2 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
3 "http://www.w3.org/TR/html4/strict.dtd">
6 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
7 <link rel="stylesheet" href="llvm.css" type="text/css">
8 <title>LLVM 2.3 Release Notes</title>
12 <div class="doc_title">LLVM 2.3 Release Notes</div>
15 <li><a href="#intro">Introduction</a></li>
16 <li><a href="#changes">Major Changes and Sub-project Status</a></li>
17 <li><a href="#whatsnew">What's New?</a></li>
18 <li><a href="GettingStarted.html">Installation Instructions</a></li>
19 <li><a href="#portability">Portability and Supported Platforms</a></li>
20 <li><a href="#knownproblems">Known Problems</a>
21 <li><a href="#additionalinfo">Additional Information</a></li>
24 <div class="doc_author">
25 <p>Written by the <a href="http://llvm.org">LLVM Team</a><p>
28 <!-- Done through Week-of-Mon-20080324.txt -->
30 <!-- *********************************************************************** -->
31 <div class="doc_section">
32 <a name="intro">Introduction</a>
34 <!-- *********************************************************************** -->
36 <div class="doc_text">
38 <p>This document contains the release notes for the LLVM compiler
39 infrastructure, release 2.3. Here we describe the status of LLVM, including
40 major improvements from the previous release and any known problems. All LLVM
41 releases may be downloaded from the <a href="http://llvm.org/releases/">LLVM
42 releases web site</a>.</p>
44 <p>For more information about LLVM, including information about the latest
45 release, please check out the <a href="http://llvm.org/">main LLVM
46 web site</a>. If you have questions or comments, the <a
47 href="http://mail.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM developer's mailing
48 list</a> is a good place to send them.</p>
50 <p>Note that if you are reading this file from a Subversion checkout or the
51 main LLVM web page, this document applies to the <i>next</i> release, not the
52 current one. To see the release notes for a specific releases, please see the
53 <a href="http://llvm.org/releases/">releases page</a>.</p>
57 <!-- *********************************************************************** -->
58 <div class="doc_section">
59 <a name="changes">Major Changes and Sub-project Status</a>
61 <!-- *********************************************************************** -->
63 <div class="doc_text">
65 <p>This is the fourteenth public release of the LLVM Compiler Infrastructure.
66 It includes a large number of features and refinements from LLVM 2.2.</p>
70 <!-- Unfinished features in 2.3:
76 <!--=========================================================================-->
77 <div class="doc_subsection">
78 <a name="changes">Major Changes in LLVM 2.3</a>
81 <div class="doc_text">
83 <p>LLVM 2.3 no longer supports llvm-gcc 4.0, it has been replaced with
86 <p>LLVM 2.3 no longer includes the <tt>llvm-upgrade</tt> tool. It was useful
87 for upgrading LLVM 1.9 files to LLVM 2.x syntax, but you can always use a
88 previous LLVM release to do this. One nice impact of this is that the LLVM
89 regression test suite no longer depends on llvm-upgrade, which makes it run
92 <p>The <tt>llvm2cpp</tt> tool has been folded into llc, use
93 <tt>llc -march=cpp</tt> instead of <tt>llvm2cpp</tt>.</p>
95 <p>LLVM API Changes:</p>
98 <li>Several core LLVM IR classes have migrated to use the
99 '<tt>FOOCLASS::Create(...)</tt>' pattern instead of '<tt>new
100 FOOCLASS(...)</tt>' (e.g. where FOOCLASS=<tt>BasicBlock</tt>). We hope to
101 standardize on <tt>FOOCLASS::Create</tt> for all IR classes in the future,
102 but not all of them have been moved over yet.</li>
103 <li>LLVM 2.3 renames the LLVMBuilder and LLVMFoldingBuilder classes to
104 <a href="http://llvm.org/doxygen/classllvm_1_1IRBuilder.html">IRBuilder</a>.
