# US 2014/0208033 A1 · Split-Word Memory

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

- Number: US 2014/0208033 A1
- Document ID: US20140208033A1
- Status: Published application
- Kind: A1
- Application number: 14/221,661
- Filed: 2014-03-21
- Published: 2014-07-24
- Earliest claimed: 2012-03-30
- Assignees: Intel Corporation
- Inventors: Lutz Naethke; Axel Borkowski; Bert Bretschneider; Kyriakos A. Stavrou; Rainer Theuer
- Web page: https://patentgrep.com/patent/US20140208033A1

## CPC

- G06F 9/3802: {Instruction prefetching}
- G06F 9/30109: {having multiple operands in a single register}
- G06F 9/30112: {comprising data of variable length}
- G06F 9/3013: {according to data content, e.g. floating-point registers, address registers}
- G06F 9/30145: {Instruction analysis, e.g. decoding, instruction word fields}
- G06F 9/30185: {according to one or more bits in the instruction, e.g. prefix, sub-opcode}
- G06F 9/30192: {according to data descriptor, e.g. dynamic data typing}
- G06F 9/3836: {Instruction issuing, e.g. dynamic instruction scheduling or out of order instruction execution}
- G06F 9/3861: {Recovery, e.g. branch miss-prediction, exception handling (error detection or correction G06F11/00)}
- G06F 9/3877: {using a secondary processor, e.g. coprocessor (peripheral processor G06F13/12)}
- G06F 9/3887: {controlled by a single instruction for multiple data lanes [SIMD]}
- G06F 9/3888: {controlled by a single instruction for multiple threads [SIMT] in parallel}
- G06F 9/3889: {controlled by multiple instructions, e.g. MIMD, decoupled access or execute}
- G06F 12/00: Accessing, addressing or allocating within memory systems or architectures (digital input from, or digital output to record carriers, e.g. to disk storage units, G06F3/06)
- G06F 12/04: Addressing variable-length words or parts of words
- G06F 12/0607: {Interleaved addressing}
- G06F 12/0846: Cache with multiple tag or data arrays being simultaneously accessible
- G06F 12/0862: with prefetch
- G06F 12/0875: with dedicated cache, e.g. instruction or stack
- G06F 15/7807: {System on chip, i.e. computer system on a single chip; System in package, i.e. computer system on one or more chips in a single package}
- G06F 15/7857: {using interleaved memory (addressing G06F12/0607)}
- G06F 2212/452: Instruction code
- G06F 2212/6028: Prefetching based on hints or prefetch instructions
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

Method, process, and apparatus to efficiently store, read, and/or process syllables of word data. A portion of a data word, which includes multiple syllables, may be read by a computer processor. The processor may read a first syllable of the data word from a first memory. The processor may read a second syllable of the data word from a second portion of memory. The second syllable may include bits which are less critical than the bits of the first syllable. The second memory may be distinct from the first memory based on one or more physical attributes.

## Claims

### Claim 1

A processor comprising:
a front end unit including an instruction fetch unit and a decode unit;
an execution engine coupled to the front end unit; and
a memory unit coupled to the execution engine and including a first cache, wherein the processor is to read a first portion of a plurality of portions of a data word from a first portion of the first cache and read a second portion of the plurality of portions from a second memory, wherein bits of the second portion are less critical than bits of the first portion, and the second memory is distinct from the first cache based on at least a physical attribute.

### Claim 2 (depends on 1)

The processor of claim 1, wherein the first cache comprise a split word cache.

### Claim 3 (depends on 1)

The processor of claim 1, wherein the first cache is to operate faster than the second memory.

### Claim 4 (depends on 1)

The processor of claim 1, wherein a first connection between the processor and the first cache includes stronger data bus repeaters than a second connection between the processor and the second memory.

### Claim 5 (depends on 1)

The processor of claim 1, wherein the first cache consumes more power than the second memory.

### Claim 6 (depends on 1)

The processor of claim 1, wherein the processor is to process the first portion, and to read the second portion dependent on the first portion processing.

### Claim 7 (depends on 1)

The processor of claim 1, wherein the processor is to process the first portion, and upon determining that the second portion is to be processed based on the first portion processing, to process the second portion.

### Claim 8 (depends on 1)

The processor of claim 1, wherein the processor is to process the less critical bits of the second portion after the bits of the first portion.

### Claim 9

A machine-readable medium having stored thereon instructions, which if performed by a machine cause the machine to perform a method comprising:
reading a first portion of a data word by a processor from a first portion of a cache memory, wherein the data word includes a plurality of portions; and
reading a second portion of the plurality of portions from a second portion of the cache memory, wherein bits of the first portion are to be accessed by a consuming unit of the processor prior to bits of the second portion of the plurality of portions.

### Claim 10 (depends on 9)

The machine-readable medium of claim 9, wherein the method further comprises reading a first set of bits corresponding to the first portion of the data word and reading a second set of bits corresponding to the second portion of the data word, wherein the first and second sets of bits are of asymmetric size.

### Claim 11 (depends on 9)

The machine-readable medium of claim 9, wherein the method further comprises:
processing the first portion, wherein the reading the second portion is dependent on the processing of the first portion.

### Claim 12 (depends on 9)

The machine-readable medium of claim 9, wherein the method further comprises:
processing the first portion;
upon determining that the second portion is to be processed based on processing the first portion, processing the second portion; and
upon determining that the second portion is not to be processed based on the processing of the first portion, discarding the second portion.

### Claim 13 (depends on 9)

The machine-readable medium of claim 9, wherein the less critical bits of the second portion are processed after processing the bits of the first portion.

### Claim 14

A system comprising:
a processor including a consuming unit to consume a portion of a data word, wherein the data word includes a plurality of portions including a first portion and a second portion;
a first memory to store the first portion; and
a second memory to store the second portion, wherein bits of the second portion are to be accessed dependent upon processing of the first portion.

### Claim 15 (depends on 14)

The system of claim 14, wherein the first memory is located closer to the processor than the second memory.

### Claim 16 (depends on 14)

The system of claim 14, further comprising:
a first connection between the processor and the first memory; and
a second connection between the processor and the second memory, wherein the first connection includes stronger data bus repeaters than the second connection.

### Claim 17 (depends on 14)

The system of claim 14, wherein the processor is further configured to process the first portion, and read the second portion if the processor determines that the second portion is to be read based on processing the first portion.

### Claim 18 (depends on 14)

The system of claim 14, wherein the processor is further configured to process the first portion, and process the second portion if the processor determines that the second portion is to be processed based on processing the first portion.
