# US 2024/0004767 A1 · Data processing systems

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

- Number: US 2024/0004767 A1
- Document ID: US20240004767A1
- Status: Published application
- Kind: A1
- Application number: 18/251,602
- Filed: 2021-11-04
- Published: 2024-01-04
- Earliest claimed: 2020-11-04
- Assignees: Arm Limited
- Applicants: Arm Limited
- Inventors: Daniel James KERRY; David Thomas GARBETT; Jussi Tuomas PENNALA; Nicholas John Nelson MURPHY
- Web page: https://patentgrep.com/patent/US20240004767A1

## CPC

- G06F 11/079: {Root cause analysis, i.e. error or fault diagnosis (in a hardware test environment G06F11/22; in a software test environment G06F11/36)}
- G06F 11/2289: {by configuration test}
- G06F 11/2236: {to test CPU or processors}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A data processing system (1) comprises a plurality of, e.g. graphics, processing units (11), and a management circuit (12) associated with the processing units and operable to configure the processing units of the plurality of processing units into respective groups of the processing units. The management circuit (12) is configured to always operate with a high level of fault protection, but the groups of the processing units can be selectively operated with either a higher level of fault protection or a lower level of fault protection, by selectively subjecting them to fault detection testing (60).

## Claims

### Claim 1

A data processing system comprising:
a plurality of processing units;
a management circuit associated with the processing units and operable to configure the processing units of the plurality of processing units into respective groups of the processing units, wherein each group of processing units comprises a set of one or more of the processing units of the plurality of processing units;
wherein:
the management circuit is configured to always operate with a higher level of fault protection;
and
groups of the processing units can be selectively operated with either of at least two modes of fault protection, with one mode providing a higher level of fault protection than the other mode.

### Claim 2

(canceled)

### Claim 3 (depends on 1)

The system of claim 1, comprising a fault detection testing mechanism that may be selectively performed for a group of processing units, whereby the groups of the processing units can be selectively operated with either a higher level of fault protection or a lower level of fault protection.

### Claim 4

(canceled)

### Claim 5 (depends on 3)

The system of claim 3, wherein the system comprises a plurality of arbiters, each arbiter being operable to control access by virtual machines that require processing operations to the processing units of a group of processing units that the arbiter has been allocated; and
the fault detection testing for a group of processing units is performed under the control of an arbiter for the group of processing units.

### Claim 6 (depends on 5)

The system of claim 5, wherein respective arbiters of the system are configured either to perform or to not perform fault detection testing for a group of processing units that they have been allocated.

### Claim 7 (depends on 6)

The system of claim 6, wherein the management circuit is operable to assign a group of processing units to an arbiter for that group of processing units, and a group of processing units is configured to be operated at a higher level of fault protection by assigning it to an arbiter that performs the fault detection testing, and a group of processing units is configured to be operated with a lower level of fault protection by assigning it to an arbiter that does not perform the fault detection testing.

### Claim 8 (depends on 3)

The system of claim 3, wherein the management circuit comprises one or more fault detection testing control interfaces that can be set to trigger fault detection testing for processing units of a group of processing units.

### Claim 9 (depends on 3)

The system of claim 3, wherein the processing units of a group of processing units are themselves able to be configured as respective partitions of processing units within the group, with each partition comprising a subset of one or more of the processing units of the group; and
the fault detection testing for a group of graphics processing units can be triggered and performed on a partition-by-partition basis.

### Claim 10

(canceled)

### Claim 11

(canceled)

### Claim 12 (depends on 1)

The system of claim 1, wherein:
each processing unit can be powered on and off independently of any of the other processing units; and/or
each processing unit can be reset independently of any of the other processing units.

### Claim 13 (depends on 1)

The system of claim 1, further comprising:
a controller operable to control the management circuit to configure the processing units of the plurality of processing units into respective groups of the processing units; and
a plurality of arbiters, each arbiter being operable to control access by virtual machines that require processing operations to the processing units of a group of processing units that the arbiter has been allocated;
wherein
the controller executes on one processor of the system, and the arbiters and the virtual machines execute on a different processor or processors to the controller.

### Claim 14 (depends on 13)

The system of claim 13, wherein the controller is configured to communicate with the management circuit over a communications bus that is not accessible to any of the virtual machines that may require processing by the processing units, nor to the arbiters for the groups of processing units.

