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 2.1 Release Notes</title>
11 <div class="doc_title">LLVM 2.1 Release Notes</div>
14 <li><a href="#intro">Introduction</a></li>
15 <li><a href="#whatsnew">What's New?</a></li>
16 <li><a href="GettingStarted.html">Installation Instructions</a></li>
17 <li><a href="#portability">Portability and Supported Platforms</a></li>
18 <li><a href="#knownproblems">Known Problems</a>
19 <li><a href="#additionalinfo">Additional Information</a></li>
22 <div class="doc_author">
23 <p>Written by the <a href="http://llvm.org">LLVM Team</a><p>
26 <!-- *********************************************************************** -->
27 <div class="doc_section">
28 <a name="intro">Introduction</a>
30 <!-- *********************************************************************** -->
32 <div class="doc_text">
34 <p>This document contains the release notes for the LLVM compiler
35 infrastructure, release 2.1. Here we describe the status of LLVM, including
36 major improvements from the previous release and any known problems. All LLVM
37 releases may be downloaded from the <a href="http://llvm.org/releases/">LLVM
38 releases web site</a>.</p>
40 <p>For more information about LLVM, including information about the latest
41 release, please check out the <a href="http://llvm.org/">main LLVM
42 web site</a>. If you have questions or comments, the <a
43 href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM developer's mailing
44 list</a> is a good place to send them.</p>
46 <p>Note that if you are reading this file from a Subversion checkout or the
47 main LLVM web page, this document applies to the <i>next</i> release, not the
48 current one. To see the release notes for a specific releases, please see the
49 <a href="http://llvm.org/releases/">releases page</a>.</p>
53 <!-- *********************************************************************** -->
54 <div class="doc_section">
55 <a name="whatsnew">What's New?</a>
57 <!-- *********************************************************************** -->
59 <div class="doc_text">
61 <p>This is the twelfth public release of the LLVM Compiler Infrastructure.
62 It includes many features and refinements from LLVM 2.0.</p>
66 <!--=========================================================================-->
67 <div class="doc_subsection">
68 <a name="frontends">New Frontends</a>
71 <div class="doc_text">
73 <p>LLVM 2.1 brings two new beta C front-ends. First, a new version of llvm-gcc
74 based on GCC 4.2, innovatively called "llvm-gcc-4.2". This promises to bring
75 FORTRAN and Ada support to LLVM as well as features like atomic builtins and
76 OpenMP. None of these actually work yet, but don't let that stop you checking
79 <p>Second, LLVM now includes its own native C and Objective-C front-end (C++ is
80 in progress, but is not very far along) code named "<a
81 href="http://clang.llvm.org/">clang</a>". This front-end has a number of great
82 features, primarily aimed at source-level analysis and speeding up compile-time.
83 At this point though, the LLVM Code Generator component is still very early in
84 development, so it's mostly useful for people looking to build source-level
85 analysis tools or source-to-source translators.</p>
89 <!--=========================================================================-->
90 <div class="doc_subsection">
91 <a name="optimizer">Optimizer Improvements</a>
94 <div class="doc_text">
96 <p>Some of the most noticable feature improvements this release have been in the
97 optimizer, speeding it up and making it more aggressive. For example:</p>
101 <li>Owen Anderson wrote the new MemoryDependenceAnalysis pass, which provides
102 a lazy, caching layer on top of <a
103 href="AliasAnalysis.html">AliasAnalysis</a>. He then used it to rewrite
104 DeadStoreElimination which resulted in significantly better compile time in
106 <li>Owen implemented the new GVN pass, which is also based on
107 MemoryDependenceAnalysis. This pass replaces GCSE/LoadVN in the standard
108 set of passes, providing more aggressive optimization at a some-what
109 improved compile-time cost.</li>
110 <li>Owen implemented GVN-PRE, a partial redundancy elimination algorithm that
111 shares some details with the new GVN pass. It is still in need of compile
112 time tuning, and is not turned on by default.</li>
113 <li>Devang merged ETForest and DomTree into a single easier to use data
114 structure. This makes it more obvious which datastructure to choose
115 (because there is only one) and makes the compiler more memory and time
116 efficient (less stuff to keep up-to-date).</li>
117 <li>Nick Lewycky improved loop trip count analysis to handle many more common
124 <!--=========================================================================-->
125 <div class="doc_subsection">
126 <a name="codegen">Code Generator Improvements</a>
129 <div class="doc_text">
131 <p>One of the main focuses of this release was performance tuning and bug
132 fixing. In addition to these, several new major changes occurred:</p>
136 <li>Dale finished up the Tail Merging optimization in the code generator, and
137 enabled it by default. This produces smaller code that is also faster in
