1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
5 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
6 <link rel="stylesheet" href="_static/llvm.css" type="text/css">
7 <title>LLVM 3.1 Release Notes</title>
11 <h1>LLVM 3.1 Release Notes</h1>
14 <img style="float:right" src="http://llvm.org/img/DragonSmall.png"
15 width="136" height="136" alt="LLVM Dragon Logo">
19 <li><a href="#intro">Introduction</a></li>
20 <li><a href="#subproj">Sub-project Status Update</a></li>
21 <li><a href="#externalproj">External Projects Using LLVM 3.1</a></li>
22 <li><a href="#whatsnew">What's New in LLVM?</a></li>
23 <li><a href="GettingStarted.html">Installation Instructions</a></li>
24 <li><a href="#knownproblems">Known Problems</a></li>
25 <li><a href="#additionalinfo">Additional Information</a></li>
28 <div class="doc_author">
29 <p>Written by the <a href="http://llvm.org/">LLVM Team</a></p>
32 <h1 style="color:red">These are in-progress notes for the upcoming LLVM 3.1
35 <a href="http://llvm.org/releases/3.0/docs/ReleaseNotes.html">LLVM 3.0
36 Release Notes</a>.</h1>
38 <!-- *********************************************************************** -->
40 <a name="intro">Introduction</a>
42 <!-- *********************************************************************** -->
46 <p>This document contains the release notes for the LLVM Compiler
47 Infrastructure, release 3.1. Here we describe the status of LLVM, including
48 major improvements from the previous release, improvements in various
49 subprojects of LLVM, and some of the current users of the code.
50 All LLVM releases may be downloaded from
51 the <a href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
53 <p>For more information about LLVM, including information about the latest
54 release, please check out the <a href="http://llvm.org/">main LLVM web
55 site</a>. If you have questions or comments,
56 the <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM
57 Developer's Mailing List</a> is a good place to send them.</p>
59 <p>Note that if you are reading this file from a Subversion checkout or the main
60 LLVM web page, this document applies to the <i>next</i> release, not the
61 current one. To see the release notes for a specific release, please see the
62 <a href="http://llvm.org/releases/">releases page</a>.</p>
67 <!-- *********************************************************************** -->
69 <a name="subproj">Sub-project Status Update</a>
71 <!-- *********************************************************************** -->
75 <p>The LLVM 3.1 distribution currently consists of code from the core LLVM
76 repository (which roughly includes the LLVM optimizers, code generators and
77 supporting tools), and the Clang repository. In
78 addition to this code, the LLVM Project includes other sub-projects that are
79 in development. Here we include updates on these subprojects.</p>
81 <!--=========================================================================-->
83 <a name="clang">Clang: C/C++/Objective-C Frontend Toolkit</a>
88 <p><a href="http://clang.llvm.org/">Clang</a> is an LLVM front end for the C,
89 C++, and Objective-C languages. Clang aims to provide a better user
90 experience through expressive diagnostics, a high level of conformance to
91 language standards, fast compilation, and low memory use. Like LLVM, Clang
92 provides a modular, library-based architecture that makes it suitable for
93 creating or integrating with other development tools. Clang is considered a
94 production-quality compiler for C, Objective-C, C++ and Objective-C++ on x86
95 (32- and 64-bit), and for Darwin/ARM targets.</p>
97 <p>In the LLVM 3.1 time-frame, the Clang team has made many improvements.
