# US 2025/0094273 A1 · Processor core fault handling method and apparatus, device, and storage medium

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

- Number: US 2025/0094273 A1
- Document ID: US20250094273A1
- Status: Published application
- Kind: A1
- Application number: 18/970,444
- Filed: 2024-12-05
- Published: 2025-03-20
- Earliest claimed: 2022-06-07
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Jun Wu
- Web page: https://patentgrep.com/patent/US20250094273A1

## CPC

- G06F 9/461: {Saving or restoring of program or task context}
- G06F 11/0793: {Remedial or corrective actions (recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40)}
- G06F 9/4856: {resumption being on a different machine, e.g. task migration, virtual machine migration (G06F9/5088 takes precedence)}
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 11/0724: {in a multiprocessor or a multi-core unit (multiprocessors per se G06F15/80)}
- G06F 11/0751: {Error or fault detection not based on redundancy (power supply failures G06F1/30; network fault management H04L41/06)}
- G06F 11/2028: {eliminating a faulty processor or activating a spare}
- G06F 11/2035: {without idle spare hardware}
- G06F 11/2043: {where the redundant components share a common memory address space}
- G06F 2209/5012: Processor sets
- G06F 2209/5018: Thread allocation
- G06F 2209/508: Monitor
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A processor core fault handling method and apparatus, a device, and a storage medium, where the method includes: performing fault detection on a plurality of cores of a processor to determine a faulty core as a target core; further, updating, based on a target task executed by the target core, tasks executed by the plurality of cores, to enable a core other than the target core in the plurality of cores to execute the target task; and controlling the core other than the target core in the plurality of cores to execute an updated task, and bringing the target core offline.

## Claims

### Claim 1

A method, comprising:
performing fault detection on a plurality of cores of a processor to determine a target core that is faulty, the target core being one of the plurality of cores;
updating, based on a target task executed by the target core, tasks executed by the plurality of cores, to enable a core other than the target core in the plurality of cores to execute the target task; and
controlling the core other than the target core in the plurality of cores to execute an updated task, and bringing the target core offline.

### Claim 2 (depends on 1)

The method according to claim 1, wherein updating, based on the target task executed by the target core, the tasks executed by the plurality of cores, to enable the core other than the target core in the plurality of cores to execute the target task comprises:
updating, based on the target task, core allocation information to indicate that the core other than the target core is executing the target task, wherein the core allocation information indicates the plurality of cores and the tasks executed by the plurality of cores; and
controlling the core other than the target core in the plurality of cores to execute the updated task, and bringing the target core offline comprises:
controlling the core other than the target core in the plurality of cores to execute the updated task based on the updated core allocation information, and bringing the target core offline.

### Claim 3 (depends on 1)

The method according to claim 1, wherein controlling the core other than the target core in the plurality of cores to execute the updated task, and bringing the target core offline comprises:
controlling the plurality of cores to end current tasks, controlling the core other than the target core in the plurality of cores to execute the updated task, and bringing the target core offline.

### Claim 4 (depends on 2)

The method according to claim 2, wherein updating, based on the target task, the core allocation information to indicate that the core other than the target core is executing the target task comprises:
determining, based on the core allocation information, a first core, other than the target core, from the plurality of cores; and
updating, based on the first core and the target task, the core allocation information to indicate that the target task is executed by the first core.

### Claim 5 (depends on 2)

The method according to claim 2, wherein updating, based on the target task, the core allocation information to indicate that the core other than the target core is executing the target task comprises:
determining, based on the core allocation information, a second core from a target core group; and
updating, based on the second core and the target task, the core allocation information to indicate that the target task is executed by the second core, and updating an execution status, in the core allocation information, of a second task executed by the second core to non-execution status.

### Claim 6 (depends on 5)

The method according to claim 5, wherein updating, based on the second core and the target task, the core allocation information to indicate that the target task is executed by the second core, and updating the execution status, in the core allocation information, of the second task executed by the second core to non-execution status comprises one of the following:
updating, based on the second core and the target task, the core allocation information to indicate that the target task is executed by the second core by running a thread based on a target thread identifier, and updating the execution status, in the core allocation information, of the second task to non-execution status, wherein the target thread identifier is a thread identifier of a thread that is run by the target core when the target core executes the target task; or
exchanging, based on the second core and the target task, the target task and the second task in the core allocation information, updating the core allocation information to indicate that the target task is executed by the second core by running the thread based on the target thread identifier, and updating the core allocation information to indicate the second task is creating a thread based on a core identifier of the second core without executing the second task.

### Claim 7 (depends on 5)

The method according to claim 5, wherein controlling the core other than the target core in the plurality of cores to execute the task based on the updated core allocation information comprises:
creating a plurality of threads based on the updated core allocation information and core identifiers of the plurality of cores, wherein a thread identifier of a thread run by the second core is a target thread identifier; and
binding a thread corresponding to the target task in the plurality of threads to the second core, and controlling the core other than the target core in the plurality of cores to run, based on the updated core allocation information, a thread other than a thread corresponding to the second task in the plurality of threads.

