[llvm-commits] PATCH: Refactoring and simplification of inline cost computation w.r.t. alloca arguments

[Eric Christopher]

> 
> No, i've bootstrapped clang and it made no difference in either. I don't really expect it to except in relatively rare circumstances.
> 

OK.

> Is this needed before I check in? I was hoping to just look for regressions in the nightly testing rather than trying to run each incremental step through the test suite.... Hopefully doing the latter will be much easier in the future.

Probably not, but nightly testing doesn't look at overall binary size so wouldn't catch odd explosions unless it also affected runtime negatively.

> I've attached a version of the patch with a testcase now. =]
> 
> > This last patch I'm a bit hesitant about. I don't have any test cases for it, and during my changes I made several typos and bugs that didn't actually cause any tests to fail. I feel like the testing here is... rather lighter than it should be. I don't want to commit this one without test cases, can anyone provide some?
> 
> Sorry for confusing wording, this was referring to the second patch.
> <0002-Undo-a-previous-restriction-on-the-inline-cost-calcu.patch>

Yeah, still no real objections to the patch. I have the same "lack of testing" worries that you do here.

-eric