Gitiles
Code Review Sign In
gerrit-public.fairphone.software / fp2-dev / platform / external / llvm / 7d9b6b439ad2d580780d7388e636ba02988653e6 / . / docs / ReleaseNotes.html
blob: 385e6357636c5dab01425592dae826d01e92f26a [file] [log] [blame]
Misha Brukman500bc302003-11-22 00:38:41 +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>
Reid Spencer6454ed32004-11-18 18:38:58 +0000[diff] [blame]5 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]6 <link rel="stylesheet" href="llvm.css" type="text/css">
Dan Gohmanb44f6c62010-05-03 23:51:05 +0000[diff] [blame]7 <title>LLVM 2.8 Release Notes</title>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]8</head>
9<body>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]10
Dan Gohmanb44f6c62010-05-03 23:51:05 +0000[diff] [blame]11<div class="doc_title">LLVM 2.8 Release Notes</div>
Mikhail Glushenkovea65d7d2008-10-13 02:08:34 +0000[diff] [blame]12
Chris Lattner0e464a92010-03-17 04:02:39 +0000[diff] [blame]13<img align=right src="http://llvm.org/img/DragonSmall.png"
Gabor Greifee2187a2010-04-22 10:21:43 +0000[diff] [blame]14 width="136" height="136" alt="LLVM Dragon Logo">
Chris Lattner0e464a92010-03-17 04:02:39 +0000[diff] [blame]15
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]16<ol>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]17 <li><a href="#intro">Introduction</a></li>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]18 <li><a href="#subproj">Sub-project Status Update</a></li>
Dan Gohmanb44f6c62010-05-03 23:51:05 +0000[diff] [blame]19 <li><a href="#externalproj">External Projects Using LLVM 2.8</a></li>
20 <li><a href="#whatsnew">What's New in LLVM 2.8?</a></li>
Chris Lattner4b538b92004-04-30 22:17:12 +0000[diff] [blame]21 <li><a href="GettingStarted.html">Installation Instructions</a></li>
Dan Gohman44aa9212008-10-14 16:23:02 +0000[diff] [blame]22 <li><a href="#knownproblems">Known Problems</a></li>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]23 <li><a href="#additionalinfo">Additional Information</a></li>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]24</ol>
25
Chris Lattner7911ce22004-05-23 21:07:27 +0000[diff] [blame]26<div class="doc_author">
Dan Gohman44aa9212008-10-14 16:23:02 +0000[diff] [blame]27 <p>Written by the <a href="http://llvm.org">LLVM Team</a></p>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]28</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]29
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]30<!--
Chris Lattner7b91eda2010-04-22 05:41:35 +0000[diff] [blame]31<h1 style="color:red">These are in-progress notes for the upcoming LLVM 2.8
Jeffrey Yasskinbec48772010-01-28 01:14:43 +0000[diff] [blame]32release.<br>
33You may prefer the
Dan Gohmanb44f6c62010-05-03 23:51:05 +0000[diff] [blame]34<a href="http://llvm.org/releases/2.7/docs/ReleaseNotes.html">LLVM 2.7
35Release Notes</a>.</h1>
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]36-->
Jeffrey Yasskinbec48772010-01-28 01:14:43 +0000[diff] [blame]37
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]38<!-- *********************************************************************** -->
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]39<div class="doc_section">
40 <a name="intro">Introduction</a>
41</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]42<!-- *********************************************************************** -->
43
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]44<div class="doc_text">
45
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]46<p>This document contains the release notes for the LLVM Compiler
Dan Gohman7ae3ac82010-05-03 23:52:21 +0000[diff] [blame]47Infrastructure, release 2.8. Here we describe the status of LLVM, including
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]48major improvements from the previous release and significant known problems.
Mikhail Glushenkovf795ef02009-03-01 18:09:47 +0000[diff] [blame]49All LLVM releases may be downloaded from the <a
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]50href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
Chris Lattner19092612003-10-02 16:38:05 +0000[diff] [blame]51
Chris Lattner7506b1d2004-12-07 08:04:13 +0000[diff] [blame]52<p>For more information about LLVM, including information about the latest
Chris Lattnerc463b272005-10-29 07:07:09 +0000[diff] [blame]53release, please check out the <a href="http://llvm.org/">main LLVM
Chris Lattner47ad72c2003-10-07 21:38:31 +0000[diff] [blame]54web site</a>. If you have questions or comments, the <a
Chris Lattnerc66bfef2010-03-17 04:41:49 +0000[diff] [blame]55href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM Developer's
56Mailing List</a> is a good place to send them.</p>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]57
Mikhail Glushenkovea65d7d2008-10-13 02:08:34 +0000[diff] [blame]58<p>Note that if you are reading this file from a Subversion checkout or the
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]59main LLVM web page, this document applies to the <i>next</i> release, not the
Gabor Greiffa933f82008-10-14 11:00:32 +0000[diff] [blame]60current one. To see the release notes for a specific release, please see the
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]61<a href="http://llvm.org/releases/">releases page</a>.</p>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]62
63</div>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]64
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]65
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]66<!--
67Almost dead code.
68 include/llvm/Analysis/LiveValues.h => Dan
69 lib/Transforms/IPO/MergeFunctions.cpp => consider for 2.8.
Chris Lattner61c70e92010-08-28 04:09:24 +0000[diff] [blame]70 GEPSplitterPass
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]71-->
72
73
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]74<!-- Features that need text if they're finished for 2.9:
Chris Lattner914ce462010-04-22 06:28:20 +0000[diff] [blame]75 combiner-aa?
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]76 strong phi elim
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]77 loop dependence analysis
Chris Lattner885b6612010-08-28 16:33:36 +0000[diff] [blame]78 TBAA
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]79 CorrelatedValuePropagation
Chris Lattneracce85d2008-02-10 07:46:44 +0000[diff] [blame]80 -->
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]81
82 <!-- Announcement, lldb, libc++ -->
Chris Lattnerafa41632010-09-29 07:25:03 +0000[diff] [blame]83
Chris Lattneracce85d2008-02-10 07:46:44 +0000[diff] [blame]84
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]85<!-- *********************************************************************** -->
86<div class="doc_section">
87 <a name="subproj">Sub-project Status Update</a>
Chris Lattnerea34f642008-06-08 21:34:41 +0000[diff] [blame]88</div>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]89<!-- *********************************************************************** -->
Chris Lattnerea34f642008-06-08 21:34:41 +0000[diff] [blame]90
91<div class="doc_text">
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]92<p>
Dan Gohmanb44f6c62010-05-03 23:51:05 +0000[diff] [blame]93The LLVM 2.8 distribution currently consists of code from the core LLVM
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]94repository (which roughly includes the LLVM optimizers, code generators
95and supporting tools), the Clang repository and the llvm-gcc repository. In
96addition to this code, the LLVM Project includes other sub-projects that are in
97development. Here we include updates on these subprojects.
