Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 5f39354d5c0c4358f51394406c2e7d2f1bc9c9c1 / . / docs / ReleaseNotes.html
blob: ee0a3a6d5d77423a51afe6d9ded0d70a33fb3acd [file] [log] [blame]
Dan Gohmanf17a25c2007-07-18 16:29:46 +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 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
6 <link rel="stylesheet" href="llvm.css" type="text/css">
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]7 <title>LLVM 2.5 Release Notes</title>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]8</head>
9<body>
10
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]11<div class="doc_title">LLVM 2.5 Release Notes</div>
Mikhail Glushenkov808cd3e2008-10-13 02:08:34 +0000[diff] [blame]12
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]13<ol>
14 <li><a href="#intro">Introduction</a></li>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]15 <li><a href="#subproj">Sub-project Status Update</a></li>
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]16 <li><a href="#externalproj">External Projects Using LLVM 2.5</a></li>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]17 <li><a href="#whatsnew">What's New in LLVM 2.5?</a></li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]18 <li><a href="GettingStarted.html">Installation Instructions</a></li>
19 <li><a href="#portability">Portability and Supported Platforms</a></li>
Dan Gohman0de549c2008-10-14 16:23:02 +0000[diff] [blame]20 <li><a href="#knownproblems">Known Problems</a></li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]21 <li><a href="#additionalinfo">Additional Information</a></li>
22</ol>
23
24<div class="doc_author">
Dan Gohman0de549c2008-10-14 16:23:02 +0000[diff] [blame]25 <p>Written by the <a href="http://llvm.org">LLVM Team</a></p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]26</div>
27
28<!-- *********************************************************************** -->
29<div class="doc_section">
30 <a name="intro">Introduction</a>
31</div>
32<!-- *********************************************************************** -->
33
34<div class="doc_text">
35
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]36<p>This document contains the release notes for the LLVM Compiler
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]37Infrastructure, release 2.5. Here we describe the status of LLVM, including
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]38major improvements from the previous release and significant known problems.
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]39All LLVM releases may be downloaded from the <a
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]40href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]41
42<p>For more information about LLVM, including information about the latest
43release, please check out the <a href="http://llvm.org/">main LLVM
44web site</a>. If you have questions or comments, the <a
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]45href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM Developer's Mailing
46List</a> is a good place to send them.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]47
Mikhail Glushenkov808cd3e2008-10-13 02:08:34 +0000[diff] [blame]48<p>Note that if you are reading this file from a Subversion checkout or the
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]49main LLVM web page, this document applies to the <i>next</i> release, not the
Gabor Greif7c65c4f2008-10-14 11:00:32 +0000[diff] [blame]50current one. To see the release notes for a specific release, please see the
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]51<a href="http://llvm.org/releases/">releases page</a>.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]52
53</div>
54
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]55<!-- Unfinished features in 2.5:
Chris Lattner8fda4952008-02-10 07:46:44 +0000[diff] [blame]56 Machine LICM
57 Machine Sinking
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]58 target-specific intrinsics
59 gold lto plugin
Chris Lattner56b64702009-02-25 07:32:53 +0000[diff] [blame]60 pre-alloc splitter, strong phi elim
Chris Lattner4f076402008-10-12 19:47:48 +0000[diff] [blame]61 llc -enable-value-prop, propagation of value info (sign/zero ext info) from
62 one MBB to another
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]63 debug info for optimized code
Chris Lattner5c109932009-02-26 07:29:17 +0000[diff] [blame]64 interpreter + libffi
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]65 postalloc scheduler: anti dependence breaking, hazard recognizer?
66
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]67initial support for debug line numbers when optimization enabled, not useful in
68 2.5 but will be for 2.6.
69
Chris Lattner8fda4952008-02-10 07:46:44 +0000[diff] [blame]70 -->
Mikhail Glushenkov808cd3e2008-10-13 02:08:34 +0000[diff] [blame]71
Chris Lattner4f076402008-10-12 19:47:48 +0000[diff] [blame]72 <!-- for announcement email:
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]73 -->
Chris Lattner8fda4952008-02-10 07:46:44 +0000[diff] [blame]74
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]75<!-- *********************************************************************** -->
76<div class="doc_section">
77 <a name="subproj">Sub-project Status Update</a>
Chris Lattnerf65224f2008-06-08 21:34:41 +0000[diff] [blame]78</div>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]79<!-- *********************************************************************** -->
Chris Lattnerf65224f2008-06-08 21:34:41 +0000[diff] [blame]80
81<div class="doc_text">
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]82<p>
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]83The LLVM 2.5 distribution currently consists of code from the core LLVM
Chris Lattnera3c683b2008-10-13 18:11:54 +0000[diff] [blame]84repository (which roughly includes the LLVM optimizers, code generators and
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]85supporting tools) and the llvm-gcc repository. In addition to this code, the
86LLVM Project includes other sub-projects that are in development. The two which
Chris Lattner2c3ef982008-10-11 18:18:28 +0000[diff] [blame]87are the most actively developed are the <a href="#clang">Clang Project</a> and
Nicolas Geoffray2b139382008-10-14 19:23:04 +0000[diff] [blame]88the <a href="#vmkit">VMKit Project</a>.
