Oh, @kees_cook, that's not how undefined behavior works, you know...https://twitter.com/ScottyBauer1/status/673999260463960065 …
@jamey_sharp @ScottyBauer1 https://sourceware.org/bugzilla/show_bug.cgi?id=7075 … "undefined" rarely is. it HAS a behavior, and people end up depending on it.
-
-
@kees_cook@ScottyBauer1 Sure, people keep making this argument about UB, but trends suggest it's a losing battle. Better to find and fix. -
@jamey_sharp@ScottyBauer1 my point is it's not undefined: it is defined. either leave it as-is or fix all callers and switch the behavior -
@kees_cook@ScottyBauer1 Um. I don't think the word "undefined" means what you think it means, then? cf the comments in the bug you linked. -
@jamey_sharp@ScottyBauer1 POSIX spec is undefined. The implementation isn't. You can't change implementation without changing all callers. -
@kees_cook@ScottyBauer1 Definitional debates aside, in that case, maybe you shouldn't push back when people do try to fix all the callers? -
@jamey_sharp@ScottyBauer1 are those all the callers? there's plenty of out-of-tree stuff. maybe a runtime WARN, sure.
End of conversation
New conversation -
-
-
@kees_cook How do we protect against compilers who decided to optimize on that UB somewhere down the line? R tests sufficient? -
@ScottyBauer1 well, since people actually depend on the existing behavior, compilers changing the behavior will be seen as buggy. :(
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.