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="llvm.css" type="text/css">
7 <title>LLVM 2.4 Release Notes</title>
11 <div class="doc_title">LLVM 2.4 Release Notes</div>
14 <li><a href="#intro">Introduction</a></li>
15 <li><a href="#subproj">Sub-project Status Update</a></li>
16 <li><a href="#whatsnew">What's New in LLVM?</a></li>
17 <li><a href="GettingStarted.html">Installation Instructions</a></li>
18 <li><a href="#portability">Portability and Supported Platforms</a></li>
19 <li><a href="#knownproblems">Known Problems</a></li>
20 <li><a href="#additionalinfo">Additional Information</a></li>
23 <div class="doc_author">
24 <p>Written by the <a href="http://llvm.org">LLVM Team</a></p>
27 <!-- *********************************************************************** -->
28 <div class="doc_section">
29 <a name="intro">Introduction</a>
31 <!-- *********************************************************************** -->
33 <div class="doc_text">
35 <p>This document contains the release notes for the LLVM Compiler
36 Infrastructure, release 2.4. Here we describe the status of LLVM, including
37 major improvements from the previous release and significant known problems.
38 All LLVM releases may be downloaded from the <a
39 href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
41 <p>For more information about LLVM, including information about the latest
42 release, please check out the <a href="http://llvm.org/">main LLVM
43 web site</a>. If you have questions or comments, the <a
44 href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM Developer's Mailing
45 List</a> is a good place to send them.</p>
47 <p>Note that if you are reading this file from a Subversion checkout or the
48 main LLVM web page, this document applies to the <i>next</i> release, not the
49 current one. To see the release notes for a specific release, please see the
50 <a href="http://llvm.org/releases/">releases page</a>.</p>
54 <!-- Unfinished features in 2.4:
58 llc -enable-value-prop, propagation of value info (sign/zero ext info) from
62 <!-- for announcement email:
67 <!-- *********************************************************************** -->
68 <div class="doc_section">
69 <a name="subproj">Sub-project Status Update</a>
71 <!-- *********************************************************************** -->
73 <div class="doc_text">
75 The LLVM 2.4 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 llvm-gcc repository. In addition to this code, the
78 LLVM Project includes other sub-projects that are in development. The two which
79 are the most actively developed are the <a href="#clang">Clang Project</a> and
80 the <a href="#vmkit">VMKit Project</a>.
86 <!--=========================================================================-->
87 <div class="doc_subsection">
88 <a name="clang">Clang: C/C++/Objective-C Frontend Toolkit</a>
91 <div class="doc_text">
93 <p>The <a href="http://clang.llvm.org/">Clang project</a> is an effort to build
94 a set of new 'LLVM native' front-end technologies for the LLVM optimizer
95 and code generator. Clang is continuing to make major strides forward in all
96 areas. Its C and Objective-C parsing support is very solid, and the code
97 generation support is far enough along to build many C applications. While not
98 yet production quality, it is progressing very nicely. In addition, C++
99 front-end work has started to make significant progress.</p>
101 Clang, in conjunction with the <tt>ccc</tt> driver, is now usable as a
102 replacement for gcc for building some small- to medium-sized C applications.
103 Additionally, Clang now has code generation support for Objective-C on Mac OS X
104 platform. Major highlights include:
106 <li> Clang/ccc pass almost all of the LLVM test suite on Mac OS X and Linux
107 on the 32-bit x86 architecture. This includes significant C
108 applications such as <a href="http://www.sqlite.org">sqlite3</a>,
109 <a href="http://www.lua.org">lua</a>, and
110 <a href="http://www.clamav.net">Clam AntiVirus</a>.
112 <li> Clang can build the majority of Objective-C examples shipped with the
113 Mac OS X Developer Tools.
116 Clang code generation still needs considerable testing and development, however.
117 Some areas under active development include:
119 <li> Improved support for C and Objective-C features, for example
120 variable-length arrays, va_arg, exception handling (Obj-C), and garbage
122 <li> ABI compatibility, especially for platforms other than 32-bit x86.
