# US 2019/0317899 A1 · Independent and separate entity-based cache

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

- Number: US 2019/0317899 A1
- Document ID: US20190317899A1
- Status: Published application
- Kind: A1
- Application number: 16/394,829
- Filed: 2019-04-25
- Published: 2019-10-17
- Earliest claimed: 2017-04-10
- Assignees: Intel Corporation
- Applicants: Intel Corporation
- Inventors: Altug Koker; Joydeep Ray; James A. Valerio; Abhishek R. Appu; Vasanth Ranganathan
- Web page: https://patentgrep.com/patent/US20190317899A1

## CPC

- G06F 12/0842: for multiprocessing or multitasking
- G06F 12/0811: with multilevel cache hierarchies
- G06F 12/0831: using a bus scheme, e.g. with bus monitoring or watching means
- G06T 1/60: Memory management
- G06F 12/0875: with dedicated cache, e.g. instruction or stack
- G06F 2212/1008: Correctness of operation, e.g. memory ordering
- G06F 2212/1024: Latency reduction
- G06F 2212/455: Image or video data
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A mechanism is described for facilitating independent and separate entity-based graphics cache at computing devices. A method of embodiments, as described herein, includes facilitate hosting of a plurality of cache at a plurality of entities associated with a graphics processor, wherein each entity hosts at least one cache, and wherein an entity includes a dual sub-slice (DSS) or a streaming multiprocessor (SM).

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

An apparatus comprising:
one or more processors including a graphics processor to:
facilitate hosting of a plurality of graphics caches at a plurality of entities associated with the graphics processor, wherein each entity of the plurality of entities is to host at least one graphics cache, and wherein each entity includes a plurality of cores; and
map each of the plurality of graphics caches to an entire system cache or memory to cache data stored in the system cache or memory,
wherein each graphics cache is logically independent of and separate from other graphics caches of the plurality of graphics caches, and wherein each graphics cache has a direct and independent path to a data port associated with the system cache or memory.

### Claim 22 (depends on 21)

The apparatus of claim 21, wherein the direct and independent paths of the graphics caches are established without employing an internal fabric.

### Claim 23 (depends on 22)

The apparatus of claim 22, wherein the apparatus further includes a low bandwidth inter-cache fabric to support coherency between the plurality of entities.

### Claim 24 (depends on 23)

The apparatus of claim 23, wherein the coherency is supported when performing one or more tasks including one or more compute tasks.

### Claim 25 (depends on 24)

The apparatus of claim 24, wherein the one or more processors are further to support pixel hashing across the plurality of entities without employing an inter-entity cacheline sharing for workloads including three-dimensional (3D) workloads.

### Claim 26 (depends on 21)

The apparatus of claim 21, wherein the plurality of entities includes one or more streaming multiprocessors.

### Claim 27 (depends on 21)

The apparatus of claim 21, wherein the graphics processor is co-located with an application processor on a common semiconductor package.

### Claim 28

A method comprising:
facilitating hosting of a plurality of graphics caches at a plurality of entities associated with a graphics processor, wherein each entity of the plurality of entities is to host at least one graphics cache, and wherein each entity includes a plurality of cores; and
mapping each of the plurality of graphics caches to an entire system cache or memory to cache data stored in the system cache or memory,
wherein each graphics cache is logically independent of and separate from other graphics caches of the plurality of graphics caches, and wherein each graphics cache has a direct and independent path to a data port associated with the system cache or memory.

### Claim 29 (depends on 28)

The method of claim 28, wherein the direct and independent paths of the graphics caches are established without employing an internal fabric.

### Claim 30 (depends on 29)

The method of claim 29, further comprising:
supporting coherency between the plurality of entities utilizing a low bandwidth inter-cache fabric.

### Claim 31 (depends on 30)

The method of claim 30, wherein the coherency is supported when performing one or more tasks including one or more compute tasks.

### Claim 32 (depends on 31)

The method of claim 31, further comprising supporting pixel hashing across the plurality of entities without employing an inter-entity cacheline sharing for workloads including three-dimensional (3D) workloads.

### Claim 33 (depends on 28)

The method of claim 28, wherein the plurality of entities includes one or more streaming multiprocessors.

### Claim 34 (depends on 28)

The method of claim 28, wherein the graphics processor is co-located with an application processor on a common semiconductor package.

### Claim 35

At least one non-transitory machine-readable medium comprising instructions that when executed by a computing device, cause the computing device to perform operations comprising:
facilitating hosting of a plurality of graphics caches at a plurality of entities associated with a graphics processor, wherein each entity of the plurality of entities is to host at least one graphics cache, and wherein each entity of the plurality of entities includes a plurality of cores; and
mapping each of the plurality of graphics caches to an entire system cache or memory to cache data stored in the system cache or memory,
wherein each graphics cache is logically independent of and separate from other graphics caches of the plurality of graphics caches, and wherein each graphics cache has a direct and independent path to a data port associated with the system cache or memory.

### Claim 36 (depends on 35)

The machine-readable medium of claim 35, wherein the direct and independent paths of the graphics caches are established without employing an internal fabric.

### Claim 37 (depends on 36)

The machine-readable medium of claim 36, wherein the operations further comprise:
supporting coherency between the plurality of entities utilizing a low bandwidth inter-cache fabric.

### Claim 38 (depends on 37)

The machine-readable medium of claim 37, wherein the coherency is supported when performing one or more tasks including one or more compute tasks.

### Claim 39 (depends on 38)

The machine-readable medium of claim 38, wherein the operations further comprise supporting pixel hashing across the plurality of entities without employing an inter-entity cacheline sharing for workloads including three-dimensional (3D) workloads.

### Claim 40 (depends on 35)

The machine-readable medium of claim 35, wherein the plurality of entities associated with the graphics processor includes one or more streaming multiprocessors.

### Claim 41 (depends on 35)

The machine-readable medium of claim 35, wherein the graphics processor is co-located with an application processor on a common semiconductor package.