98 Highlights include:</p>
100 <li>Greatly expanded <a href="http://clang.llvm.org/cxx_status.html">C++11
101 support</a> including lambdas, initializer lists, constexpr, user-defined
102 literals, and atomics.</li>
103 <li>A new <a href="http://clang.llvm.org/docs/Tooling.html">tooling</a>
104 library to ease building of clang-based standalone tools.</li>
105 <li>Extended support for
106 <a href="http://clang.llvm.org/docs/ObjectiveCLiterals.html">literals in
107 Objective C</a>.</li>
110 <p>For more details about the changes to Clang since the 3.0 release, see the
111 <a href="http://clang.llvm.org/docs/ReleaseNotes.html">Clang release notes</a>
115 <p>If Clang rejects your code but another compiler accepts it, please take a
116 look at the <a href="http://clang.llvm.org/compatibility.html">language
117 compatibility</a> guide to make sure this is not intentional or a known
122 <!--=========================================================================-->
124 <a name="dragonegg">DragonEgg: GCC front-ends, LLVM back-end</a>
128 <p><a href="http://dragonegg.llvm.org/">DragonEgg</a> is a
129 <a href="http://gcc.gnu.org/wiki/plugins">gcc plugin</a> that replaces GCC's
130 optimizers and code generators with LLVM's. It works with gcc-4.5 and gcc-4.6
131 (and partially with gcc-4.7), can target the x86-32/x86-64 and ARM processor
132 families, and has been successfully used on the Darwin, FreeBSD, KFreeBSD,
133 Linux and OpenBSD platforms. It fully supports Ada, C, C++ and Fortran. It
134 has partial support for Go, Java, Obj-C and Obj-C++.</p>
136 <p>The 3.1 release has the following notable changes:</p>
140 <li>Partial support for gcc-4.7. Ada support is poor, but other languages work
143 <li>Support for ARM processors. Some essential gcc headers that are needed to
144 build DragonEgg for ARM are not installed by gcc. To work around this,
145 copy the missing headers from the gcc source tree.</li>
147 <li>Better optimization for Fortran by exploiting the fact that Fortran scalar
148 arguments have 'restrict' semantics.</li>
150 <li>Better optimization for all languages by passing information about type
151 aliasing and type ranges to the LLVM optimizers.</li>
153 <li>A regression test-suite was added.</li>
159 <!--=========================================================================-->
161 <a name="compiler-rt">compiler-rt: Compiler Runtime Library</a>
166 <p>The new LLVM <a href="http://compiler-rt.llvm.org/">compiler-rt project</a>
167 is a simple library that provides an implementation of the low-level
168 target-specific hooks required by code generation and other runtime
169 components. For example, when compiling for a 32-bit target, converting a
170 double to a 64-bit unsigned integer is compiled into a runtime call to the
171 "__fixunsdfdi" function. The compiler-rt library provides highly optimized
172 implementations of this and other low-level routines (some are 3x faster than
173 the equivalent libgcc routines).</p>
179 <!--=========================================================================-->
181 <a name="lldb">LLDB: Low Level Debugger</a>
186 <p><a href="http://lldb.llvm.org">LLDB</a> is a ground-up implementation of a
187 command line debugger, as well as a debugger API that can be used from other
188 applications. LLDB makes use of the Clang parser to provide high-fidelity
189 expression parsing (particularly for C++) and uses the LLVM JIT for target
196 <!--=========================================================================-->
198 <a name="libc++">libc++: C++ Standard Library</a>
203 <p>Like compiler_rt, libc++ is now <a href="DeveloperPolicy.html#license">dual
204 licensed</a> under the MIT and UIUC license, allowing it to be used more
211 <!--=========================================================================-->
213 <a name="vmkit">VMKit</a>
218 <p>The <a href="http://vmkit.llvm.org/">VMKit project</a> is an
219 implementation of a Java Virtual Machine (Java VM or JVM) that uses LLVM for
220 static and just-in-time compilation.
222 <p>In the LLVM 3.1 time-frame, VMKit has had significant improvements on both
223 runtime and startup performance:</p>
232 <!--=========================================================================-->
234 <a name="Polly">Polly: Polyhedral Optimizer</a>
239 <p><a href="http://polly.llvm.org/">Polly</a> is an <em>experimental</em>
240 optimizer for data locality and parallelism. It currently provides high-level
241 loop optimizations and automatic parallelisation (using the OpenMP run time).
242 Work in the area of automatic SIMD and accelerator code generation was
245 <p>Within the LLVM 3.1 time-frame there were the following highlights:</p>
248 <li>Polly became an official LLVM project</li>
249 <li>Polly can be loaded directly into clang (Enabled by '-O3 -mllvm -polly'
251 <li>An automatic scheduling optimizer (derived from <a
252 href="http://pluto-compiler.sourceforge.net/">Pluto</a>) was integrated. It
253 performs loop transformations to optimize for data-locality and parallelism.
254 The transformations include, but are not limited to interchange, fusion,
255 fission, skewing and tiling.