127 <!--=========================================================================-->
128 <div class="doc_subsection">
129 <a name="clangsa">Clang Static Analyzer</a>
132 <div class="doc_text">
134 <p>The Clang project also includes an early stage static source code analysis
135 tool for <a href="http://clang.llvm.org/StaticAnalysis.html">automatically
136 finding bugs</a> in C and Objective-C programs. The tool performs a growing set
137 of checks to find bugs that occur on a specific path within a program. Examples
138 of bugs the tool finds include logic errors such as null dereferences,
139 violations of various API rules, dead code, and potential memory leaks in
140 Objective-C programs. Since its inception, public feedback on the tool has been
141 extremely positive, and conservative estimates put the number of real bugs it
142 has found in industrial-quality software on the order of thousands.</p>
144 <p>The tool also provides a simple web GUI to inspect potential bugs found by
145 the tool. While still early in development, the GUI illustrates some of the key
146 features of Clang: accurate source location information, which is used by the
147 GUI to highlight specific code expressions that relate to a bug (including those
148 that span multiple lines) and built-in knowledge of macros, which is used to
149 perform inline expansion of macros within the GUI itself.</p>
151 <p>The set of checks performed by the static analyzer is gradually expanding,
153 future plans for the tool include full source-level inter-procedural analysis
154 and deeper checks such as buffer overrun detection. There are many opportunities
155 to extend and enhance the static analyzer, and anyone interested in working on
156 this project is encouraged to get involved!</p>
160 <!--=========================================================================-->
161 <div class="doc_subsection">
162 <a name="vmkit">VMKit: JVM/CLI Virtual Machine Implementation</a>
165 <div class="doc_text">
167 The <a href="http://vmkit.llvm.org/">VMKit project</a> is an implementation of
168 a JVM and a CLI Virtual Machines (Microsoft .NET is an
169 implementation of the CLI) using the Just-In-Time compiler of LLVM.</p>
171 <p>Following LLVM 2.4, VMKit has its first release 0.24 that you can find on its
172 <a href="http://vmkit.llvm.org/releases/">webpage</a>. The release includes
173 bug fixes, cleanup and new features. The major changes are:</p>
177 <li> Support for generics in the .Net virtual machine.
178 <li> Initial support for the Mono class libraries.
179 <li> Support for MacOSX/x86, following LLVM's support for exceptions in
181 <li> A new vmkit driver: a program to run java or .net applications. The
182 driver supports llvm command line arguments including the new "-fast" option.
183 <li> A new memory allocation scheme in the JVM that makes unloading a
184 class loader very fast.
185 <li> VMKit now follows the LLVM Makefile machinery.
