🕵️ SicherheitslückenHak5: Hackers Just Poisoned the Rust Supply Chain | Threat Wire(01.09.2026 um 14:00 Uhr)
🕵️ SicherheitslückenHak5: Hackers Found a Way Into Humanoid Robots | Threat Wire(04.09.2026 um 15:04 Uhr)
🔧 AI Nachrichten Bits und so #1021 (Passwort für Laufwerk)(31.08.2026 um 22:15 Uhr)
🔧 AI Nachrichten Bits und so #1022 (Wie Weißbier)(06.09.2026 um 20:39 Uhr)
🍏 iOS / Mac OSHue-App 6.0 ist da: das sind die Neuerungen(07.09.2026 um 17:21 Uhr)
🕵️ SicherheitslückenHak5: Hackers Just Poisoned the Rust Supply Chain | Threat Wire(01.09.2026 um 14:00 Uhr)
🕵️ SicherheitslückenHak5: Hackers Found a Way Into Humanoid Robots | Threat Wire(04.09.2026 um 15:04 Uhr)
🔧 AI Nachrichten Bits und so #1021 (Passwort für Laufwerk)(31.08.2026 um 22:15 Uhr)
🔧 AI Nachrichten Bits und so #1022 (Wie Weißbier)(06.09.2026 um 20:39 Uhr)
🍏 iOS / Mac OSHue-App 6.0 ist da: das sind die Neuerungen(07.09.2026 um 17:21 Uhr)

🎥 Künstliche Intelligenz Videos 🕛 kürzlich 7 Min Lesezeit
0

Integrating Arm Virtual Hardware with the TensorFlow Lite Micro Continuous Integration Infrastructure

↗ Quelle (blog.tensorflow.org)
🗣️ Stimme:
📑 Inhaltsübersicht

A guest post by Matthias Hertel and Annie Tallund of Arm

.

Additionally, with well-established technologies like with almost no integration effort. Combining these two offers a great variety of platforms and configurations. In this article, we will describe how we have collaborated with the TFLM team to use Arm Virtual Hardware (AVH) as part of the TFLM projects open-source continuous integration (CI) framework to verify many Arm-based processors with TFLM. This enables developers to test their projects on Arm intellectual property IP without the additional complexity of maintaining hardware.

Arm Virtual Hardware - Models for all Cortex-M microcontrollers

, ,

  • , . The full list of supported platforms can be found ) can be used to run code on an AWS EC2 instance, which has Arm IP installed. This workflow builds the TFLM project with Corstone-300 as a target, and runs the unit tests using both GCC and armclang, displaying the results directly in the GitHub UI via a hierarchical process as visualized below.

    Demo game play in ‘Plane Strike’

    The workflow contains one or more jobs, which points to a file containing steps. The steps are defined in a separate file (), which is sent using an Arm-provided API ( using pip into any environment running Python. From there it can offload any compilation and test job onto Arm Virtual Hardware. For our Corstone-300 example, it is installed on the GitHub Actions runner by adding a pip install to the workflow file.


        - name: Install AVH Client for Python

        run: |
          pip install git+https://github.com/ARM-software/[email protected] 

    The AWS credentials are configured to allow the AVH client to connect to the AWS EC2 instance, though there are various other ways to authenticate with AWS services. These include adding the AWS keypair onto GitHub secrets, or using an allow-listed GitHub repository to ordinate a predefined role, as shown here.

    - name: Configure AWS Credentials
        uses: aws-actions/configure-aws-credentials@v1
        with:
          role-to-assume: arn:aws:iam::720528183931:role/Proj-vht-assume-role
                aws-region: eu-west-1 

    Defining and executing a workload

    Finally, let’s look at how the workload itself is executed using the AVH client. In this example, the AVH workload is described in a YAML file which we point to in the Github workflow file.

