<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Thu, May 26, 2016 at 12:41 AM, David Majnemer <span dir="ltr"><<a href="mailto:david.majnemer@gmail.com" target="_blank">david.majnemer@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-style:solid;border-left-color:rgb(204,204,204);padding-left:1ex"><span>In <a href="http://reviews.llvm.org/D18035#434095" rel="noreferrer" target="_blank">http://reviews.llvm.org/D18035#434095</a>, @DmitryPolukhin wrote:<br>
<br>
> One more friendly ping...... :(<br><br>
</span>I think the best way to make progress on this is to refactor this patch along the lines @rsmith suggested back on May 3.<br></blockquote><div><br></div><div>To be honest I still don't understand what is the key difference between these two approaches. Both assume that compiler mangles something twice: one for collecting implicit abi_tags without outputting name to real stream and once more time mangling name for real with implicit tags collected on previous stage. I tested my patch on Richard's example and explained why it works as it should be.</div></div></div></div>