International Business Machines Corporation

Dhivya Jeganathan

Dung Q Nguyen

Sundeep Chadha

David R Terry

Joshua W Bowman

Brian William Thompto

Michael Joseph Genden

Jose Angel Paredes

Cliff Kucharski

Jeffrey Carl Brownscheidle

Salim Ahmed Shah

Eula A Tolentino

Susan Elizabeth Eisen

Salma Ayub

Brian David Barrick

Steven J Battle

Kenneth Lundy Ward

Maureen Anne Delaney

Josh Bowman

Guy L Guthrie

Bryan J Lloyd

Christian Zoellin

Kurt Alan Feiste

Tu-An T Nguyen

Christopher M Mueller

Glenn Owen Kincaid

Sam G Chu

Luke Murray

Kent Li

A first instruction for processing by a processor core is received. Whether the instruction is a larx is determined. Responsive to determining the instruction is a larx, whether a cacheline associated with the larx is locked is determined. Responsive to determining the cacheline associated with the larx is not locked, the cacheline associated with the larx is locked and a counter associated with a first thread of the processor core is started. The first thread is processing the first instruction.

Prioritization of threads in a simultaneous multithreading processor core

In at least one embodiment, a processor includes architected and non-architected register files for buffering operands. The processor additionally includes an instruction fetch unit that fetches instructions to be executed and at least one execution unit. The at least one execution unit is configured to execute a first class of instructions that access operands in the architected register file and a second class of instructions that access operands in the non-architected register file. The processor also includes a mapper circuit that assigns physical registers to the instructions for buffering of operands. The processor additionally includes a dispatch circuit configured, based on detection of an instruction in one of the first and second classes of instructions for which correct operands do not reside in a respective one of the architected and non-architected register files, to automatically initiate transfer of operands between the architected and non-architected register files.

Processor providing intelligent management of values buffered in overlaid architected and non-architected register files

A computer system includes a dispatch stage configured to dispatch a plurality of instructions in a program order, and an issue stage configured to issue at least one instruction among the plurality of instructions. The computer system further includes an execution stage configured to execute the at least one instruction to generate a finish report and to determine the at least one instruction is one of an exception-free instruction or an exception instruction. In response to determining the exception-free instruction, a first finish report associated with the exception-free instruction is output to a completion stage. In response to determining the exception instruction, a second finish report associated with the exception instruction is output to an exception unit so as to halt output of the second finish report to the completion stage.

Finish exception handling of an instruction completion table

Embodiments herein describe a reservation station (RS) in a processor that merges control data from multiple sources into a merged control data value. Before an instruction issues, the RS gathers and saves control data indicating how the instruction is to be executed. This control data may be saved in control registers. An instruction, however, can update many different types of status control bits in these registers. As such, the RS may store different types of control data for an instruction. Instead of the RS containing multiple registers and data paths for every type of control data, the embodiments herein describe merge logic in the RS that permits control data from different sources to be merged into a single control data value. Once the instruction is issued, the RS passes the merged control data value to an execution unit for processing.

Merging status and control data in a reservation station

Techniques disclosed herein describe a variable latency pipe for interleaving instruction tags in a processor. According to one embodiment presented herein, an instruction tag is associated with an instruction upon issue of the instruction from the issue queue. One of a plurality of positions in the latency pipe is determined. The pipe stores one or more instruction tags, each associated with a respective instruction. The pipe also stores the instruction tags in a respective position based on the latency of each respective instruction. The instruction tag is stored at the determined position in the pipe.

Variable latency pipe for interleaving instruction tags in a microprocessor

Embodiments include systems, methods, and computer program products for on-demand error detection and correction of registers in a processor. One method includes detecting, before a first instruction is dispatched to an issue queue in the processor, an error in data, associated with the first instruction, stored in an entry of a register file in the processor. The method also includes, after detecting the error, halting the dispatch of the first instruction to the issue queue, and determining whether the entry of the register file has completed. The method further includes determining whether to perform error correction on the register file based on the determination of whether the entry of the register file has completed.

On-demand GPR ECC error detection and scrubbing for a multi-slice microprocessor

Methods and apparatus for transmitting data between execution slices of a multi-slice processor including receiving, by an execution slice, a broadcast message comprising an instruction tag (ITAG) for a producer instruction, a latency, and a source identifier; determining that an issue queue in the execution slice comprises an ITAG for a consumer instruction, wherein the consumer instruction depends on result data from the producer instruction; calculating a cycle countdown using the latency and the source identifier; determining that the cycle countdown has expired; and in response to determining that the cycle countdown has expired, reading the result data from the producer instruction.

Broadcasting messages between execution slices for issued instructions indicating when execution results are ready

Embodiments described herein include a computing system that permits partial writes into a memory element&#x2014;e.g., a register on a processor. For example, the data to be written into the memory element may be spread across multiple sources. The register may receive data from two different sources at different times and perform two separate partial write commands to store the data. Embodiments herein generate an ECC value for each of the partial writes. That is, when storing the data of the first partial write, the computing system generates a first ECC value for the data in the first partial write and stores this value in the memory element. Later, when performing the second partial write, the computing system generates a second ECC value for this data which is also stored in the memory element.

Partial ECC mechanism for a byte-write capable register

Operation of a multi-slice processor that includes execution slices and a dispatch network of the multi-slice processor implementing a hardware level mechanism to overcome a system hang. Such a multi-slice processor includes a plurality of execution slices and a dispatch network of the multi-slice processor implementing a hardware level mechanism to overcome a system hang. Operation of such a multi-slice processor includes, storing, in one or more logical units of a plurality of logical units of an age array, a logical value representing a relative age between instructions; propagating, in response to a current instruction being in a hang state, a hang signal to the plurality of logical units of the age array; in response to the hang signal, generating, from the plurality of logical units, a plurality of logical output values indicating a next instruction ready for execution; and issuing the next instruction for execution.

Operation of a multi-slice processor implementing a mechanism to overcome a system hang

Generating ECC values for byte-write capable registers

Operation of a multi-slice processor that includes execution slices implementing dynamic switching of instruction issuance order. Such a multi-slice processor includes determining a current issuance order for a plurality of instructions and a change in an operating condition of the multi-slice processor; responsive to determining the change in the operating condition, determining an alternate issuance order for the plurality of instructions; and responsive to determining the alternate issuance order, switching from the current issuance order for the plurality of instructions to the alternate issuance order for the plurality of instructions.

Operation of a multi-slice processor implementing dynamic switching of instruction issuance order

An approach is provided is provided in which a computing system matches a writeback instruction tag (ITAG) to an entry instruction tag (ITAG) included in an issue queue entry. The writeback ITAG is provided by a first of multiple load store units. The issue queue entry includes multiple ready bits, each of which corresponds to one of the multiple load store units. In response to matching the writeback ITAG to the entry ITAG, the computer system sets a first ready bit corresponding to the first load store unit. In turn, the computing system issues an instruction corresponding to the entry ITAG based upon detecting that each of the multiple ready bits is set.

Speculative load data in byte-write capable register file and history buffer for a multi-slice microprocessor

Operation of a multi-slice processor that includes a plurality of execution slices and a plurality of load/store slices coupled via a results bus includes: retrieving, from the results bus into an entry of a register file of an execution slice, speculative result data of a load instruction generated by a load/store slice; and determining, from the load/store slice after expiration of a predetermined period of time, whether the result data is valid.

Growing community of inventors

Austin, TX, United States of America

Dhivya Jeganathan