documentation: Unclear what the difference is between TFAIL and TBROK
Reading the documentations for TFAIL and TBROK its not clear that the difference is.
https://linux-test-project.readthedocs.io/en/latest/developers/api_c_tests.html#core-ltp-api
If that can be expanded a bit more, it would be great.
Unfortunately we haven't finished transition from Github wiki to readthedocs. Github wiki had more info, which is as backup in https://github.com/linux-test-project/ltp/blob/master/doc/old/, e.g.:
https://github.com/linux-test-project/ltp/blob/fc0b2b471bf232bdd6f46bf43c0e118f3778da3b/doc/old/C-Test-API.asciidoc
TBROK Something has failed in test preparation phase.
We should finish converting the docs soon.
I don't know what the current stance is, but back when I did more programming in this space TFAIL (to me) was considered an actual invariants failure, whereas TBROK (to me) represented a failure where the exact result of a test could not be known due to unexpected failed preconditions, e.g., failure to create a directory, load a module, etc. In more modern/traditional test frameworks, TBROK might be expressed with TSKIP and variously scoped fixtures, e.g., package, module, suite, or test, if you are familiar with GoogleTest or pytest.
I've send a patch that hopefully clarifies it:
http://patchwork.ozlabs.org/project/ltp/list/?series=425706