140 <li>Christopher Lamb implemented support for virtual register sub-registers,
141 which can be used to better model many forms of subregisters. As an example
142 use, he modified the X86 backend to use this to model truncates and
143 extends more accurately (leading to better code).</li>
145 <li>Dan Gohman changed the way we represent vectors before legalization,
146 significantly simplifying the SelectionDAG representation for these and
147 making the code generator faster for vector code.</li>
149 <li>Evan contributed a new target independent if-converter. While it is
150 target independent, so far only the ARM backend uses it.</li>
152 <li>Evan rewrote the way the register allocator handles rematerialization,
153 allowing it to be much more effective on two-address targets like X86,
154 and taught it to fold loads away when possible (also a big win on X86).</li>
156 <li>Dan Gohman contributed support for better alignment and volatility handling
157 in the code generator, and significantly enhanced alignment analysis for SSE
158 load/store instructions. With his changes, an insufficiently-aligned SSE
159 load instruction turns into <tt>movups</tt>, for example.</li>
161 <li>Duraid Madina contributed a new "bigblock" register allocator, and Roman
162 Levenstein contributed several big improvements. BigBlock is optimized for
163 code that uses very large basic blocks. It is slightly slower than the
164 "local" allocator, but produces much better code.</li>
166 <li>David Greene refactored the register allocator to split coalescing out from
167 allocation, making coalescers pluggable.</li>
174 <!--=========================================================================-->
175 <div class="doc_subsection">
176 <a name="targetspecific">Target Specific Improvements</a>
179 <div class="doc_text">
180 <p>New features include:
184 <li>Bruno Cardoso Lopes contributed initial MIPS support. It is sufficient to
185 run many small programs, but is still incomplete and is not yet
186 fully performant.</li>
188 <li>Bill Wendling added SSSE3 support to the X86 backend.</li>
190 <li>Nicholas Geoffray contributed improved linux/ppc ABI and JIT support.</li>
192 <li>Dale Johannesen rewrote handling of 32-bit float values in the X86 backend
193 when using the floating point stack, fixing several nasty bugs.</li>
195 <li>Dan contributed rematerialization support for the X86 backend, in addition
196 to several X86-specific micro optimizations.</li>
202 <!--=========================================================================-->
203 <div class="doc_subsection">
204 <a name="llvmgccimprovements">llvm-gcc Improvements</a>
207 <div class="doc_text">
208 <p>New features include:
212 <li>Duncan and Anton made significant progress chasing down a number of problems
213 with C++ Zero-Cost exception handling in llvm-gcc 4.0 and 4.2. It is now at
214 the point where it "just works" on linux/X86-32 and has partial support on
217 <li>Devang and Duncan fixed a huge number of bugs relating to bitfields, pragma
218 pack, and variable sized fields in structures.</li>
220 <li>Tanya implemented support for <tt>__attribute__((noinline))</tt> in
221 llvm-gcc, and added support for generic variable annotations which are
222 propagated into the LLVM IR, e.g.
223 "<tt>int X __attribute__((annotate("myproperty")));</tt>".</li>
225 <li>Sheng Zhou and Christopher Lamb implemented alias analysis support for
226 "restrict" pointer arguments to functions.</li>
228 <li>Duncan contributed support for trampolines (taking the address of a nested
229 function). Currently this is only supported on the X86-32 target.</li>
231 <li>Lauro Ramos Venancio contributed support to encode alignment info in
232 load and store instructions, the foundation for other alignment-related
239 <!--=========================================================================-->
240 <div class="doc_subsection">
241 <a name="coreimprovements">LLVM Core Improvements</a>
244 <div class="doc_text">
245 <p>New features include:
249 <li>Neil Booth contributed a new "APFloat" class, which ensures that floating
250 point representation and constant folding is not dependent on the host
251 architecture that builds the application. This support is the foundation
252 for "long double" support that will be wrapped up in LLVM 2.2.</li>
254 <li>Based on the APFloat class, Dale redesigned the internals of the ConstantFP
255 class and has been working on extending the core and optimizer components to
256 support various target-specific 'long double's. We expect this work to be
257 completed in LLVM 2.2.</li>
259 <li>LLVM now provides an LLVMBuilder class, which makes it significantly easier
260 to create LLVM IR instructions.</li>
262 <li>Reid contributed support for intrinsics that take arbitrary integer typed
263 arguments. Dan Gohman and Chandler extended it to support arbitrary
264 floating point arguments and vectors.</li>
269 <!--=========================================================================-->
270 <div class="doc_subsection">
271 <a name="otherimprovements">Other Improvements</a>
274 <div class="doc_text">
275 <p>New features include:
279 <li>Sterling Stein contributed a new BrainF frontend, located in llvm/examples.
