# US 2005/0240644 A1 · Scalar/vector processor

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2005/0240644 A1
- Document ID: US20050240644A1
- Status: Published application
- Kind: A1
- Application number: 10/515,451
- Filed: 2003-05-22
- Published: 2005-10-27
- Earliest claimed: 2002-05-24
- Assignees: ST-Ericsson SA
- Inventors: Cornelis Van Berkel; Patrick Meuwissen; Nur Engin
- Web page: https://patentgrep.com/patent/US20050240644A1

## CPC

- G06F 9/3853: {of compound instructions}
- G06F 9/30025: {Format conversion instructions, e.g. Floating-Point to Integer, decimal conversion}
- G06F 9/30032: {Movement instructions, e.g. MOVE, SHIFT, ROTATE, SHUFFLE}
- G06F 9/30036: {Instructions to perform operations on packed data, e.g. vector, tile or matrix operations}
- G06F 9/3824: {Operand accessing}
- G06F 9/3828: {with global bypass, e.g. between pipelines, between clusters}
- G06F 9/3834: {Maintaining memory consistency}
- G06F 9/3877: {using a secondary processor, e.g. coprocessor (peripheral processor G06F13/12)}
- G06F 9/3885: {using a plurality of independent parallel functional units}
- G06F 9/3887: {controlled by a single instruction for multiple data lanes [SIMD]}
- G06F 9/3893: {controlled in tandem, e.g. multiplier-accumulator}
- G06F 15/8061: {Details on data memory access}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A scalar/vector processor includes a plurality of functional units (252, 260, 262, 264, 266, 268, 270). At least one of the functional units includes a vector section (210) for operating on at least one vector and a scalar section (220) for operating on at least one scalar. The vector section and scalar section of the functional unit co-operate by the scalar section being arranged to provide and/or consume at least one scalar required by and/or supplied by the vector section of the functional unit.

## Claims

### Claim 1

A scalar/vector processor including a plurality of functional units, at least one of the functional units including a vector section for operating on at least one vector and a scalar section for operating on at least one scalar, the vector section and scalar section of the functional unit co-operating by the scalar section being arranged to provide and/or consume at least one scalar required by and/or supplied by the vector section of the functional unit.

### Claim 2 (depends on 1)

A scalar/vector processor as claimed in claim 1, wherein a plurality of the functional units include a respective vector section; the vector sections being arranged in a first pipeline.

### Claim 3 (depends on 1)

A scalar/vector processor as claimed in claim 1, wherein a plurality of the functional units include a respective scalar section; the scalar sections being arranged in a second pipeline.

### Claim 4 (depends on 2)

A scalar/vector processor as claimed in claim 2, where the first and second pipeline are independently configurable.

### Claim 5 (depends on 2)

A scalar/vector processor as claimed in claim 2, wherein at least one of the pipelines is configurable on a processor instruction-by-instruction basis.

### Claim 6 (depends on 1)

A scalar/vector processor as claimed in claim 1, wherein the scalar/vector processor is controlled by a VLIW instruction including separate instruction segments for each functional unit.

### Claim 7 (depends on 6)

A scalar/vector processor as claimed in claim 6, wherein a instruction segment for a functional unit with both a vector section and a scalar section includes respective instructions for the vector section and for the scalar section of the functional unit.

### Claim 8 (depends on 5)

A scalar/vector processor as claimed in claim 5, wherein the VLIW instruction includes the configuration of the first and/or second pipeline.

### Claim 9 (depends on 7)

A scalar/vector processor as claimed in claim 7, wherein the first pipeline includes for each vector section attached to the first pipeline a respective associated pipeline path for transferring a vector and/or the second pipeline includes for each scalar section attached to the second pipeline a respective associated pipeline path for transferring a scalar.

### Claim 10 (depends on 8)

A scalar/vector processor as claimed in claim 8, wherein the instruction for the vector section of a functional unit specifies a pipeline path associated with a different one of the functional units for transfer of a vector and/or the instruction for the scalar section of a functional unit specifies a pipeline path associated with a different one of the functional units for transfer of a scalar.

### Claim 11 (depends on 1)

A scalar/vector processor as claimed in claim 1, wherein at least one of the functional units is a shift unit; a vector section of the shift unit being operative to shift a vector over at least one scalar; a scalar section of the shift unit being operative to supply to and/or receive from the vector section the at least one scalar that is being shifted in or out of the vector.

### Claim 12 (depends on 1)

A processing system including a scalar processor and a scalar/vector processor as claimed in claim 1, wherein the scalar/vector processor is arranged as a co-processor to the scalar processor and the scalar processor is arranged to control the scalar/vector processor; the scalar section of the scalar/vector processor being arranged for performing in-loop scalar processing and the scalar processor being arranged for performing irregular, out-of-loop scalar processing.