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]89</p>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]90
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]91</div>
92
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]93
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]94<!--=========================================================================-->
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]95<div class="doc_subsection">
Chris Lattnera3c683b2008-10-13 18:11:54 +0000[diff] [blame]96<a name="clang">Clang: C/C++/Objective-C Frontend Toolkit</a>
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]97</div>
98
99<div class="doc_text">
100
101<p>The <a href="http://clang.llvm.org/">Clang project</a> is an effort to build
102a set of new 'LLVM native' front-end technologies for the LLVM optimizer
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]103and code generator. While Clang is not included in the LLVM 2.5 release, it
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]104is continuing to make major strides forward in all areas. Its C and Objective-C
105parsing and code generation support is now very solid. For example, it is
106capable of successfully building many real applications for X86-32 and X86-64,
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]107including <a href="http://wiki.freebsd.org/BuildingFreeBSDWithClang">the FreeBSD
108kernel</a>. C++ is also making <a
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]109href="http://clang.llvm.org/cxx_status.html">incredible progress</a>, and work
110on templates has recently started.</p>
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]111
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]112<p>While Clang is not yet production quality, it is progressing very nicely and
113is quite usable for building many C and Objective-C applications. If you are
114interested in fast compiles and good diagnostics, we encourage you to try it out
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]115by <a href="http://clang.llvm.org/get_started.html">building from mainline</a>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]116and reporting any issues you hit to the <a
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]117href="http://lists.cs.uiuc.edu/mailman/listinfo/cfe-dev">Clang front-end mailing
118list</a>.</p>
119
120<p>In the LLVM 2.5 time-frame, the Clang team has made many improvements:</p>
Bill Wendling385b0d32008-10-27 09:27:33 +0000[diff] [blame]121
Daniel Dunbarf3e35782008-10-14 23:25:09 +0000[diff] [blame]122<ul>
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]123<li>Clang now has a new driver, which is focused on providing a GCC-compatible
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]124 interface.</li>
125<li>The X86-64 ABI is now supported.</li>
126<li>Precompiled header support is now implemented.</li>
Chris Lattnerc97507b2009-02-25 05:09:54 +0000[diff] [blame]127<li>Objective-C support is significantly improved beyond LLVM 2.4, supporting
128 many features, such as Objective-C Garbage Collection.</li>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]129<li>Many many bugs are fixed and many features have been added.</li>
Chris Lattnera3c683b2008-10-13 18:11:54 +0000[diff] [blame]130</div>
131
132<!--=========================================================================-->
133<div class="doc_subsection">
134<a name="clangsa">Clang Static Analyzer</a>
135</div>
136
137<div class="doc_text">
138
Ted Kremeneka0d77d72009-02-27 07:01:32 +0000[diff] [blame]139<p>Previously announced in the last LLVM release, the Clang project also
140includes an early stage static source code analysis tool for <a
141href="http://clang.llvm.org/StaticAnalysis.html">automatically finding bugs</a>
142in C and Objective-C programs. The tool performs a growing set of checks to find
143bugs that occur on a specific path within a program.</p>
Chris Lattnerc2d84672008-10-14 00:52:49 +0000[diff] [blame]144
Ted Kremeneka0d77d72009-02-27 07:01:32 +0000[diff] [blame]145<p>In the LLVM 2.5 time-frame there have been many significant improvements to
146the analyzer's core path simulation engine and machinery for generating
147path-based bug reports to end-users. Particularly noteworthy improvements
148include experimental support for full field-sensitivity and reasoning about heap
149objects as well as an improved value-constraints subengine that does a much
150better job of reasoning about inequality relationships (e.g., <tt>x &gt; 2</tt>)
151between variables and constants.
Ted Kremenek0437b2c2008-10-14 05:14:21 +0000[diff] [blame]152
Ted Kremeneka0d77d72009-02-27 07:01:32 +0000[diff] [blame]153<p>The set of checks performed by the static analyzer continue to expand, and
154future plans for the tool include full source-level inter-procedural analysis
155and deeper checks such as buffer overrun detection. There are many opportunities
156to extend and enhance the static analyzer, and anyone interested in working on
157this project is encouraged to get involved!</p>
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]158
159</div>
160
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]161<!--=========================================================================-->
162<div class="doc_subsection">
Nicolas Geoffray2b139382008-10-14 19:23:04 +0000[diff] [blame]163<a name="vmkit">VMKit: JVM/CLI Virtual Machine Implementation</a>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]164</div>
165
166<div class="doc_text">
167<p>
Nicolas Geoffray2b139382008-10-14 19:23:04 +0000[diff] [blame]168The <a href="http://vmkit.llvm.org/">VMKit project</a> is an implementation of
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]169a JVM and a CLI Virtual Machines (Microsoft .NET is an
170implementation of the CLI) using the Just-In-Time compiler of LLVM.</p>
171
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]172<p>Following LLVM 2.5, VMKit has its first release that you can find on its
Nicolas Geoffray20cfdd42008-10-15 20:25:04 +0000[diff] [blame]173<a href="http://vmkit.llvm.org/releases/">webpage</a>. The release includes
174bug fixes, cleanup and new features. The major changes are:</p>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]175
Nicolas Geoffray2b139382008-10-14 19:23:04 +0000[diff] [blame]176<ul>
177
Chris Lattner47bcf612009-02-26 05:20:36 +0000[diff] [blame]178<li>Ahead of Time compiler: compiles .class files to llvm .bc. VMKit uses this
179functionality to native compile the standard classes (eg java.lang.String).
180Users can compile AOT .class files into dynamic libraries and run them with the
181help of VMKit.</li>
182
183<li>New exception model: the dwarf exception model is very slow for
184exception-intensive applications, so the JVM has had a new implementation of
185exceptions which check at each function call if an exception happened. There is
186a low performance penalty on applications without exceptions, but it is a big
187gain for exception-intensive applications. For example the jack benchmark in
188Spec JVM98 is 6x faster (performance gain of 83%).</li>
189
190<li>New support for OSX/X64, Linux/X64 (with the Boehm GC), Linux/ppc32.</li>
Nicolas Geoffray2b139382008-10-14 19:23:04 +0000[diff] [blame]191
192</ul>
Chris Lattner78a130b2008-10-13 18:01:01 +0000[diff] [blame]193</div>
194
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]195<!-- *********************************************************************** -->
196<div class="doc_section">
197 <a name="externalproj">External Projects Using LLVM 2.5</a>
198</div>
199<!-- *********************************************************************** -->
200
201<!--=========================================================================-->
202<div class="doc_subsection">
203<a name="pure">Pure</a>
204</div>
205
206<div class="doc_text">
207<p>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]208<a href="http://pure-lang.googlecode.com/">Pure</a>
209is an algebraic/functional programming language based on term rewriting.
