<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 Feb 10, 2015, at 1:22 PM, Eric Christopher <<a href="mailto:echristo@gmail.com">echristo@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div dir="ltr">Cautious is fine, but adding more register pressure heuristics to the middle end just gets us to a point where everything is poorly done via heuristics.<br></div></blockquote><div><br></div>What do you think would be the right metric to control this optimization?</div><div>My inclination right now was to use the same, if possible, heuristic as the vectorizer, which IIRC, uses some register pressure heuristic.</div><div><br></div><div>I am not saying this is good, I am saying that like every unrolling technic we need to find a good balance between code size explosion and performance improvements. Exposing too much ILP is probably not a good idea.</div><div><br></div><div>-Quentin</div><div><blockquote type="cite"><div dir="ltr"><div><br></div><div>-eric</div></div><br><div class="gmail_quote">On Tue Feb 10 2015 at 11:02:04 AM Quentin Colombet <<a href="mailto:qcolombet@apple.com">qcolombet@apple.com</a>> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">> FWIW I'm not such a huge fan of using register pressure heuristics in the middle end<br>
<br>
<br>
FWIW, I agree, but this is not what LSR currently does and since the backends do not expect that yet, I would prefer moving with cautious.<br>
<br>
Q.<br>
<br>
<br>
<a href="http://reviews.llvm.org/D7128" target="_blank">http://reviews.llvm.org/D7128</a><br>
<br>
EMAIL PREFERENCES<br>
<a href="http://reviews.llvm.org/settings/panel/emailpreferences/" target="_blank">http://reviews.llvm.org/<u></u>settings/panel/<u></u>emailpreferences/</a><br>
<br>
<br>
</blockquote></div>
</blockquote></div><br></body></html>