280 This shows a some of the more modern APIs for building a front-end, and
281 demonstrates JIT compiler support.</li>
283 <li>David Green contributed a new <tt>--enable-expensive-checks</tt> configure
284 option which enables STL checking, and fixed several bugs exposed by
290 <!-- *********************************************************************** -->
291 <div class="doc_section">
292 <a name="portability">Portability and Supported Platforms</a>
294 <!-- *********************************************************************** -->
296 <div class="doc_text">
298 <p>LLVM is known to work on the following platforms:</p>
301 <li>Intel and AMD machines running Red Hat Linux, Fedora Core and FreeBSD
302 (and probably other unix-like systems).</li>
303 <li>PowerPC and X86-based Mac OS X systems, running 10.2 and above in 32-bit and
305 <li>Intel and AMD machines running on Win32 using MinGW libraries (native)</li>
306 <li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
307 support is available for native builds with Visual C++).</li>
308 <li>Sun UltraSPARC workstations running Solaris 8.</li>
309 <li>Alpha-based machines running Debian GNU/Linux.</li>
310 <li>Itanium-based machines running Linux and HP-UX.</li>
313 <p>The core LLVM infrastructure uses
314 <a href="http://www.gnu.org/software/autoconf/">GNU autoconf</a> to adapt itself
315 to the machine and operating system on which it is built. However, minor
316 porting may be required to get LLVM to work on new platforms. We welcome your
317 portability patches and reports of successful builds or error messages.</p>
321 <!-- *********************************************************************** -->
322 <div class="doc_section">
323 <a name="knownproblems">Known Problems</a>
325 <!-- *********************************************************************** -->
327 <div class="doc_text">
329 <p>This section contains all known problems with the LLVM system, listed by
330 component. As new problems are discovered, they will be added to these
331 sections. If you run into a problem, please check the <a
332 href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
333 there isn't already one.</p>
337 <!-- ======================================================================= -->
338 <div class="doc_subsection">
339 <a name="experimental">Experimental features included with this release</a>
342 <div class="doc_text">
344 <p>The following components of this LLVM release are either untested, known to
345 be broken or unreliable, or are in early development. These components should
346 not be relied on, and bugs should not be filed against them, but they may be
347 useful to some people. In particular, if you would like to work on one of these
348 components, please contact us on the <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
351 <li>The <tt>-cee</tt> pass is known to be buggy, and may be removed in a
353 <li>The MSIL backend is experimental.</li>
354 <li>The IA64 code generator is experimental.</li>
355 <li>The Alpha backend is experimental.</li>
356 <li>"<tt>-filetype=asm</tt>" (the default) is the only supported value for the
357 <tt>-filetype</tt> llc option.</li>
362 <!-- ======================================================================= -->
363 <div class="doc_subsection">
364 <a name="x86-be">Known problems with the X86 back-end</a>
367 <div class="doc_text">
370 <li>The X86 backend does not yet support <a href="http://llvm.org/PR879">inline
371 assembly that uses the X86 floating point stack</a>.</li>
372 <li>The X86 backend occasionally has <a href="http://llvm.org/PR1649">alignment
373 problems</a> on operating systems that don't require 16-byte stack alignment
374 (including most non-darwin OS's like linux).</li>
379 <!-- ======================================================================= -->
380 <div class="doc_subsection">
381 <a name="ppc-be">Known problems with the PowerPC back-end</a>
384 <div class="doc_text">
387 <li><a href="http://llvm.org/PR642">PowerPC backend does not correctly
388 implement ordered FP comparisons</a>.</li>
389 <li>The Linux PPC32/ABI support needs testing for the interpreter and static
390 compilation, and lacks support for debug information.</li>
395 <!-- ======================================================================= -->
396 <div class="doc_subsection">
397 <a name="arm-be">Known problems with the ARM back-end</a>
400 <div class="doc_text">
403 <li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
404 processors, thumb programs can crash or produce wrong
405 results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
406 <li>Compilation for ARM Linux OABI (old ABI) is supported, but not fully tested.
