1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
5 <title>Open LLVM Projects</title>
6 <link rel="stylesheet" href="llvm.css" type="text/css">
10 <div class="doc_title">
15 <li><a href="#what">What is this?</a></li>
16 <li><a href="#improving">Improving the current system</a>
18 <li><a href="#glibc">Port glibc to LLVM</a></li>
19 <li><a href="#NightlyTest">Improving the Nightly Tester</a></li>
20 <li><a href="#programs">Compile programs with the LLVM Compiler</a></li>
21 <li><a href="#llvm_ir">Extend the LLVM intermediate representation</a></li>
22 <li><a href="#misc_imp">Miscellaneous Improvements</a></li>
25 <li><a href="#new">Adding new capabilities to LLVM</a>
27 <li><a href="#pointeranalysis">Pointer and Alias Analysis</a></li>
28 <li><a href="#profileguided">Profile Guided Optimization</a></li>
29 <li><a href="#xforms">New Transformations and Analyses</a></li>
30 <li><a href="#x86be">X86 Back-end Improvements</a></li>
31 <li><a href="#misc_new">Miscellaneous Additions</a></li>
35 <!-- *********************************************************************** -->
36 <div class="doc_section">
37 <a name="what">What is this?</a>
39 <!-- *********************************************************************** -->
41 <div class="doc_text">
43 <p>This document is meant to be a sort of "big TODO list" for LLVM. Each
44 project in this document is something that would be useful for LLVM to have, and
45 would also be a great way to get familiar with the system. Some of these
46 projects are small and self-contained, which may be implemented in a couple of
47 days, others are larger. Several of these projects may lead to interesting
48 research projects in their own right. In any case, we welcome all
51 <p>If you are thinking about tackling one of these projects, please send a mail
52 to the <a href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM
53 Developer's</a> mailing list, so that we know the project is being worked on.
54 Additionally this is a good way to get more information about a specific project
55 or to suggest other projects to add to this page.
58 <p>The projects in this page are open-ended. More specific projects are
59 filed as unassigned enhancements in the <a href="http://llvm.cs.uiuc.edu/bugs/">
60 LLVM bug tracker</a>. See the <a href="http://llvm.cs.uiuc.edu/bugs/buglist.cgi?keywords_type=allwords&keywords=&bug_status=NEW&bug_status=ASSIGNED&bug_status=REOPENED&bug_severity=enhancement&emailassigned_to1=1&emailtype1=substring&email1=unassigned">list of currently outstanding issues</a> if you wish to help improve LLVM.</p>
64 <!-- *********************************************************************** -->
65 <div class="doc_section">
66 <a name="improving">Improving the current system</a>
68 <!-- *********************************************************************** -->
70 <div class="doc_text">
72 <p>Improvements to the current infrastructure are always very welcome and tend
73 to be fairly straight-forward to implement. Here are some of the key areas that
74 can use improvement...</p>
78 <!-- ======================================================================= -->
79 <div class="doc_subsection">
80 <a name="glibc">Port glibc to LLVM</a>
83 <div class="doc_text">
85 <p>It would be very useful to <a
86 href="http://www.gnu.org/software/libc/porting.html">port</a> <a
87 href="http://www.gnu.org/software/glibc/">glibc</a> to LLVM. This would allow a
88 variety of interprocedural algorithms to be much more effective in the face of
89 library calls. The most important pieces to port are things like the string
90 library and the <tt>stdio</tt> related functions... low-level system calls like
91 '<tt>read</tt>' should stay unimplemented in LLVM.</p>
95 <!-- ======================================================================= -->
96 <div class="doc_subsection">
97 <a name="NightlyTest">Improving the Nightly Tester</a>
100 <div class="doc_text">
102 <p>The <a href="/testresults/">Nightly Tester</a> is a simple perl script
103 (located in <tt>utils/NightlyTest.pl</tt>) which runs every night to generate a
104 daily report. It could use the following improvements:</p>
107 <li>Regression tests - We should run the regression tests in addition to the
108 program tests...</li>
113 <!-- ======================================================================= -->
114 <div class="doc_subsection">
115 <a name="programs">Compile programs with the LLVM Compiler</a>
118 <div class="doc_text">
120 <p>We are always looking for new testcases and benchmarks for use with LLVM. In
121 particular, it is useful to try compiling your favorite C source code with LLVM.
122 If it doesn't compile, try to figure out why or report it to the <a
123 href="http://mail.cs.uiuc.edu/pipermail/llvmbugs/">llvm-bugs</a> list. If you
124 get the program to compile, it would be extremely useful to convert the build
125 system to be compatible with the LLVM Programs testsuite so that we can check it
126 into CVS and the automated tester can use it to track progress of the
129 <p>When testing a code, try running it with a variety of optimizations, and with
130 all the back-ends: CBE, llc, and lli.</p>
134 <!-- ======================================================================= -->
135 <div class="doc_subsection">
136 <a name="llvm_ir">Extend the LLVM intermediate representation</a>
139 <div class="doc_text">
142 <li>Add support for platform-independent prefetch support. The GCC <a
143 href="http://gcc.gnu.org/projects/prefetch.html">prefetch project</a> page
144 has a good survey of the prefetching capabilities of a variety of modern
151 <!-- ======================================================================= -->
152 <div class="doc_subsection">
153 <a name="misc_imp">Miscellaneous Improvements</a>
156 <div class="doc_text">
159 <li>Someone needs to look into getting the <tt>ranlib</tt> tool to index LLVM
160 bytecode files, so that linking in .a files is not hideously slow. They
161 would also then have to implement the reader for this index in
164 <li>Rework the PassManager to be more flexible</li>
166 <li>Some transformations and analyses only work on reducible flow graphs. It
167 would be nice to have a transformation which could be "required" by these passes
168 which makes irreducible graphs reducible. This can easily be accomplished
169 through code duplication. See <a
170 href="http://citeseer.nj.nec.com/janssen97making.html">Making Graphs Reducible
171 with Controlled Node Splitting</a> and perhaps <a
172 href="http://doi.acm.org/10.1145/262004.262005">Nesting of Reducible and
173 Irreducible Loops</a>.</li>
179 <!-- *********************************************************************** -->
180 <div class="doc_section">
181 <a name="new">Adding new capabilities to LLVM</a>
183 <!-- *********************************************************************** -->
185 <div class="doc_text">
187 <p>Sometimes creating new things is more fun than improving existing things.