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]210Programs are collections of equations which are used to evaluate expressions in
211a symbolic fashion. Pure offers dynamic typing, eager and lazy evaluation,
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]212lexical closures, a hygienic macro system (also based on term rewriting),
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]213built-in list and matrix support (including list and matrix comprehensions) and
214an easy-to-use C interface. The interpreter uses LLVM as a backend to
215 JIT-compile Pure programs to fast native code.</p>
216
217<p>In addition to the usual algebraic data structures, Pure also has
218MATLAB-style matrices in order to support numeric computations and signal
219processing in an efficient way. Pure is mainly aimed at mathematical
220applications right now, but it has been designed as a general purpose language.
221The dynamic interpreter environment and the C interface make it possible to use
222it as a kind of functional scripting language for many application areas.
223</p>
224</div>
225
226
227<!--=========================================================================-->
228<div class="doc_subsection">
229<a name="ldc">LLVM D Compiler</a>
230</div>
231
232<div class="doc_text">
233<p>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]234<a href="http://www.dsource.org/projects/ldc">LDC</a> is an implementation of
235the D Programming Language using the LLVM optimizer and code generator.
236LDC project works great with the LLVM 2.5 release. General improvmenets in this
237cycle have included new inline asm constraint handling, better debug info
238support, general bugfixes, and better x86-64 support. This has allowed
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]239some major improvements in LDC, getting us much closer to being as
240fully featured as the original DMD compiler from DigitalMars.
241</p>
242</div>
243
Chris Lattner1ad599a2009-02-28 18:58:01 +0000[diff] [blame]244<!--=========================================================================-->
245<div class="doc_subsection">
246<a name="RoadsendPHP">Roadsend PHP</a>
247</div>
248
249<div class="doc_text">
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]250<p><a href="http://code.roadsend.com/rphp">Roadsend PHP</a> (rphp) is an open
251source compiler for the PHP programming language that uses LLVM for its
252optimizer, JIT, and static compiler. This is a reimplementation of an earlier
253project that is now based on the LLVM.</p>
Chris Lattner1ad599a2009-02-28 18:58:01 +0000[diff] [blame]254</div>
Chris Lattnerb2cc5022009-02-26 22:33:38 +0000[diff] [blame]255
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]256
257<!-- *********************************************************************** -->
258<div class="doc_section">
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]259 <a name="whatsnew">What's New in LLVM 2.5?</a>
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]260</div>
261<!-- *********************************************************************** -->
262
Chris Lattner5cce7a52008-06-08 22:59:35 +0000[diff] [blame]263<div class="doc_text">
264
Bill Wendling385b0d32008-10-27 09:27:33 +0000[diff] [blame]265<p>This release includes a huge number of bug fixes, performance tweaks, and
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]266minor improvements. Some of the major improvements and new features are listed
267in this section.
Chris Lattner5cce7a52008-06-08 22:59:35 +0000[diff] [blame]268</p>
269</div>
270
Chris Lattnerf6a0d382008-06-08 21:58:17 +0000[diff] [blame]271<!--=========================================================================-->
Chris Lattnerf65224f2008-06-08 21:34:41 +0000[diff] [blame]272<div class="doc_subsection">
Chris Lattnere0798082008-02-06 18:00:06 +0000[diff] [blame]273<a name="majorfeatures">Major New Features</a>
274</div>
275
276<div class="doc_text">
277
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]278<p>LLVM 2.5 includes several major new capabilities:</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]279
280<ul>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]281<li>LLVM 2.5 includes a brand new <a
282href="http://en.wikipedia.org/wiki/XCore">XCore</a> backend.</li>
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]283
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]284<li>llvm-gcc now generally supports the GFortan front-end, and the precompiled
285release binaries now support Fortran, even on Mac OS/X.</li>
Chris Lattnera53f4972009-02-25 06:34:50 +0000[diff] [blame]286
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]287<li>CMake is now used by the <a href="GettingStartedVS.html">LLVM build process
288on Windows</a>. It automatically generates Visual Studio project files (and
289more) from a set of simple text files. This makes it much easier to
290maintain. In time, we'd like to standardize on CMake for everything.</li>
Chris Lattner5c109932009-02-26 07:29:17 +0000[diff] [blame]291
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]292<li>LLVM 2.5 now uses (and includes) Google Test for unit testing.</li>
Chris Lattner5c109932009-02-26 07:29:17 +0000[diff] [blame]293
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]294<li>The LLVM native code generator now supports arbitrary precision integers.
295Types like <tt>i33</tt> have long been valid in the LLVM IR, but were previously
296only supported by the interpreter. Note that the C backend still does not
297support these.</li>
Chris Lattner56b64702009-02-25 07:32:53 +0000[diff] [blame]298
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]299<li>LLVM 2.5 no longer uses 'bison', so it is easier to build on Windows.</li>
Chris Lattner5ba16862008-02-10 08:18:42 +0000[diff] [blame]300</ul>
301
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]302</div>
303
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]304
305<!--=========================================================================-->
306<div class="doc_subsection">
Chris Lattnerf65224f2008-06-08 21:34:41 +0000[diff] [blame]307<a name="llvm-gcc">llvm-gcc 4.2 Improvements</a>
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]308</div>
309
310<div class="doc_text">
311
Chris Lattner4352cc82008-10-13 20:47:20 +0000[diff] [blame]312<p>LLVM fully supports the llvm-gcc 4.2 front-end, which marries the GCC
313front-ends and driver with the LLVM optimizer and code generator. It currently
314includes support for the C, C++, Objective-C, Ada, and Fortran front-ends.</p>
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]315
Chris Lattner5cce7a52008-06-08 22:59:35 +0000[diff] [blame]316<ul>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]317<li>In this release, the GCC inliner is completely disabled. Previously the GCC
318inliner was used to handle always-inline functions and other cases. This caused
319problems with code size growth, and it is completely disabled in this
320release.</li>
321
322<li>llvm-gcc (and LLVM in general) now support code generation for stack
323canaries, which is an effective form of <a
324href="http://en.wikipedia.org/wiki/Stack-smashing_protection">buffer overflow
325protection</a>. llvm-gcc supports this with the <tt>-fstack-protector</tt>
326command line option (just like GCC). In LLVM IR, you can request code
327generation for stack canaries with function attributes.
