Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / c0f376dba2b51aacf06cfefbc1c46120bb8b725d / . / docs / OpenProjects.html
blob: 2acb7279c21357f297fa343edde5cfed8c4f3a24 [file] [log] [blame]
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
5 <title>Open LLVM Projects</title>
6 <link rel="stylesheet" href="llvm.css" type="text/css">
7</head>
8<body>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]9
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]10<div class="doc_title">
11 Open LLVM Projects
12</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]13
14<ul>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]15 <li><a href="#what">What is this?</a></li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]16 <li><a href="#improving">Improving the current system</a>
17 <ol>
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]18 <li><a href="#code-cleanups">Implementing Code Cleanup bugs</a></li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]19 <li><a href="#glibc">Port glibc to LLVM</a></li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]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>
Misha Brukman8bf7c282004-11-07 01:03:59 +0000[diff] [blame]22 <li><a href="#target">Target backend improvements</a></li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]23 <li><a href="#misc_imp">Miscellaneous Improvements</a></li>
24 </ol></li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]25
26 <li><a href="#new">Adding new capabilities to LLVM</a>
27 <ol>
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]28 <li><a href="#newfeaturebugs">Implementing new feature PRs</a></li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]29 <li><a href="#pointeranalysis">Pointer and Alias Analysis</a></li>
Misha Brukman05bcca92004-09-06 22:35:43 +0000[diff] [blame]30 <li><a href="#profileguided">Profile-Guided Optimization</a></li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]31 <li><a href="#xforms">New Transformations and Analyses</a></li>
32 <li><a href="#x86be">X86 Back-end Improvements</a></li>
33 <li><a href="#misc_new">Miscellaneous Additions</a></li>
34 </ol></li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]35</ul>
36
Chris Lattner97d49ef2004-05-23 21:06:21 +0000[diff] [blame]37<div class="doc_author">
38 <p>Written by the <a href="http://llvm.cs.uiuc.edu/">LLVM Team</a></p>
39</div>
40
41
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]42<!-- *********************************************************************** -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]43<div class="doc_section">
44 <a name="what">What is this?</a>
45</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]46<!-- *********************************************************************** -->
47
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]48<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]49
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]50<p>This document is meant to be a sort of "big TODO list" for LLVM. Each
51project in this document is something that would be useful for LLVM to have, and
52would also be a great way to get familiar with the system. Some of these
53projects are small and self-contained, which may be implemented in a couple of
54days, others are larger. Several of these projects may lead to interesting
55research projects in their own right. In any case, we welcome all
56contributions.</p>
57
58<p>If you are thinking about tackling one of these projects, please send a mail
59to the <a href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]60Developer's</a> mailing list, so that we know the project is being worked on.
61Additionally this is a good way to get more information about a specific project
Alkis Evlogimenos45ff0272004-03-11 02:50:35 +0000[diff] [blame]62or to suggest other projects to add to this page.
63</p>
64
Misha Brukman836db502004-05-12 18:57:54 +0000[diff] [blame]65<p>The projects in this page are open-ended. More specific projects are
66filed as unassigned enhancements in the <a href="http://llvm.cs.uiuc.edu/bugs/">
Chris Lattner9f57b182004-05-23 20:29:35 +0000[diff] [blame]67LLVM bug tracker</a>. See the <a href="http://llvm.cs.uiuc.edu/bugs/buglist.cgi?keywords_type=allwords&amp;keywords=&amp;bug_status=NEW&amp;bug_status=ASSIGNED&amp;bug_status=REOPENED&amp;bug_severity=enhancement&amp;emailassigned_to1=1&amp;emailtype1=substring&amp;email1=unassigned">list of currently outstanding issues</a> if you wish to help improve LLVM.</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]68
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]69</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]70
71<!-- *********************************************************************** -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]72<div class="doc_section">
73 <a name="improving">Improving the current system</a>
74</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]75<!-- *********************************************************************** -->
76
