[llvm-dev] [cfe-dev] Adding sanity to the Atomics implementation

[Joerg Sonnenberger via llvm-dev]

On Thu, Jan 28, 2016 at 08:32:31AM -0800, Reid Kleckner via llvm-dev wrote:
> I think Clang should continue to duplicate this information, the same way
> we duplicate target datalayout strings. Other than that, sure, we can let
> LLVM expand IR operations to libcalls. I don't immediately see a problem
> with that.

Note that a libcall doesn't necessarily mean using locks. With one
exception, NetBSD provides lock-free CAS on all platforms for natural
argument sizes. The intersection between platforms with MP support and
platforms without hardware CAS is SPARCv8. Everything else works fine,
even the libcall version.

Joerg