### Claim 8 (depends on 2)

The method according to claim 2, wherein updating, based on the target task, the core allocation information to indicate that the core other than the target core is executing the target task comprises:
when the target core belongs to a target core group, updating an execution status of the target task in the core allocation information to non-execution status.

### Claim 9 (depends on 1)

The method according to claim 1, wherein performing fault detection on the plurality of cores of the processor to determine the target core that is faulty comprises:
collecting statistics on cache errors reported by the plurality of cores, to obtain first fault detection results of the plurality of cores, wherein the first fault detection results indicates a quantity of cache errors reported by the target core; and
determining, when the first fault detection results of the target core indicates that the quantity of cache errors reported by the target core reaches a target threshold, that the target core is faulty.

### Claim 10 (depends on 1)

The method according to claim 1, wherein performing fault detection on the plurality of cores of the processor to determine the target core that is faulty comprises:
testing at least one performance indicator for each of the plurality of cores, to obtain second fault detection results of the plurality of cores, wherein the second fault detection results indicates a test result corresponding to the at least one performance indicator of the target core; and
determining, when the second fault detection results of the target core indicates that the test result corresponding to the at least one performance indicator of the target core is inconsistent with a target result, that the target core is faulty.

### Claim 11 (depends on 1)

The method according to claim 1, wherein updating, based on the target task executed by the target core, the tasks executed by the plurality of cores, to enable the core other than the target core in the plurality of cores to execute the target task comprises:
updating, based on the target task executed by the target core, when a quantity of offline cores is less than an offline quantity threshold, the tasks executed by the plurality of cores, to enable the core other than the target core in the plurality of cores to execute the target task.

### Claim 12

A computing device comprising at least one processor and a memory, the at least one processor comprising a plurality of cores, the plurality of cores including a target core, and the memory storing a computer program that, when executed by the at least one processor, causes the computing device to perform:
updating, based on a target task executed by the target core, tasks executed by the plurality of cores, to enable a core other than the target core in the plurality of cores to execute the target task; and
controlling the core other than the target core in the plurality of cores to execute an updated task, and bringing the target core offline.

### Claim 13 (depends on 12)

The computing device according to claim 12, wherein the at least one processor is configured to perform:
updating, based on the target task, core allocation information to indicate that the core other than the target core is executing the target task, wherein the core allocation information indicates the plurality of cores and the tasks executed by the plurality of cores; and
controlling the core other than the target core in the plurality of cores to execute an updated task, and bringing the target core offline comprises:
controlling the core other than the target core in the plurality of cores to execute the updated task based on the updated core allocation information, and bringing the target core offline.

### Claim 14 (depends on 12)

The computing device according to claim 12, wherein the at least one processor is configured to perform:
controlling the plurality of cores to end current tasks, controlling the core other than the target core in the plurality of cores to execute the updated task, and bringing the target core offline.

### Claim 15 (depends on 13)

The computing device according to claim 13, wherein the at least one processor is configured to perform:
determining, based on the core allocation information, a first core other than the target core, from the plurality of cores; and
updating, based on the first core and the target task, the core allocation information to indicate that the target task is executed by the first core.

### Claim 16 (depends on 13)

The computing device according to claim 13, wherein the at least one processor is configured to perform:
determining, based on the core allocation information, a second core from a target core group; and
updating, based on the second core and the target task, the core allocation information to indicate that the target task is executed by the second core, and updating an execution status, in the core allocation information, of a second task executed by the second core to non-execution status.

### Claim 17 (depends on 13)

The computing device according to claim 13, wherein the at least one processor is configured to perform
when the target core belongs to a target core group, updating an execution status of the target task in the core allocation information to non-execution status.

### Claim 18 (depends on 12)

The computing device according to claim 12, wherein the at least one processor is configured to perform
collecting statistics on cache errors reported by the plurality of cores, to obtain first fault detection results of the plurality of cores, wherein the first fault detection results indicates a quantity of cache errors reported by the target core; and
determining, when the first fault detection results of the target core indicates that the quantity of cache errors reported by the target core reaches a target threshold, that the target core is faulty.

### Claim 19 (depends on 12)

The computing device according to claim 12, wherein the at least one processor is configured to perform
testing at least one performance indicator for each of the plurality of cores, to obtain second fault detection results of the plurality of cores, wherein the second fault detection results indicates a test result corresponding to the at least one performance indicator of the target core; and
determining, when the second fault detection results of the target core indicates that the test result corresponding to the at least one performance indicator of the target core is inconsistent with a target result, that the target core is faulty.

### Claim 20

A non-transitory computer readable medium comprising program instructions that, when run on a computing device, cause the computing device to perform:
performing fault detection on a plurality of cores of a processor to determine a target core that is faulty, the target core being one of the plurality of cores;
updating, based on a target task executed by the target core, tasks executed by the plurality of cores, to enable a core other than the target core in the plurality of cores to execute the target task; and
controlling the core other than the target core in the plurality of cores to execute an updated task, and bringing the target core offline.
