docs/ReleaseNotes.html

   1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
   2                       "http://www.w3.org/TR/html4/strict.dtd">
   3 <html>
   4 <head>
   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 2.8 Release Notes</title>
   8 </head>
   9 <body>
  10
  11 <div class="doc_title">LLVM 2.8 Release Notes</div>
  12
  13 <img align=right src="http://llvm.org/img/DragonSmall.png"
  14     width="136" height="136" alt="LLVM Dragon Logo">
  15
  16 <ol>
  17   <li><a href="#intro">Introduction</a></li>
  18   <li><a href="#subproj">Sub-project Status Update</a></li>
  19   <li><a href="#externalproj">External Projects Using LLVM 2.8</a></li>
  20   <li><a href="#whatsnew">What's New in LLVM 2.8?</a></li>
  21   <li><a href="GettingStarted.html">Installation Instructions</a></li>
  22   <li><a href="#portability">Portability and Supported Platforms</a></li>
  23   <li><a href="#knownproblems">Known Problems</a></li>
  24   <li><a href="#additionalinfo">Additional Information</a></li>
  25 </ol>
  26
  27 <div class="doc_author">
  28   <p>Written by the <a href="http://llvm.org">LLVM Team</a></p>
  29 </div>
  30
  31 <h1 style="color:red">These are in-progress notes for the upcoming LLVM 2.8
  32 release.<br>
  33 You may prefer the
  34 <a href="http://llvm.org/releases/2.7/docs/ReleaseNotes.html">LLVM 2.7
  35 Release Notes</a>.</h1>
  36
  37 <!-- *********************************************************************** -->
  38 <div class="doc_section">
  39   <a name="intro">Introduction</a>
  40 </div>
  41 <!-- *********************************************************************** -->
  42
  43 <div class="doc_text">
  44
  45 <p>This document contains the release notes for the LLVM Compiler
  46 Infrastructure, release 2.8.  Here we describe the status of LLVM, including
  47 major improvements from the previous release and significant known problems.
  48 All LLVM releases may be downloaded from the <a
  49 href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
  50
  51 <p>For more information about LLVM, including information about the latest
  52 release, please check out the <a href="http://llvm.org/">main LLVM
  53 web site</a>.  If you have questions or comments, the <a
  54 href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVM Developer's
  55 Mailing List</a> is a good place to send them.</p>
  56
  57 <p>Note that if you are reading this file from a Subversion checkout or the
  58 main LLVM web page, this document applies to the <i>next</i> release, not the
  59 current one.  To see the release notes for a specific release, please see the
  60 <a href="http://llvm.org/releases/">releases page</a>.</p>
  61
  62 </div>
  63
  64
  65 <!--
  66 Almost dead code.
  67   include/llvm/Analysis/LiveValues.h => Dan
  68   lib/Transforms/IPO/MergeFunctions.cpp => consider for 2.8.
  69   llvm/Analysis/PointerTracking.h => Edwin wants this, consider for 2.8.
  70   GEPSplitterPass
  71 -->
  72
  73
  74 <!-- Features that need text if they're finished for 2.8:
  75   combiner-aa?
  76   strong phi elim
  77   llvm.dbg.value: variable debug info for optimized code
  78   loop dependence analysis
  79   TBAA
  80  -->
  81
  82  <!-- for announcement email:
  83  Logo web page.
  84  Many new papers added to /pubs/
  85    -->
  86
  87 <!-- *********************************************************************** -->
  88 <div class="doc_section">
  89   <a name="subproj">Sub-project Status Update</a>
  90 </div>
  91 <!-- *********************************************************************** -->
  92
  93 <div class="doc_text">
  94 <p>
  95 The LLVM 2.8 distribution currently consists of code from the core LLVM
  96 repository (which roughly includes the LLVM optimizers, code generators
  97 and supporting tools), the Clang repository and the llvm-gcc repository.  In
  98 addition to this code, the LLVM Project includes other sub-projects that are in
  99 development.  Here we include updates on these subprojects.
