[llvm-commits] [PATCH] Atomic NAND options

[Chris Lattner]

On Jul 14, 2011, at 10:51 AM, David Blaikie wrote:

> 
> Yes, I think that is very reasonable.  It is a very rare builtin for people to use.  Having clang warn unconditionally (with a -W flag) seems like the right answer.
> 
> That sounds like it'd be a little weird/annoying for new use cases, wouldn't it? (unless the old behavior is the obvious/expected behavior even for new users)
> Is this the sort of thing you'd keep around for one release & then remove with the following one?

Yes, we'd keep it around for a while, then remove it eventually.

-Chris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20110714/886abdf9/attachment.html>