Bill Wendling63d8c552009-03-02 04:28:57 +0000[diff] [blame]98</p>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]99
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]100</div>
101
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]102
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]103<!--=========================================================================-->
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]104<div class="doc_subsection">
Chris Lattnerfb97b2d2008-10-13 18:11:54 +0000[diff] [blame]105<a name="clang">Clang: C/C++/Objective-C Frontend Toolkit</a>
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]106</div>
107
108<div class="doc_text">
109
Chris Lattner095539f2010-04-26 17:42:18 +0000[diff] [blame]110<p><a href="http://clang.llvm.org/">Clang</a> is an LLVM front end for the C,
111C++, and Objective-C languages. Clang aims to provide a better user experience
112through expressive diagnostics, a high level of conformance to language
113standards, fast compilation, and low memory use. Like LLVM, Clang provides a
114modular, library-based architecture that makes it suitable for creating or
115integrating with other development tools. Clang is considered a
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]116production-quality compiler for C, Objective-C, C++ and Objective-C++ on x86
117(32- and 64-bit), and for darwin-arm targets.</p>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]118
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]119<p>In the LLVM 2.8 time-frame, the Clang team has made many improvements:</p>
Bill Wendling741748a2008-10-27 09:27:33 +0000[diff] [blame]120
Daniel Dunbar13739432008-10-14 23:25:09 +0000[diff] [blame]121<ul>
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]122<li>Surely these guys have done something</li>
Chris Lattner4ba2b652010-09-30 16:31:33 +0000[diff] [blame]123<li>X86-64 abi improvements? Did they make it in?</li>
Bill Wendling6bc15282009-03-02 04:28:18 +0000[diff] [blame]124</ul>
Chris Lattnerfb97b2d2008-10-13 18:11:54 +0000[diff] [blame]125</div>
126
127<!--=========================================================================-->
128<div class="doc_subsection">
129<a name="clangsa">Clang Static Analyzer</a>
130</div>
131
132<div class="doc_text">
133
Chris Lattner8cdd7932010-04-22 06:38:11 +0000[diff] [blame]134<p>The <a href="http://clang-analyzer.llvm.org/">Clang Static Analyzer</a>
135 project is an effort to use static source code analysis techniques to
136 automatically find bugs in C and Objective-C programs (and hopefully <a
137 href="http://clang-analyzer.llvm.org/dev_cxx.html">C++ in the
138 future</a>!). The tool is very good at finding bugs that occur on specific
139 paths through code, such as on error conditions.</p>
Chris Lattnercc042612008-10-14 00:52:49 +0000[diff] [blame]140
Chris Lattner2e38c7f2010-09-30 00:34:43 +0000[diff] [blame]141<p>The LLVM 2.8 release fixes a number of bugs and slightly improves precision
142 over 2.7, but there are no major new features in the release.
Chris Lattner8cdd7932010-04-22 06:38:11 +0000[diff] [blame]143</p>
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]144
145</div>
146
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]147<!--=========================================================================-->
148<div class="doc_subsection">
Nicolas Geoffray99a4d302008-10-14 19:23:04 +0000[diff] [blame]149<a name="vmkit">VMKit: JVM/CLI Virtual Machine Implementation</a>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]150</div>
151
152<div class="doc_text">
153<p>
Nicolas Geoffray99a4d302008-10-14 19:23:04 +0000[diff] [blame]154The <a href="http://vmkit.llvm.org/">VMKit project</a> is an implementation of
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]155a JVM and a CLI Virtual Machine (Microsoft .NET is an
156implementation of the CLI) using LLVM for static and just-in-time
157compilation.</p>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]158
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]159<p>With the release of LLVM 2.8, ...</p>
Chris Lattner96a445e2008-10-13 18:01:01 +0000[diff] [blame]160
161</div>
162
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]163
164<!--=========================================================================-->
165<div class="doc_subsection">
166<a name="compiler-rt">compiler-rt: Compiler Runtime Library</a>
167</div>
168
169<div class="doc_text">
170<p>
171The new LLVM <a href="http://compiler-rt.llvm.org/">compiler-rt project</a>
172is a simple library that provides an implementation of the low-level
173target-specific hooks required by code generation and other runtime components.
174For example, when compiling for a 32-bit target, converting a double to a 64-bit
175unsigned integer is compiled into a runtime call to the "__fixunsdfdi"
176function. The compiler-rt library provides highly optimized implementations of
177this and other low-level routines (some are 3x faster than the equivalent
178libgcc routines).</p>
179
180<p>
181All of the code in the compiler-rt project is available under the standard LLVM
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]182License, a "BSD-style" license. New in LLVM 2.8, compiler_rt now supports
183soft floating point (for targets that don't have a real floating point unit),
184and includes an extensive testsuite for the "blocks" language feature and the
185blocks runtime included in compiler_rt.</p>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]186
187</div>
188
189<!--=========================================================================-->
190<div class="doc_subsection">
Duncan Sands749fd832010-04-02 09:23:15 +0000[diff] [blame]191<a name="dragonegg">DragonEgg: llvm-gcc ported to gcc-4.5</a>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]192</div>
193
194<div class="doc_text">
195<p>
Duncan Sands749fd832010-04-02 09:23:15 +0000[diff] [blame]196<a href="http://dragonegg.llvm.org/">DragonEgg</a> is a port of llvm-gcc to
Duncan Sands4b1da2b2010-09-30 17:37:34 +0000[diff] [blame]197gcc-4.5. Unlike llvm-gcc, dragonegg in theory does not require any gcc-4.5
198modifications whatsoever (currently one small patch is needed) thanks to the
199new <a href="http://gcc.gnu.org/wiki/plugins">gcc plugin architecture</a>.
200DragonEgg is a gcc plugin that makes gcc-4.5 use the LLVM optimizers and code
201generators instead of gcc's, just like with llvm-gcc.
Duncan Sands749fd832010-04-02 09:23:15 +0000[diff] [blame]202</p>
203
204<p>
Duncan Sands4b1da2b2010-09-30 17:37:34 +0000[diff] [blame]205DragonEgg is still a work in progress, but it is able to compile a lot of code,
206for example all of gcc, LLVM and clang. Currently Ada, C, C++ and Fortran work
207well, while all other languages either don't work at all or only work poorly.
208For the moment only the x86-32 and x86-64 targets are supported, and only on
209linux and darwin (darwin may need additional gcc patches).
Duncan Sands749fd832010-04-02 09:23:15 +0000[diff] [blame]210</p>
211
212<p>
Duncan Sands4b1da2b2010-09-30 17:37:34 +0000[diff] [blame]213The 2.8 release has the following notable changes:
214<ul>
215<li>The plugin loads faster due to exporting fewer symbols.</li>
216<li>Additional vector operations such as addps256 are now supported.</li>
217<li>Ada global variables with no initial value are no longer zero initialized,
218resulting in better optimization.</li>
219<li>The '-fplugin-arg-dragonegg-enable-gcc-optzns' flag now runs all gcc
220optimizers, rather than just a handful.</li>
221<li>Fortran programs using common variables now link correctly.</li>
222<li>GNU OMP constructs no longer crash the compiler.</li>
223</ul>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]224</p>
225
226</div>
227
228
229<!--=========================================================================-->
230<div class="doc_subsection">
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]231<a name="lldb">LLDB: Low Level Debugger</a>
232</div>
233
234<div class="doc_text">
235<p>
236<a href="http://lldb.llvm.org/">LLDB</a> is</p>
237
238<p>
239</p>
240
241<p>
2422.8 status here.