106 <li>MRegisterInfo was renamed to
107 <a href="http://llvm.org/doxygen/classllvm_1_1TargetRegisterInfo.html">
108 TargetRegisterInfo</a>.</li>
109 <li>The MappedFile class is gone, please use
110 <a href="http://llvm.org/doxygen/classllvm_1_1MemoryBuffer.html">
111 MemoryBuffer</a> instead.</li>
112 <li>The '<tt>-enable-eh</tt>' flag to llc has been removed. Now code should
113 encode whether it is safe to omit unwind information for a function by
114 tagging the Function object with the '<tt>nounwind</tt>' attribute.</li>
119 <!--=========================================================================-->
120 <div class="doc_subsection">
121 <a name="otherprojects">Other LLVM Sub-Projects</a>
124 <div class="doc_text">
126 The core LLVM 2.3 distribution currently consists of code from the core LLVM
127 repository (which roughly contains the LLVM optimizer, code generators and
128 supporting tools) and the llvm-gcc repository. In addition to this code, the
129 LLVM Project includes other sub-projects that are in development. The two which
130 are the most actively developed are the new <a href="#vmkit">vmkit Project</a>
131 and the <a href="#clang">Clang Project</a>.
135 <!--=========================================================================-->
136 <div class="doc_subsubsection">
137 <a name="vmkit">vmkit</a>
140 <div class="doc_text">
142 The "vmkit" project is a new addition to the LLVM family. It is an
143 implementation of a JVM and a CLI Virtual Machines (Microsoft .NET is an
144 implementation of the CLI) using the Just-In-Time compiler of LLVM.</p>
146 <p>The JVM, called JnJVM, executes real-world applications such as Apache
147 projects (e.g. Felix and Tomcat) and the SpecJVM98 benchmark. It uses the GNU
148 Classpath project for the base classes. The CLI implementation, called N3, is
149 its in early stages but can execute simple applications and the "pnetmark"
150 benchmark. It uses the pnetlib project as its core library.</p>
152 <p>The 'vmkit' VMs compare in performance with industrial and top open-source
153 VMs on scientific applications. Besides the JIT, the VMs use many features of
154 the LLVM framework, including the standard set of optimizations, atomic
155 operations, custom function provider and memory manager for JITed methods, and
156 specific virtual machine optimizations. vmkit is not an official part of LLVM
157 2.3 release. It is publicly available under the LLVM license and can be
162 <tt>svn co http://llvm.org/svn/llvm-project/vmkit/trunk vmkit</tt>
167 <!--=========================================================================-->
168 <div class="doc_subsubsection">
169 <a name="clang">Clang</a>
172 <div class="doc_text">
174 <p>The <a href="http://clang.llvm.org/">Clang project</a> is an effort to build
175 a set of new 'LLVM native' front-end technologies for the LLVM optimizer
176 and code generator. Clang is continuing to make major strides forward in all
177 areas. Its C and Objective-C parsing support is very solid, and the code
178 generation support is far enough along to build many C applications. While not
179 yet production quality, it is progressing very nicely. In addition, C++
180 front-end work has started to make significant progress.</p>
182 <p>At this point, Clang is most useful if you are interested in source-to-source
183 transformations (such as refactoring) and other source-level tools for C and
184 Objective-C. Clang now also includes tools for turning C code into pretty HTML,
185 and includes a new <a href="http://clang.llvm.org/StaticAnalysis.html">static
186 analysis tool</a> in development. This tool is automatically focused on finding
187 bugs in C and Objective-C code.</p>
192 <!-- *********************************************************************** -->
193 <div class="doc_section">
194 <a name="whatsnew">What's New?</a>
196 <!-- *********************************************************************** -->
198 <div class="doc_text">
200 <p>LLVM 2.3 includes a huge number of bug fixes, performance tweaks and minor
201 improvements. Some of the major improvements and new features are listed in
206 <!--=========================================================================-->
207 <div class="doc_subsection">
208 <a name="majorfeatures">Major New Features</a>
211 <div class="doc_text">
213 <p>LLVM 2.3 includes several major new capabilities:</p>
216 <li><p>The biggest change in LLVM 2.3 is Multiple Return Value (MRV) support.