328</li>
Gabor Greif0e49cc82008-06-09 06:06:18 +0000[diff] [blame]329</ul>
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]330
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]331</div>
332
333
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]334<!--=========================================================================-->
335<div class="doc_subsection">
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]336<a name="coreimprovements">LLVM IR and Core Improvements</a>
Chris Lattnerc3649452008-02-10 08:17:19 +0000[diff] [blame]337</div>
338
339<div class="doc_text">
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]340<p>LLVM IR has several new features that are used by our existing front-ends and
341can be useful if you are writing a front-end for LLVM:</p>
Chris Lattnerc3649452008-02-10 08:17:19 +0000[diff] [blame]342
Chris Lattnerc5a4e312008-06-05 06:25:56 +0000[diff] [blame]343<ul>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]344<li>The <a href="LangRef.html#i_shufflevector">shufflevector</a> instruction
345has been generalized to allow different shuffle mask width than its input
346vectors. This allows you to use shufflevector to combine two
347"&lt;4 x float&gt;" vectors into a "&lt;8 x float&gt;" for example.</li>
348
349<li>LLVM IR now supports new intrinsics for computing and acting on <a
350href="LangRef.html#int_overflow">overflow of integer operations</a>. This allows
351efficient code generation for languages that must trap or throw an exception on
352overflow. While these intrinsics work on all targets, they only generate
353efficient code on X86 so far.</li>
354
355<li>LLVM IR now supports a new <a href="LangRef.html#linkage">private
356linkage</a> type to produce labels that are stripped by the assembler before it
357produces a .o file (thus they are invisible to the linker).</li>
358
359<li>LLVM IR supports two new attributes for better alias analysis. The <a
360href="LangRef.html#paramattrs">noalias</a> attribute can now be used on the
361return value of a function to indicate that it returns new memory (e.g.
362'malloc', 'calloc', etc).</li>
363
364<li>The new <a href="LangRef.html#paramattrs">nocapture</a> attribute can be
365used on pointer arguments to functions that access through but do not return the
366pointer in a data structure that out lives the call (e.g. 'strlen', 'memcpy',
367and many others). The simplifylibcalls pass applies these attributes to
368standard libc functions.</li>
369
370<li>The parser for ".ll" files in lib/AsmParser is now completely rewritten as a
371recursive descent parser. This parser produces better error messages (including
372caret diagnostics) is less fragile (less likely to crash on strange things) does
373not leak memory, is more efficient, and eliminates LLVM's last use of the
374'bison' tool.</li>
375
376<li>Debug information representation and manipulation internals have been
377 consolidated to use a new set of classes in
378 <tt>llvm/Analysis/DebugInfo.h</tt> classes. These routines are more
379 efficient, robust, and extensible and replace the older mechanisms.
380 llvm-gcc, clang, and the code generator now use them to create and process
381 debug information.</li>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]382
Chris Lattnerc3649452008-02-10 08:17:19 +0000[diff] [blame]383</ul>
Mikhail Glushenkov808cd3e2008-10-13 02:08:34 +0000[diff] [blame]384
Chris Lattnerc3649452008-02-10 08:17:19 +0000[diff] [blame]385</div>
386
387<!--=========================================================================-->
388<div class="doc_subsection">
Chris Lattner8fda4952008-02-10 07:46:44 +0000[diff] [blame]389<a name="optimizer">Optimizer Improvements</a>
390</div>
391
392<div class="doc_text">
393
Chris Lattnerea9c2242008-10-13 21:50:36 +0000[diff] [blame]394<p>In addition to a huge array of bug fixes and minor performance tweaks, this
395release includes a few major enhancements and additions to the optimizers:</p>
Chris Lattner8fda4952008-02-10 07:46:44 +0000[diff] [blame]396
397<ul>
398
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]399<li>The loop optimizer now improves floating point induction variables in
400several ways, including adding shadow induction variables to avoid
401"integer &lt;-&gt; floating point" conversions in loops when safe.</li>
402
403<li>The "-mem2reg" pass is now much faster on code with huge basic blocks.</li>
404
405<li>The "-jump-threading" pass is more powerful: it is iterative
406 and handles threading based on values with fully and partially redundant
407 loads.</li>
408
409<li>The "-memdep" memory dependence analysis pass (used by GVN and memcpyopt) is
410 both faster and more aggressive.</li>
411
412<li>The "-scalarrepl" scalar replacement of aggregates pass is more aggressive
413 about promoting unions to registers.</li>
414</li>
Chris Lattnerea9c2242008-10-13 21:50:36 +0000[diff] [blame]415
Chris Lattner0eef6e42008-06-08 02:45:07 +0000[diff] [blame]416</ul>
417
418</div>
419
420<!--=========================================================================-->
421<div class="doc_subsection">
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]422<a name="codegen">Target Independent Code Generator Improvements</a>
Chris Lattner0eef6e42008-06-08 02:45:07 +0000[diff] [blame]423</div>
424
425<div class="doc_text">
426
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]427<p>We have put a significant amount of work into the code generator
428infrastructure, which allows us to implement more aggressive algorithms and make
429it run faster:</p>
Chris Lattner0eef6e42008-06-08 02:45:07 +0000[diff] [blame]430
431<ul>
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]432<li>The <a href="WritingAnLLVMBackend.html">Writing an LLVM Compiler
433Backend</a> document has been greatly expanded and is substantially more
434complete.</li>
435
436<li>The SelectionDAG type legalization logic has been completely rewritten, is
437now more powerful (it supports arbitrary precision integer types for example),
438and more correct in several corner cases. The type legalizer converts
439operations on types that are not natively supported by the target machine into
440equivalent code sequences that only use natively supported types. The old type
441legalizer is still available (for now) and will be used if
442<tt>-disable-legalize-types</tt> is passed to the code generator.
