Add more granular pipelines for mobile configurations

[steveisok]

When trying to validate a specific mobile change, we'll manually trigger runtime-extra-platforms to run all tests. This results in running a large swath of configurations that, more often than not, do not need to run. The goal of this issue is to offer more granular pipelines by configuration and determine if we should go deeper (libraries vs runtime tests, for example).

All libraries and runtime tests

• Android emulators
• Android devices
• iOS/tvOS simulators
• iOS/tvOS devices
• MacCatalyst

#74859 can be used for inspiration

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

Tagging subscribers to this area: @directhex
See info in area-owners.md if you want to be subscribed.

Issue Details

---

When trying to validate a specific mobile change, we'll manually trigger runtime-extra-platforms to run all tests. This results in running a large swath of configurations that, more often than not, do not need to run. The goal of this issue is to offer more granular pipelines by configuration and determine if we should go deeper (libraries vs runtime tests, for example).

All libraries and runtime tests

• Android emulators
• Android devices
• iOS/tvOS simulators
• iOS/tvOS devices
• MacCatalyst

#74859 can be used for inspiration

Author:	steveisok
Assignees:	directhex
Labels:	area-Infrastructure-mono
Milestone:	-

[directhex]

So we want

a) not smoke-tests-only. These new pipelines should run all tests. or only run all tests when called directly, otherwise run smoke tests when run as a template include into runtime-extra-platforms?

b) only move the things we're testing in runtime-extra-platforms (e.g. interp vs not, sandbox vs not, AOT vs not, etc etc etc) - don't touch things we're testing in other pipelines like runtime.yml?

[directhex]

@steveisok ^^ ?

[steveisok]

For mobile, we want to take what's in runtime-extra-platforms and split each configuration into different pipelines.

To answer your questions directly:

a.) yes, the pipelines should run all tests when triggered manually. We don't want to run these on a schedule.

b.) correct

[carlossanlop]

@dotnet/area-infrastructure-libraries should we consider doing something similar for runtime libraries legs?