191 <!-- *********************************************************************** -->
192 <div class="doc_section">
193 <a name="whatsnew">What's New in LLVM?</a>
195 <!-- *********************************************************************** -->
197 <div class="doc_text">
199 <p>This release includes a huge number of bug fixes, performance tweaks and
200 minor improvements. Some of the major improvements and new features are listed
205 <!--=========================================================================-->
206 <div class="doc_subsection">
207 <a name="majorfeatures">Major New Features</a>
210 <div class="doc_text">
212 <p>LLVM 2.4 includes several major new capabilities:</p>
215 <li><p>The most visible end-user change in LLVM 2.4 is that it includes many
216 optimizations and changes to make -O0 compile times much faster. You should see
217 improvements on the order of 30% (or more) faster than LLVM 2.3. There are many
218 pieces to this change, described in more detail below. The speedups and new
219 components can also be used for JIT compilers that want fast compilation as
222 <li><p>The biggest change to the LLVM IR is that Multiple Return Values (which
223 were introduced in LLVM 2.3) have been generalized to full support for "First
224 Class Aggregate" values in LLVM 2.4. This means that LLVM IR supports using
225 structs and arrays as values in a function. This capability is mostly useful
226 for front-end authors, who prefer to treat things like complex numbers, simple
227 tuples, dope vectors, etc as Value*'s instead of as a tuple of Value*'s or as
228 memory values. Bitcode files from LLVM 2.3 will automatically migrate to the
229 general representation.</p></li>
231 <li><p>LLVM 2.4 also includes an initial port for the PIC16 microprocessor. This
232 is the LLVM target that only has support for 8 bit registers, and a number of
233 other crazy constraints. While the port is still in early development stages,
234 it shows some interesting things you can do with LLVM.</p></li>
241 <!--=========================================================================-->
242 <div class="doc_subsection">
243 <a name="llvm-gcc">llvm-gcc 4.2 Improvements</a>
246 <div class="doc_text">
248 <p>LLVM fully supports the llvm-gcc 4.2 front-end, which marries the GCC
249 front-ends and driver with the LLVM optimizer and code generator. It currently
250 includes support for the C, C++, Objective-C, Ada, and Fortran front-ends.</p>
253 <li>LLVM 2.4 supports the full set of atomic <tt>__sync_*</tt> builtins. LLVM
254 2.3 only supported those used by OpenMP, but 2.4 supports them all. While
255 llvm-gcc supports all of these builtins, note that not all targets do. X86
256 support them all in both 32-bit and 64-bit mode and PowerPC supports them all
257 except for the 64-bit operations when in 32-bit mode.</li>
259 <li>llvm-gcc now supports an <tt>-flimited-precision</tt> option, which tells
260 the compiler that it is ok to use low-precision approximations of certain libm
261 functions (like tan, log, etc). This allows you to get high performance if you
262 only need (say) 14-bits of precision.</li>
264 <li>llvm-gcc now supports a C language extension known as "<a
265 href="http://lists.cs.uiuc.edu/pipermail/cfe-dev/2008-August/002670.html">Blocks</a>".
266 This feature is similar to nested functions and closures, but does not
267 require stack trampolines (with most ABIs) and supports returning closures
268 from functions that define them. Note that actually <em>using</em> Blocks
269 requires a small runtime that is not included with llvm-gcc.</li>
271 <li>llvm-gcc now supports a new <tt>-flto</tt> option. On systems that support
272 transparent Link Time Optimization (currently Darwin systems with Xcode 3.1 and
273 later) this allows the use of LTO with other optimization levels like -Os.
274 Previously, LTO could only be used with -O4, which implied optimizations in
275 -O3 that can increase code size.</li>
281 <!--=========================================================================-->
282 <div class="doc_subsection">
283 <a name="coreimprovements">LLVM Core Improvements</a>
286 <div class="doc_text">
287 <p>New features include:
291 <li>A major change to the <tt>Use</tt> class landed, which shrank it by 25%. Since
292 this is a pervasive part of the LLVM, it ended up reducing the memory use of
293 LLVM IR in general by 15% for most programs.</li>
295 <li>Values with no names are now pretty printed by <tt>llvm-dis</tt> more
296 nicely. They now print as "<tt>%3 = add i32 %A, 4</tt>" instead of
297 "<tt>add i32 %A, 4 ; <i32>:3</tt>", which makes it much easier to read.
