# US 6,732,257 B1 · Reducing the length of lower level instructions by splitting and recombining an immediate

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

- Number: US 6,732,257 B1
- Document ID: US6732257B1
- Status: Granted
- Kind: B1
- Application number: 09/675,819
- Filed: 2000-09-29
- Granted: 2004-05-04
- Earliest claimed: 2000-09-29
- Assignees: Intel Corporation
- Inventors: Gad S. Sheaffer
- Web page: https://patentgrep.com/patent/US6732257B1

## CPC

- G06F 9/30167: {of immediate specifier, e.g. constants}
- G06F 9/3016: {Decoding the operand specifier, e.g. specifier format}
- G06F 9/3017: {Runtime instruction translation, e.g. macros}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method is disclosed in which a higher level instruction having an immediate is read from memory and translated into two lower level instructions. The first is to move a first portion of the immediate to a register, and the second includes a pointer to the register as well as a second portion of the immediate.

## Claims

### Claim 1

A machine-implemented method comprising:
reading from memory a first higher level instruction having a first immediate; and
translating the first higher level instruction into first and second lower level instructions, the first lower level instruction is to move a first portion of the first immediate to a register, and the second lower level instruction includes a) a pointer to the register and b) a second portion of the first immediate.

### Claim 2 (depends on 1)

The method of claim 1 further comprising:
storing the first and second lower level instructions in a cache.

### Claim 3 (depends on 2)

The method of claim 2 further comprising:
the reading the first lower level instruction from the cache and, in response, storing the first portion of the first immediate in the register.

### Claim 4 (depends on 3)

The method of claim 3 further comprising:
reading the second lower level instruction from the cache and, in response, reading the register to obtain the first portion of the first immediate; and then
merging the first and second portions of the first immediate.

### Claim 5 (depends on 4)

The method of claim 4 wherein the first higher level instruction further includes an opcode and a pointer to a further register, and the translating further provides the second lower level instruction with a translated opcode and a translated register pointer.

### Claim 6 (depends on 3)

The method of claim 3 wherein the first portion of the first immediate is one or more most significant bits of the first immediate and the second portion is the remaining bits of the first immediate.

### Claim 7 (depends on 3)

The method of claim 3 further comprising:
reading a second higher level instruction having a second immediate; and
translating the second higher level instruction into a third lower level instruction if the second immediate has a first portion that has the same value as a content of the register, the third lower level instruction to include a pointer to the register and a second portion of the second immediate.

### Claim 8 (depends on 7)

The method of claim 7 further comprising:
storing the third lower level instruction in the cache.

### Claim 9 (depends on 8)

The method of claim 8 further comprising:
reading the third lower level instruction from the cache and, in response, reading the register to obtain the first portion; and then
merging the just read first portion with the second immediate's second portion.

### Claim 10 (depends on 9)

The method of claim 9 further comprising:
marking the first, second, and third lower level instructions as part of an atomic trace sequence.

### Claim 11

A processor comprising:
a register; and
a decoder to translate a higher level instruction having an immediate into first and second lower level instructions, the first lower level instruction is to move a first portion of the immediate to the register, and the second lower level instruction includes a) a pointer to the register and b) a second portion of the immediate.

### Claim 12 (depends on 11)

The processor of claim 11 further comprising:
an instruction cache to store the first and second lower level instructions.

### Claim 13 (depends on 12)

The processor of claim 12 further comprising:
logic to read the first lower level instruction from the instruction cache in response to a cache hit, and in response cause the first portion of the immediate to be stored in the register.

### Claim 14 (depends on 13)

The processor of claim 13 wherein the logic is to further read in response to the cache hit the second lower level instruction from the instruction cache, and in response read the register to obtain the first portion of the immediate, and merge the first and second portions of the immediate.

### Claim 15

A computer system comprising:
a memory to store higher level instructions; and
a processor coupled to the memory to execute the higher level instructions, the processor having a register and a decoder to translate a higher level instruction having an immediate into first and second lower level instructions, the first lower level instruction is to move a first portion of the immediate to the register, and the second lower level instruction includes a) a pointer to the register and b) a second portion of the immediate.

### Claim 16 (depends on 15)

The system of claim 15 wherein the processor further comprises an instruction cache to store the first and second lower level instructions.

### Claim 17 (depends on 16)

The system of claim 16 wherein the processor further comprises logic to read the first lower level instruction from the instruction cache in response to a cache hit, and in response cause the first portion of the immediate to be stored in the register.

### Claim 18 (depends on 17)

The system of claim 17 wherein in the processor, the logic is to further read in response to the cache hit the second lower level instruction from the instruction cache, and in response read the register to obtain the first portion of the immediate, and merge the first and second portions of the immediate.
