US5613080A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Multiple execution unit dispatch with instruction shifting between first and second instruction buffers based upon data dependency
Assignee: International Business Machines Corp
•
Filing: 1996-08-08
•
Granted: 1997-03-18
•
Expires: 2016-08-08
Abstract
Abstract A multiple execution unit processing system is provided wherein each execution unit has an associated instruction buffer and all instruction are executed in order. The first execution unit (unit 0) will always contain the oldest instruction and the second unit (unit 1) the newest. Processor instructions, such as load, store, add and the like are provided to each of the instruction buffers (0,1) from an instruction cache buffer. The first instruction (oldest) is placed in buffer 0 and the next (second) instruction is stored in buffer 1. It is determined during the decode stage whether the instructions are dependent on an immediately preceding instruction. If both instructions are independent of other instructions, then they can execute in parallel. However, if the second instruction is dependent on the first, then (subsequent to the first instruction being executed) it is laterally shifted to the first instruction buffer. Instructions are also defined as being dependent on an unavailable resource. In most cases these "unavailable" instructions are allowed to executed in parallel on the execution units.
Inventors
- 👤 David S. Ray
- 👤 Larry E. Thatcher
- 👤 Henry S. Warren, Jr.
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-04 03:45
CPC Classifications
G; G06; G06F; G06F9/00; G06F9/06; G06F9/30; G06F9/38; G06F9/3885; G06F9/3836; G06F9/3854; G06F9/3856; G06F9/3858
Claims Analysis
Total Claims: 22
Claim 1
Independent Claim
181 words
A computer processing system including a first and second execution unit, comprising: first and second instruction buffers, which receive instructions in a particular sequence from an instruction cache, corresponding to respective said first and second execution units for providing said instructions thereto; means for interpreting whether said instructions in said first and second instruction buffers are independent instructions; means for concurrently providing said instructions in said first instruction buffer and said instructions in said second instruction buffer to said first and second execution units, respectively, when said instructions in said first and second instruction buffers are determined to be independent; and means for providing one of said instructions in said first instruction buffer to said first execution unit and concurrently moving one of said instructions directly from said second instruction buffer to said first instruction buffer and another of said instructions from said instruction cache to said second instruction buffer when said instruction in said second instruction buffer is dependent on the instruction in said first instruction buffer; wherein said particular sequence of said instructions is maintained and executed in order.
Claim 2
Dependent Claim
181 words
A computer processing system including a first and second execution unit, comprising: first and second instruction buffers, which receive instructions in a particular sequence from an instruction cache, corresponding to respective said first and second execution units for providing said instructions thereto; means for interpreting whether said instructions in said first and second instruction buffers are independent instructions; means for concurrently providing said instructions in said first instruction buffer and said instructions in said second instruction buffer to said first and second execution units, respectively, when said instructions in said first and second instruction buffers are determined to be independent; and means for providing one of said instructions in said first instruction buffer to said first execution unit and concurrently moving one of said instructions directly from said second instruction buffer to said first instruction buffer and another of said instructions from said instruction cache to said second instruction buffer when said instruction in said second instruction buffer is dependent on the instruction in said first instruction buffer; wherein said particular sequence of said instructions is maintained and executed in order.
Claim 3
Dependent Claim
25 words
A system according to claim 1 wherein said data dependent instructions comprise dependent and unavailable instructions, and said other non-data dependent instructions comprise independent instructions.
Claim 4
Dependent Claim
52 words
A system according to claim 2 further comprising a three port adder used to concurrently add three values stored in three distinct storage locations and cause said means for interpreting to designate said instructions for adding said three values as independent instructions which can be executed independent of any other said instructions.
Claim 5
Dependent Claim
43 words
A system according to claim 3 wherein a decode logic circuit is included in said means for interpreting and causes said independent instructions to be shifted in parallel from said first and second instruction buffers to said respective first and second execution units.
Claim 6
Dependent Claim
35 words
A system according to claim 4 wherein said decode control circuit causes said unavailable instructions to be shifted in parallel from said first and second instruction buffers to said respective first and second execution units.
Claim 7
Dependent Claim
49 words
A system according to claim 5 wherein said decode control circuit causes one of said dependent instructions in said first instruction buffer to be shifted to said first execution unit and one of said dependent instructions in said second instruction buffer to be moved to said first instruction buffer.
Claim 8
Dependent Claim
46 words
A system according to claim 6 wherein said decode control circuit causes one of said unavailable instructions in said first instruction buffer and one of said independent instructions in said second instruction buffer to be shifted in parallel to said respective first and second execution units.
Claim 9
Dependent Claim
49 words
A system according to claim 7 wherein said decode control circuit causes one of said independent instructions in said first instruction buffer to be shifted to said first execution unit and one of said unavailable instructions in said second instruction buffer to be moved to said first instruction buffer.
Claim 10
Dependent Claim
49 words
A system according to claim 8 wherein said decode control circuit causes one of said independent instructions in said first instruction buffer to be shifted to said first execution unit and one of said dependent instructions in said second instruction buffer to be moved to said first instruction buffer.
Claim 11
Dependent Claim
94 words
A system according to claim 9 further comprising: means for causing a single successive instruction to be input to said second instruction buffer subsequent to said instruction currently in said second instruction buffer being moved, by said decode control circuit, to said first instruction buffer; and means for causing first and second successive instructions to be input to said first and second instruction buffers, respectively, when said instructions currently in both said first and second instruction buffers are shifted in parallel by said decode control circuit to said first and second execution units, respectively.