408 <li>There is a bug in QEMU-ARM (<= 0.9.0) which causes it to incorrectly execute
409 programs compiled with LLVM. Please use more recent versions of QEMU.</li>
414 <!-- ======================================================================= -->
415 <div class="doc_subsection">
416 <a name="sparc-be">Known problems with the SPARC back-end</a>
419 <div class="doc_text">
422 <li>The SPARC backend only supports the 32-bit SPARC ABI (-m32), it does not
423 support the 64-bit SPARC ABI (-m64).</li>
428 <!-- ======================================================================= -->
429 <div class="doc_subsection">
430 <a name="alpha-be">Known problems with the Alpha back-end</a>
433 <div class="doc_text">
437 <li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
438 appropriate nops inserted to ensure restartability.</li>
443 <!-- ======================================================================= -->
444 <div class="doc_subsection">
445 <a name="ia64-be">Known problems with the IA64 back-end</a>
448 <div class="doc_text">
452 <li>C++ programs are likely to fail on IA64, as calls to <tt>setjmp</tt> are
453 made where the argument is not 16-byte aligned, as required on IA64. (Strictly
454 speaking this is not a bug in the IA64 back-end; it will also be encountered
455 when building C++ programs using the C back-end.)</li>
457 <li>The C++ front-end does not use <a href="http://llvm.org/PR406">IA64
458 ABI compliant layout of v-tables</a>. In particular, it just stores function
459 pointers instead of function descriptors in the vtable. This bug prevents
460 mixing C++ code compiled with LLVM with C++ objects compiled by other C++
463 <li>There are a few ABI violations which will lead to problems when mixing LLVM
464 output with code built with other compilers, particularly for floating-point
467 <li>Defining vararg functions is not supported (but calling them is ok).</li>
469 <li>The Itanium backend has bitrotted somewhat.</li>
474 <!-- ======================================================================= -->
475 <div class="doc_subsection">
476 <a name="c-be">Known problems with the C back-end</a>
479 <div class="doc_text">
482 <li><a href="http://llvm.org/PR802">The C backend does not support inline
483 assembly code</a>.</li>
484 <li><a href="http://llvm.org/PR1126">The C backend does not support vectors
486 <li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
487 C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
488 C++ code compiled with LLC or native compilers.</li>
494 <!-- ======================================================================= -->
495 <div class="doc_subsection">
496 <a name="c-fe">Known problems with the C front-end</a>
499 <!-- _______________________________________________________________________ -->
500 <div class="doc_subsubsection">Bugs</div>
502 <div class="doc_text">
504 <p>llvm-gcc4 does not currently support <a href="http://llvm.org/PR869">Link-Time
505 Optimization</a> on most platforms "out-of-the-box". Please inquire on the
506 llvmdev mailing list if you are interested.</p>
510 <!-- _______________________________________________________________________ -->
511 <div class="doc_subsubsection">
515 <div class="doc_text">
518 <li><p>"long double" is silently transformed by the front-end into "double". There
519 is no support for floating point data types of any size other than 32 and 64
522 <li><p>llvm-gcc does <b>not</b> support <tt>__builtin_apply</tt> yet.
523 See <a href="http://gcc.gnu.org/onlinedocs/gcc/Constructing-Calls.html#Constructing%20Calls">Constructing Calls</a>: Dispatching a call to another function.</p>
526 <li><p>llvm-gcc <b>partially</b> supports these GCC extensions:</p>
528 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Nested-Functions.html#Nested%20Functions">Nested Functions</a>:
