Add add_classifier() method to context. Allow plugins to add classifiers
to the current job, or the run as a whole. This will ensure that the new
classifiers are propagated to all relevant current and future artifacts
and metrics.
- Add modules entry to TargetInfo
- When retrieving TargetInfo from cache, make sure info modules match
those for the current target, otherwise mark info as stale and
re-generate.
To provide the same user experience of accessing a directory
artifact from a standard `wa_output` when attempting to retrieve the
path of the artifact extract the stored tar file and extract it to a
temporary location on the host returning the path.
Adds a "requirements.txt" to the project. This will not be used during a
standard installation however will be used to indicate which are known
working packages in cases of conflicts.
Update README and documentaion to reflect this.
As per https://github.com/ARM-software/workload-automation/issues/968
the current documentation for detailing the naming scheme for an revent
recording is unclear. Reword the descriptions focusing on the typical
usecase rather then based on a customized target class.
Although WA supports automatic updating during parsing of a serialized
`kernel_config` from devlib, if the installed versions of WA and devlib
have become out of sync where WA has "updated" the old implementation it
will not attempt to update it again when devlib is later updated to use
the new implementation and therefore will not trigger the existing
checks that are in place.
Allow for specifying a list of supported APK versions for a workload. If
a specific version is no specified then attempt to a resolve any valid
version for the workload.
Update the instructions for installing WA from git not to use
pip as this method does not process dependency_links correctly and
results in an incompatible version of devlib being installed.
Now allows for specifying a `group` value for each section which will
cross product the sections within that group with the sections in each
other group. Additionally classifiers will automatically be added to
each job spec with the relevant group information.
- Update 'setup.py' package name to 'wlauto' as 'wa' is already taken on
PyPi and update documentation accordingly.
- Update old references to Python 2.7
Clarification of the meanings of and the distinction between the
output-related terms iteration, job, workload and section for those
unfamiliar
- Add entries to the glossary and references to the glossary
- Added section on Configuration parameter hierarchy
- Added section on the number and names of directories generated for
jobs in the run output folder
