IP Library Granted Patent US 7,356,675
Granted Patent B2
US 7,356,675 · App. 11/482,062 · Granted Apr 8, 2008

Data processor

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,356,675
App. No.
11/482,062
Granted
Apr 8, 2008
Kind
B2
Abstract

A data processor having: an instruction fetch unit for acquiring an instruction; and an instruction execution unit for execution, by pipeline processing, the instruction acquired by the instruction fetch unit. In the data processor, the instruction execution unit includes an execution pipeline pipelined into two or more stages for execution of the execution instruction, and an execution pipeline control unit capable of changing a stage to arrange an operation by the execution unit in according to the number of wait cycles until data required for execution of the execution instruction is fixed. The stage, where an operation by the execution unit is arranged, is changed according to the number of wait cycles until the data is fixed, whereby the number of input-fixing wait cycles increased by pipelining cache access can be reduced.

Claims (28)

1. A data processor comprising:

an instruction fetch unit for acquiring an instruction; and

an instruction execution unit for execution, by pipeline processing, the instruction acquired by the instruction fetch unit,

wherein the instruction execution unit includes

an execution unit for realizing an execution function of an execution instruction,

an execution pipeline pipelined into two or more stages for execution of the execution instruction, and

an execution pipeline control unit capable of changing a stage to arrange an operation by the execution unit according to a number of wait cycles until data required for execution of the execution instruction in the execution pipeline is fixed.

2. A data processor comprising:

an instruction fetch unit for acquiring an instruction; and

an instruction execution unit for execution, by pipeline processing, the instruction acquired by the instruction fetch unit,

wherein the instruction includes a load instruction for directing load of data, and

the instruction execution unit includes

an execution unit for realizing an execution function of an execution instruction,

an execution pipeline pipelined into two or more stages for execution of the execution instruction, and

an execution pipeline control unit capable of changing a point where data loaded by execution of the load instruction is used in an operation by the execution unit according to a number of wait cycles until the data is fixed in the execution pipeline.

3. The data processor of claim 1 , wherein the execution pipeline control unit accepts a result of access to a data cache as data required for execution of the execution instruction in the execution pipeline.

4. The data processor of claim 3 , wherein the execution pipeline control unit arranges an operation by the execution unit in a stage such that the number of wait cycles until the data is fixed is minimum.

5. The data processor of claim 3 , wherein when an attempt to arrange an operation by the execution unit in a stage used in execution of another execution instruction is made, the execution pipeline control unit arranges the operation by the execution unit in a stage further behind the stage in use.

6. The data processor of claim 3 , wherein a number of cycles required for execution of the execution instruction using the result of access to a data cache can be varied depending on an instruction executed prior to the execution instruction.

7. The data processor of claim 3 , wherein the execution pipeline includes

two or more inter-stage registers, each capable of holding a result of an execution in the execution unit, and

a selector capable of making a selection from among outputs of the two or more inter-stage registers,

wherein the selection from among the outputs of the inter-stage registers by the selector enables change of a stage where an operation by the execution unit is arranged.

8. The data processor of claim 3 , wherein the execution pipeline control unit includes:

a first pipeline capable of holding a signal showing validity of register writing and information on a stage where an operation by the execution unit is arranged;

a second pipeline capable of holding a register number of a register written into the register according to the instruction;

a third pipeline capable of holding a register number of a register read out from the register according to the instruction; and

a read register number generation circuit for selectively supplying an output of the third pipeline to the execution pipeline based on an putout of the first pipeline.

Assignments (4)
CHANGE OF ADDRESS Recorded Nov 29, 2017
From: RENESAS ELECTRONICS CORPORATION
To: RENESAS ELECTRONICS CORPORATION
Reel/Frame 044928/0001 →
CHANGE OF NAME Recorded Sep 10, 2010
From: NEC ELECTRONICS CORPORATION
To: RENESAS ELECTRONICS CORPORATION
Reel/Frame 024982/0123 →
MERGER - EFFECTIVE DATE 04/01/2010 Recorded Sep 10, 2010
From: RENESAS TECHNOLOGY CORP.
To: NEC ELECTRONICS CORPORATION
Reel/Frame 024982/0198 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 7, 2006
From: OZAWA, MOTOKAZU
To: RENESAS TECHNOLOGY CORP.
Reel/Frame 018086/0420 →