530 As in Algol and Pascal, lexical scoping of functions.
531 Nested functions are supported, but llvm-gcc does not support
532 taking the address of a nested function (except on the X86-32 target)
533 or non-local gotos.</li>
535 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Function-Attributes.html#Function%20Attributes">Function Attributes</a>:
537 Declaring that functions have no side effects or that they can never
540 <b>Supported:</b> <tt>alias</tt>, <tt>always_inline</tt>, <tt>cdecl</tt>,
541 <tt>constructor</tt>, <tt>destructor</tt>,
542 <tt>deprecated</tt>, <tt>fastcall</tt>, <tt>format</tt>,
543 <tt>format_arg</tt>, <tt>non_null</tt>, <tt>noinline</tt>, <tt>noreturn</tt>, <tt>regparm</tt>
544 <tt>section</tt>, <tt>stdcall</tt>, <tt>unused</tt>, <tt>used</tt>,
545 <tt>visibility</tt>, <tt>warn_unused_result</tt>, <tt>weak</tt><br>
547 <b>Ignored:</b> <tt>pure</tt>, <tt>const</tt>, <tt>nothrow</tt>,
548 <tt>malloc</tt>, <tt>no_instrument_function</tt></li>
552 <li><p>llvm-gcc supports the vast majority of GCC extensions, including:</p>
555 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Pragmas.html#Pragmas">Pragmas</a>: Pragmas accepted by GCC.</li>
556 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Local-Labels.html#Local%20Labels">Local Labels</a>: Labels local to a block.</li>
557 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Other-Builtins.html#Other%20Builtins">Other Builtins</a>:
558 Other built-in functions.</li>
559 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Variable-Attributes.html#Variable%20Attributes">Variable Attributes</a>:
560 Specifying attributes of variables.</li>
561 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Type-Attributes.html#Type%20Attributes">Type Attributes</a>: Specifying attributes of types.</li>
562 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Thread_002dLocal.html">Thread-Local</a>: Per-thread variables.</li>
563 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Variable-Length.html#Variable%20Length">Variable Length</a>:
564 Arrays whose length is computed at run time.</li>
565 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Labels-as-Values.html#Labels%20as%20Values">Labels as Values</a>: Getting pointers to labels and computed gotos.</li>
566 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Statement-Exprs.html#Statement%20Exprs">Statement Exprs</a>: Putting statements and declarations inside expressions.</li>
567 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Typeof.html#Typeof">Typeof</a>: <code>typeof</code>: referring to the type of an expression.</li>
568 <li><a href="http://gcc.gnu.org/onlinedocs/gcc-3.4.0/gcc/Lvalues.html#Lvalues">Lvalues</a>: Using <code>?:</code>, "<code>,</code>" and casts in lvalues.</li>
569 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Conditionals.html#Conditionals">Conditionals</a>: Omitting the middle operand of a <code>?:</code> expression.</li>
570 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Long-Long.html#Long%20Long">Long Long</a>: Double-word integers.</li>
571 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Complex.html#Complex">Complex</a>: Data types for complex numbers.</li>
572 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Hex-Floats.html#Hex%20Floats">Hex Floats</a>:Hexadecimal floating-point constants.</li>
573 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html#Zero%20Length">Zero Length</a>: Zero-length arrays.</li>
574 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Empty-Structures.html#Empty%20Structures">Empty Structures</a>: Structures with no members.</li>
575 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Variadic-Macros.html#Variadic%20Macros">Variadic Macros</a>: Macros with a variable number of arguments.</li>
576 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Escaped-Newlines.html#Escaped%20Newlines">Escaped Newlines</a>: Slightly looser rules for escaped newlines.</li>
577 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html#Extended%20Asm">Extended Asm</a>: Assembler instructions with C expressions as operands.</li>
578 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Constraints.html#Constraints">Constraints</a>: Constraints for asm operands.</li>
579 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Asm-Labels.html#Asm%20Labels">Asm Labels</a>: Specifying the assembler name to use for a C symbol.</li>
580 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Explicit-Reg-Vars.html#Explicit%20Reg%20Vars">Explicit Reg Vars</a>: Defining variables residing in specified registers.</li>
581 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Vector-Extensions.html#Vector%20Extensions">Vector Extensions</a>: Using vector instructions through built-in functions.</li>
582 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Target-Builtins.html#Target%20Builtins">Target Builtins</a>: Built-in functions specific to particular targets.</li>
583 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Subscripting.html#Subscripting">Subscripting</a>: Any array can be subscripted, even if not an lvalue.</li>
584 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Pointer-Arith.html#Pointer%20Arith">Pointer Arith</a>: Arithmetic on <code>void</code>-pointers and function pointers.</li>
585 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Initializers.html#Initializers">Initializers</a>: Non-constant initializers.</li>
586 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Compound-Literals.html#Compound%20Literals">Compound Literals</a>: Compound literals give structures, unions,
587 or arrays as values.</li>