 100 </p>
 101
 102 </div>
 103
 104
 105 <!--=========================================================================-->
 106 <div class="doc_subsection">
 107 <a name="clang">Clang: C/C++/Objective-C Frontend Toolkit</a>
 108 </div>
 109
 110 <div class="doc_text">
 111
 112 <p><a href="http://clang.llvm.org/">Clang</a> is an LLVM front end for the C,
 113 C++, and Objective-C languages. Clang aims to provide a better user experience
 114 through expressive diagnostics, a high level of conformance to language
 115 standards, fast compilation, and low memory use. Like LLVM, Clang provides a
 116 modular, library-based architecture that makes it suitable for creating or
 117 integrating with other development tools. Clang is considered a
 118 production-quality compiler for C and Objective-C on x86 (32- and 64-bit).</p>
 119
 120 <p>In the LLVM 2.8 time-frame, the Clang team has made many improvements:</p>
 121
 122 <ul>
 123
 124 </ul>
 125 </div>
 126
 127 <!--=========================================================================-->
 128 <div class="doc_subsection">
 129 <a name="clangsa">Clang Static Analyzer</a>
 130 </div>
 131
 132 <div class="doc_text">
 133
 134 <p>The <a href="http://clang-analyzer.llvm.org/">Clang Static Analyzer</a>
 135    project is an effort to use static source code analysis techniques to
 136    automatically find bugs in C and Objective-C programs (and hopefully <a
 137    href="http://clang-analyzer.llvm.org/dev_cxx.html">C++ in the
 138    future</a>!).  The tool is very good at finding bugs that occur on specific
 139    paths through code, such as on error conditions.</p>
 140
 141 <p>In the LLVM 2.8 time-frame,
 142 </p>
 143
 144 </div>
 145
 146 <!--=========================================================================-->
 147 <div class="doc_subsection">
 148 <a name="vmkit">VMKit: JVM/CLI Virtual Machine Implementation</a>
 149 </div>
 150
 151 <div class="doc_text">
 152 <p>
 153 The <a href="http://vmkit.llvm.org/">VMKit project</a> is an implementation of
 154 a JVM and a CLI Virtual Machine (Microsoft .NET is an
 155 implementation of the CLI) using LLVM for static and just-in-time
 156 compilation.</p>
 157
 158 <p>With the release of LLVM 2.8, ...</p>
 159
 160 </div>
 161
 162
 163 <!--=========================================================================-->
 164 <div class="doc_subsection">
 165 <a name="compiler-rt">compiler-rt: Compiler Runtime Library</a>
 166 </div>
 167
 168 <div class="doc_text">
 169 <p>
 170 The new LLVM <a href="http://compiler-rt.llvm.org/">compiler-rt project</a>
 171 is a simple library that provides an implementation of the low-level
 172 target-specific hooks required by code generation and other runtime components.
 173 For example, when compiling for a 32-bit target, converting a double to a 64-bit
 174 unsigned integer is compiled into a runtime call to the "__fixunsdfdi"
 175 function. The compiler-rt library provides highly optimized implementations of
 176 this and other low-level routines (some are 3x faster than the equivalent
 177 libgcc routines).</p>
 178
 179 <p>
 180 All of the code in the compiler-rt project is available under the standard LLVM
 181 License, a "BSD-style" license.  New in LLVM 2.8:
 182
 183 Soft float support
 184 </p>
 185
 186 </div>
 187
 188 <!--=========================================================================-->
 189 <div class="doc_subsection">
 190 <a name="dragonegg">DragonEgg: llvm-gcc ported to gcc-4.5</a>
 191 </div>
 192
 193 <div class="doc_text">
 194 <p>
 195 <a href="http://dragonegg.llvm.org/">DragonEgg</a> is a port of llvm-gcc to
 196 gcc-4.5.  Unlike llvm-gcc, which makes many intrusive changes to the underlying
 197 gcc-4.2 code, dragonegg in theory does not require any gcc-4.5 modifications
 198 whatsoever (currently one small patch is needed).  This is thanks to the new
 199 <a href="http://gcc.gnu.org/wiki/plugins">gcc plugin architecture</a>, which
 200 makes it possible to modify the behaviour of gcc at runtime by loading a plugin,
 201 which is nothing more than a dynamic library which conforms to the gcc plugin
 202 interface.  DragonEgg is a gcc plugin that causes the LLVM optimizers to be run
 203 instead of the gcc optimizers, and the LLVM code generators instead of the gcc
 204 code generators, just like llvm-gcc.  To use it, you add
 205 "-fplugin=path/dragonegg.so" to the gcc-4.5 command line, and gcc-4.5 magically
 206 becomes llvm-gcc-4.5!
 207 </p>
 208
 209 <p>
 210 DragonEgg is still a work in progress.  Currently C works very well, while C++,
 211 Ada and Fortran work fairly well.  All other languages either don't work at all,
 212 or only work poorly.  For the moment only the x86-32 and x86-64 targets are
 213 supported, and only on linux and darwin (darwin needs an additional gcc patch).
 214 </p>
 215
 216 <p>
 217 2.8 status here.
 218 </p>
 219
 220 </div>
 221
 222
 223 <!--=========================================================================-->
 224 <div class="doc_subsection">
 225 <a name="mc">llvm-mc: Machine Code Toolkit</a>
 226 </div>
 227
 228 <div class="doc_text">
 229 <p>
 230 The LLVM Machine Code (aka MC) sub-project of LLVM was created to solve a number
 231 of problems in the realm of assembly, disassembly, object file format handling,
 232 and a number of other related areas that CPU instruction-set level tools work
 233 in. It is a sub-project of LLVM which provides it with a number of advantages
 234 over other compilers that do not have tightly integrated assembly-level tools.
