<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div><blockquote type="cite" class=""><div class="">On Feb 9, 2015, at 11:48 AM, Fiona Glaser <<a href="mailto:fglaser@apple.com" class="">fglaser@apple.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><meta http-equiv="Content-Type" content="text/html charset=utf-8" class=""><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><br class=""><div class=""><blockquote type="cite" class=""><div class="">On Feb 9, 2015, at 11:31 AM, Chandler Carruth <<a href="mailto:chandlerc@google.com" class="">chandlerc@google.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div dir="ltr" class=""><div class="gmail_extra"><br class=""><div class="gmail_quote">On Mon, Feb 9, 2015 at 11:30 AM, Fiona Glaser <span dir="ltr" class=""><<a href="mailto:fglaser@apple.com" target="_blank" class="">fglaser@apple.com</a>></span> wrote:<br class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">I get the message ;-) Already redoing it in instcombine; turns out there’s actually already code for it, but it only triggers when TypeA == TypeB. Will have a patch in a few moments.</blockquote></div><br class="">Oh, sorry, I just totally missed Duncan's email. Sorry for the noise!</div></div>
</div></blockquote></div><br class=""><div class="">No problem, thanks for the suggestion. I will eventually figure out the intuition for whether a fold belongs in Instcombine or DAG!</div><div class=""><br class=""></div><div class="">Here’s a new patch that works in instcombine instead.</div></div></div></blockquote><div><br class=""></div><div><br class=""></div><div>Are there some guidelines to find out what should be done only in InstCombine and not in the DAG?</div><div><br class=""></div><div>Target-specific expansion or combine can end-up with this situation when it was not catched by ints-combine.</div><div><div><br class=""></div><div>So I believe we want it to be done *both* in inst-combine *and* in the DAG.</div><div><br class=""></div><div>— </div><div>Mehdi</div><div><br class=""></div></div></div></body></html>