<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Jan 14, 2015 at 5:08 PM, Hal Finkel <span dir="ltr"><<a href="mailto:hfinkel@anl.gov" target="_blank">hfinkel@anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><span class="">> 1. A series of APIs of TargetTransformInfo that compute the cost of<br>
> instructions of a particular type (e.g. getArithmeticInstrCost and<br>
> getShuffleCost)<br>
<br>
</span>These are used by the vectorizers, and roughly, return costs in reciprocal throughput (the higher the throughput the lower the cost).<br></blockquote><div><br></div><div>Do you think we could replace this with data from the CPU scheduling models (if it exists in a particular CPU's model)?<br><br></div><div>One number for all CPUs may not always be adequate:<br></div><div><a href="http://llvm.org/bugs/show_bug.cgi?id=21356">http://llvm.org/bugs/show_bug.cgi?id=21356</a><br><br></div><div> <br></div></div></div></div>