243</p>
244
245</div>
246
247<!--=========================================================================-->
248<div class="doc_subsection">
249<a name="libc++">libc++: C++ Standard Library</a>
250</div>
251
252<div class="doc_text">
253<p>
254<a href="http://libc++.llvm.org/">libc++</a> is</p>
255
256<p>
257</p>
258
259<p>
2602.8 status here.
261</p>
262
263</div>
264
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]265
Chris Lattnerab68e9e2009-02-26 22:33:38 +0000[diff] [blame]266<!-- *********************************************************************** -->
267<div class="doc_section">
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]268 <a name="externalproj">External Open Source Projects Using LLVM 2.8</a>
Chris Lattnerab68e9e2009-02-26 22:33:38 +0000[diff] [blame]269</div>
270<!-- *********************************************************************** -->
271
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]272<div class="doc_text">
273
274<p>An exciting aspect of LLVM is that it is used as an enabling technology for
275 a lot of other language and tools projects. This section lists some of the
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]276 projects that have already been updated to work with LLVM 2.8.</p>
Chris Lattner7c8e7962010-04-26 17:38:10 +0000[diff] [blame]277</div>
278
Chris Lattner2e38c7f2010-09-30 00:34:43 +0000[diff] [blame]279<!--=========================================================================-->
280<div class="doc_subsection">
281<a name="tce">TTA-based Codesign Environment (TCE)</a>
282</div>
283
284<div class="doc_text">
285<p>
286<a href="http://tce.cs.tut.fi/">TCE</a> is a toolset for designing
287application-specific processors (ASP) based on the Transport triggered
288architecture (TTA). The toolset provides a complete co-design flow from C/C++
289programs down to synthesizable VHDL and parallel program binaries. Processor
290customization points include the register files, function units, supported
291operations, and the interconnection network.</p>
292
293<p>TCE uses llvm-gcc/Clang and LLVM for C/C++ language support, target
294independent optimizations and also for parts of code generation. It generates
295new LLVM-based code generators "on the fly" for the designed TTA processors and
296loads them in to the compiler backend as runtime libraries to avoid per-target
297recompilation of larger parts of the compiler chain.</p>
298
299</div>
300
301<!--=========================================================================-->
302<div class="doc_subsection">
303<a name="Horizon">Horizon Bytecode Compiler</a>
304</div>
305
306<div class="doc_text">
307<p>
308<a href="http://www.quokforge.org/projects/horizon">Horizon</a> is a bytecode
309language and compiler written on top of LLVM, intended for producing
310single-address-space managed code operating systems that
311run faster than the equivalent multiple-address-space C systems.
312More in-depth blurb is available on <a
313href="http://www.quokforge.org/projects/horizon/wiki/Wiki">the wiki</a>.</p>
314
315</div>
316
317<!--=========================================================================-->
318<div class="doc_subsection">
319<a name="clamav">Clam AntiVirus</a>
320</div>
321
322<div class="doc_text">
323<p>
324<a href=http://www.clamav.net>Clam AntiVirus</a> is an open source (GPL)
325anti-virus toolkit for UNIX, designed especially for e-mail scanning on mail
326gateways. Since version 0.96 it has <a
327href="http://vrt-sourcefire.blogspot.com/2010/09/introduction-to-clamavs-low-level.html">bytecode
328signatures</a> that allow writing detections for complex malware. It
329uses LLVM's JIT to speed up the execution of bytecode on
330X86,X86-64,PPC32/64, falling back to its own interpreter otherwise.
331The git version was updated to work with LLVM 2.8
332</p>
333
334<p>The <a
335href="http://git.clamav.net/gitweb?p=clamav-bytecode-compiler.git;a=blob_plain;f=docs/user/clambc-user.pdf">
336ClamAV bytecode compiler</a> uses Clang and LLVM to compile a C-like
337language, insert runtime checks, and generate ClamAV bytecode.</p>
338
339</div>
340
341<!--=========================================================================-->
342<div class="doc_subsection">
343<a name="pure">Pure</a>
344</div>
345
346<div class="doc_text">
347<p>
348<a href="http://pure-lang.googlecode.com/">Pure</a>
349is an algebraic/functional
350programming language based on term rewriting. Programs are collections
351of equations which are used to evaluate expressions in a symbolic
352fashion. Pure offers dynamic typing, eager and lazy evaluation, lexical
353closures, a hygienic macro system (also based on term rewriting),
354built-in list and matrix support (including list and matrix
355comprehensions) and an easy-to-use C interface. The interpreter uses
356LLVM as a backend to JIT-compile Pure programs to fast native code.</p>
357
358<p>Pure versions 0.44 and later have been tested and are known to work with
359LLVM 2.8 (and continue to work with older LLVM releases >= 2.5).</p>
360
361</div>
362
363<!--=========================================================================-->
364<div class="doc_subsection">
365<a name="GHC">Glasgow Haskell Compiler (GHC)</a>
366</div>
367
368<div class="doc_text">
369<p>
370<a href="http://www.haskell.org/ghc/">GHC</a> is an open source,
371state-of-the-art programming suite for
372Haskell, a standard lazy functional programming language. It includes
373an optimizing static compiler generating good code for a variety of
374platforms, together with an interactive system for convenient, quick
375development.</p>
376
377<p>In addition to the existing C and native code generators, GHC 7.0 now
378supports an <a
379href="http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/Backends/LLVM">LLVM
380code generator</a>. GHC supports LLVM 2.7 and later.</p>
381
382</div>
383
384<!--=========================================================================-->
385<div class="doc_subsection">
386<a name="Clay">Clay Programming Language</a>
387</div>
388
389<div class="doc_text">
390<p>
Chris Lattner97fe6452010-09-30 01:12:09 +0000[diff] [blame]391<a href="http://tachyon.in/clay/">Clay</a> is a new systems programming
Chris Lattner2e38c7f2010-09-30 00:34:43 +0000[diff] [blame]392language that is specifically designed for generic programming. It makes
393generic programming very concise thanks to whole program type propagation. It
394uses LLVM as its backend.</p>
395
396</div>
Chris Lattner3a1d4cf2010-04-22 21:34:16 +0000[diff] [blame]397
Chris Lattnere0518442010-10-01 06:34:49 +0000[diff] [blame]398<!--=========================================================================-->
399<div class="doc_subsection">
400<a name="llvm-py">llvm-py Python Bindings for LLVM</a>
401</div>
402
403<div class="doc_text">
404<p>
405<a href="http://www.mdevan.org/llvm-py/">llvm-py</a> has been updated to work
406with LLVM 2.8. llvm-py provides Python bindings for LLVM, allowing you to write a
407compiler backend or a VM in Python.</p>
408
409</div>
410
411
412
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]413<!-- *********************************************************************** -->
414<div class="doc_section">
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]415 <a name="whatsnew">What's New in LLVM 2.8?</a>
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]416</div>
417<!-- *********************************************************************** -->
418
Chris Lattnerf8e0b4e2008-06-08 22:59:35 +0000[diff] [blame]419<div class="doc_text">
420
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]421<p>This release includes a huge number of bug fixes, performance tweaks and
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]422minor improvements. Some of the major improvements and new features are listed
423in this section.
