Distribution of injected data among caches of data processing system

GB2623912BActive Publication Date: 2024-11-13INTERNATIONAL BUSINESS MACHINE CORPORATION
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
GB2024001002
Authority / Receiving Office
GB · GB
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-08-04
Filing Date
2022-08-04
Publication Date
2024-11-13
Estimated Expiration
2042-08-04

AI Technical Summary

Technical Problem

Current digital circuit design and verification processes are inefficient due to the complexity of hardware description languages like Verilog, which lack formal verification methods for ensuring correctness and completeness, leading to errors and increased design cycles.

Method used

The integration of formal verification techniques into the design flow using model checking and theorem proving to validate digital circuits described in Verilog, ensuring correctness and completeness by automatically detecting errors before physical implementation.

Benefits of technology

This approach significantly reduces design cycles and improves accuracy by automatically identifying and correcting errors early in the design process, enhancing the reliability and efficiency of digital circuit development.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000001_0000
    Figure 00000001_0000
  • Figure 00000002_0000
    Figure 00000002_0000
  • Figure 00000003_0000
    Figure 00000003_0000
Patent Text Reader

Abstract

A data processing system includes a plurality of processor cores each supported by a respective one of a plurality of vertical cache hierarchies. Based on receiving on a system fabric a cache injection request requesting injection of a data into a cache line identified by a target real address, the data is written into a cache in a first vertical cache hierarchy among the plurality of vertical cache hierarchies. Based on a value in a field of the cache injection request, a distribute field is set in a directory entry of the first vertical cache hierarchy. Upon eviction of the cache line the first vertical cache hierarchy, a determination is made whether the distribute field is set. Based on determining the distribute field is set, a lateral castout of the cache line from the first vertical cache hierarchy to a second vertical cache hierarchy is performed.
Need to check novelty before this filing date? Find Prior Art

Description

The following terms are registered trade ma: an should be rea assuc w they occur in this document; Verilog

Claims

1. A method of data processing in a data processing system including a plurality of processor cores each supported by a respective one of a plurality of vertical cache hierarchies, the method comprising:based on receiving on a system fabric of the data processing system a cache injection request requesting injection of a data into a cache line identified by a target real address, writing the data into a cache in a first vertical cache hierarchy among the plurality of vertical cache hierarchies;based on a value in a field of the cache injection request, setting a distribute field in a directory entry of the first vertical cache hierarchy;upon eviction of the cache line from a cache memory in the first vertical cache hierarchy, determining whether the distribute field is set; andbased on determining the distribute field is set, performing a lateral castout of the cache line from the first vertical cache hierarchy to a second vertical cache hierarchy among the plurality of vertical cache hierarchies.

2. The method of Claim 1, wherein performing a lateral castout based on determining the distribute field is set comprises performing a lateral castout based on determining the distribute field is set only if the cache line is held by the cache memory in a coherence state providing write authority.

3. The method of Claim 1, wherein:the method further comprises evicting the cache line from an upper level cache in the first vertical cache hierarchy;performing the lateral castout includes a lower level cache in the first vertical cache hierarchy receiving the cache line upon eviction from the upper level cache and issuing a lateral castout request targeting the second vertical cache hierarchy without installing the cache line in a data array of the lower level cache.

4. The method of Claim 1, wherein:a processor core among the plurality of processor cores is supported by the first vertical cache hierarchy;the method further comprises:prior to receiving the cache injection request at the first vertical cache hierarchy, executing, by the processor core, an instruction to cause the cache line to be installed in the first vertical cache hierarchy in a coherence state providing write authority.

5. The method of Claim 1, wherein performing the lateral castout includes:transmitting, in a lateral castout request on the system fabric of the data processing system, a distribute field that is set;transmitting the cache line on the system fabric; andinstalling the cache line in a data array in the second vertical cache hierarchy and, based on the distributefield in the lateral castout request being set, setting a distribute field in a directory entry of the second vertical cache hierarchy.

6. The method of Claim 1, wherein:the data processing system includes a system memory;the method further comprising:thereafter, casting out the data of the cache line from the second vertical cache hierarchy to the system memory and storing the distribute field in the system memory in association with the data.

