Snaipe

I'm starting to understand the issue. Criterion is behaving correctly: it transfers your parameters verbatim as they are specified in ParameterizedTestParameters to your test function. Unfortunately, this means passing the...

Criterion does exec itself (actually it would be more accurate to say that [BoxFort](https://github.com/diacritic/BoxFort) does). It didn't use to be true, but we had a lot of problems with libraries...

Marking this as wontfix for the moment. I'll leave the issue open to remind me to find something more intuitive for #204.

I think that if such an option was to be included, it shouldn't depend on the ordering. Criterion doesn't guarantee any ordering for tests between versions, and parallelized tests aren't...

Depending on the ordering would also be prevending us from implementing test randomization or run-order optimization in the future, so I'd say implementing a `--rerun-failed` would be the way to...

> As far as I know, there is no standardized way to get the start and the end of the data section, and the change could therefore not make the...

> No, completely eliminating false positives is impossible. Assuming the magic number is 2 bytes, and the checksum 2 bytes, there would we a theoretical chance of around 1 in...

I think going on forward the best bet we have is to scan dynamic symbols for special names (maybe starting with `criterion_user_`) in the executable. This approach will have the...

Interesting. This should be standard POSIX shell scripting, so I'm not sure what changes on OS X's /bin/sh. I don't have an OS X machine handy right now, does their...

Oh, I stand corrected: `local` was proposed for POSIX, but isn't currently part of the standard, so I would wager that this is the problem.