    - name: Execute test suite on Arm Virtual Hardware at AWS
        run: |
          avhclient -b aws execute --specfile   ./tensorflow/lite/micro/tools/github/arm_virtual_hardware/cortex_m_generic_avh.yml 


    This is where we define a list of steps to be executed. The steps will point to an inventory of files to be transferred, like the TFLM repository itself. Additionally, we define the code that we want to execute using these files, which can be done through the script that we provided earlier.

    steps:
      - run: |
      git clone https://github.com/tensorflow/tflite-micro.git 
      mv ./tflite-micro/TensorFlow/ .
      TensorFlow/lite/micro/tools/ci_build/test_cortex_m_corstone_300.sh armclang &> ./corstone300.log 


    Next, we set up a list of files to copy back to the GitHub Actions runner. For the TFLM unit test, a complete command line log will be written to a file, corstone300.log – that is returned to the GitHub Actions runner to analyze the test run outcome:

    - name: Fetch results from Arm Virtual Hardware
        run: |
          cat ./tensorflow/lite/micro/tools/github/arm_virtual_hardware/cortex_m_generic.log 


    You can find a detailed explanation of avhclient and its usage on the .

    Expanding the toolbox by adding more hardware targets

    Using AVH it is easy to extend tests to all available Arm platforms. You can also avoid a negative impact on the overall CI workflow execution time by hosting through cloud services like AWS and spawning an arbitrary number of AVH instances in parallel.

    Virtual Hardware targets like the Corstone-310 demonstrate how software validation is feasible even before silicon is available. This will make well-tested software stacks available for new Cortex-M devices from day one and we plan to expand the support. The introduction of Corstone-1000 will extend the range of tested architectures into the world of Cortex-A application processors, including , Cortex-A53.

    Wrapping up

    To summarize: by providing a workflow file, a use-case file, and a workload (in our case, a test script), we have enabled running all the TFLM unit tests on the Corstone-300 and will work to extend it to all AVH targets available.

    Thanks to the AVH integration, CI flows with virtual hardware targets open up new possibilities. Choosing the right architecture, integrating, and verifying has never been easier. We believe it is an important step in making embedded ML more accessible and that it will pave the way for future applications.

    Thank you for reading!

    Acknowledgements

    We would like to acknowledge a number of our colleagues at Arm who have contributed to this project, including Samuel Peligrinello Caipers, Fredrik Knutsson, and Måns Nilsson.

    We would also like to thank Advait Jain from Google and John Withers of Berkeley Design Technology, Inc. for architecting a continuous integration system using GitHub Actions that has enabled the Arm Virtual Hardware integration described in this article.
    Vollständiger Original-Bericht
    Ausführliche Details, Code-Beispiele & Hersteller-Stellungnahme auf blog.tensorflow.org.
    ↗ Original-Artikel auf blog.tensorflow.org lesen
    Wie bewertest du diesen Beitrag?
    1 Klick Feedback
    Teilen mit Netzwerk & Team:

    Community-Analysen & Experten-Meinungen 0

    Verfasse deine eigene Analyse, teile Workarounds oder diskutiere diesen Vorfall im Blog.
    Noch keine Community-Analyse verfasst. Markiere einen Textabschnitt oder klicke oben auf Eigene Analyse verfassen“!
    Community Pulse: Relevanz-Einschätzung
    1 Klick Experten-Votum
    🔴 Akute Relevanz 0%
    🟡 In Evaluierung 0%
    🟢 Keine Auswirkung 0%
    Spannende Innovation 0%
    Verwandte Story-Cluster & Quellen (Vektor-KI)
    Port 8095 Engine
    1 Quelle
    Hackers Just Poisoned the Rust Supply Chain | Threat Wire
    1 Quelle
    Hackers Found a Way Into Humanoid Robots | Threat Wire
    1 Quelle
    Bits und so #1021 (Passwort für Laufwerk)
    Ähnliche Beiträge
    🔍 Verwandte News

    Auch interessante Nachrichten Integrating Arm Virtual Hardware with the TensorFlow Lite Micro Continuous Integration Infrastructure

    Thematisch verwandte Begriffe: Integrating, Virtual, Hardware, with · 6 Treffer

    Laden...

    Videos werden geladen ...

    Laden...

    Beiträge werden geladen ...

    Laden...

    Videos werden geladen ...

    Laden...

    Beiträge werden geladen ...

    Laden...

    Videos werden geladen ...

    Laden...

    Beiträge werden geladen ...

    Laden...

    Videos werden geladen ...

    Laden...

    Beiträge werden geladen ...

    Laden...

    Videos werden geladen ...