263 <!-- *********************************************************************** -->
265 <a name="externalproj">External Open Source Projects Using LLVM 3.1</a>
267 <!-- *********************************************************************** -->
271 <p>An exciting aspect of LLVM is that it is used as an enabling technology for
272 a lot of other language and tools projects. This section lists some of the
273 projects that have already been updated to work with LLVM 3.1.</p>
279 <p><a href="http://faust.grame.fr/">FAUST</a> is a compiled language for
280 real-time audio signal processing. The name FAUST stands for Functional
281 AUdio STream. Its programming model combines two approaches: functional
282 programming and block diagram composition. In addition with the C, C++, Java,
283 JavaScript output formats, the Faust compiler can generate LLVM bitcode, and
284 works with LLVM 2.7-3.1.</p>
288 <h3>Glasgow Haskell Compiler (GHC)</h3>
292 <p><a href="http://www.haskell.org/ghc/">GHC</a> is an open source compiler and
293 programming suite for Haskell, a lazy functional programming language. It
294 includes an optimizing static compiler generating good code for a variety of
295 platforms, together with an interactive system for convenient, quick
298 <p>GHC 7.0 and onwards include an LLVM code generator, supporting LLVM 2.8 and
303 <h3>LLVM D Compiler</h3>
307 <p><a href="https://github.com/ldc-developers/ldc">LLVM D Compiler</a> (LDC) is
308 a compiler for the D programming Language. It is based on the DMD frontend
309 and uses LLVM as backend.</p>
313 <h3>Open Shading Language</h3>
317 <p><a href="https://github.com/imageworks/OpenShadingLanguage/">Open Shading
318 Language (OSL)</a> is a small but rich language for programmable shading in
319 advanced global illumination renderers and other applications, ideal for
320 describing materials, lights, displacement, and pattern generation. It uses
321 LLVM to JIT complex shader networks to x86 code at runtime.</p>
323 <p>OSL was developed by Sony Pictures Imageworks for use in its in-house
324 renderer used for feature film animation and visual effects, and is
325 distributed as open source software with the "New BSD" license.</p>
329 <h3>Portable OpenCL (pocl)</h3>
333 <p>In addition to producing an easily portable open source OpenCL
334 implementation, another major goal of <a href="http://pocl.sourceforge.net/">
335 pocl</a> is improving performance portability of OpenCL programs with
336 compiler optimizations, reducing the need for target-dependent manual
337 optimizations. An important part of pocl is a set of LLVM passes used to
338 statically parallelize multiple work-items with the kernel compiler, even in
339 the presence of work-group barriers. This enables static parallelization of
340 the fine-grained static concurrency in the work groups in multiple ways
341 (SIMD, VLIW, superscalar,...).</p>
349 <p><a href="http://pure-lang.googlecode.com/">Pure</a> is an
350 algebraic/functional programming language based on term rewriting. Programs
351 are collections of equations which are used to evaluate expressions in a
352 symbolic fashion. The interpreter uses LLVM as a backend to JIT-compile Pure
353 programs to fast native code. Pure offers dynamic typing, eager and lazy
354 evaluation, lexical closures, a hygienic macro system (also based on term
355 rewriting), built-in list and matrix support (including list and matrix
356 comprehensions) and an easy-to-use interface to C and other programming
357 languages (including the ability to load LLVM bitcode modules, and inline C,
358 C++, Fortran and Faust code in Pure programs if the corresponding
359 LLVM-enabled compilers are installed).</p>
361 <p>Pure version 0.54 has been tested and is known to work with LLVM 3.1 (and
362 continues to work with older LLVM releases >= 2.5).</p>
366 <h3>TTA-based Co-design Environment (TCE)</h3>
370 <p><a href="http://tce.cs.tut.fi/">TCE</a> is a toolset for designing
371 application-specific processors (ASP) based on the Transport triggered
372 architecture (TTA). The toolset provides a complete co-design flow from C/C++
373 programs down to synthesizable VHDL/Verilog and parallel program binaries.
374 Processor customization points include the register files, function units,
375 supported operations, and the interconnection network.</p>
377 <p>TCE uses Clang and LLVM for C/C++ language support, target independent
378 optimizations and also for parts of code generation. It generates new
379 LLVM-based code generators "on the fly" for the designed TTA processors and
380 loads them in to the compiler backend as runtime libraries to avoid
381 per-target recompilation of larger parts of the compiler chain.</p>
387 <!-- *********************************************************************** -->
389 <a name="whatsnew">What's New in LLVM 3.1?</a>
391 <!-- *********************************************************************** -->
395 <p>This release includes a huge number of bug fixes, performance tweaks and
396 minor improvements. Some of the major improvements and new features are
397 listed in this section.</p>
399 <!--=========================================================================-->
401 <a name="majorfeatures">Major New Features</a>
406 <!-- Features that need text if they're finished for 3.1:
410 loop dependence analysis
411 CorrelatedValuePropagation
412 lib/Transforms/IPO/MergeFunctions.cpp => consider for 3.1.