Chris Lattnerf8e0b4e2008-06-08 22:59:35 +0000[diff] [blame]424</p>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]425
Chris Lattner914ce462010-04-22 06:28:20 +0000[diff] [blame]426</div>
427
428<!--=========================================================================-->
429<div class="doc_subsection">
430<a name="orgchanges">LLVM Community Changes</a>
431</div>
432
433<div class="doc_text">
434
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]435<p>In addition to changes to the code, between LLVM 2.7 and 2.8, a number of
Chris Lattnerb7c85b42010-04-21 05:17:40 +0000[diff] [blame]436organization changes have happened:
437</p>
438
439<ul>
Chris Lattnere07043c2010-09-29 05:30:03 +0000[diff] [blame]440<li>libc++ and lldb are new</li>
Chris Lattnerafa41632010-09-29 07:25:03 +0000[diff] [blame]441<li>Debugging optimized code support.</li>
Chris Lattnerb7c85b42010-04-21 05:17:40 +0000[diff] [blame]442</ul>
Chris Lattnerf8e0b4e2008-06-08 22:59:35 +0000[diff] [blame]443</div>
444
Chris Lattner8348b472008-06-08 21:58:17 +0000[diff] [blame]445<!--=========================================================================-->
Chris Lattnerea34f642008-06-08 21:34:41 +0000[diff] [blame]446<div class="doc_subsection">
Chris Lattner252b83d2008-02-06 18:00:06 +0000[diff] [blame]447<a name="majorfeatures">Major New Features</a>
448</div>
449
450<div class="doc_text">
451
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]452<p>LLVM 2.8 includes several major new capabilities:</p>
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]453
454<ul>
Chris Lattner702f2d42010-09-06 19:14:40 +0000[diff] [blame]455<li>llvm-diff</li>
Chris Lattner8170c102008-02-10 08:18:42 +0000[diff] [blame]456</ul>
Chris Lattnerdc910082010-03-17 06:41:58 +0000[diff] [blame]457
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]458</div>
459
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]460<!--=========================================================================-->
461<div class="doc_subsection">
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]462<a name="coreimprovements">LLVM IR and Core Improvements</a>
Chris Lattnerf304ffc2008-02-10 08:17:19 +0000[diff] [blame]463</div>
464
465<div class="doc_text">
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]466<p>LLVM IR has several new features for better support of new targets and that
467expose new optimization opportunities:</p>
Chris Lattnerf304ffc2008-02-10 08:17:19 +0000[diff] [blame]468
Chris Lattner791f77b2008-06-05 06:25:56 +0000[diff] [blame]469<ul>
Gabor Greif17b59fe2010-07-21 10:20:08 +0000[diff] [blame]470
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]471 memcpy, memmove, and memset now take address space qualified pointers + volatile.
472 per-instruction debug info metadata is much faster and uses less space (new DebugLoc class).
473 New "trap values" concept: http://llvm.org/docs/LangRef.html#trapvalues
474 New linker_private_weak and linker_private_weak_def_auto linkage types
475 Triples are now stored in normalized form. Triple::normalize.
476
477
Gabor Greif17b59fe2010-07-21 10:20:08 +0000[diff] [blame]478<li>LLVM 2.8 changes the internal order of operands in <a
479 href="http://llvm.org/doxygen/classllvm_1_1InvokeInst.html"><tt>InvokeInst</tt></a>
480 and <a href="http://llvm.org/doxygen/classllvm_1_1CallInst.html"><tt>CallInst</tt></a>.
481 To be portable across releases, resort to <tt>CallSite</tt> and the
Gabor Greif50fb3302010-07-21 10:22:41 +0000[diff] [blame]482 high-level accessors, such as <tt>getCalledValue</tt> and <tt>setUnwindDest</tt>.
Gabor Greif17b59fe2010-07-21 10:20:08 +0000[diff] [blame]483</li>
Gabor Greif05006e22010-07-26 18:48:07 +0000[diff] [blame]484<li>
485 You can no longer pass use_iterators directly to cast<> (and similar), because
486 these routines tend to perform costly dereference operations more than once. You
487 have to dereference the iterators yourself and pass them in.
488</li>
Torok Edwin7c46cf02010-08-04 12:43:22 +0000[diff] [blame]489<li>
490 llvm.memcpy.*, llvm.memset.*, llvm.memmove.* (and possibly other?) intrinsics
491 take an extra parameter now (i1 isVolatile), totaling 5 parameters.
492 If you were creating these intrinsic calls and prototypes yourself (as opposed
493 to using Intrinsic::getDeclaration), you can use UpgradeIntrinsicFunction/UpgradeIntrinsicCall
494 to be portable accross releases.
495 Note that you cannot use Intrinsic::getDeclaration() in a backwards compatible
496 way (needs 2/3 types now, in 2.7 it needed just 1).
497</li>
498<li>
499 SetCurrentDebugLocation takes a DebugLoc now instead of a MDNode.
500 Change your code to use
501 SetCurrentDebugLocation(DebugLoc::getFromDILocation(...)).
502</li>
503<li>
504 VISIBILITY_HIDDEN is gone.
505</li>
506<li>
Owen Anderson87ebbc02010-08-04 18:27:08 +0000[diff] [blame]507 The <tt>RegisterPass</tt> and <tt>RegisterAnalysisGroup</tt> templates are
508 considered deprecated, but continue to function in LLVM 2.8. Clients are
509 strongly advised to use the upcoming <tt>INITIALIZE_PASS()</tt> and
510 <tt>INITIALIZE_AG_PASS()</tt> macros instead.
511<li>
Torok Edwin7c46cf02010-08-04 12:43:22 +0000[diff] [blame]512 SMDiagnostic takes different parameters now. //FIXME: how to upgrade?
513</li>
514<li>
Duncan Sands12881e72010-08-30 10:57:54 +0000[diff] [blame]515 The constructor for the Triple class no longer tries to understand odd triple
516 specifications. Frontends should ensure that they only pass valid triples to
517 LLVM. The Triple::normalize utility method has been added to help front-ends
518 deal with funky triples.
519<li>
Torok Edwin7c46cf02010-08-04 12:43:22 +0000[diff] [blame]520 Some APIs got renamed:
521 <ul>
522 <li>llvm_report_error -&gt; report_fatal_error</li>
523 <li>llvm_install_error_handler -&gt; install_fatal_error_handler</li>
524 <li>llvm::DwarfExceptionHandling -&gt; llvm::JITExceptionHandling</li>
525 </ul>
526</li>
Chris Lattnerf304ffc2008-02-10 08:17:19 +0000[diff] [blame]527</ul>
Mikhail Glushenkovea65d7d2008-10-13 02:08:34 +0000[diff] [blame]528
Chris Lattnerf304ffc2008-02-10 08:17:19 +0000[diff] [blame]529</div>
530
531<!--=========================================================================-->
532<div class="doc_subsection">
Chris Lattneracce85d2008-02-10 07:46:44 +0000[diff] [blame]533<a name="optimizer">Optimizer Improvements</a>
534</div>
535
536<div class="doc_text">
537
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]538<p>In addition to a large array of minor performance tweaks and bug fixes, this
Chris Lattnerf3013872008-10-13 21:50:36 +0000[diff] [blame]539release includes a few major enhancements and additions to the optimizers:</p>
Chris Lattneracce85d2008-02-10 07:46:44 +0000[diff] [blame]540
541<ul>
542
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]543<li></li>
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]544 Preliminary work on TBAA but not usable in 2.8.