217 MRVs allow LLVM IR to directly represent functions that return multiple
218 values without having to pass them "by reference" in the LLVM IR. This
219 allows a front-end to generate more efficient code, as MRVs are generally
220 returned in registers if a target supports them. See the <a
221 href="LangRef.html#i_getresult">LLVM IR Reference</a> for more details.</p>
223 <p>MRVs are fully supported in the LLVM IR, but are not yet fully supported in
224 on all targets. However, it is generally safe to return up to 2 values from
225 a function: most targets should be able to handle at least that. MRV
226 support is a critical requirement for X86-64 ABI support, as X86-64 requires
227 the ability to return multiple registers from functions, and we use MRVs to
228 accomplish this in a direct way.</p></li>
230 <li><p>LLVM 2.3 includes a complete reimplementation of the "<tt>llvmc</tt>"
231 tool. It is designed to overcome several problems with the original
232 <tt>llvmc</tt> and to provide a superset of the features of the
233 '<tt>gcc</tt>' driver.</p>
235 <p>The main features of <tt>llvmc2</tt> are:
237 <li>Extended handling of command line options and smart rules for
238 dispatching them to different tools.</li>
239 <li>Flexible (and extensible) rules for defining different tools.</li>
240 <li>The different intermediate steps performed by tools are represented
241 as edges in the abstract graph.</li>
242 <li>The 'language' for driver behavior definition is tablegen and thus
243 it's relatively easy to add new features.</li>
244 <li>The definition of driver is transformed into set of C++ classes, thus
245 no runtime interpretation is needed.</li>
249 <li><p>LLVM 2.3 includes a completely rewritten interface for <a
250 href="LinkTimeOptimization.html">Link Time Optimization</a>. This interface
251 is written in C, which allows for easier integration with C code bases, and
252 incorporates improvements we learned about from the first incarnation of the
255 <li><p>The <a href="tutorial/LangImpl1.html">Kaleidoscope tutorial</a> now
256 includes a "port" of the tutorial that <a
257 href="tutorial/OCamlLangImpl1.html">uses the Ocaml bindings</a> to implement
258 the Kaleidoscope language.</p></li>
265 <!--=========================================================================-->
266 <div class="doc_subsection">
267 <a name="llvm-gcc">llvm-gcc 4.2 Improvements</a>
270 <div class="doc_text">
272 <p>LLVM 2.3 fully supports the llvm-gcc 4.2 front-end, and includes support
273 for the C, C++, Objective-C, Ada, and Fortran front-ends.</p>
277 <li>llvm-gcc 4.2 includes numerous fixes to better support the Objective-C
278 front-end. Objective-C now works very well on Mac OS/X.</li>
280 <li>Fortran EQUIVALENCEs are now supported by the gfortran front-end.</li>
282 <li>llvm-gcc 4.2 includes many other fixes which improve conformance with the
283 relevant parts of the GCC testsuite.</li>
290 <!--=========================================================================-->
291 <div class="doc_subsection">
292 <a name="coreimprovements">LLVM Core Improvements</a>
295 <div class="doc_text">
296 <p>New features include:
300 <li>LLVM IR now directly represents "common" linkage, instead of representing it
301 as a form of weak linkage.</li>
303 <li>LLVM IR now has support for atomic operations, and this functionality can
304 be accessed through the llvm-gcc "__sync_synchronize",
305 "__sync_val_compare_and_swap", and related builtins. Support for atomics are
306 available in the Alpha, X86, X86-64, and PowerPC backends.</li>
308 <li>The C and Ocaml bindings have extended to cover pass managers, several
309 transformation passes, iteration over the LLVM IR, target data, and parameter
310 attribute lists.</li>
315 <!--=========================================================================-->
316 <div class="doc_subsection">
317 <a name="optimizer">Optimizer Improvements</a>
320 <div class="doc_text">
322 <p>In addition to a huge array of bug fixes and minor performance tweaks, the
323 LLVM 2.3 optimizers support a few major enhancements:</p>
327 <li><p>Loop index set splitting on by default.