413 Integrated assembler on by default for arm/thumb?
418 Analysis/RegionInfo.h + Dom Frontiers
419 SparseBitVector: used in LiveVar.
420 llvm/lib/Archive - replace with lib object?
423 <p>LLVM 3.1 includes several major changes and big features:</p>
426 <li><a href="../tools/clang/docs/AddressSanitizer.html">AddressSanitizer</a>,
427 a fast memory error detector.</li>
428 <li><a href="CodeGenerator.html#machineinstrbundle">MachineInstr Bundles</a>,
429 Support to model instruction bundling / packing.</li>
430 <li><a href="#armintegratedassembler">ARM Integrated Assembler</a>,
431 A full featured assembler and direct-to-object support for ARM.</li>
432 <li><a href="#blockplacement">Basic Block Placement</a>
433 Probability driven basic block placement.</li>
440 <!--=========================================================================-->
442 <a name="coreimprovements">LLVM IR and Core Improvements</a>
447 <p>LLVM IR has several new features for better support of new targets and that
448 expose new optimization opportunities:</p>
451 <li>IR support for half float</li>
452 <li>IR support for vectors of pointers, including vector GEPs.</li>
453 <li>Module flags have been introduced. They convey information about the
454 module as a whole to LLVM subsystems.</li>
455 <li>Loads can now have range metadata attached to them to describe the
456 possible values being loaded.</li>
457 <li>Inline cost heuristics have been completely overhauled and now closely
458 model constant propagation through call sites, disregard trivially dead
459 code costs, and can model C++ STL iterator patterns.</li>
464 <!--=========================================================================-->
466 <a name="optimizer">Optimizer Improvements</a>
471 <p>In addition to many minor performance tweaks and bug fixes, this
472 release includes a few major enhancements and additions to the
476 <li>The loop unroll pass now is able to unroll loops with run-time trip counts.
477 This feature is turned off by default, and is enabled with the
478 <code>-unroll-runtime</code> flag.</li>
479 <li>A new basic-block autovectorization pass is available. Pass
480 <code>-vectorize</code> to run this pass along with some associated
481 post-vectorization cleanup passes. For more information, see the EuroLLVM
482 2012 slides: <a href="http://llvm.org/devmtg/2012-04-12/Slides/Hal_Finkel.pdf">
483 Autovectorization with LLVM</a>.</li>
489 <!--=========================================================================-->
491 <a name="mc">MC Level Improvements</a>
496 <p>The LLVM Machine Code (aka MC) subsystem was created to solve a number of
497 problems in the realm of assembly, disassembly, object file format handling,
498 and a number of other related areas that CPU instruction-set level tools work
499 in. For more information, please see
500 the <a href="http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html">Intro
501 to the LLVM MC Project Blog Post</a>.</p>
509 <!--=========================================================================-->
511 <a name="codegen">Target Independent Code Generator Improvements</a>
516 <p>We have changed the way that the Type Legalizer legalizes vectors. The type
517 legalizer now attempts to promote integer elements. This enabled the
518 implementation of vector-select. Additionally, we see a performance boost on
519 workloads which use vectors of chars and shorts, since they are now promoted
520 to 32-bit types, which are better supported by the SIMD instruction set.
521 Floating point types are still widened as before.</p>
524 <p>We have put a significant amount of work into the code generator
525 infrastructure, which allows us to implement more aggressive algorithms and
526 make it run faster:</p>
529 <li>TableGen can now synthesize register classes that are only needed to
530 represent combinations of constraints from instructions and sub-registers.