Duncan Sands335d5e42009-02-25 11:50:08 +0000[diff] [blame]443</li>
Chris Lattner0eef6e42008-06-08 02:45:07 +0000[diff] [blame]444
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]445<li>The code generator now supports widening illegal vectors to larger legal
446ones (for example, converting operations on &lt;3 x float&gt; to work on
447&lt;4 x float&gt;) which is very important for common graphics
448applications.</li>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]449
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]450<li>The assembly printers for each target are now split out into their own
451libraries that are separate from the main code generation logic. This reduces
452code size of JIT compilers by not requiring them to be linked in.</li>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]453
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]454<li>The 'fast' instruction selection path (used at -O0 and for fast JIT
455 compilers) now supports accelerating codegen for code that uses exception
456 handling constructs.</li>
457
458<li>The optional PBQP register allocator now supports register coalescing.</li>
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]459</ul>
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]460</div>
461
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]462<!--=========================================================================-->
463<div class="doc_subsection">
464<a name="x86">X86-32 and X86-64 Target Improvements</a>
465</div>
466
467<div class="doc_text">
468<p>New features of the X86 target include:
469</p>
470
471<ul>
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]472<li>The "<a href="LangRef.html#int_returnaddress">llvm.returnaddress</a>"
473intrinsic (which is used to implement "__builtin_return_address") now supports
474non-zero stack depths on X86.</li>
475
476<li>The X86 backend now supports code generation of vector shift operations
477using SSE instructions.</li>
478
479<li>X86-64 code generation now takes advantage of red zone (unless
480-mno-red-zone option is specified).</li>
481
482<li>The X86 backend now supports using address space #256 in LLVM IR as a way of
483performing memory references off the GS segment register. This allows a
484front-end to take advantage of very low-level programming techniques when
485targetting X86 CPUs. See test/CodeGen/X86/movgs.ll for a simple example.</li>
486
487<li>The X86 backend now supports a <tt>-disable-mmx</tt> command line option to
488 prevent use of MMX even on chips that support it. This is important for cases
489 where code does not contain the proper "llvm.x86.mmx.emms" intrinsics.</li>
490
491<li>The X86 JIT now detects the new Intel "<a
492 href="http://en.wikipedia.org/wiki/Intel_Core_i7">Core i7</a>" and <a
493 href="http://en.wikipedia.org/wiki/Intel_Atom">Atom</a>" chips,
494 auto-configuring itself appropriately for the features of these chips.</li>
495
496<li>The JIT now supports exception handling constructs on Linux/X86-64 and
497 Darwin/x86-64.</li>
498<li>The JIT supports Thread Local Storage (TLS) on Linux/X86-32 but not yet on
499 X86-64.</li>
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]500</ul>
501
502
503</div>
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]504
505<!--=========================================================================-->
506<div class="doc_subsection">
Chris Lattnere1fbd752009-02-26 07:32:11 +0000[diff] [blame]507<a name="pic16">PIC16 Target Improvements</a>
508</div>
509
510<div class="doc_text">
511<p>New features of the PIC16 target include:
512</p>
513
514<ul>
515<li>Both direct and indirect load/stores work now.</li>
516<li>Logical, bitwise and conditional operations now work for integer data
517types.</li>
518<li>Function calls involving basic types work now.</li>
519<li>Support for integer arrays.</li>
520<li>Compiler can now emit libcalls for operations not support by m/c insns.</li>
521<li>Support for both data and rom address spaces.</li>
522</li>
523</ul>
524
525<p>Things not yet supported:</p>
526
527<ul>
528<li>Floating point.</li>
529<li>Passing/returning aggregate types to/from functions.</li>
530<li>Variable arguments.</li>
531<li>Indirect function calls.</li>
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]532<li>Interrupts/programs.</li>
Chris Lattnere1fbd752009-02-26 07:32:11 +0000[diff] [blame]533<li>Debug info.</li>
534</li>
535</ul>
536
537</div>
538
539
540<!--=========================================================================-->
541<div class="doc_subsection">
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]542<a name="llvmc">Improvements in LLVMC</a>
543</div>
544
545<div class="doc_text">
546<p>New features include:</p>
547
548<ul>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]549<li>Beginning with LLVM 2.5, <tt>llvmc2</tt> is known as
550 just <tt>llvmc</tt>. The old <tt>llvmc</tt> driver was removed.</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]551
552<li>The Clang plugin was substantially improved and is now enabled
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]553 by default. The command <tt>llvmc --clang</tt> can be now used as a
554 synonym to <tt>ccc</tt>.</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]555
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]556<li>There is now a <tt>--check-graph</tt> option which is supposed to catch
557 common errors like multiple default edges, mismatched output/input language
558 names and cycles. In general, these checks can't be done at compile-time
559 because of the need to support plugins.</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]560
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]561<li>Plugins are now more flexible and can refer to compilation graph nodes and
562 options defined in other plugins. To manage dependencies, a priority-sorting
563 mechanism was introduced. This change affects the TableGen file syntax; see the
564 documentation for details.</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]565
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]566<li>Hooks can now be provided with arguments. The syntax is "<tt>$CALL(MyHook,
567 'Arg1', 'Arg2', 'Arg #3')</tt>".</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]568
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]569<li>A new option type: multi-valued option, for options that take more than one
570 argument (for example, "<tt>-foo a b c</tt>").</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]571
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]572<li>New option properties: '<tt>one_or_more</tt>', '<tt>zero_or_more</tt>',
573'<tt>hidden</tt>' and '<tt>really_hidden</tt>'.</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]574
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]575<li>The '<tt>case</tt>' expression gained an '<tt>error</tt>' action and
576 an '<tt>empty</tt>' test (equivalent to "<tt>(not (not_empty ...))</tt>").</li>
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]577
578<li>Documentation now looks more consistent to the rest of the LLVM
579 docs. There is also a man page now.</li>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]580
Chris Lattner7ccc40d2009-03-01 02:30:21 +0000[diff] [blame]581</ul>
582
583</div>
584
Chris Lattnere45f18b2008-06-05 08:02:49 +0000[diff] [blame]585
586<!--=========================================================================-->
587<div class="doc_subsection">
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]588<a name="otherimprovements">Other Improvements</a>
589</div>
590
591<div class="doc_text">
592<p>New features include:
593</p>
594
595<ul>
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]596<li>?</li>
Tanya Lattnerc79f7542008-11-04 18:40:27 +0000[diff] [blame]597
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]598</ul>
Mikhail Glushenkov808cd3e2008-10-13 02:08:34 +0000[diff] [blame]599
Chris Lattnerb0d6c502007-09-21 03:54:09 +0000[diff] [blame]600</div>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]601