### Claim 15 (depends on 1)

The system of claim 1, comprising:
a first processor cluster that is configured to operate in a safety critical manner and that executes a safety-critical arbiter and a group of one or more safety critical virtual machines, and that communicates with the processing units via a first communications bus; and
a second processor cluster that is other than configured to operate in a safety-critical manner, and that executes a non-safety-critical arbiter and a group of one or more non-safety-critical virtual machines, and that communicates with the processing units via a second, different communications bus.

### Claim 16

A method of operating a data processing system, the data processing system comprising:
a plurality of processing units; and
a management circuit associated with the processing units and operable to configure the processing units of the plurality of processing units into respective groups of the processing units, wherein each group of processing units comprises a set of one or more of the processing units of the plurality of processing units;
the method comprising:
operating the management circuit with a higher level of fault protection; and
the management circuit configuring the groups of the processing units such that:
at least one of the groups of the processing units will be operated with a higher level of fault protection; and
at least one other group of the processing units will be operated with a lower level of fault protection.

### Claim 17

(canceled)

### Claim 18 (depends on 16)

The method of claim 16, further comprising selectively performing fault detection testing for the groups of processing units, whereby the groups of the processing units can be selectively operated with either a higher level of fault protection or a lower level of fault protection.

### Claim 19

(canceled)

### Claim 20 (depends on 18)

The method of claim 18, wherein the system comprises a plurality of arbiters, each arbiter being operable to control access by virtual machines that require processing operations to the processing units of a group of processing units that the arbiter has been allocated; and
the fault detection testing for a group of processing units is performed under the control of an arbiter for the group of processing units.

### Claim 21 (depends on 20)

The method of claim 20, wherein:
respective arbiters of the system are configured either to perform or to not perform fault detection testing for a group of processing units that they have been allocated;
and the management circuit configures a group of processing units to be operated at a higher level of fault protection by assigning it to an arbiter that performs the fault detection testing, and configures a group of processing units to be operated with a lower level of fault protection by assigning it to an arbiter that does not perform the fault detection testing.

### Claim 22

(canceled)

### Claim 23 (depends on 18)

The method of claim 18, wherein the processing units of a group of processing units are themselves able to be configured as respective partitions of processing units within the group, with each partition comprising a subset of one or more of the processing units of the group;
and the method comprises performing the fault detection testing for a group of graphics processing units on a partition-by-partition basis.

### Claim 24 (depends on 23)

The method of claim 23, comprising, for a processing unit partition that is to be tested, making the partition unavailable to any virtual machines for processing operations, then testing the partition, and then returning the partition to being available for use by virtual machines for processing operations.

### Claim 25 (depends on 16)

The method of claim 16, wherein:
the data processing system comprises:
a controller operable to control the management circuit to configure the processing units of the plurality of processing units into respective groups of the processing units; and
a plurality of arbiters, each arbiter being operable to control access by virtual machines that require processing operations to the processing units of a group of processing units that the arbiter has been allocated;
and the method comprises:
the controller communicating with the management circuit over a communications bus that is not accessible to any of the virtual machines that may require processing by the processing units, nor to the arbiters for the groups of processing units.

### Claim 26 (depends on 16)

The method of claim 16, wherein:
the data processing system comprises:
a first processor cluster that is configured to operate in a safety critical manner; and
a second processor cluster that is other than configured to operate in a safety-critical manner;
and the method comprises:
assigning one group of processing units to the first processor cluster and one group of processing units to the second processor cluster; and
operating the group of processing units assigned to the first processor cluster with a higher level of fault protection, but operating the group of processing units assigned to the second processor cluster with a lower level of fault protection.

### Claim 27

A non-transitory computer readable storage medium storing computer software code which when executing on one or more processors performs a method of operating a data processing system, the data processing system comprising: a plurality of processing units; and a management circuit associated with the processing units and operable to configure the processing units of the plurality of processing units into respective groups of the processing units, wherein each group of processing units comprises a set of one or more of the processing units of the plurality of processing units; the method comprising:
operating the management circuit with a higher level of fault protection; and
the management circuit configuring the groups of the processing units such that at least one of the groups of the processing units will be operated with a higher level of fault protection;
and at least one other group of the processing units will be operated with a lower level of fault protection.