545 New CorrelatedValuePropagation pass, not on by default in 2.8 yet.
546 JumpThreading much more aggressive about implied value relations.
547 New RegionInfo pass "opt -regions analyze" or "opt -view-regions".
548 Improved trip count analysis for <= and >= loops, and uses sign overflow info.
549 llvm.dbg.value: variable debug info for optimized code
550 Now iterate function passes when a cgsccpassmanager detects a devirtualization
551 Atomic lowering patch: -loweratomic (see Passes.html#loweratomic)
Chris Lattnera54c1f72010-04-21 06:42:24 +0000[diff] [blame]552
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]553</ul>
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]554
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]555</div>
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]556
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]557
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]558<!--=========================================================================-->
559<div class="doc_subsection">
560<a name="executionengine">Interpreter and JIT Improvements</a>
561</div>
Chris Lattnerf3013872008-10-13 21:50:36 +0000[diff] [blame]562
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]563<div class="doc_text">
564
565<ul>
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]566<li></li>
Chris Lattnerb7c85b42010-04-21 05:17:40 +0000[diff] [blame]567
Chris Lattner0b832202008-06-08 02:45:07 +0000[diff] [blame]568</ul>
569
570</div>
571
572<!--=========================================================================-->
573<div class="doc_subsection">
Chris Lattner4ba2b652010-09-30 16:31:33 +0000[diff] [blame]574<a name="mc">MC Level Improvements</a>
575</div>
576
577<div class="doc_text">
578<p>
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]579The LLVM Machine Code (aka MC) subsystem was created to solve a number
Chris Lattner4ba2b652010-09-30 16:31:33 +0000[diff] [blame]580of problems in the realm of assembly, disassembly, object file format handling,
581and a number of other related areas that CPU instruction-set level tools work
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]582in.</p>
583
584<p>The MC subproject has made great leaps in LLVM 2.8. For example, support for
585 directly writing .o files from LLC (and clang) now works reliably for
586 darwin/x86[-64] (including inline assembly support) and the integrated
587 assembler is turned on by default in Clang for these targets. This provides
588 improved compile times among other things.</p>
589
590<ul>
591<li>The entire compiler has converted over to using the MCStreamer assembler API
592 instead of writing out a .s file textually.</li>
593<li>The "assembler parser" is far more mature than in 2.7, supporting a full
594 complement of directives, now supports assembler macros, etc.</li>
595<li>The "assembler backend" has been completed, including support for relaxation
596 relocation processing and all the other things that an assembler does.</li>
597<li>The MachO file format support is now fully functional and works.</li>
598<li>The MC disassembler now fully supports ARM and Thumb. ARM assembler support
599 is still in early development though.</li>
600<li>The X86 MC assembler now supports the X86 AES and AVX instruction set.</li>
601<li>Work on ELF and COFF support is well underway, but isn't useful yet in LLVM
602 2.8. Please contact the llvmdev mailing list if you're interested in
603 this.</li>
604</ul>
605
606<p>For more information, please see the <a
Chris Lattner4ba2b652010-09-30 16:31:33 +0000[diff] [blame]607href="http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html">Intro to the
608LLVM MC Project Blog Post</a>.
609</p>
610
Chris Lattner4ba2b652010-09-30 16:31:33 +0000[diff] [blame]611</div>
612
613
614
615<!--=========================================================================-->
616<div class="doc_subsection">
Chris Lattner511433e2009-03-02 03:24:11 +0000[diff] [blame]617<a name="codegen">Target Independent Code Generator Improvements</a>
Chris Lattner0b832202008-06-08 02:45:07 +0000[diff] [blame]618</div>
619
620<div class="doc_text">
621
Mikhail Glushenkovf795ef02009-03-01 18:09:47 +0000[diff] [blame]622<p>We have put a significant amount of work into the code generator
623infrastructure, which allows us to implement more aggressive algorithms and make
624it run faster:</p>
Chris Lattner0b832202008-06-08 02:45:07 +0000[diff] [blame]625
626<ul>
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]627<li></li>
628
629 MachineCSE tuned and on by default.
630
631 Rewrote tblgen's type inference for backends to be more consistent and
632 diagnose more target bugs. This also allows limited support for writing
633 patterns for instructions that return multiple results, e.g. a virtual
634 register and a flag result. Stuff that used 'parallel' before should use
635 this.
636
637 New -regalloc=fast, =local got removed
638 New -regalloc=default option that chooses a register allocator based on the -O optimization level.
639 New SubRegIndex tblgen class for targets -> jakob
640
641 Bottom up fast isel. Simple Load reuse. No more machinedce.
642 IR ABI: <3 x float> is passed as <4 x float> instead of 3 floats.
643
644 New COPY instruction. copyRegToReg -> copyPhysReg, isMoveInstr is gone.
645 RenderMachineFunction: -rendermf
646 SplitKit?
647 Evan: Teach bottom up pre-ra scheduler to track register pressure. Work in progress.
648 Evan: Add an ILP scheduler. On x86_64, this is a win for all tests in CFP2000. It also sped up 256.bzip2 by 16%.
649
650 New OptimizeExts+OptimizeCmps -> PeepholeOptimizer pass
651 New LocalStackSlotAllocation.cpp pass (jimg)
652 Atomics now get legalized when not natively supported (jim g)
653
654 -ffunction-sections and -fdata-sections are supported on ELF targets.
655 -momit-leaf-frame-pointer now supported.
656
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]657</ul>
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]658</div>
659
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]660<!--=========================================================================-->
661<div class="doc_subsection">
662<a name="x86">X86-32 and X86-64 Target Improvements</a>
663</div>
664
665<div class="doc_text">
666<p>New features of the X86 target include:
667</p>
668
669<ul>
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]670<li>The X86 backend now supports holding X87 floating point stack values
671 in registers across basic blocks, dramatically improving performance of code
672 that uses long double, and when targetting CPUs that don't support SSE.</li>
Chris Lattner511433e2009-03-02 03:24:11 +0000[diff] [blame]673
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]674 New SSEDomainFix pass:
675 On Nehalem and newer CPUs there is a 2 cycle latency penalty on using a
676 register in a different domain than where it was defined. Some instructions
677 have equvivalents for different domains, like por/orps/orpd. The
678 SSEDomainFix pass tries to minimize the number of domain crossings by
679 changing between equvivalent opcodes where possible.
680
681 X86 backend attempts to promote 16-bit integer operations to 32-bits to avoid
682 0x66 prefixes, which are slow on some microarchitectures and bloat the code
683 on others.