 235 For a gentle introduction, please see the <a
 236 href="http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html">Intro to the
 237 LLVM MC Project Blog Post</a>.
 238 </p>
 239
 240 <p>2.8 status here</p>
 241 </div>
 242
 243
 244 <!-- *********************************************************************** -->
 245 <div class="doc_section">
 246   <a name="externalproj">External Open Source Projects Using LLVM 2.8</a>
 247 </div>
 248 <!-- *********************************************************************** -->
 249
 250 <div class="doc_text">
 251
 252 <p>An exciting aspect of LLVM is that it is used as an enabling technology for
 253    a lot of other language and tools projects.  This section lists some of the
 254    projects that have already been updated to work with LLVM 2.8.</p>
 255 </div>
 256
 257
 258 <!-- *********************************************************************** -->
 259 <div class="doc_section">
 260   <a name="whatsnew">What's New in LLVM 2.8?</a>
 261 </div>
 262 <!-- *********************************************************************** -->
 263
 264 <div class="doc_text">
 265
 266 <p>This release includes a huge number of bug fixes, performance tweaks and
 267 minor improvements.  Some of the major improvements and new features are listed
 268 in this section.
 269 </p>
 270
 271 </div>
 272
 273 <!--=========================================================================-->
 274 <div class="doc_subsection">
 275 <a name="orgchanges">LLVM Community Changes</a>
 276 </div>
 277
 278 <div class="doc_text">
 279
 280 <p>In addition to changes to the code, between LLVM 2.7 and 2.8, a number of
 281 organization changes have happened:
 282 </p>
 283
 284 <ul>
 285 </ul>
 286 </div>
 287
 288 <!--=========================================================================-->
 289 <div class="doc_subsection">
 290 <a name="majorfeatures">Major New Features</a>
 291 </div>
 292
 293 <div class="doc_text">
 294
 295 <p>LLVM 2.8 includes several major new capabilities:</p>
 296
 297 <ul>
 298 <li>atomic lowering pass.</li>
 299 <li>RegionInfo pass: opt -regions analyze" or "opt -view-regions".
 300 <!-- Tobias Grosser --></li>
 301 <li>ARMGlobalMerge: <!-- Anton --> </li>
 302 <li>llvm-diff</li>
 303 </ul>
 304
 305 </div>
 306
 307 <!--=========================================================================-->
 308 <div class="doc_subsection">
 309 <a name="coreimprovements">LLVM IR and Core Improvements</a>
 310 </div>
 311
 312 <div class="doc_text">
 313 <p>LLVM IR has several new features for better support of new targets and that
 314 expose new optimization opportunities:</p>
 315
 316 <ul>
 317
 318 <li>LLVM 2.8 changes the internal order of operands in <a
 319   href="http://llvm.org/doxygen/classllvm_1_1InvokeInst.html"><tt>InvokeInst</tt></a>
 320   and <a href="http://llvm.org/doxygen/classllvm_1_1CallInst.html"><tt>CallInst</tt></a>.
 321   To be portable across releases, resort to <tt>CallSite</tt> and the
 322   high-level accessors, such as <tt>getCalledValue</tt> and <tt>setUnwindDest</tt>.
 323 </li>
 324 <li>
 325   You can no longer pass use_iterators directly to cast<> (and similar), because
 326   these routines tend to perform costly dereference operations more than once. You
 327   have to dereference the iterators yourself and pass them in.
 328 </li>
 329 <li>
 330   llvm.memcpy.*, llvm.memset.*, llvm.memmove.* (and possibly other?) intrinsics
 331   take an extra parameter now (i1 isVolatile), totaling 5 parameters.
 332   If you were creating these intrinsic calls and prototypes yourself (as opposed
 333   to using Intrinsic::getDeclaration), you can use UpgradeIntrinsicFunction/UpgradeIntrinsicCall
 334   to be portable accross releases.
 335   Note that you cannot use Intrinsic::getDeclaration() in a backwards compatible
 336   way (needs 2/3 types now, in 2.7 it needed just 1).
 337 </li>
 338 <li>
 339   SetCurrentDebugLocation takes a DebugLoc now instead of a MDNode.
 340   Change your code to use
 341   SetCurrentDebugLocation(DebugLoc::getFromDILocation(...)).
 342 </li>
 343 <li>
 344   VISIBILITY_HIDDEN is gone.
 345 </li>
 346 <li>
 347   The <tt>RegisterPass</tt> and <tt>RegisterAnalysisGroup</tt> templates are
 348   considered deprecated, but continue to function in LLVM 2.8.  Clients are
 349   strongly advised to use the upcoming <tt>INITIALIZE_PASS()</tt> and
 350   <tt>INITIALIZE_AG_PASS()</tt> macros instead.
