mirror of
https://github.com/ARM-software/workload-automation.git
synced 2025-11-14 13:56:02 +00:00
0656e61bc2e38fd2a2d652e423397c932aa46406
The Corporate version is a specialised version of Geekbench. It has
different package names and does not require dismissing a EULA. The
new corporate version is added as a distinct
benchmark ("geekbench-corporate" vs "geekbench").
Note that this changes the wait-for-results UiAutomator snippet from
looking for "Running Benchmarks..." to "*Running*". This is because
the version I've tested updates the text widget with the name of each
benchmark phase as it is run. However I don't know if this is a
feature of the Corporate version or simply of Geekbench 4.1.0.
Workload Automation
+++++++++++++++++++
Workload Automation (WA) is a framework for executing workloads and collecting
measurements on Android and Linux devices. WA includes automation for nearly 50
workloads (mostly Android), some common instrumentation (ftrace, ARM
Streamline, hwmon). A number of output formats are supported.
Workload Automation is designed primarily as a developer tool/framework to
facilitate data driven development by providing a method of collecting
measurements from a device in a repeatable way.
Workload Automation is highly extensible. Most of the concrete functionality is
implemented via plug-ins, and it is easy to write new plug-ins to support new
device types, workloads, instrumentation or output processing.
Requirements
============
- Python 2.7
- Linux (should work on other Unixes, but untested)
- Latest Android SDK (ANDROID_HOME must be set) for Android devices, or
- SSH for Linux devices
Installation
============
To install::
python setup.py sdist
sudo pip install dist/wlauto-*.tar.gz
Please refer to the `installation section <./doc/source/installation.rst>`_
in the documentation for more details.
Basic Usage
===========
Please see the `Quickstart <./doc/source/quickstart.rst>`_ section of the
documentation.
Documentation
=============
You can view pre-built HTML documentation `here <http://pythonhosted.org/wlauto/>`_.
Documentation in reStructuredText format may be found under ``doc/source``. To
compile it into cross-linked HTML, make sure you have `Sphinx
<http://sphinx-doc.org/install.html>`_ installed, and then ::
cd doc
make html
License
=======
Workload Automation is distributed under `Apache v2.0 License
<http://www.apache.org/licenses/LICENSE-2.0>`_. Workload automation includes
binaries distributed under differnt licenses (see LICENSE files in specfic
directories).
Feedback, Contrubutions and Support
===================================
- Please use the GitHub Issue Tracker associated with this repository for
feedback.
- ARM licensees may contact ARM directly via their partner managers.
- We welcome code contributions via GitHub Pull requests. Please see
"Contributing Code" section of the documentation for details.
Description
Languages
Python
76.1%
Java
15.7%
C
5.4%
Shell
2.4%
Dockerfile
0.3%