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]602<!--=========================================================================-->
603<div class="doc_subsection">
604<a name="changes">Major Changes and Removed Features</a>
605</div>
606
607<div class="doc_text">
608
Chris Lattnercc64a322008-10-13 22:06:31 +0000[diff] [blame]609<p>If you're already an LLVM user or developer with out-of-tree changes based
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]610on LLVM 2.4, this section lists some "gotchas" that you may run into upgrading
Chris Lattnercc64a322008-10-13 22:06:31 +0000[diff] [blame]611from the previous release.</p>
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]612
613<ul>
614
Chris Lattner5f393542009-03-02 03:24:11 +0000[diff] [blame^]615<li>llvm-gcc defaults to -fno-math-errno on all X86 targets.</li>
Chris Lattnercc64a322008-10-13 22:06:31 +0000[diff] [blame]616</ul>
617
618
619<p>In addition, many APIs have changed in this release. Some of the major LLVM
620API changes are:</p>
621
622<ul>
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]623<li>?</li>
Devang Patel0e3da1a2008-10-14 20:03:43 +0000[diff] [blame]624</ul>
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]625
Chris Lattner95d25072009-02-25 04:41:31 +0000[diff] [blame]626<li>?</li>
Chris Lattner92eacf82008-10-13 17:57:36 +0000[diff] [blame]627</ul>
628
629</div>
630
631
632
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]633<!-- *********************************************************************** -->
634<div class="doc_section">
635 <a name="portability">Portability and Supported Platforms</a>
636</div>
637<!-- *********************************************************************** -->
638
639<div class="doc_text">
640
641<p>LLVM is known to work on the following platforms:</p>
642
643<ul>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]644<li>Intel and AMD machines (IA32, X86-64, AMD64, EMT-64) running Red Hat
Chris Lattnerb1ad47c2008-11-10 05:40:34 +0000[diff] [blame]645Linux, Fedora Core and FreeBSD (and probably other unix-like systems).</li>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]646<li>PowerPC and X86-based Mac OS X systems, running 10.3 and above in 32-bit
Chris Lattnerb1ad47c2008-11-10 05:40:34 +0000[diff] [blame]647and 64-bit modes.</li>
Chris Lattnere0798082008-02-06 18:00:06 +0000[diff] [blame]648<li>Intel and AMD machines running on Win32 using MinGW libraries (native).</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]649<li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
650 support is available for native builds with Visual C++).</li>
Gabor Greif5e75e652008-06-05 18:39:01 +0000[diff] [blame]651<li>Sun UltraSPARC workstations running Solaris 10.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]652<li>Alpha-based machines running Debian GNU/Linux.</li>
Gabor Greif5e75e652008-06-05 18:39:01 +0000[diff] [blame]653<li>Itanium-based (IA64) machines running Linux and HP-UX.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]654</ul>
655
Chris Lattnerba948522008-06-05 06:57:39 +0000[diff] [blame]656<p>The core LLVM infrastructure uses GNU autoconf to adapt itself
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]657to the machine and operating system on which it is built. However, minor
658porting may be required to get LLVM to work on new platforms. We welcome your
659portability patches and reports of successful builds or error messages.</p>
660
661</div>
662
663<!-- *********************************************************************** -->
664<div class="doc_section">
665 <a name="knownproblems">Known Problems</a>
666</div>
667<!-- *********************************************************************** -->
668
669<div class="doc_text">
670
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]671<p>This section contains significant known problems with the LLVM system,
Chris Lattnerb1ad47c2008-11-10 05:40:34 +0000[diff] [blame]672listed by component. If you run into a problem, please check the <a
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]673href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
674there isn't already one.</p>
675
676</div>
677
678<!-- ======================================================================= -->
679<div class="doc_subsection">
680 <a name="experimental">Experimental features included with this release</a>
681</div>
682
683<div class="doc_text">
684
685<p>The following components of this LLVM release are either untested, known to
686be broken or unreliable, or are in early development. These components should
687not be relied on, and bugs should not be filed against them, but they may be
688useful to some people. In particular, if you would like to work on one of these
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]689components, please contact us on the <a
690href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]691
692<ul>
Gabor Greifaa53e562008-10-14 11:31:14 +0000[diff] [blame]693<li>The MSIL, IA64, Alpha, SPU, MIPS, and PIC16 backends are experimental.</li>
Duncan Sandsa73561c2008-06-08 20:18:35 +0000[diff] [blame]694<li>The llc "<tt>-filetype=asm</tt>" (the default) is the only supported
Chris Lattnerbd633de92008-02-06 06:30:34 +0000[diff] [blame]695 value for this option.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]696</ul>
697
698</div>
699
700<!-- ======================================================================= -->
701<div class="doc_subsection">
702 <a name="x86-be">Known problems with the X86 back-end</a>
703</div>
704
705<div class="doc_text">
706
707<ul>
Anton Korobeynikov77d15e22008-06-08 10:24:13 +0000[diff] [blame]708 <li>The X86 backend does not yet support
709 all <a href="http://llvm.org/PR879">inline assembly that uses the X86
710 floating point stack</a>. It supports the 'f' and 't' constraints, but not
711 'u'.</li>
Chris Lattnerd95b5962008-06-08 21:19:07 +0000[diff] [blame]712 <li>The X86 backend generates inefficient floating point code when configured
713 to generate code for systems that don't have SSE2.</li>
Duncan Sands8b4f3772008-06-08 19:38:43 +0000[diff] [blame]714 <li>Win64 code generation wasn't widely tested. Everything should work, but we
Anton Korobeynikov77d15e22008-06-08 10:24:13 +0000[diff] [blame]715 expect small issues to happen. Also, llvm-gcc cannot build mingw64 runtime
716 currently due
717 to <a href="http://llvm.org/PR2255">several</a>
Chris Lattnerf06141a2008-06-08 23:12:47 +0000[diff] [blame]718 <a href="http://llvm.org/PR2257">bugs</a> due to lack of support for the
719 'u' inline assembly constraint and X87 floating point inline assembly.</li>
Dan Gohman3882ecb2008-06-08 23:05:11 +0000[diff] [blame]720 <li>The X86-64 backend does not yet support the LLVM IR instruction
Chris Lattner47202c32009-03-02 02:37:32 +0000[diff] [blame]721 <tt>va_arg</tt>. Currently, the llvm-gcc and front-ends support variadic
Dan Gohman3882ecb2008-06-08 23:05:11 +0000[diff] [blame]722 argument constructs on X86-64 by lowering them manually.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]723</ul>
724
725</div>
726
727<!-- ======================================================================= -->
728<div class="doc_subsection">
729 <a name="ppc-be">Known problems with the PowerPC back-end</a>
730</div>
731
732<div class="doc_text">
733
734<ul>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]735<li>The Linux PPC32/ABI support needs testing for the interpreter and static
736compilation, and lacks support for debug information.</li>
737</ul>
738
739</div>
740
741<!-- ======================================================================= -->
742<div class="doc_subsection">
743 <a name="arm-be">Known problems with the ARM back-end</a>
744</div>
745
746<div class="doc_text">
747
748<ul>
749<li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
Duncan Sands1827fa62007-09-26 15:59:54 +0000[diff] [blame]750processors, thumb programs can crash or produce wrong
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]751results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
752<li>Compilation for ARM Linux OABI (old ABI) is supported, but not fully tested.
