Snaipe
lldb-server doesn't always ship with lldb, and some other systems have clang+gdb without lldb. It would be better to fallback to gdbserver if lldb-server isn't available.
The windows implementation is lacking some job control to kill child processes on exit. There should also be a workaround for POSIX systems without PR_SET_PDEATHSIG.
BoxFort should propagate `SIGTSTP` and `SIGCONT` to the child workers.
This is a catch-all issue to track the inclusion of Criterion on various distributions, and any user-maintained alternatives in the meantime. It's a major headache to maintain myself unofficial/user packages...
Using the `stream` and `mem` tags to test against files is a bit of a hassle currently. It's not unusable, but it forces users to write boilerplate code, which Criterion...
On the topic of shaving preprocessor branches, I think the simplest thing to do here, since we only expose these format specifiers through cr_asprintf which in turn uses whatever vsnprintf...
pb_decode takes a lot of the running time in a criterion run. This used to make cr_assert worse (see #172). I think there's room for improvement here, which might require...
So far, Criterion spawns a whole new process per test. This makes things simple implementation-wise, but it's otherwise not that great on a performance standpoint. execve()s are pretty fast these...
The current API is a bit of a macro hell, which is somewhat a necessary evil to get a nice UX. I've been thinking of ways to keep things sane...
In hindsight I think that ParameterizedTest was kind of a mistake. I think that the concept should be merged with Test directly, with more options wrt parameter generation: - criterion...