531 The synthetic register classes inherit most of their properties form their
532 closest user-defined super-class.</li>
533 <li><code>MachineRegisterInfo</code> now allows the reserved registers to be
534 frozen when register allocation starts. Target hooks should use the
535 <code>MRI->canReserveReg(FramePtr)</code> method to avoid accidentally
536 disabling frame pointer elimination during register allocation.</li>
537 <li>A new kind of <code>MachineOperand</code> provides a compact
538 representation of large clobber lists on call instructions. The register
539 mask operand references a bit mask of preserved registers. Everything else
543 <p> We added new TableGen infrastructure to support bundling for
544 Very Long Instruction Word (VLIW) architectures. TableGen can now
545 automatically generate a deterministic finite automaton from a VLIW
546 target's schedule description which can be queried to determine
547 legal groupings of instructions in a bundle.</p>
549 <p> We have added a new target independent VLIW packetizer based on the
550 DFA infrastructure to group machine instructions into bundles.</p>
555 <a name="blockplacement">Basic Block Placement</a>
558 <p>A probability based block placement and code layout algorithm was added to
559 LLVM's code generator. This layout pass supports probabilities derived from
560 static heuristics as well as source code annotations such as
561 <code>__builtin_expect</code>.</p>
564 <!--=========================================================================-->
566 <a name="x86">X86-32 and X86-64 Target Improvements</a>
571 <p>New features and major changes in the X86 target include:</p>
574 <li>Bug fixes and improved support for AVX1</li>
575 <li>Support for AVX2 (still incomplete at this point)</li>
576 <li>Call instructions use the new register mask operands for faster compile
577 times and better support for different calling conventions. The old WINCALL
578 instructions are no longer needed.</li>
579 <li>DW2 Exception Handling is enabled on Cygwin and MinGW.</li>
580 <li>Support for implicit TLS model used with MS VC runtime</li>
585 <!--=========================================================================-->
587 <a name="ARM">ARM Target Improvements</a>
592 <p>New features of the ARM target include:</p>
595 <li>The constant island pass now supports basic block and constant pool entry
596 alignments greater than 4 bytes.</li>
597 <li>On Darwin, the ARM target now has a full-featured integrated assembler.
602 <a name="armintegratedassembler">ARM Integrated Assembler</a>
605 <p>The ARM target now includes a full featured macro assembler, including
606 direct-to-object module support for clang. The assembler is currently enabled
607 by default for Darwin only pending testing and any additional necessary
608 platform specific support for Linux.</p>
610 <p>Full support is included for Thumb1, Thumb2 and ARM modes, along with
611 subtarget and CPU specific extensions for VFP2, VFP3 and NEON.</p>
613 <p>The assembler is Unified Syntax only (see ARM Architecural Reference Manual
614 for details). While there is some, and growing, support for pre-unfied (divided)
615 syntax, there are still significant gaps in that support.</p>
619 <!--=========================================================================-->
621 <a name="MIPS">MIPS Target Improvements</a>
626 <p>This release has seen major new work on just about every aspect of the MIPS
627 backend. Some of the major new features include:</p>
634 <!--=========================================================================-->
636 <a name="OtherTS">Other Target Specific Improvements</a>
641 <p>Support for Qualcomm's Hexagon VLIW processor has been added.</p>
651 <!--=========================================================================-->
653 <a name="changes">Major Changes and Removed Features</a>
658 <p>If you're already an LLVM user or developer with out-of-tree changes based on
659 LLVM 3.1, this section lists some "gotchas" that you may run into upgrading
660 from the previous release.</p>
663 <li>LLVM 3.1 removes support for reading LLVM 2.9 bitcode files. Going
664 forward, we aim for all future versions of LLVM to read bitcode files and
665 <tt>.ll</tt> files produced by LLVM 3.0 and later.</li>
666 <li>The <tt>unwind</tt> instruction is now gone. With the introduction of the
667 new exception handling system in LLVM 3.0, the <tt>unwind</tt> instruction
668 became obsolete.</li>
669 <li>LLVM 3.0 and earlier automatically added the returns_twice fo functions
670 like setjmp based on the name. This functionality was removed in 3.1.
