I saw the other day that c++14 was ratified. I assume that clang support for c++14 is complete with clang 3.4 (looking at the website status page) and we are good to start using it for production work? is the “-std=c++1y” switch going to change to “-std=c++14” now?


The switch happened in r215982, and it is currently being discussed
whether that change will be merged into the 3.5 branch.


Hi James,

Take a look at r215982


great! thanks!

Is everything ready for production work? I can turn it on and expect nothing to break going forward?


Anything that isn't ready for production work is a bug and should be
reported. :slight_smile:


thanks! since c++14 is now official, I’m switching it on for my company’s stuff.

--- end quoted text ---

Hi James,
Louis Dionne, author of the boost hana metaprogramming library, has found c++14
related bugs in clang 3.4 that cause clang 3.4 to segfault. Louis says those
bugs are fixed in clang 3.5. I am not sure if he reported those bugs here, or
if the llvm/clang usage patterns would find those bugs even though they exist.
But, apparently there are some limitations to the support for the c++14 dialect
in clang 3.4.


thanks for the heads-up! As it turns out, I’m using a recent trunk build of clang and I’ll be upgrading it to 3.5 once 3.5 is released. Hopefully, those issues will be resolved by 3.5 final version. The only feature I immediately plan to use is lambda function move-capturing. I’ve been waiting for that for a while.