300 <li>LLVM 2.4 includes some changes for better vector support. First, the shift
301 operations (<tt>shl</tt>, <tt>ashr</tt>, <tt>lshr</tt>) now all support vectors
302 and do an element-by-element shift (shifts of the whole vector can be
303 accomplished by bitcasting the vector to <1 x i128> for example). Second,
304 there is initial support in development for vector comparisons with the
305 <a href="LangRef.html#i_fcmp">fcmp</a>/<a href="LangRef.html#i_icmp">icmp</a>
306 instructions. These instructions compare two vectors and return a vector of
307 i1's for each result. Note that there is very little codegen support available
308 for any of these IR features though.</li>
310 <li>A new <tt>DebugInfoBuilder</tt> class is available, which makes it much
311 easier for front-ends to create debug info descriptors, similar to the way that
312 <tt>IRBuilder</tt> makes it easier to create LLVM IR.</li>
314 <li>The <tt>IRBuilder</tt> class is now parameterized by a class responsible
315 for constant folding. The default <tt>ConstantFolder</tt> class does target independent
316 constant folding. The <tt>NoFolder</tt> class does no constant folding at all, which is
317 useful when learning how LLVM works. The <tt>TargetFolder</tt> class folds the most,
318 doing target dependent constant folding.</li>
320 <li>LLVM now supports "function attributes", which allows us to separate return
321 value attributes from function attributes. LLVM now supports attributes on a
322 function itself, a return value, and its parameters. New supported function
323 attributes include noinline/alwaysinline and the "opt-size" flag which says the
324 function should be optimized for code size.</li>
326 <li>LLVM IR now directly represents "common" linkage, instead of
327 representing it as a form of weak linkage.</li>
333 <!--=========================================================================-->
334 <div class="doc_subsection">
335 <a name="optimizer">Optimizer Improvements</a>
338 <div class="doc_text">
340 <p>In addition to a huge array of bug fixes and minor performance tweaks, this
341 release includes a few major enhancements and additions to the optimizers:</p>
345 <li>The Global Value Numbering (GVN) pass now does local Partial Redundancy
346 Elimination (PRE) to eliminate some partially redundant expressions in cases
347 where doing so won't grow code size.</li>
349 <li>LLVM 2.4 includes a new loop deletion pass (which removes output-free
350 provably-finite loops) and a rewritten Aggressive Dead Code Elimination (ADCE)
351 pass that no longer uses control dependence information. These changes speed up
352 the optimizer and also prevent it from deleting output-free infinite
355 <li>The new AddReadAttrs pass works out which functions are read-only or
356 read-none (these correspond to 'pure' and 'const' in GCC) and marks them
357 with the appropriate attribute.</li>
359 <li>LLVM 2.4 now includes a new SparsePropagation framework, which makes it
360 trivial to build lattice-based dataflow solvers that operate over LLVM IR. Using
361 this interface means that you just define objects to represent your lattice
362 values and the transfer functions that operate on them. It handles the
363 mechanics of worklist processing, liveness tracking, handling PHI nodes,
366 <li>The Loop Strength Reduction and induction variable optimization passes have
367 several improvements to avoid inserting MAX expressions, to optimize simple
368 floating point induction variables and to analyze trip counts of more
371 <li>Various helper functions (ComputeMaskedBits, ComputeNumSignBits, etc) were
372 pulled out of the Instruction Combining pass and put into a new
373 <tt>ValueTracking.h</tt> header, where they can be reused by other passes.</li>
375 <li>The tail duplication pass has been removed from the standard optimizer
376 sequence used by llvm-gcc. This pass still exists, but the benefits it once
377 provided are now achieved by other passes.</li>
383 <!--=========================================================================-->
384 <div class="doc_subsection">
385 <a name="codegen">Code Generator Improvements</a>
388 <div class="doc_text">
390 <p>We have put a significant amount of work into the code generator infrastructure,
391 which allows us to implement more aggressive algorithms and make it run
395 <li>The target-independent code generator supports (and the X86 backend
396 currently implements) a new interface for "fast" instruction selection. This
397 interface is optimized to produce code as quickly as possible, sacrificing
398 code quality to do it. This is used by default at -O0 or when using
399 "llc -fast" on X86. It is straight-forward to add support for
400 other targets if faster -O0 compilation is desired.</li>
402 <li>In addition to the new 'fast' instruction selection path, many existing
403 pieces of the code generator have been optimized in significant ways.
