Stefan Koch

@WalterBright when you remove the risky label please state why the PR is not risky anymore. The reason I added the risky label was because of Iains comment. > Genuine...

@thewilsonator not on it's own. There is a work in progress PR however that want's to use this. https://github.com/dlang/dmd/pull/11101

There's no way to mark functions compile-time only in the current language. But it's going to come.

@MoonlightSentinel It is needed for type functions. and `assert(__ctfe)` (when I revive that). That's because you're using the template in a specutlative context. The templates instance in not generated for...

It would be nice to have. The discussion is in a stalemate though I think.

I agree that an if is to be preferred to a static if whenever possible. I am not sure however if you want unreachable warnings to be ignore for static...

`src/object.d(2215): Error: function `object.ModuleInfo.name` no `return exp;` or `assert(0);` at end of function` yep. sure has unintended consequences.

so .... thinking about this again, in theory we should be able to constant-fold everything around `__ctfe` in cases where marking the scope can be done and/or simplifying the expression...

@ibuclaw since you don't like ctfe to be another langauge, we could "fake" a malloc call at ctfe.

This needs an explanation, what's the reasoning behind this?