<div dir="ltr"><div><div><div>Hi John,<br><br>Thanks for your email. I was looking for simply an estimation of the performance loss/gain when each of the current LLVM passes is disabled/enabled. Is there such data for recent versions of LLVM ? I could not see such quantitative data in the outcomes of the search link you suggested. <br><br></div><div>Alternatively, is their any sort of regressions that runs with every update to LLVM source code and provides the impact on performance for a set of benchmarks (or a set of test cases) ? <br><br></div><div>Best regards,<br></div><div>Ahmed<br></div><div><br><br></div></div></div><div><div><br></div></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Jun 19, 2015 at 3:56 PM, John Regehr <span dir="ltr"><<a href="mailto:regehr@cs.utah.edu" target="_blank">regehr@cs.utah.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I was wondering if there is a paper or a technical report that documents the performance impact<br>
of the different optimizations passes on a some set of benchmarks. Is something like this<br>
available ?<br>
</blockquote>
<br></span>
A lot of reserach has been done on this generic topic, a search like this will get you some of it and you can no doubt find the rest by following references:<br>
<br>
<a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__www.google.com_search-3Fq-3Dllvm-2B-2522phase-2Bordering-2522-2Bfiletype-3Apdf&d=AwMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=Mfk2qtn1LTDThVkh6-oGglNfMADXfJdty4_bhmuhMHA&m=m3GpB9_W6-0uRS0unaNLgG_Cn2XOVxfkTwFBcGNPNMw&s=LePovt0wIhOGOrlPvTpX9HGnDMWvdnjhFu6rLTdNuuA&e=" rel="noreferrer" target="_blank">https://www.google.com/search?q=llvm+%22phase+ordering%22+filetype:pdf</a><span class="HOEnZb"><font color="#888888"><br>
<br>
John<br>
</font></span></blockquote></div><br></div>