News for hackers. Stuff that matters.
Re:The C standard formalized in Coq by JabavuAdams on Monday November 30, 2015 @10:01AM (#10649692)
> Surprisingly, correctly describing the set of programs that have undefined behavior in a formal manner is the hardest part of formalizing C.

Isn't this undecidable?

Re:The C standard formalized in Coq by mikeash on Monday November 30, 2015 @11:34AM (#10650204)
Yes, but that doesn't mean you can't describe it.

The halting problem is undecidable, but it's really easy to describe the set of programs that halt. I just described it right there!

That quote is talking about defining the set, not determining whether a program is a member of the set (which is indeed impossible in the general case).

Re:The C standard formalized in Coq by obastani on Monday November 30, 2015 @10:33AM (#10649864)
I think that statement is referring to simply defining "undefined behavior", not actually determining whether a given program has undefined behavior.