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]77<div class="doc_text">
78
79<p>Improvements to the current infrastructure are always very welcome and tend
80to be fairly straight-forward to implement. Here are some of the key areas that
81can use improvement...</p>
82
83</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]84
85<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]86<div class="doc_subsection">
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]87 <a name="code-cleanups">Implementing Code Cleanup bugs</a>
88</div>
89
90<div class="doc_text">
91
92<p>
93The <a href="http://llvm.cs.uiuc.edu/bugs/">LLVM bug tracker</a> occasionally
Reid Spencer11d45b62004-11-01 20:41:01 +0000[diff] [blame]94has <a
95 href="http://llvm.cs.uiuc.edu/bugs/buglist.cgi?short_desc_type=allwordssubstr&amp;short_desc=&amp;long_desc_type=allwordssubstr&amp;long_desc=&amp;bug_file_loc_type=allwordssubstr&amp;bug_file_loc=&amp;status_whiteboard_type=allwordssubstr&amp;status_whiteboard=&amp;keywords_type=allwords&amp;keywords=code-cleanup&amp;bug_status=NEW&amp;bug_status=ASSIGNED&amp;bug_status=REOPENED&amp;emailassigned_to1=1&amp;emailtype1=substring&amp;email1=&amp;emailassigned_to2=1&amp;emailreporter2=1&amp;emailcc2=1&amp;emailtype2=substring&amp;email2=&amp;bugidtype=include&amp;bug_id=&amp;votes=&amp;changedin=&amp;chfieldfrom=&amp;chfieldto=Now&amp;chfieldvalue=&amp;cmdtype=doit&amp;order=Bug+Number&amp;field0-0-0=noop&amp;type0-0-0=noop&amp;value0-0-0=">"code-cleanup" bugs</a> filed in it. Taking one of these and fixing it is a good
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]96way to get your feet wet in the LLVM code and discover how some of its components
97work.
98</p>
99
100</div>
101
102<!-- ======================================================================= -->
103<div class="doc_subsection">
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]104 <a name="glibc">Port glibc to LLVM</a>
105</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]106
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]107<div class="doc_text">
108
109<p>It would be very useful to <a
Tanya Lattner09cf73c2004-06-22 04:24:55 +0000[diff] [blame]110href="http://www.gnu.org/software/libc/manual/html_node/Porting.html">port</a> <a
111href="http://www.gnu.org/software/libc/">glibc</a> to LLVM. This would allow a
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]112variety of interprocedural algorithms to be much more effective in the face of
113library calls. The most important pieces to port are things like the string
114library and the <tt>stdio</tt> related functions... low-level system calls like
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]115'<tt>read</tt>' should stay unimplemented in LLVM.</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]116
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]117</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]118
119<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]120<div class="doc_subsection">
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]121 <a name="programs">Compile programs with the LLVM Compiler</a>
122</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]123
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]124<div class="doc_text">
125
126<p>We are always looking for new testcases and benchmarks for use with LLVM. In
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]127particular, it is useful to try compiling your favorite C source code with LLVM.
128If it doesn't compile, try to figure out why or report it to the <a
129href="http://mail.cs.uiuc.edu/pipermail/llvmbugs/">llvm-bugs</a> list. If you
130get the program to compile, it would be extremely useful to convert the build
131system to be compatible with the LLVM Programs testsuite so that we can check it
Chris Lattner1b360042003-02-07 19:25:35 +0000[diff] [blame]132into CVS and the automated tester can use it to track progress of the
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]133compiler.</p>
Chris Lattner1b360042003-02-07 19:25:35 +0000[diff] [blame]134
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]135<p>When testing a code, try running it with a variety of optimizations, and with
136all the back-ends: CBE, llc, and lli.</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]137
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]138</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]139
140<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]141<div class="doc_subsection">
142 <a name="llvm_ir">Extend the LLVM intermediate representation</a>
143</div>
144
145<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]146
147<ol>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]148<li>Add support for platform-independent prefetch support. The GCC <a
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]149 href="http://gcc.gnu.org/projects/prefetch.html">prefetch project</a> page
150 has a good survey of the prefetching capabilities of a variety of modern
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]151 processors.</li>