404 SelectionDAG's are now pool allocated and use better algorithms in many
405 places, the ".s" file printers now use <tt>raw_ostream</tt> to emit text much faster,
406 etc. The end result of these improvements is that the compiler also takes
407 substantially less time to generate code that is just as good (and often
408 better) than before.</li>
410 <li>Each target has been split to separate the ".s" file printing logic from the
411 rest of the target. This enables JIT compilers that don't link in the
412 (somewhat large) code and data tables used for printing a ".s" file.</li>
414 <li>The code generator now includes a "stack slot coloring" pass, which packs
415 together individual spilled values into common stack slots. This reduces
416 the size of stack frames with many spills, which tends to increase L1 cache
419 <li>Various pieces of the register allocator (e.g. the coalescer and two-address
420 operation elimination pass) now know how to rematerialize trivial operations
421 to avoid copies and include several other optimizations.</li>
423 <li>The <a href="CodeGenerator.html#selectiondag_process">graphs</a> produced by
424 the <tt>llc -view-*-dags</tt> options are now significantly prettier and
427 <li>LLVM 2.4 includes a new register allocator based on Partitioned Boolean
428 Quadratic Programming (PBQP). This register allocator is still in
429 development, but is very simple and clean.</li>
436 <!--=========================================================================-->
437 <div class="doc_subsection">
438 <a name="targetspecific">Target Specific Improvements</a>
441 <div class="doc_text">
442 <p>New target-specific features include:
446 <li>Exception handling is supported by default on Linux/x86-64.</li>
447 <li>Position Independent Code (PIC) is now supported on Linux/x86-64.</li>
448 <li>@llvm.frameaddress now supports getting the frame address of stack frames
449 > 0 on x86/x86-64.</li>
450 <li>MIPS floating point support? [BRUNO]</li>
451 <li>The PowerPC backend now supports trampolines.</li>
457 <!--=========================================================================-->
458 <div class="doc_subsection">
459 <a name="otherimprovements">Other Improvements</a>
462 <div class="doc_text">
463 <p>New features include:
467 <li><tt>llvmc2</tt> (the generic compiler driver) gained plugin
468 support. It is now easier to experiment with <tt>llvmc2</tt> and
469 build your own tools based on it.</li>
471 <li>LLVM 2.4 includes a number of new generic algorithms and data structures,
472 include a scoped hash table, 'immutable' data structures, a simple
473 free-list manager, and a <tt>raw_ostream</tt> class.
474 The <tt>raw_ostream</tt> class and
475 <tt>format</tt> allow for efficient file output, and various pieces of LLVM
476 have switched over to use it. The eventual goal is to eliminate
477 std::ostream in favor of it.</li>
483 <!--=========================================================================-->
484 <div class="doc_subsection">
485 <a name="changes">Major Changes and Removed Features</a>
488 <div class="doc_text">
490 <p>If you're already an LLVM user or developer with out-of-tree changes based
491 on LLVM 2.3, this section lists some "gotchas" that you may run into upgrading
492 from the previous release.</p>
496 <li>The LLVM IR generated by llvm-gcc no longer names all instructions. This
497 makes it run faster, but may be more confusing to some people. If you
498 prefer to have names, the '<tt>opt -instnamer</tt>' pass will add names to
499 all instructions.</li>
501 <li>The LoadVN and GCSE passes have been removed from the tree. They are
502 obsolete and have been replaced with the GVN and MemoryDependence passes.
507 <p>In addition, many APIs have changed in this release. Some of the major LLVM
512 <li>Now, function attributes and return value attributes are managed
513 separately. Interface exported by <tt>ParameterAttributes.h</tt> header is now
514 experted by <tt>Attributes.h</tt> header. The new attributes interface changes are:
516 <li><tt>getParamAttrs</tt> method is now replaced by
517 <tt>getParamAttributes</tt>, <tt>getRetAttributes</tt> and
518 <tt>getFnAttributes</tt> methods.</li>
519 <li> Return value attributes are stored at index 0. Function attributes are
520 stored at index ~0U. Parameter attributes are stored at index that matches
521 parameter number.</li>
522 <li> <tt>ParamAttr</tt> namespace is now renamed as <tt>Attribute</tt>.</li>
523 <li> The name of the class that manages reference count of opaque
524 attributes is changed from <tt>PAListPtr</tt> to <tt>AttrListPtr</tt>.</li>
525 <li> <tt>ParamAttrsWithIndex</tt> is now renamed as <tt>AttributeWithIndex</tt>.
