Memory compression method and apparatus for heterogeneous processor architectures in an information handling system

• US 7,849,241 B2
• Filed: 03/23/2006
• Issued: 12/07/2010
• Est. Priority Date: 03/23/2006
• Status: Expired due to Fees

• Current Assignee: LinkedIn Corporation (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: Gschwind, Michael Karl, Minor, Barry L
• Primary Examiner(s): Tsai; Henry W
• Assistant Examiner(s): Sun; Michael
• Application Number:
  US11/387,999
• Publication Number:
  US 20070245097A1
• Time in Patent Office: 1,720 Days
• Field of Search: 710/68
• US Class Current: 710/68
• CPC Class Codes: G06F 12/04 Addressing variable-length words or parts of words
• CPC Class Codes: G06F 12/08 in hierarchically structured memory systems, e.g. virtual memory systems
• CPC Class Codes: G06F 2212/251 Local memory within processor subsystem
• CPC Class Codes: G06F 2212/253 Centralized memory
• CPC Class Codes: G06F 2212/401 Compressed data
• CPC Class Codes: G06F 9/30178 of compressed or encrypted instructions

Abstract

The disclosed heterogeneous processor compresses information to more efficiently store the information in a system memory coupled to the processor. The heterogeneous processor includes a general purpose processor core coupled to one or more processor cores that exhibit an architecture different from the architecture of the general purpose processor core. In one embodiment, the processor dedicates a processor core other than the general purpose processor core to memory compression and decompression tasks. In another embodiment, system memory stores both compressed information and uncompressed information.

Claims

1. A method of compressing and decompressing information in a heterogeneous multi-core processor, the method comprising:

processing information by a first processor core exhibiting a first architecture, the first processor core including a first data path exhibiting a first data width;

processing information by a second processor core exhibiting a second architecture, the first processor core and the second processor core being in the same heterogeneous multi-core processor, the second processor core including a second data path, the second data path exhibiting a second data width that is wider than the width of the first data path;

compressing, by the second processor core, information to be sent by the heterogeneous multi-core processor to a system memory for storage therein as compressed information;

decompressing, by the second processor core, compressed information received from the system memory for use as uncompressed information by the heterogeneous processor, the second processor core being dedicated to compressing and decompressing operations;

storing, by the heterogeneous multi-core processor, both compressed information and uncompressed information in the system memory, the system memory including a compressed memory portion that stores the compressed information and an uncompressed memory portion that stores the uncompressed information, the system memory acting as a shared memory for both the compressed information and the uncompressed information, the system memory including a movable boundary between the compressed memory portion and the uncompressed memory portion, the movable boundary enabling changing of the ratio of the amount of compressed information in the compressed memory portion to the amount of uncompressed information in the uncompressed memory portion;

receiving, by the second processor core, a request to decompress compressed information stored in the compressed memory portion;

testing to determine if memory space is available in the uncompressed memory portion to store resultant decompressed information generated by performing the request to decompress;

compressing, by the second processor core, uncompressed information in the uncompressed memory portion to provide resultant compressed information,

transferring the resultant compressed information to the compressed memory portion for storage if the testing step determines that sufficient memory space is not available in the uncompressed memory portion for the resultant decompressed information;

performing, by the second processor core, the request to decompress to generate the resultant decompressed information; and

storing the resultant decompressed information in the uncompressed memory portion.

2. The method of claim 1, wherein the first architecture is a general purpose processor core and the second architecture is a special purpose processor core that performs tasks assigned by the first processor core.

3. The method of claim 1, wherein a memory flow controller in the second processor core communicates compressed information and uncompressed information to and from the system memory.

4. The method of claim 1, wherein the second processor core speculatively decompresses information in the system memory that may be needed by the first processor core, and the second processor core speculatively compresses information for storage in the system memory.

5. The method of claim 1, wherein the first processor core performs other tasks while the second processor core is compressing or decompressing information.

6. An information handling system (IHS) comprising:

a heterogeneous processor including;

a first processor core exhibiting a first architecture, the first processor core including a first data path exhibiting a first data width;

a second processor core exhibiting a second architecture that compresses information to provide compressed information and that decompresses compressed information to provide uncompressed information, the first processor core and the second processor core being situated in the same heterogeneous multi-core processor, the second processor core including a second data path, the second data path exhibiting a second data width that is wider than the width of the first data path;

a bus coupling the first processor core to the second processor core; and

a system memory, coupled to the heterogeneous processor, that stores the compressed information and the uncompressed information, the system memory acting as a shared memory that stores both the compressed information and the uncompressed information, the system memory including a compressed memory portion that stores the compressed information and an uncompressed memory portion that stores the uncompressed information, the system memory including a movable boundary between the compressed memory portion and the uncompressed memory portion, the movable boundary enabling changing of the ratio of the amount of compressed information in the compressed memory portion to the amount of uncompressed information in the uncompressed memory portion ;

the second processor core being dedicated to compressing and decompressing operations, the second processor core being configured to;

receive a request to decompress compressed information stored in the compressed memory portion;

test to determine if memory space is available in the uncompressed memory portion to store resultant decompressed information generated by performing the request to decompress;

compress uncompressed information in the uncompressed memory portion to provide resultant compressed information and transfer the resultant compressed information to the compressed memory portion for storage if the test determines that sufficient memory space is not available in the uncompressed memory portion for the resultant decompressed information;

perform the request to decompress to generate the resultant decompressed information; and

store the resultant decompressed information in the uncompressed memory portion.

7. The IHS of claim 6, wherein the first architecture is a general purpose processor core and the second architecture is a special purpose processor core that performs tasks assigned by the first processor core.

8. The IHS of claim 6, wherein the second processor core speculatively decompresses information in the system memory that may be needed by the first processor core, and the second processor core speculatively compresses information for storage in the system memory.