328 This transformation hoists conditions from loop bodies and reduces a loop's
329 iteration space to improve performance. For example,</p>
332 for (i = LB; i < UB; ++i)
337 <p>is transformed into:</p>
341 for (i = LB; i < NUB; ++i)
346 <li>LLVM now includes a new <tt>memcpy</tt> optimization pass which removes
347 dead <tt>memcpy</tt> calls, unneeded copies of aggregates, and performs
348 return slot optimization. The LLVM optimizer now notices long sequences of
349 consecutive stores and merges them into <tt>memcpy</tt>'s where profitable.</li>
351 <li>Alignment detection for vector memory references and for <tt>memcpy</tt> and
352 <tt>memset</tt> is now more aggressive.</li>
354 <li>The Aggressive Dead Code Elimination (ADCE) optimization has been rewritten
355 to make it both faster and safer in the presence of code containing infinite
356 loops. Some of its prior functionality has been factored out into the loop
357 deletion pass, which <em>is</em> safe for infinite loops. The new ADCE pass is
358 no longer based on control dependence, making it run faster.</li>
360 <li>The 'SimplifyLibCalls' pass, which optimizes calls to libc and libm
361 functions for C-based languages, has been rewritten to be a FunctionPass
362 instead a ModulePass. This allows it to be run more often and to be
363 included at -O1 in llvm-gcc. It was also extended to include more
364 optimizations and several corner case bugs were fixed.</li>
366 <li>LLVM now includes a simple 'Jump Threading' pass, which attempts to simplify
367 conditional branches using information about predecessor blocks, simplifying
368 the control flow graph. This pass is pretty basic at this point, but
369 catches some important cases and provides a foundation to build on.</li>
371 <li>Several corner case bugs which could lead to deleting volatile memory
372 accesses have been fixed.</li>
374 <li>Several optimizations have been sped up, leading to faster code generation
375 with the same code quality.</li>
381 <!--=========================================================================-->
382 <div class="doc_subsection">
383 <a name="codegen">Code Generator Improvements</a>
386 <div class="doc_text">
388 <p>We put a significant amount of work into the code generator infrastructure,
389 which allows us to implement more aggressive algorithms and make it run
393 <li>The code generator now has support for carrying information about memory
394 references throughout the entire code generation process, via the
395 <a href="http://llvm.org/doxygen/classllvm_1_1MachineMemOperand.html">
396 MachineMemOperand</a> class. In the future this will be used to improve
397 both pre-pass and post-pass scheduling, and to improve compiler-debugging
400 <li>The target-independent code generator infrastructure now uses LLVM's
401 <a href="http://llvm.org/doxygen/classllvm_1_1APInt.html">APInt</a>
402 class to handle integer values, which allows it to support integer types
403 larger than 64 bits. Note that support for such types is also dependent on
404 target-specific support. Use of APInt is also a step toward support for
405 non-power-of-2 integer sizes.</li>
407 <li>LLVM 2.3 includes several compile time speedups for code with large basic
408 blocks, particularly in the instruction selection phase, register
409 allocation, scheduling, and tail merging/jump threading.</li>
411 <li>Several improvements which make llc's <tt>--view-sunit-dags</tt>
412 visualization of scheduling dependency graphs easier to understand.</li>
414 <li>The code generator allows targets to write patterns that generate subreg
415 references directly in .td files now.</li>
417 <li><tt>memcpy</tt> lowering in the backend is more aggressive, particularly for
418 <tt>memcpy</tt> calls introduced by the code generator when handling
419 pass-by-value structure argument copies.</li>
421 <li>Inline assembly with multiple register results now returns those results
422 directly in the appropriate registers, rather than going through memory.
423 Inline assembly that uses constraints like "ir" with immediates now use the
424 'i' form when possible instead of always loading the value in a register.