530 <li>The <tt>DbgStopPointInst</tt> methods <tt>getDirectory</tt> and
531 <tt>getFileName</tt> now return <tt>Value*</tt> instead of strings. These can be
532 converted to strings using <tt>llvm::GetConstantStringInfo</tt> defined via
533 "<tt>llvm/Analysis/ValueTracking.h</tt>".</li>
535 <li>The APIs to create various instructions have changed from lower case
536 "create" methods to upper case "Create" methods (e.g.
537 <tt>BinaryOperator::create</tt>). LLVM 2.4 includes both cases, but the
538 lower case ones are removed in mainline, please migrate.</li>
540 <li>Various header files like "<tt>llvm/ADT/iterator</tt>" were given a ".h" suffix.
541 Change your code to #include "<tt>llvm/ADT/iterator.h</tt>" instead.</li>
543 <li>In the code generator, many <tt>MachineOperand</tt> predicates were renamed to be
544 shorter (e.g. <tt>isFrameIndex()</tt> -> <tt>isFI()</tt>),
545 <tt>SDOperand</tt> was renamed to <tt>SDValue</tt> (and the "<tt>Val</tt>"
546 member was changed to be the <tt>getNode()</tt> accessor), and the
547 <tt>MVT::ValueType</tt> enum has been replaced with an "<tt>MVT</tt>"
548 struct. The <tt>getSignExtended</tt> and <tt>getValue</tt> methods in the
549 ConstantSDNode class were renamed to <tt>getSExtValue</tt> and
550 <tt>getZExtValue</tt> respectively, to be more consistent with
551 the <tt>ConstantInt</tt> class.</li>
558 <!-- *********************************************************************** -->
559 <div class="doc_section">
560 <a name="portability">Portability and Supported Platforms</a>
562 <!-- *********************************************************************** -->
564 <div class="doc_text">
566 <p>LLVM is known to work on the following platforms:</p>
569 <li>Intel and AMD machines (IA32) running Red Hat Linux, Fedora Core and FreeBSD
570 (and probably other unix-like systems).</li>
571 <li>PowerPC and X86-based Mac OS X systems, running 10.3 and above in 32-bit and
573 <li>Intel and AMD machines running on Win32 using MinGW libraries (native).</li>
574 <li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
575 support is available for native builds with Visual C++).</li>
576 <li>Sun UltraSPARC workstations running Solaris 10.</li>
577 <li>Alpha-based machines running Debian GNU/Linux.</li>
578 <li>Itanium-based (IA64) machines running Linux and HP-UX.</li>
581 <p>The core LLVM infrastructure uses GNU autoconf to adapt itself
582 to the machine and operating system on which it is built. However, minor
583 porting may be required to get LLVM to work on new platforms. We welcome your
584 portability patches and reports of successful builds or error messages.</p>
588 <!-- *********************************************************************** -->
589 <div class="doc_section">
590 <a name="knownproblems">Known Problems</a>
592 <!-- *********************************************************************** -->
594 <div class="doc_text">
596 <p>This section contains all known problems with the LLVM system, listed by
597 component. As new problems are discovered, they will be added to these
598 sections. If you run into a problem, please check the <a
599 href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
600 there isn't already one.</p>
604 <!-- ======================================================================= -->
605 <div class="doc_subsection">
606 <a name="experimental">Experimental features included with this release</a>
609 <div class="doc_text">
611 <p>The following components of this LLVM release are either untested, known to
612 be broken or unreliable, or are in early development. These components should
613 not be relied on, and bugs should not be filed against them, but they may be
614 useful to some people. In particular, if you would like to work on one of these
615 components, please contact us on the <a
616 href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
619 <li>The MSIL, IA64, Alpha, SPU, MIPS, and PIC16 backends are experimental.</li>
620 <li>The llc "<tt>-filetype=asm</tt>" (the default) is the only supported
621 value for this option.</li>
626 <!-- ======================================================================= -->
627 <div class="doc_subsection">
628 <a name="x86-be">Known problems with the X86 back-end</a>
631 <div class="doc_text">
634 <li>The X86 backend does not yet support