753</li>
Chris Lattnerd95b5962008-06-08 21:19:07 +0000[diff] [blame]754<li>There is a bug in QEMU-ARM (&lt;= 0.9.0) which causes it to incorrectly
755 execute
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]756programs compiled with LLVM. Please use more recent versions of QEMU.</li>
757</ul>
758
759</div>
760
761<!-- ======================================================================= -->
762<div class="doc_subsection">
763 <a name="sparc-be">Known problems with the SPARC back-end</a>
764</div>
765
766<div class="doc_text">
767
768<ul>
769<li>The SPARC backend only supports the 32-bit SPARC ABI (-m32), it does not
770 support the 64-bit SPARC ABI (-m64).</li>
771</ul>
772
773</div>
774
775<!-- ======================================================================= -->
776<div class="doc_subsection">
Bruno Cardoso Lopes62d45d72008-10-25 14:56:26 +0000[diff] [blame]777 <a name="mips-be">Known problems with the MIPS back-end</a>
778</div>
779
780<div class="doc_text">
781
782<ul>
783<li>The O32 ABI is not fully supported.</li>
784<li>64-bit MIPS targets are not supported yet.</li>
785</ul>
786
787</div>
788
789<!-- ======================================================================= -->
790<div class="doc_subsection">
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]791 <a name="alpha-be">Known problems with the Alpha back-end</a>
792</div>
793
794<div class="doc_text">
795
796<ul>
797
798<li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
799appropriate nops inserted to ensure restartability.</li>
800
801</ul>
802</div>
803
804<!-- ======================================================================= -->
805<div class="doc_subsection">
806 <a name="ia64-be">Known problems with the IA64 back-end</a>
807</div>
808
809<div class="doc_text">
810
811<ul>
Chris Lattnerf06141a2008-06-08 23:12:47 +0000[diff] [blame]812<li>The Itanium backend is highly experimental, and has a number of known
813 issues. We are looking for a maintainer for the Itanium backend. If you
Gabor Greif45fc8b12008-10-15 10:47:24 +0000[diff] [blame]814 are interested, please contact the LLVMdev mailing list.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]815</ul>
816
817</div>
818
819<!-- ======================================================================= -->
820<div class="doc_subsection">
821 <a name="c-be">Known problems with the C back-end</a>
822</div>
823
824<div class="doc_text">
825
826<ul>
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]827<li><a href="http://llvm.org/PR802">The C backend has only basic support for
828 inline assembly code</a>.</li>
Chris Lattnere0194462007-09-26 06:01:35 +0000[diff] [blame]829<li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
830 C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
Duncan Sandsa73561c2008-06-08 20:18:35 +0000[diff] [blame]831 C++ code compiled with llc or native compilers.</li>
Duncan Sandsbd0ad6e2008-02-10 13:40:55 +0000[diff] [blame]832<li>The C backend does not support all exception handling constructs.</li>
Duncan Sandse6e264c2009-02-25 11:51:54 +0000[diff] [blame]833<li>The C backend does not support arbitrary precision integers.</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]834</ul>
835
836</div>
837
838
839<!-- ======================================================================= -->
840<div class="doc_subsection">
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]841 <a name="c-fe">Known problems with the llvm-gcc C front-end</a>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]842</div>
843
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]844<div class="doc_text">
845
Chris Lattnere0798082008-02-06 18:00:06 +0000[diff] [blame]846<p>llvm-gcc does not currently support <a href="http://llvm.org/PR869">Link-Time
847Optimization</a> on most platforms "out-of-the-box". Please inquire on the
Gabor Greif45fc8b12008-10-15 10:47:24 +0000[diff] [blame]848LLVMdev mailing list if you are interested.</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]849
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]850<p>The only major language feature of GCC not supported by llvm-gcc is
851 the <tt>__builtin_apply</tt> family of builtins. However, some extensions
852 are only supported on some targets. For example, trampolines are only
Duncan Sandsa73561c2008-06-08 20:18:35 +0000[diff] [blame]853 supported on some targets (these are used when you take the address of a
854 nested function).</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]855
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]856<p>If you run into GCC extensions which are not supported, please let us know.
