Scalable system on a chip

Patent metadata

Field	Value
Publication number	US-11803471-B2
Application number	US-202217821312-A
Country	US
Kind code	B2
Filing date	Aug 22, 2022
Priority date	Aug 23, 2021
Publication date	Oct 31, 2023
Grant date	Oct 31, 2023

Abstract

An integrated circuit (IC) including a plurality of processor cores, a plurality of graphics processing units, a plurality of peripheral circuits, and a plurality of memory controllers is configured to support scaling of the system using a unified memory architecture. For example, the IC may include an interconnect fabric configured to provide communication between the one or more memory controller circuits and the processor cores, graphics processing units, and peripheral devices; and an off-chip interconnect coupled to the interconnect fabric and configured to couple the interconnect fabric to a corresponding interconnect fabric on another instance of the integrated circuit, wherein the interconnect fabric and the off-chip interconnect provide an interface that transparently connects the one or more memory controller circuits, the processor cores, graphics processing units, and peripheral devices in either a single instance of the integrated circuit or two or more instances of the integrated circuit.

First claim

What is claimed is:

1. An integrated circuit, comprising:
a plurality of processor cores;
a plurality of graphics processing units;
a plurality of peripheral devices distinct from the processor cores and graphics processing units;
one or more memory controller circuits configured to interface with a system memory;
an interconnect fabric configured to provide communication between the one or more memory controller circuits and the processor cores, graphics processing units, and peripheral devices; and
an off-chip interconnect coupled to the interconnect fabric and configured to couple the interconnect fabric to a corresponding interconnect fabric on another instance of the integrated circuit, wherein the interconnect fabric and the off-chip interconnect provide an interface that is configured to connect the one or more memory controller circuits, the processor cores, the graphics processing units, and the peripheral devices in either a single instance of the integrated circuit or two or more instances of the integrated circuit, and wherein the interface is further configured to couple multiple instances of the integrated circuit in a manner that the multiple instances of the integrated circuit transparently appear to software as a single system.

2. The integrated circuit of claim 1 , wherein a unified address space defined by a unified memory architecture extends across the two or more instances of the integrated circuit die transparent to software executing on the processor cores, graphics processing units, or peripheral devices.

3. The integrated circuit of claim 2 , wherein the unified address space maps to the single instance transparent to software.

4. The integrated circuit of claim 1 , further comprising a local interrupt distribution circuit to distribute interrupts among processor cores in the integrated circuit.

5. The integrated circuit of claim 4 , further comprising a global interrupt distribution circuit, wherein the local interrupt distribution circuits and the global interrupt distribution circuit implement a multi-level interrupt distribution scheme when two or more instances of the integrated circuit are used.

6. The integrated circuit of claim 5 , wherein the global interrupt distribution circuit is configured to transmit an interrupt request to the local interrupt distribution circuits in at least two instances in a sequence, and wherein the local interrupt distribution circuits are configured to transmit the interrupt request to local interrupt destinations in a sequence before replying to the interrupt request from the global interrupt distribution circuit.

7. The integrated circuit of claim 1 , wherein a given integrated circuit die comprises a power manager circuit configured to manage a local power state of the given integrated circuit die.

8. The integrated circuit of claim 7 , further comprising another power manager circuit configured to synchronize the power manager circuits on the two or more instances of the integrated circuit.

9. The integrated circuit of claim 1 , wherein the peripheral devices include one of more of: an audio processing device, a video processing device, a machine learning accelerator circuit, a matrix arithmetic accelerator circuit, a camera processing circuit, a display pipeline circuit, a nonvolatile memory controller, a peripheral component interconnect controller, a security processor, or a serial bus controller.

10. The integrated circuit of claim 1 , wherein the interconnect fabric interconnects coherent agents.

11. The integrated circuit of claim 10 , wherein an individual one of the processor cores corresponds to a coherent agent.

12. The integrated circuit of claim 10 , wherein a cluster of processor cores corresponds to a coherent agent.

13. The integrated circuit of claim 1 , wherein a given one of the peripheral devices is a non-coherent agent.

14. The integrated circuit of claim 13 , further comprising an input/output agent interposed between the given peripheral device and the interconnect fabric, wherein the input/output agent is configured to enforce coherency protocols of the interconnect fabric with respect to the given peripheral device.

15. The integrated circuit of claim 14 , wherein the input/output agent ensures the ordering of requests from the given peripheral device using the coherency protocols.

16. The integrated circuit of claim 14 , wherein the input/output agent is configured to couple a network of two or more peripheral devices to the interconnect fabric.

17. The integrated circuit of claim 1 , further comprising hashing circuitry configured to distribute memory request traffic to system memory according to a selectively programmable hashing protocol.

18. The integrated circuit of claim 17 , wherein at least one programming of the programmable hashing protocol evenly distributes a series of memory requests over a plurality of memory controllers in the system for a variety of memory requests in the series.

19. The integrated circuit of claim 18 , wherein at least one programming of the programmable hashing protocol distributes adjacent requests within the memory space, at a specified granularity, to physically distant memory interfaces.

20. A system, comprising:
a plurality of instances of an integrated circuit, the integrated circuit comprising:
a plurality of processor cores;
a plurality of graphics processing units;
a plurality of peripheral devices distinct from the processor cores and graphics processing units;
one or more memory controller circuits configured to interface with a system memory;
an interconnect fabric configured to provide communication between the one or more memory controller circuits and the processor cores, graphics processing units, and peripheral devices; and

an off-chip interconnect coupled to the interconnect fabric and configured to couple the interconnect fabric to a corresponding interconnect fabric on another one of the plurality of instances of the integrated circuit, wherein the interconnect fabric and the off-chip interconnect provide an interface that is configured to connect the one or more memory controller circuits, the processor cores, the graphics processing units, and the peripheral devices in either a single instance of the integrated circuit or two or more instances of the integrated circuit, and wherein the interface is further configured to couple the plurality of instances of the integrated circuit in a manner that the plurality of instances of the integrated circuit transparently appear to software as a single system.

21. The system as recited in claim 20 , further comprising a substrate configured to couple the off-chip interconnect from one of the plurality of instances to the off-chip interconnect of another one of the plurality of instances.

22. The system as recited in claim 20 , further comprising a network integrated circuit configured to couple to the off-chip interconnect in the plurality of instances and to route communications between the plurality of instances.

Assignees

• Apple Inc

Inventors

• Hammarlund Per H
• Zimet Lior
• Kolor Sergio
• Lahav Sagi
• Vash James
• Garg Gaurav
• Kuzi Tal
• Gonion Jeffry E
• Tucker Charles E
• Levy-Rubin Lital
• Davidov Dany
• Fishwick Steven
• Leshem Nir
• Pilip Mark
• Williams Iii Gerard R
• Kaushikkar Harshavardhan
• Sridharan Srinivasa Rangan

Classifications

• G06F2212/1016

  Performance improvement · CPC title

• G06F12/128

  adapted to multidimensional cache systems, e.g. set-associative, multicache, multiset or multilevel · CPC title

• G06F12/109

  for multiple virtual address spaces, e.g. segmentation (G06F12/1036 takes precedence) · CPC title

• G06F12/0815

  Cache consistency protocols · CPC title

• G06F12/0811

  with multilevel cache hierarchies · CPC title