US5619664A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Processor with architecture for improved pipelining of arithmetic instructions by forwarding redundant intermediate data forms
Assignee: Intel Corp
•
Filing: 1995-03-10
•
Granted: 1997-04-08
•
Expires: 2015-03-10
Abstract
Abstract A processor providing a redundant intermediate form of a result in fewer than the total number of clock cycles needed to calculate a final complete result. The redundant form is forwarded to subsequent instructions or operations capable of utilizing the redundant intermediate form to enhance the performance of the processor.
Inventors
- 👤 Andrew F. Glew
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-10 00:53
CPC Classifications
G; G06; G06F; G06F9/00; G06F9/06; G06F9/30; G06F9/30003; G06F9/30007; G06F9/3001; G06F9/30014; G06F9/38; G06F9/3824; G06F9/3826; G06F9/3828; G06F9/3867; G06F9/3875; G06F9/3885; G06F9/3893
Claims Analysis
Total Claims: 12
Claim 1
Independent Claim
146 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when the first instruction involves inter arithmetic and the second instruction involves inter arithmetic, comparison of two non-zero values, memory addressing, data storage, multiplication or division and that causes the operand bypass network to forward a non-redundant result generated during a second execution stage of the first instruction when the second instruction involves logic and zero comparison.
Claim 2
Dependent Claim
146 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when the first instruction involves inter arithmetic and the second instruction involves inter arithmetic, comparison of two non-zero values, memory addressing, data storage, multiplication or division and that causes the operand bypass network to forward a non-redundant result generated during a second execution stage of the first instruction when the second instruction involves logic and zero comparison.
Claim 3
Dependent Claim
18 words
The processor of claim 1, wherein the arithmetic execution unit comprises a plurality of carry save adder units.
Claim 4
Dependent Claim
25 words
The processor of claim 1, wherein the pipeline control logic comprises a comparison logic unit for comparing opcodes of the first and the second instructions.
Claim 5
Dependent Claim
116 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when first instruction involves logic and the second instruction involves inter arithmetic, logic, zero comparison, comparison of two non-zero values, memory addressing, data storage, multiplication or division.
Claim 6
Dependent Claim
116 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when first instruction involves logic and the second instruction involves inter arithmetic, logic, zero comparison, comparison of two non-zero values, memory addressing, data storage, multiplication or division.
Claim 7
Dependent Claim
18 words
The processor of claim 4, wherein the arithmetic execution unit comprises a plurality of carry save adder units.
Claim 8
Dependent Claim
25 words
The processor of claim 4, wherein the pipeline control logic comprises a comparison logic unit for comparing opcodes of the first and the second instructions.
Claim 9
Dependent Claim
121 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when first instruction involves AND, OR, XOR, and NOT operations and the second instruction involves inter arithmetic, logic, zero comparison, comparison of two non-zero values, memory addressing, data storage, multiplication or division.
Claim 10
Dependent Claim
121 words
A processor comprising: a decoder that decodes a first and a second instruction; a register file coupled to the decoder that stores source operand values specified by the first and second instructions; an arithmetic execution unit, coupled to the decoder, having an operand bypass network that receives the source operand values from the register file; pipeline control logic, coupled to the decoder and the operand bypass network, that causes the operand bypass network to forward a redundant intermediate result generated during a first execution stage of the first instruction when first instruction involves AND, OR, XOR, and NOT operations and the second instruction involves inter arithmetic, logic, zero comparison, comparison of two non-zero values, memory addressing, data storage, multiplication or division.
Claim 11
Dependent Claim
18 words
The processor of claim 7, wherein the arithmetic execution unit comprises a plurality of carry save adder units.
Claim 12
Dependent Claim
25 words
The processor of claim 7, wherein the pipeline control logic comprises a comparison logic unit for comparing opcodes of the first and the second instructions.
References Cited (Backward Citations)
US4156922A
Cited
US4525796A
Cited
US4916606A
Cited
US4849926A
Cited
US4873660A
Cited
US4939686A
Cited
US4953119A
Cited
US5270962A
Cited
US5258944A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-04-07 | 2000-10-07 | 2001-04-08 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-04-07 | 2004-10-07 | 2005-04-08 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-04-07 | 2008-10-07 | 2009-04-08 | $7700.00 | Window Closed / Lapsed |