Claim 12
Independent Claim
198 words
A method of executing instructions on a computer processing system including a first and second execution unit, comprising the steps of: providing first and second instruction buffers, which receive instructions in a particular sequence from an instruction cache, corresponding to said first and second execution units for providing instructions thereto; interpreting, by a decode control circuit, whether said instructions in said first and second instruction buffers are independent instructions; concurrently providing, by said decode control circuit, said instructions in said first instruction buffer and said instructions in said second instruction buffer to said first and second execution units, respectively, when said instructions in said first and second instruction buffers are determined to be independent; and providing one of said instructions in said first instruction buffer to said first execution unit and concurrently moving, by said decode control circuit, one of said instructions directly from said second instruction buffer to said first instruction buffer and another of said instructions from said instruction cache to said second instruction buffer when said instruction in said second instruction buffer is dependent on the instruction in said first instruction buffer; wherein said particular sequence of said instructions is maintained and executed in order.
Claim 13
Independent Claim
198 words
A method of executing instructions on a computer processing system including a first and second execution unit, comprising the steps of: providing first and second instruction buffers, which receive instructions in a particular sequence from an instruction cache, corresponding to said first and second execution units for providing instructions thereto; interpreting, by a decode control circuit, whether said instructions in said first and second instruction buffers are independent instructions; concurrently providing, by said decode control circuit, said instructions in said first instruction buffer and said instructions in said second instruction buffer to said first and second execution units, respectively, when said instructions in said first and second instruction buffers are determined to be independent; and providing one of said instructions in said first instruction buffer to said first execution unit and concurrently moving, by said decode control circuit, one of said instructions directly from said second instruction buffer to said first instruction buffer and another of said instructions from said instruction cache to said second instruction buffer when said instruction in said second instruction buffer is dependent on the instruction in said first instruction buffer; wherein said particular sequence of said instructions is maintained and executed in order.
Claim 14
Dependent Claim
35 words
A method according to claim 11 wherein said step of moving comprises the step of transferring said instructions from said second buffer to said first buffer via a bus interconnecting said first and second buffers.
Claim 15
Dependent Claim
23 words
A method according to claim 12 wherein said data dependent instructions comprise dependent and unavailable instructions, and said other instructions comprise independent instructions.
Claim 16
Dependent Claim
54 words
A method according to claim 13 further comprising the step of using a three port adder to concurrently add three values stored in three distinct storage locations and cause said second circuit to designate said instructions for adding said three values as independent instructions which can be executed independent of any other said instructions.
Claim 17
Dependent Claim
62 words
A method according to claim 14 further comprising the steps of: shifting, by said decode circuit, said independent instructions in parallel from said first and second instruction buffers to said respective first and second execution units; and shifting, by said decode circuit, said unavailable instructions in parallel from said first and second instruction buffers to said respective first and second execution units.
Claim 18
Dependent Claim
52 words
A method according to claim 15 further comprising the steps of: shifting, by said decode circuit, one of said dependent instructions in said first instruction buffer to said first execution unit; and moving, by said decode circuit, one of said dependent instructions in said second instruction buffer to said first instruction buffer.
Claim 19
Dependent Claim
45 words
A method according to claim 16 further comprising the step of shifting, by said decode circuit, one of said unavailable instructions in said first instruction buffer and one of said independent instructions in said second instruction buffer to said respective first and second execution units.
Claim 20
Dependent Claim
52 words
A method according to claim 17 further comprising the steps of: shifting, by said decode circuit, one of said independent instructions in said first instruction buffer to said first execution unit; and moving, by said decode circuit, one of said unavailable instructions in said second instruction buffer to said first instruction buffer.
Claim 21
Dependent Claim
52 words
A method according to claim 18 further comprising the steps of: shifting, by said decode circuit, one of said independent instructions in said first instruction buffer to said first execution unit; and moving, by said decode circuit, one of said dependent instructions in said second instruction buffer to said first instruction buffer.
Claim 22
Dependent Claim
91 words
A method according to claim 19 further comprising the steps of: causing a single successive instruction to be input to said second instruction buffer subsequent to said instruction currently in said second instruction buffer being moved, by said decode circuit, to said first instruction buffer; and causing first and second successive instructions to be input to said first and second instruction buffers, respectively, when said instructions currently in both said first and second instruction buffers are shifted in parallel by said decode circuit to said first and second execution units, respectively.
References Cited (Backward Citations)
US3771141A
Cited
US4095269A
Cited
US4130885A
Cited
US4232366A
Cited
US4376976A
Cited
US4675806A
Cited
US4626989A
Cited
EP0101596A2
Cited
EP0118830A2
Cited
US4807115A
Cited
US4630195A
Cited
US4847755A
Cited
US4890218A
Cited
US4766566A
Cited
US4928226A
Cited
US4837678A
Cited
US5133077A
Cited
US5247628A
Cited
US4974155A
Cited
US5099421A
Cited
US5075840A
Cited
US5127091A
Cited
US5113515A
Cited
US5093908A
Cited
US5197137A
Cited
US5269007A
Cited
US5251306A
Cited
US5185868A
Cited
US5051940A
Cited
US5465377A
Cited
US5502826A
Cited
US5504932A
Cited
EP0471191A2
Cited
US5381531A
Cited
US5301341A
Cited
US5257354A
Cited
US5283874A
Cited
GB2263565A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-03-17 | 2000-09-16 | 2001-03-18 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-03-17 | 2004-09-16 | 2005-03-18 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-03-17 | 2008-09-16 | 2009-03-18 | $7700.00 | Window Closed / Lapsed |