[llvm-dev] Testing End-To-End Functionality of Specific Optimization

[Eli Davis via llvm-dev]

Hello!

I am writing an instrumentation pass using LLVM. I have unit tests using 
llvm-lit in the style of the tests in llvm/test to make sure the IR 
changes show up in the correct place. I would also like to have some 
end-to-end tests that start from a .c file, compile the program, run it, 
and check the output.

I imagine I could do this as a test in the same style, but I'm wondering 
if that's the llvm-correct way of doing it. Can anyone advise?

Thank you!

Eli