Data processing method and system based on pipeline
AssigneeInternational Business Machines Corporation
InventorsXiao Tao Chang, Qiang Liui
Life of this patent
Estimating the term…
Abstract
A data processing system and method are disclosed. The system comprises an instruction-fetch stage where an instruction is fetched and a specific instruction is input into decode stage; a decode stage where said specific instruction indicates that contents of a register in a register file are used as an index, and then, the register file pointed to by said index is accessed based on said index; an execution stage where an access result of said decode stage is received, and computations are implemented according to the access result of the decode stage.
Claims6
A data processing method based on pipeline, comprising the following steps:
- fetching an instruction at an instruction-fetch stage, and inputting a specific instruction into a decode stage;
- at the decode stage, said specific instruction indicating that contents of a register in a first register file are used as an index;
- and accessing a second register file pointed to by said index based on said index;
- and
- at an execution stage, receiving an access result of said decode stage, and implementing computations according to the access result of the decode stage;
- wherein the first register file and the second register file have identical contents at the decode stage;
- and when said first register file and said second register file receive said specific instruction at the decode stage, said first register file is accessed and the contents of the register in said first register file are used as said index pointing to said second register file, and said second register file is accessed with said index.
A data processing system based on pipeline, comprising:
- an instruction-fetch stage where an instruction is fetched and a specific instruction is input into decode stage;
- a decode stage where said specific instruction indicates that contents of a register in a first register file are used as an index, and a second register file pointed to by said index is accessed based on said index;
- an execution stage where an access result of said decode stage is received, and computations are implemented according to the access result of the decode stage;
- wherein the first register file and the second register file have identical contents at the decode stage;
- and when said first register file and said second register file receive said specific instruction at the decode stage, said first register file is accessed and the contents of the register in said first register file are used as said index pointing to said second register file, and said second register file is accessed with said index.
A data processing method based on pipeline, comprising the following steps:
- fetching an instruction at an instruction-fetch stage, and inputting a specific instruction into a decode stage;
- at the decode stage, said specific instruction indicating that contents of a register in a first register file are used as an index;
- and accessing a second register file pointed to by said index based on said index;
- and
- at an execution stage, receiving an access result of said decode stage, and implementing computations according to the access result of the decode stage;
- wherein the decode stage comprises a pre-decode phase and a general-purpose decode phase, wherein at the pre-decode phase, the register in the first register file as indicated in the specific instruction is accessed and the access result is used as said index;
- and at the general-purpose decode phase, the second register file pointed to by said index is accessed with the use of the index acquired at the pre-decode phase;
- and
- wherein the decode phase further comprises a selecting operation for supporting two operation modes of the method, when the method operates in a traditional mode, the access result acquired at the pre-decode phase is discarded and said specific instruction is selected by the selecting operation to enter into the general-purpose phase as said index to access said second register file at the general-purpose phase, and when the method operates in an extension mode, the access result acquired at the pre-decode phase is selected by the selecting operation to enter into the general-purpose phase as said index to access said second register file at the general-purpose phase.
A data processing system based on pipeline, comprising:
- an instruction-fetch stage where an instruction is fetched and a specific instruction is input into decode stage;
- a decode stage where said specific instruction indicates that contents of a register in a first register file are used as an index, and a second register file pointed to by said index is accessed based on said index;
- an execution stage where an access result of said decode stage is received, and computations are implemented according to the access result of the decode stage;
- wherein the decode stage comprises a pre-decode phase and a general-purpose decode phase, wherein at the pre-decode phase, the register in the first register file as indicated in the specific instruction is accessed and the access result is used as said index;
- and at the general-purpose decode phase, the second register file pointed to by said index is accessed with the use of the index acquired at the pre-decode phase;
- and
- wherein the system is operable in two operating modes and the decode phase further comprises a selecting operation, when the system operates in a traditional mode, the access result acquired at the pre-decode phase is discarded and said specific instruction is selected by the selecting operation to enter into the general-purpose phase as said index to access said second register file at the general-purpose phase, and when the system operates in an extension mode, the access result acquired at the pre-decode phase is selected by the selecting operation to enter into the general-purpose phase as said index to access said second register file at the general-purpose phase.
Dependencies are inferred from the claim text.