# US 2008/0313422 A1 · Enhanced Single Threaded Execution in a Simultaneous Multithreaded Microprocessor

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

- Number: US 2008/0313422 A1
- Document ID: US20080313422A1
- Status: Published application
- Kind: A1
- Application number: 11/763,736
- Filed: 2007-06-15
- Published: 2008-12-18
- Earliest claimed: 2007-06-15
- Assignees: International Business Machines Corporation
- Inventors: Hung Q. Le; Dung Q. Nguyen
- Web page: https://patentgrep.com/patent/US20080313422A1

## CPC

- G06F 9/3885: {using a plurality of independent parallel functional units}
- G06F 9/30105: {Register structure}
- G06F 9/30116: {Shadow registers, e.g. coupled registers, not forming part of the register space}
- G06F 9/3012: {Organisation of register space, e.g. banked or distributed register file}
- G06F 9/30141: {Implementation provisions of register files, e.g. ports}
- G06F 9/30181: {Instruction operation extension or modification}
- G06F 9/3802: {Instruction prefetching}
- G06F 9/3814: {Implementation provisions of instruction buffers, e.g. prefetch buffer; banks}
- G06F 9/3851: {from multiple instruction streams, e.g. multistreaming}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method, system, and computer program product are provided for enhancing the execution of independent loads in a processing unit. The processing unit dispatches a first set of instructions in order from a first buffer for execution. The processing unit receives updated results from the execution of the first set of instructions. The processing unit updates, in a first register, at least one register entry associated with each instruction in the first set of instructions, with the updated results. The processing unit determines if the first set of instructions from the first buffer have completed execution. Responsive to the completed execution of the first set of instructions from the first buffer, the processing unit copies the set of entries from the first register to a second register.

## Claims

### Claim 1

A method, in a data processing system, for enhancing the execution of independent loads in a processing unit, the method comprising:
dispatching a first set of instructions in order from a first buffer for execution;
receiving updated results from the execution of the first set of instructions;
updating, in a first register, at least one register entry associated with each instruction in the first set of instructions, with the updated results;
determining if the first set of instructions from the first buffer have completed execution; and
responsive to the completed execution of the first set of instructions from the first buffer, copying the set of entries from the first register to a second register.

### Claim 2 (depends on 1)

The method of claim 1, wherein the set of entries from the first register are copied to the second register using a copy_unconditional signal.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
dispatching a second set of instructions in order from a second buffer;
receiving the updated results from the execution of the second set of instructions;
updating, in the second register, a register entry associated with each instruction in the second set of instructions with the updated results; and
setting a select_mask bit in a select_mask register for each entry in the set of entries that is updated with the updated results.

### Claim 4 (depends on 3)

The method of claim 3, wherein the first buffer and the first register are associated with a first thread, wherein the second buffer and the second register are associated with a second thread, and wherein the first set of instructions and the second set of instructions are associated with the first thread.

### Claim 5 (depends on 3)

The method of claim 3, further comprising:
determining if the second set of instructions from the second buffer have completed execution; and
responsive to the completed execution of the second set of instructions from the first buffer, copying the set of entries from the second register to the first register, wherein the set of entries from the second register are copied to the first register using a copy_using_mask signal.

### Claim 6 (depends on 5)

The method of claim 5, further comprising:
clearing each set select_mask bit from of the select_mask register upon completion of the copying the set of entries from the second register to the first register.

### Claim 7 (depends on 5)

The method of claim 5, further comprising:
dispatching a new set of instructions in order from the first buffer;
receiving the updated results from the execution of the new set of instructions;
updating, in the first register, the register entry associated with each instruction in the new set of instructions with the updated results; and
setting the select_mask bit in the select_mask register for each entry in the set of entries that is updated with the updated results.

### Claim 8 (depends on 1)

The method of claim 1, wherein the processing unit is operating in a single threaded mode and wherein the processing unit executing the single threaded mode comprises:
determining if a mode bit is set that indicates an initiation of the single threaded mode;
responsive to the mode bit being set, initiating the single threaded mode;
clearing the contents of the first buffer, the second buffer, the first vector array, the second vector array, a first register, and a second register; and
fetching instructions into the first buffer.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
fetching the first set of instructions into the first buffer in order until the first buffer is full;
responsive to filling the first buffer, fetching a second set of instructions into a second buffer in order until the second buffer is full, wherein a first initial instruction of the second set of instructions has a first_bit set; and
responsive to filling the second buffer, fetching a new set of instructions into the first buffer in order until the first buffer is full, wherein a second initial instruction of the new set of instructions has the first_bit set.