425 This saves an instruction and reduces register use.</li>
427 <li>Added support for PIC/GOT style tail calls on x86/32 and initial support
428 for tail calls on PowerPC 32 (it may also work on ppc64 but not
429 thoroughly tested).</li>
435 <!--=========================================================================-->
436 <div class="doc_subsection">
437 <a name="x86specific">X86/X86-64 Specific Improvements</a>
440 <div class="doc_text">
441 <p>New target-specific features include:
445 <li>llvm-gcc's X86-64 ABI conformance is far improved, particularly in the
446 area of passing and returning structures by value. llvm-gcc compiled code
447 now interoperates very well on X86-64 systems with other compilers.</li>
449 <li>Support for Win64 was added. This includes code generation itself, JIT
450 support and necessary changes to llvm-gcc.</li>
452 <li>The LLVM X86 backend now supports the support SSE 4.1 instruction set, and
453 the llvm-gcc 4.2 front-end supports the SSE 4.1 compiler builtins. Various
454 generic vector operations (insert/extract/shuffle) are much more efficient
455 when SSE 4.1 is enabled. The JIT automatically takes advantage of these
456 instructions, but llvm-gcc must be explicitly told to use them, e.g. with
457 <tt>-march=penryn</tt>.</li>
459 <li>The X86 backend now does a number of optimizations that aim to avoid
460 converting numbers back and forth from SSE registers to the X87 floating
463 <li>The X86 backend supports stack realignment, which is particularly useful for
464 vector code on OS's without 16-byte aligned stacks.</li>
466 <li>The X86 backend now supports the "sseregparm" options in GCC, which allow
467 functions to be tagged as passing floating point values in SSE
470 <li>Trampolines (taking the address of a nested function) now work on
473 <li><tt>__builtin_prefetch</tt> is now compiled into the appropriate prefetch
474 instructions instead of being ignored.</li>
476 <li>128-bit integers are now supported on X86-64 targets.</li>
478 <li>The register allocator can now rematerialize PIC-base computations.</li>
480 <li>The "t" and "f" inline assembly constraints for the X87 floating point stack
481 now work. However, the "u" constraint is still not fully supported.</li>
487 <!--=========================================================================-->
488 <div class="doc_subsection">
489 <a name="targetspecific">Other Target Specific Improvements</a>
492 <div class="doc_text">
493 <p>New target-specific features include:
497 <li>The LLVM C backend now supports vector code.</li>
504 <!--=========================================================================-->
505 <div class="doc_subsection">
506 <a name="otherimprovements">Other Improvements</a>
509 <div class="doc_text">
510 <p>New features include:
514 <li>LLVM now builds with GCC 4.3.</li>
515 <li>Bugpoint now supports running custom scripts (with the <tt>-run-custom</tt>
516 option) to determine how to execute the command and whether it is making
517 forward process.</li>
522 <!-- *********************************************************************** -->
523 <div class="doc_section">
524 <a name="portability">Portability and Supported Platforms</a>
526 <!-- *********************************************************************** -->
528 <div class="doc_text">
530 <p>LLVM is known to work on the following platforms:</p>
533 <li>Intel and AMD machines (IA32) running Red Hat Linux, Fedora Core and FreeBSD
534 (and probably other unix-like systems).</li>
535 <li>PowerPC and X86-based Mac OS X systems, running 10.3 and above in 32-bit and
537 <li>Intel and AMD machines running on Win32 using MinGW libraries (native).</li>
538 <li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
539 support is available for native builds with Visual C++).</li>
540 <li>Sun UltraSPARC workstations running Solaris 10.</li>
541 <li>Alpha-based machines running Debian GNU/Linux.</li>
542 <li>Itanium-based (IA64) machines running Linux and HP-UX.</li>
545 <p>The core LLVM infrastructure uses GNU autoconf to adapt itself
546 to the machine and operating system on which it is built. However, minor
547 porting may be required to get LLVM to work on new platforms. We welcome your
548 portability patches and reports of successful builds or error messages.</p>
552 <!-- *********************************************************************** -->
553 <div class="doc_section">
554 <a name="knownproblems">Known Problems</a>
556 <!-- *********************************************************************** -->
558 <div class="doc_text">
560 <p>This section contains all known problems with the LLVM system, listed by
561 component. As new problems are discovered, they will be added to these
562 sections. If you run into a problem, please check the <a
563 href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
564 there isn't already one.</p>
568 <!-- ======================================================================= -->
569 <div class="doc_subsection">
570 <a name="experimental">Experimental features included with this release</a>
573 <div class="doc_text">
575 <p>The following components of this LLVM release are either untested, known to
576 be broken or unreliable, or are in early development. These components should
577 not be relied on, and bugs should not be filed against them, but they may be
578 useful to some people. In particular, if you would like to work on one of these