635 all <a href="http://llvm.org/PR879">inline assembly that uses the X86
636 floating point stack</a>. It supports the 'f' and 't' constraints, but not
638 <li>The X86 backend generates inefficient floating point code when configured
639 to generate code for systems that don't have SSE2.</li>
640 <li>Win64 code generation wasn't widely tested. Everything should work, but we
641 expect small issues to happen. Also, llvm-gcc cannot build mingw64 runtime
643 to <a href="http://llvm.org/PR2255">several</a>
644 <a href="http://llvm.org/PR2257">bugs</a> due to lack of support for the
645 'u' inline assembly constraint and X87 floating point inline assembly.</li>
646 <li>The X86-64 backend does not yet support the LLVM IR instruction
647 <tt>va_arg</tt>. Currently, the llvm-gcc front-end supports variadic
648 argument constructs on X86-64 by lowering them manually.</li>
653 <!-- ======================================================================= -->
654 <div class="doc_subsection">
655 <a name="ppc-be">Known problems with the PowerPC back-end</a>
658 <div class="doc_text">
661 <li>The Linux PPC32/ABI support needs testing for the interpreter and static
662 compilation, and lacks support for debug information.</li>
667 <!-- ======================================================================= -->
668 <div class="doc_subsection">
669 <a name="arm-be">Known problems with the ARM back-end</a>
672 <div class="doc_text">
675 <li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
676 processors, thumb programs can crash or produce wrong
677 results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
678 <li>Compilation for ARM Linux OABI (old ABI) is supported, but not fully tested.
680 <li>There is a bug in QEMU-ARM (<= 0.9.0) which causes it to incorrectly
682 programs compiled with LLVM. Please use more recent versions of QEMU.</li>
687 <!-- ======================================================================= -->
688 <div class="doc_subsection">
689 <a name="sparc-be">Known problems with the SPARC back-end</a>
692 <div class="doc_text">
695 <li>The SPARC backend only supports the 32-bit SPARC ABI (-m32), it does not
696 support the 64-bit SPARC ABI (-m64).</li>
701 <!-- ======================================================================= -->
702 <div class="doc_subsection">
703 <a name="alpha-be">Known problems with the Alpha back-end</a>
706 <div class="doc_text">
710 <li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
711 appropriate nops inserted to ensure restartability.</li>
716 <!-- ======================================================================= -->
717 <div class="doc_subsection">
718 <a name="ia64-be">Known problems with the IA64 back-end</a>
721 <div class="doc_text">
724 <li>The Itanium backend is highly experimental, and has a number of known
725 issues. We are looking for a maintainer for the Itanium backend. If you
726 are interested, please contact the LLVMdev mailing list.</li>
731 <!-- ======================================================================= -->
732 <div class="doc_subsection">
733 <a name="c-be">Known problems with the C back-end</a>
736 <div class="doc_text">
739 <li><a href="http://llvm.org/PR802">The C backend has only basic support for
740 inline assembly code</a>.</li>
741 <li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
742 C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
743 C++ code compiled with llc or native compilers.</li>
744 <li>The C backend does not support all exception handling constructs.</li>
750 <!-- ======================================================================= -->
751 <div class="doc_subsection">
752 <a name="c-fe">Known problems with the llvm-gcc C front-end</a>
755 <div class="doc_text">
757 <p>llvm-gcc does not currently support <a href="http://llvm.org/PR869">Link-Time
758 Optimization</a> on most platforms "out-of-the-box". Please inquire on the
759 LLVMdev mailing list if you are interested.</p>
761 <p>The only major language feature of GCC not supported by llvm-gcc is
762 the <tt>__builtin_apply</tt> family of builtins. However, some extensions
763 are only supported on some targets. For example, trampolines are only
764 supported on some targets (these are used when you take the address of a
765 nested function).</p>
767 <p>If you run into GCC extensions which are not supported, please let us know.