152
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]153</ol>
154
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]155</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]156
157<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]158<div class="doc_subsection">
Misha Brukman8bf7c282004-11-07 01:03:59 +0000[diff] [blame]159 <a name="target">Target backend improvements</a>
160</div>
161
162<div class="doc_text">
163
164<ol>
165 <li>Find benchmarks either using our <a
166 href="http://llvm.cs.uiuc.edu/testresults/">test results</a> or on your own,
167 where LLVM code generators do not produce optimal code or simply where another
168 compiler produces better code. Try to minimize the test case that
169 demonstrates the issue. Then, either <a
170 href="http://llvm.cs.uiuc.edu/bugs/">submit a bug</a> with your testcase and
171 the code that LLVM produces vs. the code that it <em>should</em> produce, or
172 even better, see if you can improve the code generator and submit a
173 patch.</li>
174</ol>
175
176</div>
177
178<!-- ======================================================================= -->
179<div class="doc_subsection">
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]180 <a name="misc_imp">Miscellaneous Improvements</a>
181</div>
182
183<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]184
185<ol>
Chris Lattner8bde96f2003-05-06 20:47:11 +0000[diff] [blame]186<li>Someone needs to look into getting the <tt>ranlib</tt> tool to index LLVM
187 bytecode files, so that linking in .a files is not hideously slow. They
188 would also then have to implement the reader for this index in
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]189 <tt>gccld</tt>.</li>
Chris Lattner8bde96f2003-05-06 20:47:11 +0000[diff] [blame]190
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]191<li>Rework the PassManager to be more flexible</li>
192
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]193<li>Some transformations and analyses only work on reducible flow graphs. It
194would be nice to have a transformation which could be "required" by these passes
195which makes irreducible graphs reducible. This can easily be accomplished
196through code duplication. See <a
197href="http://citeseer.nj.nec.com/janssen97making.html">Making Graphs Reducible
198with Controlled Node Splitting</a> and perhaps <a
199href="http://doi.acm.org/10.1145/262004.262005">Nesting of Reducible and
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]200Irreducible Loops</a>.</li>
201
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]202</ol>
203
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]204</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]205
206<!-- *********************************************************************** -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]207<div class="doc_section">
208 <a name="new">Adding new capabilities to LLVM</a>
209</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]210<!-- *********************************************************************** -->
211
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]212<div class="doc_text">
213
Brian Gaeke2d261352004-04-07 15:31:23 +0000[diff] [blame]214<p>Sometimes creating new things is more fun than improving existing things.
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]215These projects tend to be more involved and perhaps require more work, but can
216also be very rewarding.</p>
217
218</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]219
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]220
221<!-- ======================================================================= -->
222<div class="doc_subsection">
223 <a name="newfeaturebugs">Implementing new feature PRs</a>
224</div>
225
226<div class="doc_text">
227
Reid Spencer11d45b62004-11-01 20:41:01 +0000[diff] [blame]228<p>Many ideas for feature requests are stored in LLVM bugzilla. Just <a
Misha Brukman8bf7c282004-11-07 01:03:59 +0000[diff] [blame]229 href="http://llvm.cs.uiuc.edu/bugs/buglist.cgi?short_desc_type=allwordssubstr&amp;short_desc=&amp;long_desc_type=allwordssubstr&amp;long_desc=&amp;bug_file_loc_type=allwordssubstr&amp;bug_file_loc=&amp;status_whiteboard_type=allwordssubstr&amp;status_whiteboard=&amp;keywords_type=allwords&amp;keywords=new-feature&amp;bug_status=UNCONFIRMED&amp;bug_status=NEW&amp;bug_status=ASSIGNED&amp;bug_status=REOPENED&amp;emailassigned_to1=1&amp;emailtype1=substring&amp;email1=&amp;emailassigned_to2=1&amp;emailreporter2=1&amp;emailcc2=1&amp;emailtype2=substring&amp;email2=&amp;bugidtype=include&amp;bug_id=&amp;votes=&amp;changedin=&amp;chfieldfrom=&amp;chfieldto=Now&amp;chfieldvalue=&amp;cmdtype=doit&amp;namedcmd=All+PRs&amp;newqueryname=&amp;order=Bug+Number&amp;field0-0-0=noop&amp;type0-0-0=noop&amp;value0-0-0=">search for bugs with a "new-feature" keyword</a>.</p>
Chris Lattner7e0281f2004-05-28 20:41:37 +0000[diff] [blame]230
231</div>
232
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]233<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]234<div class="doc_subsection">