7. A processing unit for a data processing system, comprising:a processor core;a vertical cache hierarchy coupled to the processor core and configured to be coupled to a system fabric of the data processing system, wherein the vertical cache hierarchy includes a cache having a data array and a directory and is configured to perform:based on receiving on the system fabric of the data processing system a cache injection request requesting injection of a data into a cache line identified by a target real address, writing the data into the data array in the vertical cache hierarchy;based on a value in a field of the cache injection request, setting a distribute field in a directory entry of the directory;upon eviction of the cache line from the first vertical cache hierarchy, determining whether the distribute field is set; andbased on determining the distribute field is set, performing a lateral castout of the cache line from the vertical cache hierarchy to another vertical cache hierarchy in the data processing system.

8. The processing unit of Claim 7, wherein performing a lateral castout based on determining the distribute field is set comprises performing a lateral castout based on determining the distribute field is set only if the cache line is held by the cache memory in a coherence state providing write authority.

9. The processing unit of Claim 7, wherein:the cache is an upper level cache;the vertical cache hierarchy includes a lower level cache;performing the lateral castout includes a lower level cache receiving the cache line upon eviction from the upper level cache and issuing a lateral castout request targeting said another vertical cache hierarchy without installing the cache line in the lower level cache.

10. The processing unit of Claim 7, wherein:the processor core, prior to receipt of the cache injection request, executes an instruction to cause thecache line to be installed in the first vertical cache hierarchy in a coherence state providing write authority.

11. The processing unit of Claim 7, wherein performing the lateral castout includes:transmitting the cache line on the system fabric; andtransmitting, in a lateral castout request on the system fabric of the data processing system, a distribute field that is set, wherein the distribute field being set causes a distribute field to be set in a directory entry associated with the cache line in the second vertical cache hierarchy.

12. The processing unit of Claim 7, wherein the vertical cache hierarchy is further configured to perform: thereafter, casting out the data of the cache line from the second vertical cache hierarchy to the system memory and storing the distribute field in the system memory in association with the data.

13. A data processing system, comprising:a plurality of processing units in accordance with Claim 7;the interconnect fabric coupling the plurality of processing units; anda system memory communicatively coupled to the interconnect fabric.

14. The data processing system of Claim 13, further comprising a memory controller that, responsive to a castout of the data of the cache line from the second vertical cache hierarchy, installs the data in the system memory and stores the distribute field in the system memory in association with the data.

15. A design structure tangibly embodied in a machine-readable storage medium for designing, manufacturing, or testing an integrated circuit, the design structure comprising:a processing unit for a data processing system, including:a processor core;a vertical cache hierarchy coupled to the processor core and configured to be coupled to a system fabric of the data processing system, wherein the vertical cache hierarchy includes a cache having a data array and a directory and is configured to perform:based on receiving on the system fabric of the data processing system a cache injection request requesting injection of a data into a cache line identified by a target real address, writing the data into the data array in the vertical cache hierarchy;based on a value in a field of the cache injection request, setting a distribute field in a directory entry of the directory;upon eviction of the cache line from the first vertical cache hierarchy, determining whether the distribute field is set; andbased on determining the distribute field is set, performing a lateral castout of the cache line from the vertical cache hierarchy to another vertical cache hierarchy in the data processing system.

16. The design structure of Claim 15, wherein performing a lateral castout based on determining the distribute field is set comprises performing a lateral castout based on determining the distribute field is set only if the cache line is held by the cache memory in a coherence state providing write authority.

17. The design structure of Claim 15, wherein:the cache is an upper level cache;the vertical cache hierarchy includes a lower level cache;performing the lateral castout includes a lower level cache receiving the cache line upon eviction from the upper level cache and issuing a lateral castout request targeting said another vertical cache hierarchy without installing the cache line in the lower level cache.

18. The design structure of Claim 15, wherein:the processor core, prior to receipt of the cache injection request, executes an instruction to cause the cache line to be installed in the first vertical cache hierarchy in a coherence state providing write authority.

19. The design structure of Claim 15, wherein performing the lateral castout includes:transmitting the cache line on the system fabric; andtransmitting, in a lateral castout request on the system fabric of the data processing system, a distribute field that is set, wherein the distribute field being set causes a distribute field to be set in a directory entry associated with the cache line in the second vertical cache hierarchy.

20. The design structure of Claim 15, wherein the design structure comprises a hardware description language (HDL) design structure.

Citation Information

Patent Citations

  • Hybrid lower-level cache inclusion policy for cache hierarchy having at least three caching levels

    CN111263934A

  • Selective storage of data in levels of a cache memory

    US20080059707A1

  • Victim cache lateral castout targeting

    US20100235577A1

  • Cache coherency protocol employing a read operation including a programmable flag to indicate deallocation of an intervened cache line

    US6345342B1