857</p>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]858
859</div>
860
861<!-- ======================================================================= -->
862<div class="doc_subsection">
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]863 <a name="c++-fe">Known problems with the llvm-gcc C++ front-end</a>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]864</div>
865
866<div class="doc_text">
867
868<p>The C++ front-end is considered to be fully
869tested and works for a number of non-trivial programs, including LLVM
870itself, Qt, Mozilla, etc.</p>
871
872<ul>
Anton Korobeynikovd9a17932008-10-11 18:27:16 +0000[diff] [blame]873<li>Exception handling works well on the X86 and PowerPC targets. Currently
Chris Lattnerb1ad47c2008-11-10 05:40:34 +0000[diff] [blame]874 only Linux and Darwin targets are supported (both 32 and 64 bit).</li>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]875</ul>
876
877</div>
878
Chris Lattner6c8a1b82008-10-30 03:58:13 +0000[diff] [blame]879<!-- ======================================================================= -->
880<div class="doc_subsection">
881 <a name="fortran-fe">Known problems with the llvm-gcc Fortran front-end</a>
882</div>
883
884<div class="doc_text">
Gabor Greif1fc8fbc2008-11-04 21:50:59 +0000[diff] [blame]885<ul>
Chris Lattner6c8a1b82008-10-30 03:58:13 +0000[diff] [blame]886<li>Fortran support generally works, but there are still several unresolved bugs
887 in Bugzilla. Please see the tools/gfortran component for details.</li>
Mikhail Glushenkovf61066a2009-03-01 18:09:47 +0000[diff] [blame]888
Chris Lattner6c8a1b82008-10-30 03:58:13 +0000[diff] [blame]889<li>The Fortran front-end currently does not build on Darwin (without tweaks)
890 due to unresolved dependencies on the C front-end.</li>
Gabor Greif1fc8fbc2008-11-04 21:50:59 +0000[diff] [blame]891</ul>
Chris Lattner6c8a1b82008-10-30 03:58:13 +0000[diff] [blame]892</div>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]893
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]894<!-- ======================================================================= -->
895<div class="doc_subsection">
Chris Lattner499c07c2008-06-05 06:35:40 +0000[diff] [blame]896 <a name="ada-fe">Known problems with the llvm-gcc Ada front-end</a>
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]897</div>
898
899<div class="doc_text">
900The llvm-gcc 4.2 Ada compiler works fairly well, however this is not a mature
901technology and problems should be expected.
902<ul>
Duncan Sandsa73561c2008-06-08 20:18:35 +0000[diff] [blame]903<li>The Ada front-end currently only builds on X86-32. This is mainly due
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]904to lack of trampoline support (pointers to nested functions) on other platforms,
Duncan Sandsa73561c2008-06-08 20:18:35 +0000[diff] [blame]905however it <a href="http://llvm.org/PR2006">also fails to build on X86-64</a>
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]906which does support trampolines.</li>
907<li>The Ada front-end <a href="http://llvm.org/PR2007">fails to bootstrap</a>.
Duncan Sandsf73473e2009-02-25 11:59:06 +0000[diff] [blame]908This is due to lack of LLVM support for <tt>setjmp</tt>/<tt>longjmp</tt> style
909exception handling, which is used internally by the compiler.
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]910Workaround: configure with --disable-bootstrap.</li>
Duncan Sands2dbc2e82008-10-13 17:27:23 +0000[diff] [blame]911<li>The c380004, <a href="http://llvm.org/PR2010">c393010</a>
912and <a href="http://llvm.org/PR2421">cxg2021</a> ACATS tests fail
Duncan Sandsf73473e2009-02-25 11:59:06 +0000[diff] [blame]913(c380004 also fails with gcc-4.2 mainline).
914If the compiler is built with checks disabled then <a href="http://llvm.org/PR2010">c393010</a>
915causes the compiler to go into an infinite loop, using up all system memory.</li>
Duncan Sands2dbc2e82008-10-13 17:27:23 +0000[diff] [blame]916<li>Some gcc specific Ada tests continue to crash the compiler.</li>
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]917<li>The -E binder option (exception backtraces)
918<a href="http://llvm.org/PR1982">does not work</a> and will result in programs
919crashing if an exception is raised. Workaround: do not use -E.</li>
920<li>Only discrete types <a href="http://llvm.org/PR1981">are allowed to start
921or finish at a non-byte offset</a> in a record. Workaround: do not pack records
922or use representation clauses that result in a field of a non-discrete type
923starting or finishing in the middle of a byte.</li>
Chris Lattnerd95b5962008-06-08 21:19:07 +0000[diff] [blame]924<li>The <tt>lli</tt> interpreter <a href="http://llvm.org/PR2009">considers
925'main' as generated by the Ada binder to be invalid</a>.
926Workaround: hand edit the file to use pointers for <tt>argv</tt> and
927<tt>envp</tt> rather than integers.</li>
928<li>The <tt>-fstack-check</tt> option <a href="http://llvm.org/PR2008">is
929ignored</a>.</li>
Chris Lattnerdaeb3232008-02-12 06:29:45 +0000[diff] [blame]930</ul>
931</div>
932
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]933<!-- *********************************************************************** -->
934<div class="doc_section">
935 <a name="additionalinfo">Additional Information</a>
936</div>
937<!-- *********************************************************************** -->
938
939<div class="doc_text">
940
941<p>A wide variety of additional information is available on the <a
942href="http://llvm.org">LLVM web page</a>, in particular in the <a
943href="http://llvm.org/docs/">documentation</a> section. The web page also
944contains versions of the API documentation which is up-to-date with the
945Subversion version of the source code.
946You can access versions of these documents specific to this release by going
947into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
948
949<p>If you have any questions or comments about LLVM, please feel free to contact
950us via the <a href="http://llvm.org/docs/#maillist"> mailing
951lists</a>.</p>
952
953</div>
954
955<!-- *********************************************************************** -->
956
957<hr>
958<address>
959 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman947321d2008-12-11 17:34:48 +0000[diff] [blame]960 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]961 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman947321d2008-12-11 17:34:48 +0000[diff] [blame]962 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Dan Gohmanf17a25c2007-07-18 16:29:46 +0000[diff] [blame]963
964 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
965 Last modified: $Date$
966</address>
967
968</body>
969</html>