188 These projects tend to be more involved and perhaps require more work, but can
189 also be very rewarding.</p>
193 <!-- ======================================================================= -->
194 <div class="doc_subsection">
195 <a name="pointeranalysis">Pointer and Alias Analysis</a>
198 <div class="doc_text">
200 <p>We have a <a href="AliasAnalysis.html">strong base for development</a> of
201 both pointer analysis based optimizations as well as pointer analyses
202 themselves. It seems natural to want to take advantage of this...</p>
205 <li>Implement a flow-sensitive context-sensitive alias analysis algorithm<br>
206 - Pick one of the somewhat efficient algorithms, but strive for maximum
209 <li>Implement a flow-sensitive context-insensitive alias analysis algorithm<br>
210 - Just an efficient local algorithm perhaps?</li>
212 <li>Implement an interface to update analyses in response to common code motion
215 <li>Implement alias-analysis-based optimizations:
217 <li>Dead store elimination</li>
224 <!-- ======================================================================= -->
225 <div class="doc_subsection">
226 <a name="profileguided">Profile Guided Optimization</a>
229 <div class="doc_text">
231 <p>We now have a unified infrastructure for writing profile-guided
232 transformations, which will work either at offline-compile-time or in the JIT,
233 but we don't have many transformations. We would welcome new profile-guided
234 transformations as well as improvements to the current profiling system.
237 <p>Ideas for profile guided transformations:</p>
240 <li>Superblock formation (with many optimizations)</li>
241 <li>Loop unrolling/peeling</li>
242 <li>Profile directed inlining</li>
247 <p>Improvements to the existing support:</p>
250 <li>The current block and edge profiling code that gets inserted is very simple
251 and inefficient. Through the use of control-dependence information, many fewer
252 counters could be inserted into the code. Also, if the execution count of a
253 loop is known to be a compile-time or runtime constant, all of the counters in
254 the loop could be avoided.</li>
256 <li>You could implement one of the "static profiling" algorithms which analyze a
257 piece of code an make educated guesses about the relative execution frequencies
258 of various parts of the code.</li>
260 <li>You could add path profiling support, or adapt the existing LLVM path
261 profiling code to work with the generic profiling interfaces.</li>
266 <!-- ======================================================================= -->
267 <div class="doc_subsection">
268 <a name="xforms">New Transformations and Analyses</a>
271 <div class="doc_text">
274 <li>Implement a Dependence Analysis Infrastructure<br>
275 - Design some way to represent and query dep analysis</li>
276 <li>Implement a strength reduction pass</li>
277 <li>Value range propagation pass</li>
278 <li>Implement an unswitching pass</li>
279 <li>Write a loop unroller, with a simple heuristic for when to unroll</li>
284 <!-- ======================================================================= -->
285 <div class="doc_section">
286 <a name="x86be">X86 Back-end Improvements</a>
289 <div class="doc_text">
292 <li>Implement a better instruction selector</li>
293 <li>Implement support for the "switch" instruction without requiring the
294 lower-switches pass.</li>
299 <!-- ======================================================================= -->
300 <div class="doc_section">
301 <a name="misc_new">Miscellaneous Additions</a>
304 <div class="doc_text">
307 <li>Port the <A HREF="http://www-sop.inria.fr/mimosa/fp/Bigloo/">Bigloo</A>
308 Scheme compiler, from Manuel Serrano at INRIA Sophia-Antipolis, to
309 output LLVM bytecode. It seems that it can already output .NET
310 bytecode, JVM bytecode, and C, so LLVM would ostensibly be another good
312 <li>Write a new frontend for some other language (Java? OCaml? Forth?)</li>
313 <li>Write a new backend for a target (IA64? MIPS? MMIX?)</li>
314 <li>Write a disassembler for machine code that would use TableGen to output
315 <tt>MachineInstr</tt>s for transformations, optimizations, etc.</li>
316 <li>Random test vector generator: Use a C grammar to generate random C code;
317 run it through llvm-gcc, then run a random set of passes on it using opt.
318 Try to crash opt. When opt crashes, use bugpoint to reduce the test case and
319 mail the result to yourself. Repeat ad infinitum.</li>
320 <li>Design a simple, recognizable logo.</li>
325 <!-- *********************************************************************** -->
329 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
330 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
331 <a href="http://validator.w3.org/check/referer"><img
332 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
334 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
335 <a href="http://llvm.cs.uiuc.edu">LLVM Compiler Infrastructure</a><br>
336 Last modified: $Date$