772 <!-- ======================================================================= -->
773 <div class="doc_subsection">
774 <a name="c++-fe">Known problems with the llvm-gcc C++ front-end</a>
777 <div class="doc_text">
779 <p>The C++ front-end is considered to be fully
780 tested and works for a number of non-trivial programs, including LLVM
781 itself, Qt, Mozilla, etc.</p>
784 <li>Exception handling works well on the X86 and PowerPC targets. Currently
785 only linux and darwin targets are supported (both 32 and 64 bit).</li>
791 <!-- ======================================================================= -->
792 <div class="doc_subsection">
793 <a name="ada-fe">Known problems with the llvm-gcc Ada front-end</a>
796 <div class="doc_text">
797 The llvm-gcc 4.2 Ada compiler works fairly well, however this is not a mature
798 technology and problems should be expected.
800 <li>The Ada front-end currently only builds on X86-32. This is mainly due
801 to lack of trampoline support (pointers to nested functions) on other platforms,
802 however it <a href="http://llvm.org/PR2006">also fails to build on X86-64</a>
803 which does support trampolines.</li>
804 <li>The Ada front-end <a href="http://llvm.org/PR2007">fails to bootstrap</a>.
805 Workaround: configure with --disable-bootstrap.</li>
806 <li>The c380004, <a href="http://llvm.org/PR2010">c393010</a>
807 and <a href="http://llvm.org/PR2421">cxg2021</a> ACATS tests fail
808 (c380004 also fails with gcc-4.2 mainline).</li>
809 <li>Some gcc specific Ada tests continue to crash the compiler.</li>
810 <li>The -E binder option (exception backtraces)
811 <a href="http://llvm.org/PR1982">does not work</a> and will result in programs
812 crashing if an exception is raised. Workaround: do not use -E.</li>
813 <li>Only discrete types <a href="http://llvm.org/PR1981">are allowed to start
814 or finish at a non-byte offset</a> in a record. Workaround: do not pack records
815 or use representation clauses that result in a field of a non-discrete type
816 starting or finishing in the middle of a byte.</li>
817 <li>The <tt>lli</tt> interpreter <a href="http://llvm.org/PR2009">considers
818 'main' as generated by the Ada binder to be invalid</a>.
819 Workaround: hand edit the file to use pointers for <tt>argv</tt> and
820 <tt>envp</tt> rather than integers.</li>
821 <li>The <tt>-fstack-check</tt> option <a href="http://llvm.org/PR2008">is
826 <!-- *********************************************************************** -->
827 <div class="doc_section">
828 <a name="additionalinfo">Additional Information</a>
830 <!-- *********************************************************************** -->
832 <div class="doc_text">
834 <p>A wide variety of additional information is available on the <a
835 href="http://llvm.org">LLVM web page</a>, in particular in the <a
836 href="http://llvm.org/docs/">documentation</a> section. The web page also
837 contains versions of the API documentation which is up-to-date with the
838 Subversion version of the source code.
839 You can access versions of these documents specific to this release by going
840 into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
842 <p>If you have any questions or comments about LLVM, please feel free to contact
843 us via the <a href="http://llvm.org/docs/#maillist"> mailing
848 <!-- *********************************************************************** -->
852 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
853 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
854 <a href="http://validator.w3.org/check/referer"><img
855 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
857 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
858 Last modified: $Date$