<div dir="ltr">I too agree with the general consensus that the benchmarks<div>are hard to get are hard to trust on. Moreover, profiling at </div><div>function level is also cumbersome.</div><div><br></div><div>However, </div><div>1. what we want to know (directly or indirectly) is how many </div><div>registers we saved in *hot* path?</div><div>2. How much time we saved in spill-refill?</div><div>3. What is causing the regression?</div><div><br></div><div>and I think as long we focus on this data, we can surely conclude that <br></div><div>the change has a sterling impact.</div><div><br></div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Jun 21, 2016 at 7:48 PM, vivek pandya via llvm-dev <span dir="ltr"><<a href="mailto:llvm-dev@lists.llvm.org" target="_blank">llvm-dev@lists.llvm.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote"><span class="">On Tue, Jun 21, 2016 at 1:59 PM, Axel Hecht <span dir="ltr"><<a href="mailto:l10n@mozilla.com" target="_blank">l10n@mozilla.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Hi,</p>
<p>the best way to move forward, and I guess the answer to your
question, is to get your work to build and test on try.</p>
<p>And now I'm going to flood you with jargon, sorry. Instead of
trying to document that all here, I suggest you ask people for
help on irc. Notably, #taskcluster, and #build. Folks there are
your target audience for compiler changes, too.<br>
</p>
<p>First read is
<a href="https://wiki.mozilla.org/ReleaseEngineering/TryServer" target="_blank">https://wiki.mozilla.org/ReleaseEngineering/TryServer</a>, just to get
the first piece of jargon out.</p>
<p>I suspect that the next steps are:</p>
<p>I speculate that you can use your version of llvm if you can
create a docker image that is able to compile Firefox. Please
confirm that with #taskcluster, and get some help there on how to
do that.</p>
<p>Then you'd want to make sure that you can actually compile/run
mozilla-central. In particular wrt to taskcluster, there's been
various file moves and changes, so working off of release is going
to make your path rockier. I understand that mozilla-central is
more of a moving target, but that shouldn't be that much of a
problem for you in practice, I hope.</p>
<p>I suspect that just focusing on linux x64 is good for you for
now?</p>
<p>So jargon-thought-train:</p>
<p>Validate my assumptions about all of this ;-) .<br>
</p>
<p>Make your setup work with mozilla-central.</p>
<p>Work in docker.</p>
<p>Make your mozilla-central's taskcluster tasks for linux x64 pick
your docker image to build on try.</p>
<p>Push to try for linux x64, with all tests and perf tests.</p>
<p>Use treeherder and perfherder (web uis) to see tests and
performance.</p>
<p><br>
</p>
<p>If you want to focus on your toolchain impact over time instead
of mozilla-central, I think you should be able to keep the same
base version of your patch and just update the docker image, and
then use perfherder to compare the results.</p>
<p></p></div></blockquote></span><div>Hello Axel,</div><div><br></div><div>This seems good plan but I think I would require time to setup all these, so I prefer to do this during next weekend.</div><div><br></div><div>Thanks,</div><div>Vivek </div><span class=""><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div text="#000000" bgcolor="#FFFFFF"><p>HTH</p>
<p>Axel<br>
</p><div><div>
<br>
<div>On 19/06/16 20:41, vivek pandya wrote:<br>
</div>
</div></div><blockquote type="cite"><div><div>
<div dir="ltr">
<div>
<div>Hello,</div>
<div><br>
</div>
<div>I build FireFox-46.0.1 source with llvm to test
interprocedural register allocation.</div>
<div>The build was successful with out any runtime faliures,
here are few stats:</div>
<div><font size="1"><br>
</font></div>
<div><font size="1">Measure <span style="white-space:pre-wrap"> </span>W/O
IPRA<span style="white-space:pre-wrap"> </span>WITH
IPRA</font></div>
<div><font size="1">=======<span style="white-space:pre-wrap"> </span>========<span style="white-space:pre-wrap"> </span>=========</font></div>
<div><font size="1">Total Build Time<span style="white-space:pre-wrap"> </span>76
mins<span style="white-space:pre-wrap"> </span>82.3
mins<span style="white-space:pre-wrap"> </span>8%
increment</font></div>
<div><font size="1">Octane v2.0 JS Benchmark Score (higher is
better)<span style="white-space:pre-wrap"> </span>18675.69 <span style="white-space:pre-wrap"> </span>19665.16<span style="white-space:pre-wrap"> </span>5%
improvement</font></div>
<div><font size="1">Kraken JS Benchmark time (lower is better)<span style="white-space:pre-wrap"> </span>1416.2
ms <span style="white-space:pre-wrap"> </span>1421.3
ms <span style="white-space:pre-wrap"> </span>0.35%
regression</font></div>
<div><font size="1">JetStream JS Benchmark Score (higer is
better)<span style="white-space:pre-wrap"> </span>110.10<span style="white-space:pre-wrap"> </span>112.88<span style="white-space:pre-wrap"> </span>2.52%
improvement</font></div>
<div><br>
</div>
<div>Any suggestions are welcome on how to effectively measure
performance improvement!</div>
<div><br>
</div>
<div>Sincerely,</div>
<div>Vivek</div>
</div>
<div><br>
</div>
</div>
<br>
<fieldset></fieldset>
<br>
</div></div><pre>_______________________________________________
firefox-dev mailing list
<a href="mailto:firefox-dev@mozilla.org" target="_blank">firefox-dev@mozilla.org</a>
<a href="https://mail.mozilla.org/listinfo/firefox-dev" target="_blank">https://mail.mozilla.org/listinfo/firefox-dev</a>
</pre>
</blockquote>
<br>
</div>
</blockquote></span></div><br></div></div>
<br>_______________________________________________<br>
LLVM Developers mailing list<br>
<a href="mailto:llvm-dev@lists.llvm.org">llvm-dev@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev" rel="noreferrer" target="_blank">http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><i style="font-size:12.8px">Disclaimer: Views, concerns, thoughts, questions, ideas expressed in this mail are of my own and my employer has no take in it. </i><br></div><div>Thank You.<br>Madhur D. Amilkanthwar<br><br></div></div></div>
</div>