235 <a name="pointeranalysis">Pointer and Alias Analysis</a>
236</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]237
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]238<div class="doc_text">
239
240<p>We have a <a href="AliasAnalysis.html">strong base for development</a> of
241both pointer analysis based optimizations as well as pointer analyses
242themselves. It seems natural to want to take advantage of this...</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]243
244<ol>
245<li>Implement a flow-sensitive context-sensitive alias analysis algorithm<br>
246 - Pick one of the somewhat efficient algorithms, but strive for maximum
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]247 precision</li>
248
Chris Lattnerf56703b2003-06-25 14:31:06 +0000[diff] [blame]249<li>Implement a flow-sensitive context-insensitive alias analysis algorithm<br>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]250 - Just an efficient local algorithm perhaps?</li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]251
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]252<li>Implement alias-analysis-based optimizations:
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]253 <ul>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]254 <li>...</li>
255 </ul></li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]256</ol>
257
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]258</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]259
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]260<!-- ======================================================================= -->
261<div class="doc_subsection">
Misha Brukman05bcca92004-09-06 22:35:43 +0000[diff] [blame]262 <a name="profileguided">Profile-Guided Optimization</a>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]263</div>
264
265<div class="doc_text">
266
Chris Lattnera5422e42004-03-08 22:29:35 +0000[diff] [blame]267<p>We now have a unified infrastructure for writing profile-guided
268transformations, which will work either at offline-compile-time or in the JIT,
269but we don't have many transformations. We would welcome new profile-guided
270transformations as well as improvements to the current profiling system.
271</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]272
Misha Brukman05bcca92004-09-06 22:35:43 +0000[diff] [blame]273<p>Ideas for profile-guided transformations:</p>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]274
275<ol>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]276<li>Superblock formation (with many optimizations)</li>
277<li>Loop unrolling/peeling</li>
278<li>Profile directed inlining</li>
279<li>Code layout</li>
280<li>...</li>
281</ol>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]282
Chris Lattnera5422e42004-03-08 22:29:35 +0000[diff] [blame]283<p>Improvements to the existing support:</p>
284
285<ol>
286<li>The current block and edge profiling code that gets inserted is very simple
287and inefficient. Through the use of control-dependence information, many fewer
288counters could be inserted into the code. Also, if the execution count of a
289loop is known to be a compile-time or runtime constant, all of the counters in
290the loop could be avoided.</li>
291
292<li>You could implement one of the "static profiling" algorithms which analyze a
293piece of code an make educated guesses about the relative execution frequencies
294of various parts of the code.</li>
295
296<li>You could add path profiling support, or adapt the existing LLVM path
297profiling code to work with the generic profiling interfaces.</li>
298</ol>
299
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]300</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]301
302<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]303<div class="doc_subsection">
304 <a name="xforms">New Transformations and Analyses</a>
305</div>
306
307<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]308
309<ol>
Chris Lattnerc2a57302004-10-01 22:25:28 +0000[diff] [blame]310<li>Implement <a href="http://www.cs.purdue.edu/homes/vandrutj/">GVN-PRE</a>, a
311 powerful and simple Partial Redundancy Elimination algorithm for SSA form</li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]312<li>Implement a Dependence Analysis Infrastructure<br>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]313 - Design some way to represent and query dep analysis</li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]314<li>Implement a strength reduction pass</li>
315<li>Value range propagation pass</li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]316</ol>
317
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]318</div>
319
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]320<!-- ======================================================================= -->
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]321<div class="doc_section">
322 <a name="x86be">X86 Back-end Improvements</a>
323</div>
324
325<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]326
