<div class="gmail_quote">On Wed, Apr 11, 2012 at 4:51 PM, David Chisnall <span dir="ltr"><<a href="mailto:csdavec@swan.ac.uk">csdavec@swan.ac.uk</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On 11 Apr 2012, at 16:46, Chandler Carruth wrote:<br>
<br>
> It seems like this should get an ext-warn...<br>
<br>
</div>In which case we'd get a load of them for anything using libc++.<br></blockquote><div><br></div><div>No? System headers suppress warnings fer exactly this reason.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">> Is there not an existing test for _Atomic as used in C++ that this could be folded into?<br>
<br>
</div>$ grep _Atomic test/CodeGenCXX/*<br>
atomicinit.cpp: _Atomic(int) i;<br>
<br>
So, no. This test should be growing a bit soon, once I fix the issue with __atomic_init() with C++ types.<br></blockquote><div><br></div><div>Ahh, maybe rename it to c11_atomics.cpp or something to clarify that its for testing the interaction between C++ and C11 atomics? Anyways, just trying to keep the growth of test files reasonable.</div>
<div><br></div><div>Thanks!</div><div>-Chandler</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class="HOEnZb"><font color="#888888"><br>
David</font></span></blockquote></div><br>