### Claim 10 (depends on 8)

The method of claim 8, wherein the first_bit being set on the first initial instruction indicates that the first initial instruction and any younger instructions in the second buffer must wait until the first set of instruction in the first buffer have dispatched and wherein the first_bit being set on the second initial instruction indicates that the second initial instruction and any younger instructions in the first buffer must wait until the second set of instruction in the second buffer have dispatched.

### Claim 11

A computer program product comprising a computer useable medium having a computer readable program, wherein the computer readable program, when executed in a data processing system, causes the data processing system to:
dispatch a first set of instructions in order from a first buffer for execution;
receive updated results from the execution of the first set of instructions;
update, in a first register, at least one register entry associated with each instruction in the first set of instructions, with the updated results;
determine if the first set of instructions from the first buffer have completed execution; and
responsive to the completed execution of the first set of instructions from the first buffer, copy the set of entries from the first register to a second register.

### Claim 12 (depends on 11)

The computer program product of claim 11, wherein the computer readable program further causes the data processing system to:
dispatch a second set of instructions in order from a second buffer;
receive the updated results from the execution of the second set of instructions;
update, in the second register, a register entry associated with each instruction in the second set of instructions with the updated results; and
set a select_mask bit in a select_mask register for each entry in the set of entries that is updated with the updated results.

### Claim 13 (depends on 12)

The computer program product of claim 12, wherein the first buffer and the first register are associated with a first thread, wherein the second buffer and the second register are associated with a second thread, and wherein the first set of instructions and the second set of instructions are associated with the first thread.

### Claim 14 (depends on 12)

The computer program product of claim 12, wherein the computer readable program further causes the data processing system to:
determine if the second set of instructions from the second buffer have completed execution; and
responsive to the completed execution of the second set of instructions from the first buffer, copy the set of entries from the second register to the first register, wherein the set of entries from the second register are copied to the first register using a copy_using_mask signal.

### Claim 15 (depends on 11)

The computer program product of claim 11, wherein the computer readable program further causes the data processing system to:
fetch the first set of instructions into the first buffer in order until the first buffer is full;
responsive to filling the first buffer, fetch a second set of instructions into a second buffer in order until the second buffer is full, wherein a first initial instruction of the second set of instructions has a first_bit set; and
responsive to filling the second buffer, fetch a new set of instructions into the first buffer in order until the first buffer is full, wherein a second initial instruction of the new set of instructions has the first_bit set.

### Claim 16

A system, comprising:
a processing unit; and
a memory coupled to the processing unit, wherein the memory comprises instructions which, when executed by the processing unit, cause the processing unit to:
dispatch a first set of instructions in order from a first buffer for execution;
receive updated results from the execution of the first set of instructions;
update, in a first register, at least one register entry associated with each instruction in the first set of instructions, with the updated results;
determine if the first set of instructions from the first buffer have completed execution; and
responsive to the completed execution of the first set of instructions from the first buffer, copy the set of entries from the first register to a second register.

### Claim 17 (depends on 16)

The system of claim 16, wherein the instructions further cause the processor to:
dispatch a second set of instructions in order from a second buffer;
receive the updated results from the execution of the second set of instructions;
update, in the second register, a register entry associated with each instruction in the second set of instructions with the updated results; and
set a select_mask bit in a select_mask register for each entry in the set of entries that is updated with the updated results.

### Claim 18 (depends on 17)

The system of claim 17, wherein the first buffer and the first register are associated with a first thread, wherein the second buffer and the second register are associated with a second thread, and wherein the first set of instructions and the second set of instructions are associated with the first thread.

### Claim 19 (depends on 17)

The system of claim 17, wherein the instructions further cause the processor to:
determine if the second set of instructions from the second buffer have completed execution; and
responsive to the completed execution of the second set of instructions from the first buffer, copy the set of entries from the second register to the first register, wherein the set of entries from the second register are copied to the first register using a copy_using_mask signal.

### Claim 20 (depends on 16)

The system of claim 16, wherein the instructions further cause the processor to:
fetch the first set of instructions into the first buffer in order until the first buffer is full;
responsive to filling the first buffer, fetch a second set of instructions into a second buffer in order until the second buffer is full, wherein a first initial instruction of the second set of instructions has a first_bit set; and
responsive to filling the second buffer, fetch a new set of instructions into the first buffer in order until the first buffer is full, wherein a second initial instruction of the new set of instructions has the first_bit set.