579 components, please contact us on the <a
580 href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
583 <li>The MSIL, IA64, Alpha, SPU, and MIPS backends are experimental.</li>
584 <li>The llc "<tt>-filetype=asm</tt>" (the default) is the only supported
585 value for this option.</li>
590 <!-- ======================================================================= -->
591 <div class="doc_subsection">
592 <a name="x86-be">Known problems with the X86 back-end</a>
595 <div class="doc_text">
598 <li>The X86 backend does not yet support
599 all <a href="http://llvm.org/PR879">inline assembly that uses the X86
600 floating point stack</a>. It supports the 'f' and 't' constraints, but not
602 <li>The X86 backend generates inefficient floating point code when configured
603 to generate code for systems that don't have SSE2.</li>
604 <li>Win64 code generation wasn't widely tested. Everything should work, but we
605 expect small issues to happen. Also, llvm-gcc cannot build mingw64 runtime
607 to <a href="http://llvm.org/PR2255">several</a>
608 <a href="http://llvm.org/PR2257">bugs</a> in FP stackifier
613 <!-- ======================================================================= -->
614 <div class="doc_subsection">
615 <a name="ppc-be">Known problems with the PowerPC back-end</a>
618 <div class="doc_text">
621 <li>The Linux PPC32/ABI support needs testing for the interpreter and static
622 compilation, and lacks support for debug information.</li>
627 <!-- ======================================================================= -->
628 <div class="doc_subsection">
629 <a name="arm-be">Known problems with the ARM back-end</a>
632 <div class="doc_text">
635 <li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
636 processors, thumb programs can crash or produce wrong
637 results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
638 <li>Compilation for ARM Linux OABI (old ABI) is supported, but not fully tested.
640 <li>There is a bug in QEMU-ARM (<= 0.9.0) which causes it to incorrectly
642 programs compiled with LLVM. Please use more recent versions of QEMU.</li>
647 <!-- ======================================================================= -->
648 <div class="doc_subsection">
649 <a name="sparc-be">Known problems with the SPARC back-end</a>
652 <div class="doc_text">
655 <li>The SPARC backend only supports the 32-bit SPARC ABI (-m32), it does not
656 support the 64-bit SPARC ABI (-m64).</li>
661 <!-- ======================================================================= -->
662 <div class="doc_subsection">
663 <a name="alpha-be">Known problems with the Alpha back-end</a>
666 <div class="doc_text">
670 <li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
671 appropriate nops inserted to ensure restartability.</li>
676 <!-- ======================================================================= -->
677 <div class="doc_subsection">
678 <a name="ia64-be">Known problems with the IA64 back-end</a>
681 <div class="doc_text">
685 <li>C++ programs are likely to fail on IA64, as calls to <tt>setjmp</tt> are
686 made where the argument is not 16-byte aligned, as required on IA64. (Strictly
687 speaking this is not a bug in the IA64 back-end; it will also be encountered
688 when building C++ programs using the C back-end.)</li>
690 <li>The C++ front-end does not use <a href="http://llvm.org/PR406">IA64
691 ABI compliant layout of v-tables</a>. In particular, it just stores function
692 pointers instead of function descriptors in the vtable. This bug prevents
693 mixing C++ code compiled with LLVM with C++ objects compiled by other C++
696 <li>There are a few ABI violations which will lead to problems when mixing LLVM
697 output with code built with other compilers, particularly for floating-point
700 <li>Defining vararg functions is not supported (but calling them is OK).</li>
702 <li>The Itanium backend has bitrotted somewhat.</li>
707 <!-- ======================================================================= -->
708 <div class="doc_subsection">
709 <a name="c-be">Known problems with the C back-end</a>
712 <div class="doc_text">
715 <li><a href="http://llvm.org/PR802">The C backend has only basic support for
716 inline assembly code</a>.</li>
717 <li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
718 C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
719 C++ code compiled with llc or native compilers.</li>
720 <li>The C backend does not support all exception handling constructs.</li>
726 <!-- ======================================================================= -->
727 <div class="doc_subsection">
728 <a name="c-fe">Known problems with the llvm-gcc C front-end</a>
731 <!-- _______________________________________________________________________ -->
732 <div class="doc_subsubsection">Bugs</div>
734 <div class="doc_text">
736 <p>llvm-gcc does not currently support <a href="http://llvm.org/PR869">Link-Time
737 Optimization</a> on most platforms "out-of-the-box". Please inquire on the
738 llvmdev mailing list if you are interested.</p>
740 <p>The only major language feature of GCC not supported by llvm-gcc is
741 the <tt>__builtin_apply</tt> family of builtins. However, some extensions
742 are only supported on some targets. For example, trampolines are only
743 supported on some targets (these are used when you take the address of a
744 nested function).</p>
746 <p>If you run into GCC extensions which are not supported, please let us know.
