# US 2019/0377493 A1 · Supporting hierarchical ordering points in a microprocessor system

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

- Number: US 2019/0377493 A1
- Document ID: US20190377493A1
- Status: Published application
- Kind: A1
- Application number: 16/006,505
- Filed: 2018-06-12
- Published: 2019-12-12
- Earliest claimed: 2018-06-12
- Assignees: Intel Corporation
- Inventors: Erik Hallnor; Matthew Erler
- Web page: https://patentgrep.com/patent/US20190377493A1

## CPC

- G06F 3/0604: {Improving or facilitating administration, e.g. storage management}
- G06F 12/0833: {in combination with broadcast means (e.g. for invalidation or updating)}
- G06F 3/0629: {Configuration or reconfiguration of storage systems}
- G06F 3/0673: {Single storage device}
- G06F 12/0811: with multilevel cache hierarchies
- G06F 2212/1008: Correctness of operation, e.g. memory ordering
- G06F 2212/1024: Latency reduction
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

Embodiment of this disclosure provides a hierarchical structure of ordering points. In some embodiments, the hierarchical structure includes a single primary ordering point (POP) and at least one (or more) auxiliary order point (AOP) of a processing device. In one implementation, the processing device includes one or more cores; and a coherency circuit, operatively coupled to the cores. The processing device is to receive a plurality of memory access requests to be ordered by a first ordering point of the processing device. The processing device determines whether to stop the first ordering point based on a system event. Responsive to determining that the first ordering point is stopped, a second ordering point of the processing device is identified. Thereupon, a memory access request of the plurality of memory access requests is provided to the second ordering point.

## Claims

### Claim 1

A processing device comprising:
one or more cores; and
a coherency circuit, operatively coupled to the cores, to:
receive a plurality of memory access requests to be ordered by a first ordering point of the processing device;
determine whether to stop the first ordering point based on a system event;
responsive to determining that the first ordering point is stopped, identify a second ordering point of the processing device; and
provide a memory access request of the plurality of memory access requests to the second ordering point.

### Claim 2 (depends on 1)

The processing device of claim 1, wherein the coherency circuit is further to, responsive to determining that the first ordering point is stopped:
complete in-flight memory access requests associated with the second ordering point; and
responsive to completing the in-flight memory access requests, write cache data associated with the first ordering point to memory.

### Claim 3 (depends on 1)

The processing device of claim 1, wherein the coherency circuit is further to:
detect that the one or more cores are shutdown in accordance with the system event.

### Claim 4 (depends on 1)

The processing device of claim 1, wherein the coherency circuit is further to:
cause the second ordering point to send the provided memory access request to memory via a memory controller.

### Claim 5 (depends on 1)

The processing device of claim 1, wherein the coherency circuit is further to:
determine that the first ordering point is restarted based on a termination of the system event; and
cause the second ordering point to expose memory data associated with one or more memory access requests to the first ordering point.

### Claim 6 (depends on 5)

The processing device of claim 5, wherein to expose the memory data, the coherency circuit is further to:
cause the second ordering point to generate a request for the memory data to be forwarded from an owner of the memory data to the first ordering point.

### Claim 7 (depends on 1)

The processing device of claim 1, wherein the coherency circuit is further to:
stop the first ordering point of the processing device; and
direct the received plurality of memory access requests to be ordered by the second ordering point of the processing device.

### Claim 8

A method comprising:
receiving by a processing device, a plurality of memory access requests to be ordered by a first ordering point of the processing device;
determining, by the processing device, whether to stop the first ordering point based on a system event;
responsive to determining that the first ordering point is stopped, identifying, by the processing device a second ordering point of the processing device; and
providing, by the processing device, a memory access request of the plurality of memory access requests to the second ordering point.

### Claim 9 (depends on 8)

The method of claim 8, further comprising, responsive to determining that the first ordering point is stopped:
completing in-flight memory access requests associated with the second ordering point; and
responsive to completing the in-flight memory access requests, writing cache data associated with the first ordering point to memory.

### Claim 10 (depends on 8)

The method of claim 8, further comprising:
detecting that the one or more cores are shutdown in accordance with the system event.

### Claim 11 (depends on 8)

The method of claim 8, further comprising:
causing the second ordering point to send the provided memory access request to memory via a memory controller.

### Claim 12 (depends on 8)

The method of claim 8, further comprising:
determining that the first ordering point is restarted based on a termination of the system event; and
causing the second ordering point to expose memory data associated with one or more memory access requests to the first ordering point.

### Claim 13 (depends on 12)

The method of claim 12, wherein exposing the memory data further comprises:
causing the second ordering point to generate a request for the memory data to be forwarded from an owner of the memory data to the first ordering point.

### Claim 14 (depends on 8)

The method of claim 8, further comprising:
stopping the first ordering point of the processing device; and
directing the received plurality of memory access requests to be ordered by the second ordering point of the processing device.

### Claim 15

A system comprising:
a memory controller to access a memory shared by a plurality of cores; and
a processing device, operatively coupled to the memory controller, to:
receive a plurality of memory access requests to be ordered by a first ordering point of the processing device;
determine whether to stop the first ordering point based on a system event;
responsive to determining that the first ordering point is stopped, identify a second ordering point of the processing device; and
provide a memory access request of the plurality of memory access requests to the second ordering point.

### Claim 16 (depends on 15)

The system of claim 15, wherein the processing device is further to, responsive to determining that the first ordering point is stopped:
complete in-flight memory access requests associated with the second ordering point; and
responsive to completing the in-flight memory access requests, write cache data associated with the first ordering point to the memory.

### Claim 17 (depends on 15)

The system of claim 15, wherein the processing device is further to:
detect that the one or more cores are shutdown in accordance with the system event.

### Claim 18 (depends on 15)

The system of claim 15, wherein the processing device is further to:
determine that the first ordering point is restarted based on a termination of the system event; and
cause the second ordering point to expose memory data associated with one or more memory access requests to the first ordering point.

### Claim 19 (depends on 18)

The system of claim 18, wherein to expose the memory data, the processing device is further to:
cause the second ordering point to generate a request for the memory data to be forwarded from an owner of the memory data to the first ordering point.

### Claim 20 (depends on 15)

The system of claim 15, wherein the processing device is further to:
stop the first ordering point of the processing device; and
direct the received plurality of memory access requests to be ordered by the second ordering point of the processing device.