684
685 New support for X86 "thiscall" calling convention (x86_thiscallcc in IR) for windows.
686
687 New llvm.x86.int intrinsic (for int $42 and int3)
688
689 Verbose assembly decodes X86 shuffle instructions, e.g.:
690 insertps $113, %xmm3, %xmm0 ## xmm0 = zero,xmm0[1,2],xmm3[1]
691 unpcklps %xmm1, %xmm0 ## xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
692 pshufd $1, %xmm1, %xmm1 ## xmm1 = xmm1[1,0,0,0]
693
694 X86 ABI: <2 x float> in IR no longer maps onto MMX, it turns into <4 x float>
695
696 new GHC calling convention
697
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]698</ul>
699
Chris Lattner917cc712009-03-02 02:37:32 +0000[diff] [blame]700</div>
Chris Lattner84977642007-09-21 03:54:09 +0000[diff] [blame]701
702<!--=========================================================================-->
703<div class="doc_subsection">
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]704<a name="ARM">ARM Target Improvements</a>
Chris Lattnerc441fb82009-03-01 02:30:21 +0000[diff] [blame]705</div>
706
707<div class="doc_text">
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]708<p>New features of the ARM target include:
709</p>
Chris Lattnerc441fb82009-03-01 02:30:21 +0000[diff] [blame]710
711<ul>
Chris Lattnerc441fb82009-03-01 02:30:21 +0000[diff] [blame]712
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]713 NEON: Better performance for QQQQ (4-consecutive Q register) instructions. New reg sequence abstraction?
714 ARM: Better scheduling (list-hybrid, hybrid?)
715 ARM: Tail call support.
716 ARM: General performance work and tuning.
717
718 ARM: Half float support through intrinsics LangRef.html#int_fp16
719<li>ARMGlobalMerge: <!-- Anton --> </li>
720
Bob Wilson5b2fb952010-09-13 17:37:55 +0000[diff] [blame]721<li>
722 All of the NEON load and store intrinsics (llvm.arm.neon.vld* and
723 llvm.arm.neon.vst*) take an extra parameter to specify the alignment in bytes
724 of the memory being accessed.
725</li>
726<li>
727 The llvm.arm.neon.vaba intrinsic (vector absolute difference and
728 accumulate) has been removed. This operation is now represented using
729 the llvm.arm.neon.vabd intrinsic (vector absolute difference) followed by a
730 vector add.
Bob Wilsone8472772010-09-13 17:39:35 +0000[diff] [blame]731</li>
Bob Wilson5b2fb952010-09-13 17:37:55 +0000[diff] [blame]732<li>
733 The llvm.arm.neon.vabdl and llvm.arm.neon.vabal intrinsics (lengthening
734 vector absolute difference with and without accumlation) have been removed.
735 They are represented using the llvm.arm.neon.vabd intrinsic (vector absolute
736 difference) followed by a vector zero-extend operation, and for vabal,
737 a vector add.
738</li>
739<li>
740 The llvm.arm.neon.vmovn intrinsic has been removed. Calls of this intrinsic
741 are now replaced by vector truncate operations.
742</li>
743<li>
744 The llvm.arm.neon.vmovls and llvm.arm.neon.vmovlu intrinsics have been
745 removed. They are now represented as vector sign-extend (vmovls) and
746 zero-extend (vmovlu) operations.
747</li>
748<li>
749 The llvm.arm.neon.vaddl*, llvm.arm.neon.vaddw*, llvm.arm.neon.vsubl*, and
750 llvm.arm.neon.vsubw* intrinsics (lengthening vector add and subtract) have
751 been removed. They are replaced by vector add and vector subtract operations
752 where one (vaddw, vsubw) or both (vaddl, vsubl) of the operands are either
753 sign-extended or zero-extended.
754</li>
755<li>
756 The llvm.arm.neon.vmulls, llvm.arm.neon.vmullu, llvm.arm.neon.vmlal*, and
757 llvm.arm.neon.vmlsl* intrinsics (lengthening vector multiply with and without
758 accumulation and subtraction) have been removed. These operations are now
759 represented as vector multiplications where the operands are either
760 sign-extended or zero-extended, followed by a vector add for vmlal or a
761 vector subtract for vmlsl. Note that the polynomial vector multiply
762 intrinsic, llvm.arm.neon.vmullp, remains unchanged.
763</li>
Bob Wilsone8472772010-09-13 17:39:35 +0000[diff] [blame]764
765</ul>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]766</div>
Chris Lattnerc441fb82009-03-01 02:30:21 +0000[diff] [blame]767
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]768<!--=========================================================================-->
769<div class="doc_subsection">
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]770<a name="newapis">New Useful APIs</a>
771</div>
772
773<div class="doc_text">
774
775<p>This release includes a number of new APIs that are used internally, which
776 may also be useful for external clients.
777</p>
778
779<ul>
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]780<li></li>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]781</ul>
782
783
784</div>
785
786<!--=========================================================================-->
787<div class="doc_subsection">
788<a name="otherimprovements">Other Improvements and New Features</a>
789</div>
790
791<div class="doc_text">
792<p>Other miscellaneous features include:</p>
793
794<ul>
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]795<li></li>
Chris Lattnerc441fb82009-03-01 02:30:21 +0000[diff] [blame]796</ul>
797
798</div>
799
Chris Lattner77d29b12008-06-05 08:02:49 +0000[diff] [blame]800
801<!--=========================================================================-->
802<div class="doc_subsection">
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]803<a name="changes">Major Changes and Removed Features</a>
804</div>
805
806<div class="doc_text">
807
Chris Lattnereeb4da02008-10-13 22:06:31 +0000[diff] [blame]808<p>If you're already an LLVM user or developer with out-of-tree changes based
Chris Lattner922d00f2010-07-21 15:57:40 +0000[diff] [blame]809on LLVM 2.7, this section lists some "gotchas" that you may run into upgrading
Chris Lattnereeb4da02008-10-13 22:06:31 +0000[diff] [blame]810from the previous release.</p>
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]811
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]812
813 renamed "Release" -> "Release+Asserts"; "Release-Asserts" -> "Release etc.
