Data processing apparatus and method for reducing issue circuitry responsibility by using a predetermined pipeline stage to schedule a next operation in a sequence of operations defined by a complex instruction

A data processing apparatus and method are provided for executing complex instructions. The data processing apparatus executes instructions defining operations to be performed by the data processing apparatus, those instructions including at least one complex instruction defining a sequence of opera...

Full description

Saved in:
Bibliographic Details
Main Authors Orion, Luc, Airaud, Cédric Denis Robert, Alvarez-Heredia, Boris Sira
Format Patent
LanguageEnglish
Published 19.07.2011
Online AccessGet full text

Cover

Loading…
More Information
Summary:A data processing apparatus and method are provided for executing complex instructions. The data processing apparatus executes instructions defining operations to be performed by the data processing apparatus, those instructions including at least one complex instruction defining a sequence of operations to be performed. The data processing apparatus comprises a plurality of execution pipelines, each execution pipeline having a plurality of pipeline stages and arranged to perform at least one associated operation. Issue circuitry interfaces with the plurality of execution pipelines and is used to schedule performance of the operations defined by the instructions. For the at least one complex instruction, the issue circuitry is arranged to schedule a first operation in the sequence, and to issue control signals to one of the execution pipelines with which that first operation is associated, those control signals including an indication of each additional operation in the sequence. Then, when performance of the first operation reaches a predetermined pipeline stage in that execution pipeline, that predetermined pipeline stage is arranged to schedule a next operation in the sequence, and to issue additional control signals to a further one of the execution pipelines with which that next operation is associated in order to cause that next operation to be performed. This has been found to provide a particularly efficient mechanism for handling the execution of complex instructions without the need to provide dedicated execution pipelines for those complex instructions, and without an increase in complexity of the issue circuitry.