Hitachi, Ltd.

Chih-Jui Peng

Margaret Rose Gearty

Naohiko Irie

Sivaram Krishnan

Lew Chua-Eoan

Tony Lee Werner

Glenn Ashley Farrall

Jackie Andrew Freeman

Sebastian Haviuj Ziesler

A method for forwarding data within a pipeline of a pipelined data processor having a plurality of execution pipeline stages where each stage accepts a plurality of operand inputs and generates a result. The result generated by each execution pipeline stage is selectively coupled to an operand input of one of the execution pipeline stages.

Mechanism for forward data in a processor pipeline using a single pipefile connected to the pipeline

In a computer system having a central processing unit (CPU) execution pipeline and a floating point unit (FPU) execution pipeline, the CPU execution pipeline including a CPU decoder pipestage and the FPU execution pipeline including an FPU decoder pipestage, the method including the steps of, (a) sending a first instruction to the CPU decoder pipestage, (b) sending the first instruction to the FPU decoder pipestage, (c) generating a signal indicating that the first instruction has been accepted by the CPU decoder pipestage, (d) generating a signal indicating that the first instruction has been accepted by the FPU decoder pipestage, (e) sending a second instruction to the CPU decoder pipestage in response to step (d), and (f) sending a second instruction to the FPU decoder pipestage in response to step (c). A corresponding apparatus is also provided.

Microcomputer/floating point processor interface and method

A computer system having a central processing unit (CPU) execution pipeline and a floating point unit (FPU) execution pipeline, the CPU pipeline including a plurality of pipestages and the FPU pipeline including a plurality of pipestages, wherein each CPU pipestage in the CPU pipeline has a corresponding pipestage in the FPU pipeline, a method of synchronizing operation of the CPU pipeline and the FPU pipeline, the method including the steps of (a) receiving an instruction in a first CPU pipestage, (b) receiving the instruction in a corresponding first FPU pipestage, (c) processing the instruction in the first CPU pipestage, (d) processing the instruction in the first FPU pipestage, (e) generating, by the first CPU pipestage, a first signal indicating that the instruction has been processed by first CPU pipestage and is ready to proceed to a second pipestage in the CPU pipeline, (f) generating by the first FPU pipestage, a second signal indicating that the instruction has been processed by the first FPU pipestage and is ready to proceed to a second pipestage in the FPU pipeline, (g) sending the instruction from the first CPU pipestage to the second pipestage in the CPU pipeline, (h) sending the instruction from the first FPU pipestage to the second pipestage in the FPU pipeline, (i) wherein the second pipestage in the CPU pipeline responds to the second signal to send the instruction to a third pipestage in the CPU pipeline, and (j) wherein the second pipestage in the FPU pipeline responds to the first signal to send the instruction to a third pipestage in the FPU pipeline. A corresponding apparatus is also provided.

Synchronized instruction advancement through CPU and FPU pipelines

According to the present invention, techniques for setting selected operand fields in pipelined architectures are provided. Methods and systems for efficiently selecting operand fields according to the present invention can be operative on a variety of computer architectures, including RISC architectures.

Method and system for selecting and using source operands in computer system instructions

A processor element, structured to execute a 32-bit fixed length instruction set architecture, is backward compatible for executing a 16-bit fixed length instruction set architecture by translating each of the 16-bit instructions into a sequence of one or more 32-bit instructions. The 32-bit instruction set architecture includes &ldquo;prepare to branch&rdquo; instructions that allow target addresses for branch instructions to be set up in advance of the branch. The 32-bit prepare to branch and branch instructions are combined to execute a 16-bit branch instruction coupled with a 16-bit Delay Slot instruction.

Emulating execution of smaller fixed-length branch/delay slot instructions with a sequence of larger fixed-length instructions

A method for low latency access to the control space. A pipeline processor executes instructions in multiple stages including a decode stage, one or more execution, stages, and a writeback stage. A control space access instruction includes a first field containing a control register specifier and a second field containing a general purpose register specifier. The decode stage is configured to decode the first and second fields and place the decoded contents on a global operand bus. The specified control register is addressed from the global operand bus while the access instruction is in decode. In the case of a read instruction, the addressed control register places its contents on the global operand bus while the instruction remains in decode. In the case of a write instruction, the general purpose register is addressed during the execution stage and its contents placed on the global operand bus during the writeback stage such that the contents of the addressed general purpose register are moved to the addressed configuration register during the writeback stage.

Mechanism for fast access to control space in a pipeline processor

A processor having an execution pipeline and a cache memory including a plurality of cache blocks with instruction words held in selected ones of the cache blocks. An ICBI address buffer is provided for holding addresses of instruction cache blocks to be invalidated by ICBI instructions pending in the processor's execution pipeline. An instruction cache controller coupled to the cache memory generates cache accesses to invalidate specified cache blocks in response to receiving buffered addresses from the ICBI address buffer. Preferably the cache accesses serve to commit ICBI instructions to the instruction cache asynchronously with respect to the processor's execution pipeline.

Mechanism for invalidating instruction cache blocks in a pipeline processor

A processor including a plurality of execution pipeline stages where each stage accepts a plurality of operand inputs and generates a result. A pipefile having at least the same number of entries as the number of execution pipeline stages is included in the processor. A pointer register is associated with each execution pipeline stage. A value is stored in at least one of the pointer registers, the value indicating a particular one of the entries in the pipefile.

Growing community of inventors

San Jose, CA, United States of America

Chih-Jui Peng