814 RegisterPass<> -> INTIALIZE_PASS()
815
816
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]817<ul>
Chris Lattner2295b5a2010-09-02 23:22:50 +0000[diff] [blame]818<li>.ll file doesn't produce #uses comments anymore, to get them, run a .bc file
819 through "llvm-dis --show-annotations".</li>
Chris Lattner885b6612010-08-28 16:33:36 +0000[diff] [blame]820<li>MSIL Backend removed.</li>
821<li>ABCD and SSI passes removed.</li>
822<li>'Union' LLVM IR feature removed.</li>
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]823<li>SCCVN pass removed.</li>
Chris Lattnerb7c85b42010-04-21 05:17:40 +0000[diff] [blame]824</ul>
Chris Lattnereeb4da02008-10-13 22:06:31 +0000[diff] [blame]825
826<p>In addition, many APIs have changed in this release. Some of the major LLVM
827API changes are:</p>
Chris Lattnereeb4da02008-10-13 22:06:31 +0000[diff] [blame]828<ul>
Devang Patelb34dd132008-10-14 20:03:43 +0000[diff] [blame]829</ul>
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]830
Chris Lattnerf6662f92008-10-13 17:57:36 +0000[diff] [blame]831</div>
832
833
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]834<!-- *********************************************************************** -->
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]835<div class="doc_section">
836 <a name="knownproblems">Known Problems</a>
837</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]838<!-- *********************************************************************** -->
839
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]840<div class="doc_text">
841
Mikhail Glushenkovf795ef02009-03-01 18:09:47 +0000[diff] [blame]842<p>This section contains significant known problems with the LLVM system,
Chris Lattnere18b32e2008-11-10 05:40:34 +0000[diff] [blame]843listed by component. If you run into a problem, please check the <a
Chris Lattnerc463b272005-10-29 07:07:09 +0000[diff] [blame]844href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
Chris Lattner5eccca42003-12-12 21:22:16 +0000[diff] [blame]845there isn't already one.</p>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]846
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]847</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]848
Chris Lattnerf5ee1702004-03-14 02:03:02 +0000[diff] [blame]849<!-- ======================================================================= -->
850<div class="doc_subsection">
851 <a name="experimental">Experimental features included with this release</a>
852</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]853
Chris Lattnerf5ee1702004-03-14 02:03:02 +0000[diff] [blame]854<div class="doc_text">
855
Misha Brukman6df9e2c2004-05-12 21:46:05 +0000[diff] [blame]856<p>The following components of this LLVM release are either untested, known to
857be broken or unreliable, or are in early development. These components should
858not be relied on, and bugs should not be filed against them, but they may be
859useful to some people. In particular, if you would like to work on one of these
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]860components, please contact us on the <a
861href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
Chris Lattnerf5ee1702004-03-14 02:03:02 +0000[diff] [blame]862
863<ul>
Chris Lattner885b6612010-08-28 16:33:36 +0000[diff] [blame]864<li>The Alpha, SPU, MIPS, PIC16, Blackfin, MSP430, SystemZ and MicroBlaze
Wesley Peck7c4a1212010-03-18 14:31:30 +0000[diff] [blame]865 backends are experimental.</li>
Chris Lattner7d9b6b42010-10-02 21:59:30 +0000[diff] [blame^]866<li><tt>llc</tt> "<tt>-filetype=obj</tt>" is experimental on all targets
867 other than darwin-i386 and darwin-x86_64.</li>
Chris Lattnerf5ee1702004-03-14 02:03:02 +0000[diff] [blame]868</ul>
869
870</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]871
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]872<!-- ======================================================================= -->
873<div class="doc_subsection">
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]874 <a name="x86-be">Known problems with the X86 back-end</a>
John Criswellc0c186d2005-11-08 21:11:33 +0000[diff] [blame]875</div>
876
877<div class="doc_text">
878
879<ul>
Anton Korobeynikova6094be2008-06-08 10:24:13 +0000[diff] [blame]880 <li>The X86 backend does not yet support
881 all <a href="http://llvm.org/PR879">inline assembly that uses the X86
882 floating point stack</a>. It supports the 'f' and 't' constraints, but not
883 'u'.</li>
Duncan Sands47eff2b2008-06-08 19:38:43 +0000[diff] [blame]884 <li>Win64 code generation wasn't widely tested. Everything should work, but we
John Criswell524a5dd2009-03-02 15:28:15 +0000[diff] [blame]885 expect small issues to happen. Also, llvm-gcc cannot build the mingw64
Chris Lattner914ce462010-04-22 06:28:20 +0000[diff] [blame]886 runtime currently due to lack of support for the 'u' inline assembly
887 constraint and for X87 floating point inline assembly.</li>
Dan Gohman8207ba92008-06-08 23:05:11 +0000[diff] [blame]888 <li>The X86-64 backend does not yet support the LLVM IR instruction
Chris Lattner914ce462010-04-22 06:28:20 +0000[diff] [blame]889 <tt>va_arg</tt>. Currently, front-ends support variadic
Dan Gohman8207ba92008-06-08 23:05:11 +0000[diff] [blame]890 argument constructs on X86-64 by lowering them manually.</li>
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]891</ul>
892
893</div>
894
895<!-- ======================================================================= -->
896<div class="doc_subsection">
897 <a name="ppc-be">Known problems with the PowerPC back-end</a>
898</div>
899
900<div class="doc_text">
901
902<ul>
Nicolas Geoffraye4285dc2007-05-15 09:21:28 +0000[diff] [blame]903<li>The Linux PPC32/ABI support needs testing for the interpreter and static
Chris Lattner57a460e2007-05-23 04:39:32 +0000[diff] [blame]904compilation, and lacks support for debug information.</li>
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]905</ul>
906
907</div>
908
909<!-- ======================================================================= -->
910<div class="doc_subsection">
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]911 <a name="arm-be">Known problems with the ARM back-end</a>
912</div>
913
914<div class="doc_text">
915
916<ul>
Chris Lattner57a460e2007-05-23 04:39:32 +0000[diff] [blame]917<li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
Duncan Sandsc90d68b2007-09-26 15:59:54 +0000[diff] [blame]918processors, thumb programs can crash or produce wrong
Chris Lattner57a460e2007-05-23 04:39:32 +0000[diff] [blame]919results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
John Criswell524a5dd2009-03-02 15:28:15 +0000[diff] [blame]920<li>Compilation for ARM Linux OABI (old ABI) is supported but not fully tested.
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]921</li>
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]922</ul>
923
924</div>
925
926<!-- ======================================================================= -->
927<div class="doc_subsection">
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]928 <a name="sparc-be">Known problems with the SPARC back-end</a>
929</div>
930
931<div class="doc_text">
932
933<ul>
John Criswell524a5dd2009-03-02 15:28:15 +0000[diff] [blame]934<li>The SPARC backend only supports the 32-bit SPARC ABI (-m32); it does not
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]935 support the 64-bit SPARC ABI (-m64).</li>
936</ul>
937
938</div>
939
940<!-- ======================================================================= -->
941<div class="doc_subsection">
Bruno Cardoso Lopesb7e1a4f2008-10-25 14:56:26 +0000[diff] [blame]942 <a name="mips-be">Known problems with the MIPS back-end</a>
943</div>
944
945<div class="doc_text">
946
947<ul>
Bruno Cardoso Lopesb7e1a4f2008-10-25 14:56:26 +0000[diff] [blame]948<li>64-bit MIPS targets are not supported yet.</li>
949</ul>
950
951</div>
952
953<!-- ======================================================================= -->
954<div class="doc_subsection">
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]955 <a name="alpha-be">Known problems with the Alpha back-end</a>
956</div>
957
958<div class="doc_text">
959
960<ul>
961
962<li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
963appropriate nops inserted to ensure restartability.</li>
964
John Criswellc0c186d2005-11-08 21:11:33 +0000[diff] [blame]965</ul>
966</div>
967
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]968<!-- ======================================================================= -->
969<div class="doc_subsection">
Chris Lattnerf3e5bc62007-05-14 06:56:09 +0000[diff] [blame]970 <a name="c-be">Known problems with the C back-end</a>
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]971</div>
972
973<div class="doc_text">
974
Chris Lattner3016ee92010-09-29 05:34:42 +0000[diff] [blame]975<p>The C backend has numerous problems and is not being actively maintained.
