Cache arrangement for graphics processing systems

Patent metadata

Field	Value
Publication number	US-10726519-B2
Application number	US-201715714037-A
Country	US
Kind code	B2
Filing date	Sep 25, 2017
Priority date	Sep 25, 2017
Publication date	Jul 28, 2020
Grant date	Jul 28, 2020

Abstract

A graphics processing system includes a cache system for transferring texture data stored in memory to a graphics processing unit for use by the graphics processing unit when generating a render output. The cache system includes a first cache operable to receive texture data from the memory system, and a second cache operable to receive texture data from the first cache and to provide texture data to the graphics processing unit for use when generating a render output, and a data processing unit intermediate the first cache and the second cache and operable to process data stored in the first cache and to store the processed data in the second cache.

First claim

What is claimed is:

1. A graphics processing system comprising:
a memory system;
a graphics processing unit; and
a cache system arranged between the memory system and the graphics processing unit and operable to transfer texture data stored in the memory system to the graphics processing unit for use by the graphics processing unit when generating a render output;
the cache system comprising:
a first cache operable to receive texture data from the memory system and operable to accept unaligned reads;
a second cache operable to receive texture data from the first cache and to provide texture data to the graphics processing unit for use when generating a render output; and
a data processing unit intermediate the first cache and the second cache and operable to process data stored in the first cache and to store the processed data in the second cache for use by the graphics processing unit when generating a render output;
wherein the cache system is operable to perform an aligned read operation from the memory system in order to provide texture data to the first cache,
wherein the data processing unit is operable to decompress texture data that is stored in a compressed form in the first cache and to then store the decompressed texture data in the second cache; and
wherein the data processing unit is able to perform an unaligned read operation from the first cache in order to provide texture data to the second cache.

2. The system of claim 1 , wherein the second cache is larger than the first cache.

3. The system of claim 1 , wherein the texture data is stored in the first cache of the cache system in the form that it is stored in the memory system, and is identified in the first cache using an identifier that is indicative of the memory address where the texture data is stored.

4. The system of claim 1 , wherein the texture data is stored in the second cache as respective texture data elements, and the texture data that is stored in the second cache is identified in the second cache using an identifier that is indicative of a position in a graphics texture associated with the texture data.

5. The system of claim 1 , wherein the data processing unit is operable to read data stored in the first cache and to store the read data in the second cache in a different arrangement to the way that that data was stored in the first cache.

6. The system of claim 1 , wherein the data processing unit is operable to: reorder texture data elements in a given cache line; distribute texture data elements over plural cache lines in the second cache where they are stored in a single cache line in the first cache; and/or store texture data elements that are stored in different cache lines in the first cache in the same cache line in the second cache.

7. The system of claim 1 , wherein the cache system is further operable so as to be able to store texture data from the memory system in the second cache without first storing the texture data in the first cache.

8. A cache system for use in a graphics processing system that comprises a memory system and a graphics processing unit;
the cache system comprising:
a first cache operable to receive texture data from a memory system and operable to accept unaligned reads;
a second cache operable to receive texture data from the first cache and to provide texture data to a graphics processing unit for use when generating a render output; and
a data processing unit intermediate the first cache and the second cache and operable to process data stored in the first cache and to store the processed data in the second cache for use by the graphics processing unit when generating a render output;
wherein the cache system is operable to perform an aligned read operation from the memory system in order to provide texture data to the first cache,
wherein the data processing unit is operable to decompress texture data that is stored in a compressed form in the first cache and to then store the decompressed texture data in the second cache; and
wherein the data processing unit is able to perform an unaligned read operation from the first cache in order to provide texture data to the second cache.

9. A method of operating a graphics processing system, the graphics processing system comprising:
a memory system;
a graphics processing unit; and
a cache system arranged between the memory system and the graphics processing unit and operable to transfer texture data stored in the memory system to the graphics processing unit for use by the graphics processing unit when generating a render output;
the cache system comprising:
a first cache operable to receive texture data from the memory system and operable to accept unaligned reads;
a second cache operable to receive texture data from the first cache and to provide texture data to the graphics processing unit for use when generating a render output; and
a data processing unit intermediate the first cache and the second cache and operable to process data stored in the first cache and to store the processed data in the second cache for use by the graphics processing unit when generating a render output;
wherein the cache system is operable to perform an aligned read operation from the memory system in order to provide texture data to the first cache,
wherein the data processing unit is operable to decompress texture data that is stored in a compressed form in the first cache and to then store the decompressed texture data in the second cache; and
wherein the data processing unit is able to perform an unaligned read operation from the first cache in order to provide texture data to the second cache;
the method comprising:
performing an aligned read operation in order to load texture data from the memory system into the first cache of the cache system;
the data processing unit processing the texture data loaded into the first cache and storing the processed data in the second cache; and
the graphics processing unit using the processed texture data stored in the second cache to perform a texturing operation;
wherein the method further comprises:
loading compressed texture data from the memory system into the first cache of the cache system; and
the data processing unit:
performing an unaligned read operation to read compressed texture data from the first cache, decompressing the compressed texture data read from the first cache, and storing the decompressed texture data in a cache line of the second cache.

10. The method of claim 9 , wherein the second cache is larger than the first cache.

11. The method of claim 9 , wherein the texture data is stored in the first cache of the cache system in the form that it is stored in the memory system, and is identified in the first cache using an identifier that is indicative of the memory address where the texture data is stored.

12. The method of claim 9 , wherein the texture data is stored in the second cache as texture data elements, and the texture data that is stored in the second cache is identified in the second cache using an identifier that is indicative of a position in a graphics texture associated with the texture data.

13. The method of claim 9 comprising:
loading texture data from the memory system into the first cache so as to be stored in a first arrangement in the first cache; and
the data processing unit:
reading the data from the first cache and storing the data in the second cache in a different arrangement to the first arrangement.

14. The method of claim 9 , comprising the data processing unit:
reordering texture data elements in a given cache line;
distributing texture data element

Assignees

• Advanced Risc Mach Ltd

Inventors

• Fielding Edvard
• Fries Jakob Axel

Classifications

• Y02D10/00

  Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title

• G06T15/04

  Texture mapping · CPC title

• G06T15/005

  General purpose rendering architectures · CPC title

• G06T1/60 Primary

  Memory management · CPC title

• G06F12/0811

  with multilevel cache hierarchies · CPC title