Add hot cold splitting to GSOC
diff --git a/OpenProjects.html b/OpenProjects.html
index 8db1c77..536a3f9 100755
--- a/OpenProjects.html
+++ b/OpenProjects.html
@@ -17,6 +17,7 @@
<li><a href="#llvm_dbg_invariant">Make LLVM passes debug info invariant</a></li>
<li><a href="#llvm_mergesim">Improve MergeFunctions to incorporate MergeSimilarFunction patches and ThinLTO Support</a></li>
<li><a href="#llvm_dwarf_yaml2obj">Add DWARF support to yaml2obj</a></li>
+ <li><a href="#llvm_holtcold">Improve hot cold splitting to outline maximal SESE/SEME regions</a></li>
</ul>
<li><a href="http://clang.llvm.org/"><b>Clang</b></a>
<ul>
@@ -447,6 +448,78 @@
</div>
<!-- *********************************************************************** -->
+
+<div class="www_subsubsection">
+ <a name="llvm_hotcold">Improve hot cold splitting to outline maximal SESE/SEME regions</a>
+</div>
+<!-- *********************************************************************** -->
+
+ <div class="www_text">
+ <p><b>Description of the project:</b>Hot Cold Splitting in LLVM is an IR level
+ function splitting transformation. The goal of hot/cold splitting is to improve
+ the memory locality of code and helps reduce startup working set. The splitting pass
+ does this by identifying cold blocks and moving them into separate functions. Because it
+ is implemented at the IR level all the back end target benefit from it.
+
+ It is a relatively new optimization and it was recently presented at
+ the <a href="https://llvm.org/devmtg/2019-10/talk-abstracts.html#tech8">LLVM Dev Meeting in
+ 2019</a> and the slides are <a href="https://llvm.org/devmtg/2019-10/slides/Kumar-HotColdSplitting.pdf">here</a>
+ Currently, hot cold splitting works as a greedy algorithm where a region found
+ by first cold basic block is given preference. When a new region is found which
+ intersects with an existing region, the new region is dropped. One approach
+ would be to first run an analysis pass to find a set of SESE/SEME regions and do
+ some bookkeeping to detect the most profitable ones. The goal should be to not
+ regress and keep the compile time linear. There are fast algorithms to detect
+ SESE regions as illustrated in
+ (http://impact.gforge.inria.fr/impact2016/papers/impact2016-kumar.pdf), we can
+ additionally leverage the existing RegionInfo as well if that has acceptable
+ compilation time complexity.
+
+ </p>
+ <p><b>Preparation resources:</b>
+ <ul>
+ <li>
+ <a href="http://lists.llvm.org/pipermail/llvm-dev/2019-January/129606.html">Update on hot cold splitting</a>
+ </li>
+ <ul>
+ The following two papers provide earlier work on hot cold splitting. While these papers are a good start, LLVM's
+ HCS has completely different implementation in two aspects a) It is implemented at IR level and outlines basic
+ blocks as function rather than naked branches. b) It is based on regions and outlines a set of basic blocks.
+ <li>
+ <a href="http://pages.cs.wisc.edu/~fischer/cs701.f05/code.positioning.pdf">Original paper on hot cold splitting by
+ Pettis and Hansen.</a>Section 5 on procedure splitting is interesting one. It has nice examples ;) to help
+ understand why HCS works.
+ </li>
+ <li>
+ <a href="https://www.cs.cmu.edu/afs/cs/academic/class/15745-s07/www/papers/p80-cohn.pdf">Paper on hot cold
+ splitting</a> The paper provides some details on one approach to split functions. This is helpful to get a
+ different perspective and may help get new ideas.
+ </li>
+ <li>
+ <a href="https://llvm.org/devmtg/2019-10/talk-abstracts.html#tech8">Video and slides of the presentation</a>
+ </li>
+ </ul>
+ </p>
+ <p><b>Expected results:</b>
+ <ul>
+ <li>
+ Improve Hot Cold Splitting to detect maximal SEME regions without incurring super linear compile time overhead.
+ In case compile time overhead becomes quadratic, come up with a cost model to detect when quadratic behavior
+ gets triggered and bail out based on a compiler flag.
+ </li>
+ </ul>
+ </p>
+
+ <p><b>Confirmed Mentors:</b>Aditya Kumar</p>
+
+ <p><b>Desirable skills:</b>
+ Course on compiler design, SSA Representation,
+ Intermediate knowledge of C++, Familiarity with LLVM Core.
+ </p>
+</div>
+
+
+<!-- *********************************************************************** -->
<div class="www_subsection">
<a>MLIR</a>
</div>
