<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;"><br><div><div>On Jun 17, 2013, at 5:16 AM, "Demikhovsky, Elena" <<a href="mailto:elena.demikhovsky@intel.com">elena.demikhovsky@intel.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">Benjamin already explained me that it was the IR optimization.<br>I'm proposing to do the same on DAG level because<br>1) We have the this optimization for scalar and don't have it for vector<br>2) The MUL node can be produced on DAG-level transformations of other items.<br><br>- Elena<br><br></div></blockquote><div dir="auto"><br></div><div dir="auto">This makes sense to me. This is going to be target-independent, right ?</div></div></body></html>