# US 9,928,183 B2 · Priority framework for a computing device

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

- Number: US 9,928,183 B2
- Document ID: US9928183B2
- Status: Granted
- Kind: B2
- Application number: 14/497,619
- Filed: 2014-09-26
- Published: 2016-03-31
- Granted: 2018-03-27
- Earliest claimed: 2014-09-26
- Assignees: Ampere Computing LLC
- Applicants: Applied Micro Circuits Corporation
- Inventors: Kjeld Svendsen; Millind Mittal; Gaurav Singh
- Web page: https://patentgrep.com/patent/US9928183B2

## CPC

- G06F 13/18: based on priority control (G06F13/1605 takes precedence)
- 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

Proving for a framework for propagating priorities to a memory subsystem in a computing system environment is disclosed herein. By way of example, a memory access handler is provided for managing memory access requests and determining associated priorities. The memory access handler includes logic configured for propagating memory requests and the associated priorities to lower levels of a computer hierarchy. A memory subsystem receives the memory access requests and the priorities.

## Claims

### Claim 1

A system for managing priority to a memory subsystem, comprising:
a first level memory access handling unit configured to generate memory access requests in accordance with corresponding instructions originating from multiple levels of a computing system architecture, the multiple levels comprising at least an application level, a compiler level, an operating system level, and a micro-architectural level, wherein the first level memory access handling unit is of a higher level of the computing system architecture relative to the memory subsystem, and the instructions generated by the multiple levels, have associated priority indicators assigned in accordance with prioritization schemes respectively associated with the multiple levels, the priority indicators comprising first priority indicators assigned to a first subset of the instructions originating at the application level based on user-defined application priorities, second priority indicators assigned to a second subset of the instructions originating at the compiler level based on a determination of whether the second subset of the instructions are associated with shared data or private data, and third priority indicators assigned to a third subset of the instructions originating at the operating system level based on a determination of whether the third subset of the instructions are associated with an interrupt at the operating system level; and
a memory controller unit of the memory subsystem configured to receive the memory access requests and the priority indicators, and to assign respective memory access priorities of the memory access requests based on the priority indicators.

### Claim 2 (depends on 1)

The system of claim 1, wherein the memory controller unit is further configured to prioritize the memory access requests in an input queue of the memory controller unit based on the respective memory access priorities.

### Claim 3 (depends on 1)

The system of claim 1, wherein the priority indicators accord to a bit level priority protocol.

### Claim 4 (depends on 1)

The system of claim 1, further comprising a plurality of other memory access handling units at respective different levels of the computing system architecture and configured to generate other memory access requests and respective priority indicators.

### Claim 5 (depends on 1)

The system of claim 1, wherein the memory controller unit of the memory subsystem is further configured to alter one or more of the memory access priorities of one or more of the memory access requests based on a fairness scheduling process.

### Claim 6 (depends on 1)

The system of claim 1, wherein the first level memory access handling unit is further configured to determine one of the first priority indicators of one of the first subset of the instructions based on the user-defined application priorities of respective applications executing on the application level that generated the one of the first subset of the instructions.

### Claim 7 (depends on 1)

The system of claim 1, wherein the first level memory access handling unit is further configured to determine one of the first priority indicators of one of the first subset of the instructions further based on a type of application that generated the one of the first subset of the instructions.

### Claim 8 (depends on 1)

The system of claim 1, wherein the first level memory access handling unit is further configured to render, on a display associated with the system in response to a right-click operation performed on a graphic of an application executing on the application level, an interface display configured to receive, via manual input, application priority information that defines one of the user-defined application priorities for the application, and at least one of the first priority indicators is assigned to one of the first subset of the instructions generated by the application based on the application priority information.

