JLink

The jlink boot method is used to flash and boot devices using the J-Link debug probe from SEGGER. It supports flashing single or multiple images and can interact with devices that have multiple cores.

yaml - boot: method: jlink commands: - loadfile {shell} 0x0 prompts: - 'SHELL>>' timeout: minutes: 2

Note

Images passed to JLink commands must be first deployed using the tmpfs deploy action.

Installation

jlink must be installed manually on the LAVA worker since it is not provided as a dependency by LAVA. See JLink Software and Documentation Pack for installing the tool.

Configuration

For adding a new device type, see JLink device type.

For adding a new device, refer to JLink device configuration.

commands

Optional. A list of JLink commands to execute. These can include loadfile or any other valid JLink command. Variables in the commands reference images deployed during the prior tmpfs deploy action.

Flashing signal image

When only a single image deployed with tmpfs deployment method and the commands is not specified in this boot method, LAVA flashes the single image to the default load address 0x0.

yaml - boot: method: jlink timeout: minutes: 2

Flashing multiple images

yaml - boot: method: jlink commands: - loadfile {shell} 0x0 - loadfile {hello_world} 0x0

In this example, shell and hello_world are the image keys used in the tmpfs deploy action.

coretype

For devices with multiple cores, specifies which core to connect to. The available core types are defined by supported_core_types in the device type template. If not specified, the first core in the list is used.

yaml - boot: method: jlink coretype: M7 commands: - loadfile {test} prompts: - 'SHELL>>' timeout: minutes: 2

prompts

See prompts.

timeout

See timeouts.

Example job

```yaml device_type: frdm-kw36zj job_name: health-check

timeouts: job: minutes: 10 action: minutes: 3

priority: medium visibility: public

actions: - deploy: to : tmpfs images : boot : url: https://example.com/frdm-kw36-shell.bin

• boot: method: jlink

• test: monitors:

  • name: tests start: Running test suite common_test end: PROJECT EXECUTION SUCCESSFUL pattern: '(?P(PASS|FAIL)) - (?P.*).' fixupdict: PASS: pass FAIL: fail ```

Using multiple MCUs

To set up a test requiring multiple MCUs, such as testing Bluetooth connectivity, you can use a multinode job. In this setup, MCU 1 and MCU 2 are placed on the same worker, and a Docker device is used to execute tests once both MCUs have completed their deployment and boot processes.

The test setup consists of:

• MCU 1 and MCU 2: These are the target devices to be tested.
• Docker Device: A Docker container is deployed to execute the tests. This Docker device is responsible for coordinating the test execution once both MCUs are ready.

Here’s how the process works:

1. Deployment and Boot: The necessary images are deployed to both MCUs using tmpfs. Once deployed, the MCUs are booted to initiate the test process.
2. Test Execution: Once both MCUs have completed their boot processes, the Docker device executes the test suite.
3. Results Analysis: The test results are collected and analyzed by a LAVA test to verify the functionality of the MCUs and ensure successful connectivity.

Example job:

```yaml job_name: Multiple MCUs

timeouts: job: minutes: 5 action: minutes: 5 connection: minutes: 2

priority: medium visibility: public

protocols: lava-multinode: roles: host: device_type: rw610bga count: 1 tags: - rw610bga-fr01 timeout: minutes: 30 guest: device_type: rw610bga tags: - rw610bga-fr02 count: 1 timeout: minutes: 30 docker: device_type: docker count: 1

actions: - deploy: role: - host - guest to : tmpfs images : boot : url: https://example.com/bt_ble_cli.out cpu1 : url: https://example.com/rw61xw_raw_cpu1_a1.bin cpu2 : url: https://example.com/rw61xn_raw_cpu2_ble_a1.bin combo : url: https://example.com/rw61xn_combo_raw_cpu2_ble_15_4_combo_a1.bin

• deploy: role:

  • docker to: docker os: ubuntu image: name: connectivity_test local: true

• boot: role:

  • host
  • guest method: jlink commands :
  • loadfile {boot}
  • loadfile {cpu1} 0x8400000
  • loadfile {cpu2} 0x8540000
  • loadfile {combo} 0x85E0000 timeout: minutes: 2

• boot: role:

  • docker method: docker command: bash prompts:
  • 'root@\w:(.)#' timeout: minutes: 2 ```