on a first source operand and a second source operand associated with the first instruction and write a result of the operation to a destination location comprising a third source operand. The write may be a partial write of the destination location to maintain an unmodified portion of the third source operand. Other embodiments are described and claimed.

Performing partial register write operations in a processor

Methods and apparatuses relating to a fusion manager to fuse instructions are described. In one embodiment, a hardware processor includes a hardware binary translator to translate an instruction stream into a translated instruction stream, a hardware fusion manager to fuse multiple instructions of the translated instruction stream into a single fused instruction, a hardware decode unit to decode the single fused instruction into a decoded, single fused instruction, and a hardware execution unit to execute the decoded, single fused instruction.

Hardware apparatuses and methods to fuse instructions

In one implementation, a processing device is provided that includes a memory to store instructions and a processor core to execute the instructions. The processor core is to receive a sequence of instructions reordered by a binary translator for execution. A first load of the sequence of instructions is identified. The first load references a memory location that stores a data item to be loaded. An occurrence of a second load is detected. The second load to access the memory location subsequent to an execution of the first load instruction. A protection field in the first load is enabled based on the detected occurrence of the second load. The enabled protection field indicates that the first load is to be checked for an aliasing associated with the memory location with respect to a subsequent store instruction. The second load is eliminated based on the enabled of the protection field.

Supporting binary translation alias detection in an out-of-order processor

A processing device includes a store instruction identification unit to identify a pair of store instructions among a plurality of instructions in an instruction queue. The pair of store instructions include a first store instruction and a second store instruction. The first data of the first store instruction corresponds to a first memory region adjacent to a second memory region, and a second data of the second store instruction corresponds to the second memory region. The processing device to include a store instruction fusion unit to fuse the first store instruction with the second store instruction resulting in a fused store instruction.

Fused adjacent memory stores

A method and apparatus are described for efficient register reclamation. For example, one embodiment of an apparatus comprises: single usage detection and tagging logic to examine a sequence of instructions to detect logical registers used by the sequence of instructions that have a single use and to tag an instruction as a single usage instruction if the instruction is a consumer of a logical register that has a single use; an allocator to allocate processor resources to execute the sequence of instructions, the processor resources including physical registers mapped to logical registers to execute the sequence of instructions; and register reclamation logic to free up a logical to physical mapping of a single use register in response to detecting the tag provided by the instruction tagging logic.

Apparatus and method for efficient register allocation and reclamation

A processor includes a front end and a scheduler. The front end includes circuitry to determine whether to apply an acyclical or cyclical thread assignment scheme to code received at the processor, and to, based upon a determined thread assignment scheme, assign code to a static logical thread and to a rotating logical thread. The scheduler includes circuitry to assign the static logical thread to the same physical thread upon a subsequent control flow execution of the static logical thread, and to assign the rotating logical thread to different physical threads upon different executions of instructions in the rotating logical thread.

Instruction and logic for scheduling instructions

Apparatus and method for detecting and recovering from incorrect memory dependence speculation in an out-of-order processor are described herein. For example, one embodiment of a method comprises: executing a first load instruction; detecting when the first load instruction experiences a bad store-to-load forwarding event during execution; tracking the occurrences of bad store-to-load forwarding event experienced by the first load instruction during execution; controlling enablement of an S-bit in the first load instruction based on the tracked occurrences; generating a plurality of load operations responsive to an enabled S-bit in first load instruction, wherein execution of the plurality of load operations produces a result equivalent to that from the execution of the first load instruction.

Method and apparatus for recovering from bad store-to-load forwarding in an out-of-order processor

A processor and method are described for alias detection. For example, one embodiment of an apparatus comprises: reordering logic to receive a set of read and write operations in a program order and to responsively reorder the read and write operations; adjustment information attachment logic to associate adjustment information with one or more of the set of read and write operations, wherein for a read operation the adjustment information is to indicate a number of write operations which the read operation has bypassed and for a write operation the adjustment information is to indicate a number of read operations which have bypassed the write operation; and out-of-order processing logic to determine whether execution of the reordered read and write operations will result in a conflict based, at least in part, on the adjustment information associated with the one or more reads and writes.

Method and apparatus for memory aliasing detection in an out-of-order instruction execution platform

A system includes a processor with a front end to receive an instruction stream reordered by a software scheduler and including a plurality of memory operations and alias information indicating how a given memory operation may be evaluated. Furthermore, the processor includes a hardware scheduler to reorder, in hardware, the instruction stream for out-of-order execution. In addition, the processor includes a calculation module to determine, for a given memory operation and based upon the alias information, a checking range of memory atoms subsequent to the given memory operation and a virtual order of the memory operation. The virtual order indicates an original ordering of the instructions. The processor also includes an alias unit to reorder the instruction stream, determine whether the hardware reordering caused an error, and determine whether the software reordering caused an error based upon the checking range and the virtual order.

Instruction and logic for memory disambiguation in an out-of-order processor

Growing community of inventors

Los Altos, CA, United States of America

Sebastian Winkel