### Claim 9

A computer-implemented method for managing priority to a memory subsystem, comprising:
assigning, by a system comprising a processor, priority indicators to respective instructions originating from respective multiple levels of a computing system architecture, wherein the multiple levels comprise at least an application level, a compiler level, an operating system level, and a micro-architectural level, and the assigning comprises
assigning first priority indicators to a first subset of the instructions originating at the application level based on relative priorities of applications,
assigning second priority indicators to a second subset of the instructions originating at the compiler level based on a determination of whether the second subset of the instructions are associated with shared data or private data, and
assigning third priority indicators to a third subset of the instructions originating at the operating system level based on a determination of whether the third subset of the instructions are associated with an interrupt at the operating system level;
generating, by the system, memory access requests to a memory device, the memory access requests respectively corresponding to the instructions;
determining, by the system, respective priorities for processing the memory access requests based on the priority indicators;
propagating, by the system, the priorities and the memory access requests to the memory subsystem of the computing system architecture; and
prioritizing, by the system, processing of the memory requests based on the priorities.

### Claim 10 (depends on 9)

The computer-implemented method of claim 9, wherein the propagating the priorities and the memory access requests comprises translating one of the instructions originating at a level higher than the memory subsystem to a memory access request.

### Claim 11 (depends on 9)

The computer-implemented method of claim 9, wherein the prioritizing the processing of the memory requests comprises altering an input queue of the memory subsystem based on the priorities to cause a memory request having a first priority to be serviced before other memory requests stored in the input queue that have respective second priorities that are lower than the first priority.

### Claim 12 (depends on 9)

The computer-implemented method of claim 9, wherein the propagating the priorities comprises generating at least one memory access request associated with at least one of the instructions, and wherein one or more bits of the at least one memory access request indicate a priority of the at least one memory access request.

### Claim 13 (depends on 9)

The method of claim 9, further comprising:
in response to a right-click operation performed on an application graphic of one of the applications executing on the application level of the computing system architecture, rendering, by the system, an interface display configured to receive application priority information for the one of the applications; and
receiving, by the system via interaction with the interface display, the application priority information for the one of the applications, the application priority information defining one of the relative priorities for the one of the applications.

### Claim 14 (depends on 9)

The computer-implemented method of claim 9, further comprising altering, by the system, one or more of the priorities for processing one or more of the memory access requests based on a fairness scheduling process.

### Claim 15

A non-transitory computer-readable medium having stored thereon executable instructions that, in response to execution by a processor, causes a system to perform operations, the operations comprising:
receiving instructions that require a read request from a memory device, the instructions originating from multiple levels of a computing architecture, the multiple levels comprising at least an application level, a compiler level, an operating system level, and a micro-architectural level, wherein the instructions have associated priority indicators assigned in accordance with prioritization schemes respectively associated with the multiple levels, and the priority indicators comprise first priority indicators assigned to a first subset of the instructions originating at the application level based on relative priorities of applications that generate the first subset of the instructions, second priority indicators assigned to a second subset of the instructions originating at the compiler level based on a determination of whether the second subset of the instructions are associated with shared data or private data, and third priority indicators assigned to a third subset of the instructions originating at the operating system level based on a determination of whether the third subset of the instructions are associated with an interrupt at the operating system level;
determining respective priorities of memory access requests associated with the instructions based on the priority indicators;
propagating priority information identifying the respective priorities and the memory access requests from higher levels of the computing architecture to lower levels of the computing architecture; and
propagating the priority information and the memory access requests to a memory controller.

### Claim 16 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise:
altering a schedule for processing the memory access requests based on the priority information.

### Claim 17 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the determining the respective priorities comprises:
receiving input representing a level of priority for a computer process that generates one or more of the instructions.

### Claim 18 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the propagating the priority information comprises:
generating at least one of the priority indicators based on a type of application that generated a corresponding at least one of the instructions.

### Claim 19 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein
the operations further comprise rendering an interface display in response to a right-click operation performed on a graphic associated with an application, of the applications, executing on the application level of the computing architecture, the interface display configured to receive application priority information for the application that defines a relative priority for the application, and
one of the first priority indicators is assigned to one of the first subset of the instructions originating from the application based on the application priority information.

### Claim 20 (depends on 15)

The non-transitory computer-readable medium of claim 15, wherein the operations further comprise altering one or more of the priorities of one or more of the memory access requests based on a fairness scheduling process.
