<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:12.0pt;
font-family:"Times New Roman",serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:#1F497D;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">OMP_WAIT_POLICY=ACTIVE is equivalent to KMP_LIBRARY=turnaround.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">If KMP_LIBRARY=throughput, then each thread pauses / releases its timeslice in spin-wait loops.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">If KMP_LIBRARY=turnaround, the threads only pause it they know that the machine is oversubscribed.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">KMP_BLOCKTIME controls the blocking, not the pausing. The default value is 200 ms. If a thread spins for longer than 200 ms (actually, some value between 200-400
ms), then it goes to sleep.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">If KMP_BLOCKTIME=0, then the thread does a single check to see if it can proceed, then immediately goes to sleep if it cannot.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">If KMP_BLOCKTIME=infinite (which is implemented as 2^31-1), then the threads will never block at a barrier.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">KMP_BLOCKTIME=infinite also disables a lot of checks (and corresponding cache misses) in the barrier spin-wait loop, and can result in a 2x-3x speedup over KMP_BLOCKTIME=2^31-2
on EPCC parallel, for a machine with many procs.<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">FYI<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">Maybe we should have mapped OMP_WAIT_POLICY=ACTIVE to KMP_BLOCKTIME=infinite, and not KMP_LIBRARY=turnaround. I don’t know…<o:p></o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"><o:p> </o:p></span></p>
<p class="MsoNormal"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D">-bb<o:p></o:p></span></p>
<p class="MsoNormal"><a name="_MailEndCompose"><span style="font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D"><o:p> </o:p></span></a></p>
<div>
<div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0in 0in 0in">
<p class="MsoNormal"><b><span style="font-size:11.0pt;font-family:"Calibri",sans-serif">From:</span></b><span style="font-size:11.0pt;font-family:"Calibri",sans-serif"> Openmp-dev [mailto:openmp-dev-bounces@lists.llvm.org]
<b>On Behalf Of </b>John Mellor-Crummey via Openmp-dev<br>
<b>Sent:</b> Tuesday, August 18, 2015 2:18 PM<br>
<b>To:</b> César<br>
<b>Cc:</b> openmp-dev@lists.llvm.org<br>
<b>Subject:</b> Re: [Openmp-dev] Performance slowdown<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<div>
<p class="MsoNormal">My guess is that you are blocking rather than spinning. Using OMP_WAIT_POLICY=active doesn't seem to be enough with Intel's runtime to turn off all blocking. As I recall, there is a KMP_xxx flag that applies as well. You can grep the barrier
code in the runtime or hope that someone from intel responds to your inquiry.<br>
<br>
--<o:p></o:p></p>
<div>
<p class="MsoNormal">John Mellor-Crummey<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">(sent from my phone)<o:p></o:p></p>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-bottom:12.0pt"><br>
On Aug 18, 2015, at 1:14 PM, César via Openmp-dev <<a href="mailto:openmp-dev@lists.llvm.org">openmp-dev@lists.llvm.org</a>> wrote:<o:p></o:p></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<div>
<p class="MsoNormal">Hello,<o:p></o:p></p>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">I don't know if this is the correct list to talk about this - I did not find a better place..<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">I am doing performance experiments with a few OpenMP implementations (IOMP, GOMP and our private impl.) and I am seeing a severe slowdown when I use IOMP (GOMP and others are performing well). <o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">The benchmarks I am using are these ones: <a href="http://kastors.gforge.inria.fr/#!index.md">http://kastors.gforge.inria.fr/#!index.md</a><br clear="all">
<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">Really, the slowdown is huge. For one of the programs (plasma/dpotrf_taskdep -n 8192 -b 64 -i 1 -c) the serial version executes in ~28s and the parallel one executes in ~110s. I did some profiling and found that most of the time is being
spent on synchronization barriers and dependence tracking (see attached image). Before digging deeper I would like to hear back from you if I am doing something wrong here:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal">- I tested with the last version of the repository: <a href="http://llvm.org/svn/llvm-project/openmp/trunk">http://llvm.org/svn/llvm-project/openmp/trunk</a><o:p></o:p></p>
</div>
<div>
<div>
<div>
<p class="MsoNormal">- I am using Ubuntu 14.10.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- I have tested on more than one machine, the results above are from a Intel i7-3770<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- The runtime itself is compiled using: make compiler=gcc os_omp=linux arch=32e<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- The version of GCC that I am using is: 4.9.1<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal">- The version of Clang that I am using to compile the benchmarks: 3.5.0<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal"><br>
<br>
César.<o:p></o:p></p>
</div>
</div>
</div>
</div>
<p class="MsoNormal">!DSPAM:8504,55d37631260061683114033! <o:p></o:p></p>
</div>
</blockquote>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal"><pic1.png><o:p></o:p></p>
</div>
</blockquote>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal"><pic2.png><o:p></o:p></p>
</div>
</blockquote>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal"><pic3.png><o:p></o:p></p>
</div>
</blockquote>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal">_______________________________________________<br>
Openmp-dev mailing list<br>
<a href="mailto:Openmp-dev@lists.llvm.org">Openmp-dev@lists.llvm.org</a><br>
<a href="http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev">http://lists.llvm.org/cgi-bin/mailman/listinfo/openmp-dev</a><br>
<br>
!DSPAM:8504,55d37631260061683114033!<o:p></o:p></p>
</div>
</blockquote>
</div>
</body>
</html>