671 This affects Clang users, if -ffreestanding is used.</li>
677 <!--=========================================================================-->
679 <a name="api_changes">Internal API Changes</a>
684 <p>In addition, many APIs have changed in this release. Some of the major
685 LLVM API changes are:</p>
688 <li>Target specific options have been moved from global variables to members
689 on the new <code>TargetOptions</code> class, which is local to each
690 <code>TargetMachine</code>. As a consequence, the associated flags will
691 no longer be accepted by <tt>clang -mllvm</tt>. This includes:
693 <li><code>llvm::PrintMachineCode</code></li>
694 <li><code>llvm::NoFramePointerElim</code></li>
695 <li><code>llvm::NoFramePointerElimNonLeaf</code></li>
696 <li><code>llvm::DisableFramePointerElim(const MachineFunction &)</code></li>
697 <li><code>llvm::LessPreciseFPMADOption</code></li>
698 <li><code>llvm::LessPrecideFPMAD()</code></li>
699 <li><code>llvm::NoExcessFPPrecision</code></li>
700 <li><code>llvm::UnsafeFPMath</code></li>
701 <li><code>llvm::NoInfsFPMath</code></li>
702 <li><code>llvm::NoNaNsFPMath</code></li>
703 <li><code>llvm::HonorSignDependentRoundingFPMathOption</code></li>
704 <li><code>llvm::HonorSignDependentRoundingFPMath()</code></li>
705 <li><code>llvm::UseSoftFloat</code></li>
706 <li><code>llvm::FloatABIType</code></li>
707 <li><code>llvm::NoZerosInBSS</code></li>
708 <li><code>llvm::JITExceptionHandling</code></li>
709 <li><code>llvm::JITEmitDebugInfo</code></li>
710 <li><code>llvm::JITEmitDebugInfoToDisk</code></li>
711 <li><code>llvm::GuaranteedTailCallOpt</code></li>
712 <li><code>llvm::StackAlignmentOverride</code></li>
713 <li><code>llvm::RealignStack</code></li>
714 <li><code>llvm::DisableJumpTables</code></li>
715 <li><code>llvm::EnableFastISel</code></li>
716 <li><code>llvm::getTrapFunctionName()</code></li>
717 <li><code>llvm::EnableSegmentedStacks</code></li>
719 <li>The MDBuilder class has been added to simplify the creation of
726 <!--=========================================================================-->
728 <a name="tools_changes">Tools Changes</a>
733 <p>In addition, some tools have changed in this release. Some of the changes
738 <li>llvm-stress is a command line tool for generating random .ll files to fuzz
739 different LLVM components. </li>
740 <li>llvm-ld has been removed. Use llvm-link or Clang instead.</li>
751 <!--=========================================================================-->
753 <a name="python">Python Bindings</a>
758 <p>Officially supported Python bindings have been added! Feature support is far
759 from complete. The current bindings support interfaces to:</p>
761 <li>Object File Interface</li>
762 <li>Disassembler</li>
765 <p>Using the Object File Interface, it is possible to inspect binary object files.
766 Think of it as a Python version of readelf or llvm-objdump.</p>
768 <p>Support for additional features is currently being developed by community
769 contributors. If you are interested in shaping the direction of the Python
770 bindings, please express your intent on IRC or the developers list.</p>
776 <!-- *********************************************************************** -->
778 <a name="knownproblems">Known Problems</a>
780 <!-- *********************************************************************** -->
784 <p>LLVM is generally a production quality compiler, and is used by a broad range
785 of applications and shipping in many products. That said, not every
786 subsystem is as mature as the aggregate, particularly the more obscure
787 targets. If you run into a problem, please check the <a
788 href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
789 there isn't already one or ask on the <a
790 href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev
793 <p>Known problem areas include:</p>
796 <li>The Alpha, Blackfin, CellSPU, MSP430, PTX, SystemZ and
797 XCore backends are experimental, and the Alpha, Blackfin and SystemZ
798 targets have already been removed from mainline.</li>
800 <li>The integrated assembler, disassembler, and JIT is not supported by
801 several targets. If an integrated assembler is not supported, then a
802 system assembler is required. For more details, see the <a
803 href="CodeGenerator.html#targetfeatures">Target Features Matrix</a>.
806 <li>The C backend has numerous problems and is not being actively maintained.
807 Depending on it for anything serious is not advised.</li>
812 <!-- *********************************************************************** -->
814 <a name="additionalinfo">Additional Information</a>
816 <!-- *********************************************************************** -->
820 <p>A wide variety of additional information is available on
821 the <a href="http://llvm.org/">LLVM web page</a>, in particular in
822 the <a href="http://llvm.org/docs/">documentation</a> section. The web page
823 also contains versions of the API documentation which is up-to-date with the
824 Subversion version of the source code. You can access versions of these
825 documents specific to this release by going into the "<tt>llvm/doc/</tt>"
826 directory in the LLVM tree.</p>
828 <p>If you have any questions or comments about LLVM, please feel free to contact
829 us via the <a href="http://llvm.org/docs/#maillist"> mailing lists</a>.</p>
833 <!-- *********************************************************************** -->
837 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
838 src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
839 <a href="http://validator.w3.org/check/referer"><img
840 src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
842 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
843 Last modified: $Date$