327<ol>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]328<li>Implement a better instruction selector</li>
Chris Lattner99df25f2003-04-28 03:42:31 +0000[diff] [blame]329<li>Implement support for the "switch" instruction without requiring the
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]330 lower-switches pass.</li>
Reid Spencer750c4252004-07-02 06:27:12 +0000[diff] [blame]331<li>Implement interprocedural register allocation. The CallGraphSCCPass can be
332 used to implement a bottom-up analysis that will determine the *actual*
333 registers clobbered by a function. Use the pass to fine tune register usage
334 in callers based on *actual* registers used by the callee.</li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]335</ol>
336
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]337</div>
338
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]339<!-- ======================================================================= -->
Misha Brukman3497ae92004-03-11 21:26:29 +0000[diff] [blame]340<div class="doc_section">
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]341 <a name="misc_new">Miscellaneous Additions</a>
342</div>
343
344<div class="doc_text">
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]345
346<ol>
Misha Brukman85595942004-07-12 21:10:47 +0000[diff] [blame]347<li>Port the <a href="http://www-sop.inria.fr/mimosa/fp/Bigloo/">Bigloo</A>
Brian Gaeke2d261352004-04-07 15:31:23 +0000[diff] [blame]348Scheme compiler, from Manuel Serrano at INRIA Sophia-Antipolis, to
349output LLVM bytecode. It seems that it can already output .NET
350bytecode, JVM bytecode, and C, so LLVM would ostensibly be another good
351candidate.</li>
Misha Brukman85595942004-07-12 21:10:47 +0000[diff] [blame]352<li>Write a new frontend for C/C++ <b>in</b> C++, giving us the ability to
353directly use LLVM C++ classes from within a compiler rather than use
354C-based wrapper functions a la llvm-gcc. One possible starting point is the <a
355href="http://www.parashift.com/c++-faq-lite/compiler-dependencies.html#faq-37.11">C++
356yacc grammar by Ed Willink</a>.</li>
Brian Gaeke2d261352004-04-07 15:31:23 +0000[diff] [blame]357<li>Write a new frontend for some other language (Java? OCaml? Forth?)</li>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]358<li>Write a new backend for a target (IA64? MIPS? MMIX?)</li>
Misha Brukmandec8cfb2004-04-16 16:55:30 +0000[diff] [blame]359<li>Write a disassembler for machine code that would use TableGen to output
360<tt>MachineInstr</tt>s for transformations, optimizations, etc.</li>
Brian Gaekea2245a32004-03-10 19:15:50 +0000[diff] [blame]361<li>Random test vector generator: Use a C grammar to generate random C code;
362run it through llvm-gcc, then run a random set of passes on it using opt.
363Try to crash opt. When opt crashes, use bugpoint to reduce the test case and
364mail the result to yourself. Repeat ad infinitum.</li>
Misha Brukman1a49bc82004-03-10 19:22:29 +0000[diff] [blame]365<li>Design a simple, recognizable logo.</li>
Misha Brukman6d84e0d2004-10-01 18:36:22 +0000[diff] [blame]366<li>Improve the usefulness and utility of the Skeleton target backend:
367<ul>
368 <li>Convert the non-functional Skeleton target to become an abstract machine
369 target (choose some simple instructions, a register set, etc). This will
370 become a much more useful example of a backend since it would be a simple
Misha Brukmanb8cbd9f2004-10-04 17:36:35 +0000[diff] [blame]371 but <em>functional</em> backend. Examples of such architectures include MIX,
372 MMIX, <a
373 href="http://www.cs.cinvestav.mx/SC/prof_personal/adiaz/vhdl/DLX/">DLX</a>,
374 or come up with your own!</li>
Misha Brukman6d84e0d2004-10-01 18:36:22 +0000[diff] [blame]375 <li>Use the new Skeleton backend in the Interpreter: compile LLVM to Skeleton
376 target, and then interpret that code instead of LLVM. Performance win would
377 be the primary goal, as the number of registers would be a small constant
378 instead of unbounded, for example.</li>
379</ul></li>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]380</ol>
381
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]382</div>
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]383
384<!-- *********************************************************************** -->
Chris Lattnerb7131092003-02-06 19:39:21 +0000[diff] [blame]385
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]386<hr>
Misha Brukman836db502004-05-12 18:57:54 +0000[diff] [blame]387<address>
388 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
389 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
390 <a href="http://validator.w3.org/check/referer"><img
391 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
392
393 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
394 <a href="http://llvm.cs.uiuc.edu">LLVM Compiler Infrastructure</a><br>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]395 Last modified: $Date$
Misha Brukman836db502004-05-12 18:57:54 +0000[diff] [blame]396</address>
Misha Brukman076438b2003-11-14 17:34:38 +0000[diff] [blame]397
398</body>
399</html>