751 <!-- ======================================================================= -->
752 <div class="doc_subsection">
753 <a name="c++-fe">Known problems with the llvm-gcc C++ front-end</a>
756 <div class="doc_text">
758 <p>The C++ front-end is considered to be fully
759 tested and works for a number of non-trivial programs, including LLVM
760 itself, Qt, Mozilla, etc.</p>
763 <li>Exception handling works well on the X86 and PowerPC targets, including
764 X86-64 darwin. This works when linking to a libstdc++ compiled by GCC. It is
765 supported on X86-64 linux, but that is disabled by default in this release.</li>
771 <!-- ======================================================================= -->
772 <div class="doc_subsection">
773 <a name="ada-fe">Known problems with the llvm-gcc Ada front-end</a>
776 <div class="doc_text">
777 The llvm-gcc 4.2 Ada compiler works fairly well, however this is not a mature
778 technology and problems should be expected.
780 <li>The Ada front-end currently only builds on X86-32. This is mainly due
781 to lack of trampoline support (pointers to nested functions) on other platforms,
782 however it <a href="http://llvm.org/PR2006">also fails to build on X86-64</a>
783 which does support trampolines.</li>
784 <li>The Ada front-end <a href="http://llvm.org/PR2007">fails to bootstrap</a>.
785 Workaround: configure with --disable-bootstrap.</li>
786 <li>The c380004 and <a href="http://llvm.org/PR2010">c393010</a> ACATS tests
787 fail (c380004 also fails with gcc-4.2 mainline). When built at -O3, the
788 <a href="http://llvm.org/PR2421">cxg2021</a> ACATS test also fails.</li>
789 <li>Some gcc specific Ada tests continue to crash the compiler. The testsuite
790 reports most tests as having failed even though they pass.</li>
791 <li>The -E binder option (exception backtraces)
792 <a href="http://llvm.org/PR1982">does not work</a> and will result in programs
793 crashing if an exception is raised. Workaround: do not use -E.</li>
794 <li>Only discrete types <a href="http://llvm.org/PR1981">are allowed to start
795 or finish at a non-byte offset</a> in a record. Workaround: do not pack records
796 or use representation clauses that result in a field of a non-discrete type
797 starting or finishing in the middle of a byte.</li>
798 <li>The <tt>lli</tt> interpreter <a href="http://llvm.org/PR2009">considers
799 'main' as generated by the Ada binder to be invalid</a>.
800 Workaround: hand edit the file to use pointers for <tt>argv</tt> and
801 <tt>envp</tt> rather than integers.</li>
802 <li>The <tt>-fstack-check</tt> option <a href="http://llvm.org/PR2008">is
807 <!-- *********************************************************************** -->
808 <div class="doc_section">
809 <a name="additionalinfo">Additional Information</a>
811 <!-- *********************************************************************** -->
813 <div class="doc_text">
815 <p>A wide variety of additional information is available on the <a
816 href="http://llvm.org">LLVM web page</a>, in particular in the <a
817 href="http://llvm.org/docs/">documentation</a> section. The web page also
818 contains versions of the API documentation which is up-to-date with the
819 Subversion version of the source code.
820 You can access versions of these documents specific to this release by going
821 into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
823 <p>If you have any questions or comments about LLVM, please feel free to contact
824 us via the <a href="http://llvm.org/docs/#maillist"> mailing
829 <!-- *********************************************************************** -->
833 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
834 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
835 <a href="http://validator.w3.org/check/referer"><img
836 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
838 <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
839 Last modified: $Date$