588 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Designated-Inits.html#Designated%20Inits">Designated Inits</a>: Labeling elements of initializers.</li>
589 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Cast-to-Union.html#Cast%20to%20Union">Cast to Union</a>: Casting to union type from any member of the union.</li>
590 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html#Case%20Ranges">Case Ranges</a>: `case 1 ... 9' and such.</li>
591 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Mixed-Declarations.html#Mixed%20Declarations">Mixed Declarations</a>: Mixing declarations and code.</li>
592 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Function-Prototypes.html#Function%20Prototypes">Function Prototypes</a>: Prototype declarations and old-style definitions.</li>
593 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/C_002b_002b-Comments.html#C_002b_002b-Comments">C++ Comments</a>: C++ comments are recognized.</li>
594 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Dollar-Signs.html#Dollar%20Signs">Dollar Signs</a>: Dollar sign is allowed in identifiers.</li>
595 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Character-Escapes.html#Character%20Escapes">Character Escapes</a>: <code>\e</code> stands for the character <ESC>.</li>
596 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Alignment.html#Alignment">Alignment</a>: Inquiring about the alignment of a type or variable.</li>
597 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Inline.html#Inline">Inline</a>: Defining inline functions (as fast as macros).</li>
598 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Alternate-Keywords.html#Alternate%20Keywords">Alternate Keywords</a>:<code>__const__</code>, <code>__asm__</code>, etc., for header files.</li>
599 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Incomplete-Enums.html#Incomplete%20Enums">Incomplete Enums</a>: <code>enum foo;</code>, with details to follow.</li>
600 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Function-Names.html#Function%20Names">Function Names</a>: Printable strings which are the name of the current function.</li>
601 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Return-Address.html#Return%20Address">Return Address</a>: Getting the return or frame address of a function.</li>
602 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Unnamed-Fields.html#Unnamed%20Fields">Unnamed Fields</a>: Unnamed struct/union fields within structs/unions.</li>
603 <li><a href="http://gcc.gnu.org/onlinedocs/gcc/Attribute-Syntax.html#Attribute%20Syntax">Attribute Syntax</a>: Formal syntax for attributes.</li>
608 <p>If you run into GCC extensions which have not been included in any of these
609 lists, please let us know (also including whether or not they work).</p>
613 <!-- ======================================================================= -->
614 <div class="doc_subsection">
615 <a name="c++-fe">Known problems with the C++ front-end</a>
618 <div class="doc_text">
620 <p>The C++ front-end is considered to be fully
621 tested and works for a number of non-trivial programs, including LLVM
622 itself, Qt, Mozilla, etc.</p>
625 <li>Exception handling only works well on the linux/X86-32 target.
626 In some cases, illegally throwing an exception does not result
627 in a call to terminate.</li>
631 <li>Destructors for local objects are not always run when a <tt>longjmp</tt> is
632 performed. In particular, destructors for objects in the <tt>longjmp</tt>ing
633 function and in the <tt>setjmp</tt> receiver function may not be run.
634 Objects in intervening stack frames will be destroyed, however (which is
635 better than most compilers).</li>
637 <li>The LLVM C++ front-end follows the <a
638 href="http://www.codesourcery.com/cxx-abi">Itanium C++ ABI</a>.
639 This document, which is not Itanium specific, specifies a standard for name
640 mangling, class layout, v-table layout, RTTI formats, and other C++
641 representation issues. Because we use this API, code generated by the LLVM
642 compilers should be binary compatible with machine code generated by other
643 Itanium ABI C++ compilers (such as G++, the Intel and HP compilers, etc).
644 <i>However</i>, the exception handling mechanism used by llvm-gcc3 is very
645 different from the model used in the Itanium ABI, so <b>exceptions will not
646 interact correctly</b>. </li>
654 <!-- *********************************************************************** -->
655 <div class="doc_section">
656 <a name="additionalinfo">Additional Information</a>
658 <!-- *********************************************************************** -->
660 <div class="doc_text">
662 <p>A wide variety of additional information is available on the <a
663 href="http://llvm.org">LLVM web page</a>, in particular in the <a
664 href="http://llvm.org/docs/">documentation</a> section. The web page also
665 contains versions of the API documentation which is up-to-date with the
666 Subversion version of the source code.
667 You can access versions of these documents specific to this release by going
668 into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
670 <p>If you have any questions or comments about LLVM, please feel free to contact
671 us via the <a href="http://llvm.org/docs/#maillist"> mailing
676 <!-- *********************************************************************** -->
680 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
681 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
682 <a href="http://validator.w3.org/check/referer"><img
683 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
685 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
686 Last modified: $Date$