976Depending on it for anything serious is not advised.</p>
977
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]978<ul>
Chris Lattner5733b272008-06-05 06:35:40 +0000[diff] [blame]979<li><a href="http://llvm.org/PR802">The C backend has only basic support for
980 inline assembly code</a>.</li>
Chris Lattner725a0d82007-09-26 06:01:35 +0000[diff] [blame]981<li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
982 C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
Gabor Greif4906abe2009-03-02 12:02:51 +0000[diff] [blame]983 C++ code compiled with <tt>llc</tt> or native compilers.</li>
Duncan Sandsf74c0cc2008-02-10 13:40:55 +0000[diff] [blame]984<li>The C backend does not support all exception handling constructs.</li>
Duncan Sands50723a92009-02-25 11:51:54 +0000[diff] [blame]985<li>The C backend does not support arbitrary precision integers.</li>
Chris Lattner26299222006-11-18 07:51:14 +0000[diff] [blame]986</ul>
987
988</div>
John Criswellc0c186d2005-11-08 21:11:33 +0000[diff] [blame]989
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]990
991<!-- ======================================================================= -->
992<div class="doc_subsection">
Chris Lattner3016ee92010-09-29 05:34:42 +0000[diff] [blame]993 <a name="llvm-gcc">Known problems with the llvm-gcc front-end</a>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]994</div>
Chris Lattner47588f92003-10-02 05:07:23 +0000[diff] [blame]995
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]996<div class="doc_text">
Chris Lattnerc5d658a2006-03-03 00:34:26 +0000[diff] [blame]997
Chris Lattner3016ee92010-09-29 05:34:42 +0000[diff] [blame]998<p>llvm-gcc is generally very stable for the C family of languages. The only
999 major language feature of GCC not supported by llvm-gcc is the
1000 <tt>__builtin_apply</tt> family of builtins. However, some extensions
1001 are only supported on some targets. For example, trampolines are only
1002 supported on some targets (these are used when you take the address of a
1003 nested function).</p>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1004
Chris Lattner3016ee92010-09-29 05:34:42 +0000[diff] [blame]1005<p>Fortran support generally works, but there are still several unresolved bugs
1006 in <a href="http://llvm.org/bugs/">Bugzilla</a>. Please see the
1007 tools/gfortran component for details. Note that llvm-gcc is missing major
1008 Fortran performance work in the frontend and library that went into GCC after
1009 4.2. If you are interested in Fortran, we recommend that you consider using
1010 <a href="#dragonegg">dragonegg</a> instead.</p>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1011
Chris Lattner3016ee92010-09-29 05:34:42 +0000[diff] [blame]1012<p>The llvm-gcc 4.2 Ada compiler has basic functionality. However, this is not a
1013mature technology, and problems should be expected. For example:</p>
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]1014<ul>
Duncan Sands27aff872008-06-08 20:18:35 +0000[diff] [blame]1015<li>The Ada front-end currently only builds on X86-32. This is mainly due
John Criswell524a5dd2009-03-02 15:28:15 +0000[diff] [blame]1016to lack of trampoline support (pointers to nested functions) on other platforms.
1017However, it <a href="http://llvm.org/PR2006">also fails to build on X86-64</a>
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]1018which does support trampolines.</li>
1019<li>The Ada front-end <a href="http://llvm.org/PR2007">fails to bootstrap</a>.
Duncan Sands326a4982009-02-25 11:59:06 +0000[diff] [blame]1020This is due to lack of LLVM support for <tt>setjmp</tt>/<tt>longjmp</tt> style
1021exception handling, which is used internally by the compiler.
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]1022Workaround: configure with <tt>--disable-bootstrap</tt>.</li>
Duncan Sands978bcee2008-10-13 17:27:23 +0000[diff] [blame]1023<li>The c380004, <a href="http://llvm.org/PR2010">c393010</a>
1024and <a href="http://llvm.org/PR2421">cxg2021</a> ACATS tests fail
Duncan Sands326a4982009-02-25 11:59:06 +0000[diff] [blame]1025(c380004 also fails with gcc-4.2 mainline).
1026If the compiler is built with checks disabled then <a href="http://llvm.org/PR2010">c393010</a>
1027causes the compiler to go into an infinite loop, using up all system memory.</li>
Duncan Sandsdd3e6722009-03-02 16:35:57 +0000[diff] [blame]1028<li>Some GCC specific Ada tests continue to crash the compiler.</li>
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]1029<li>The <tt>-E</tt> binder option (exception backtraces)
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]1030<a href="http://llvm.org/PR1982">does not work</a> and will result in programs
Chris Lattner61358ab2009-10-13 17:48:04 +0000[diff] [blame]1031crashing if an exception is raised. Workaround: do not use <tt>-E</tt>.</li>
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]1032<li>Only discrete types <a href="http://llvm.org/PR1981">are allowed to start
1033or finish at a non-byte offset</a> in a record. Workaround: do not pack records
1034or use representation clauses that result in a field of a non-discrete type
1035starting or finishing in the middle of a byte.</li>
Chris Lattnere6e1b352008-06-08 21:19:07 +0000[diff] [blame]1036<li>The <tt>lli</tt> interpreter <a href="http://llvm.org/PR2009">considers
1037'main' as generated by the Ada binder to be invalid</a>.
1038Workaround: hand edit the file to use pointers for <tt>argv</tt> and
1039<tt>envp</tt> rather than integers.</li>
1040<li>The <tt>-fstack-check</tt> option <a href="http://llvm.org/PR2008">is
1041ignored</a>.</li>
Chris Lattner2b659ef2008-02-12 06:29:45 +0000[diff] [blame]1042</ul>
1043</div>
1044
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1045<!-- *********************************************************************** -->
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1046<div class="doc_section">
1047 <a name="additionalinfo">Additional Information</a>
1048</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1049<!-- *********************************************************************** -->
1050
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1051<div class="doc_text">
1052
Chris Lattner416db102005-05-16 17:13:10 +0000[diff] [blame]1053<p>A wide variety of additional information is available on the <a
Chris Lattnerb4b0ce72007-05-18 00:44:29 +0000[diff] [blame]1054href="http://llvm.org">LLVM web page</a>, in particular in the <a
1055href="http://llvm.org/docs/">documentation</a> section. The web page also
Reid Spencer669ed452007-07-09 08:04:31 +0000[diff] [blame]1056contains versions of the API documentation which is up-to-date with the
1057Subversion version of the source code.
Misha Brukman109d9e82005-03-30 19:14:24 +0000[diff] [blame]1058You can access versions of these documents specific to this release by going
1059into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1060
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1061<p>If you have any questions or comments about LLVM, please feel free to contact
Chris Lattnerc463b272005-10-29 07:07:09 +0000[diff] [blame]1062us via the <a href="http://llvm.org/docs/#maillist"> mailing
Chris Lattner5eccca42003-12-12 21:22:16 +0000[diff] [blame]1063lists</a>.</p>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1064
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1065</div>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1066
1067<!-- *********************************************************************** -->
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1068
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1069<hr>
Misha Brukman2061e892003-11-22 01:23:39 +0000[diff] [blame]1070<address>
Misha Brukman38847d52003-12-21 22:53:21 +0000[diff] [blame]1071 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]1072 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
Misha Brukman2061e892003-11-22 01:23:39 +0000[diff] [blame]1073 <a href="http://validator.w3.org/check/referer"><img
Misha Brukman44408702008-12-11 17:34:48 +0000[diff] [blame]1074 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
Misha Brukman2061e892003-11-22 01:23:39 +0000[diff] [blame]1075
Chris Lattnerb4b0ce72007-05-18 00:44:29 +0000[diff] [blame]1076 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1077 Last modified: $Date$
Misha Brukman2061e892003-11-22 01:23:39 +0000[diff] [blame]1078</address>
Chris Lattner79c3fe12003-10-02 04:57:28 +0000[diff] [blame]1079
Misha Brukman500bc302003-11-22 00:38:41 +0000[diff] [blame]1080</body>
1081</html>