 351 <li>
 352   SMDiagnostic takes different parameters now. //FIXME: how to upgrade?
 353 </li>
 354 <li>
 355   The constructor for the Triple class no longer tries to understand odd triple
 356   specifications.  Frontends should ensure that they only pass valid triples to
 357   LLVM.  The Triple::normalize utility method has been added to help front-ends
 358   deal with funky triples.
 359 <li>
 360   Some APIs got renamed:
 361   <ul>
 362       <li>llvm_report_error -&gt; report_fatal_error</li>
 363       <li>llvm_install_error_handler -&gt; install_fatal_error_handler</li>
 364       <li>llvm::DwarfExceptionHandling -&gt; llvm::JITExceptionHandling</li>
 365   </ul>
 366 </li>
 367 </ul>
 368
 369 </div>
 370
 371 <!--=========================================================================-->
 372 <div class="doc_subsection">
 373 <a name="optimizer">Optimizer Improvements</a>
 374 </div>
 375
 376 <div class="doc_text">
 377
 378 <p>In addition to a large array of minor performance tweaks and bug fixes, this
 379 release includes a few major enhancements and additions to the optimizers:</p>
 380
 381 <ul>
 382
 383 <li></li>
 384
 385 </ul>
 386
 387 </div>
 388
 389
 390 <!--=========================================================================-->
 391 <div class="doc_subsection">
 392 <a name="executionengine">Interpreter and JIT Improvements</a>
 393 </div>
 394
 395 <div class="doc_text">
 396
 397 <ul>
 398 <li></li>
 399
 400 </ul>
 401
 402 </div>
 403
 404 <!--=========================================================================-->
 405 <div class="doc_subsection">
 406 <a name="codegen">Target Independent Code Generator Improvements</a>
 407 </div>
 408
 409 <div class="doc_text">
 410
 411 <p>We have put a significant amount of work into the code generator
 412 infrastructure, which allows us to implement more aggressive algorithms and make
 413 it run faster:</p>
 414
 415 <ul>
 416 <li>MachO writer works.</li>
 417 </ul>
 418 </div>
 419
 420 <!--=========================================================================-->
 421 <div class="doc_subsection">
 422 <a name="x86">X86-32 and X86-64 Target Improvements</a>
 423 </div>
 424
 425 <div class="doc_text">
 426 <p>New features of the X86 target include:
 427 </p>
 428
 429 <ul>
 430 <li>The X86 backend now supports holding X87 floating point stack values
 431     in registers across basic blocks, dramatically improving performance of code
 432     that uses long double, and when targetting CPUs that don't support SSE.</li>
 433
 434 </ul>
 435
 436 </div>
 437
 438 <!--=========================================================================-->
 439 <div class="doc_subsection">
 440 <a name="ARM">ARM Target Improvements</a>
 441 </div>
 442
 443 <div class="doc_text">
 444 <p>New features of the ARM target include:
 445 </p>
 446
 447 <ul>
 448
 449 <li>
 450   All of the NEON load and store intrinsics (llvm.arm.neon.vld* and
 451   llvm.arm.neon.vst*) take an extra parameter to specify the alignment in bytes
 452   of the memory being accessed.
 453 </li>
 454 <li>
 455   The llvm.arm.neon.vaba intrinsic (vector absolute difference and
 456   accumulate) has been removed.  This operation is now represented using
 457   the llvm.arm.neon.vabd intrinsic (vector absolute difference) followed by a
 458   vector add.
 459 </li>
 460 <li>
 461   The llvm.arm.neon.vabdl and llvm.arm.neon.vabal intrinsics (lengthening
 462   vector absolute difference with and without accumlation) have been removed.
 463   They are represented using the llvm.arm.neon.vabd intrinsic (vector absolute
 464   difference) followed by a vector zero-extend operation, and for vabal,
 465   a vector add.
 466 </li>
 467 <li>
 468   The llvm.arm.neon.vmovn intrinsic has been removed.  Calls of this intrinsic
 469   are now replaced by vector truncate operations.
 470 </li>
 471 <li>
 472   The llvm.arm.neon.vmovls and llvm.arm.neon.vmovlu intrinsics have been
 473   removed.  They are now represented as vector sign-extend (vmovls) and
 474   zero-extend (vmovlu) operations.
 475 </li>
 476 <li>
 477   The llvm.arm.neon.vaddl*, llvm.arm.neon.vaddw*, llvm.arm.neon.vsubl*, and
 478   llvm.arm.neon.vsubw* intrinsics (lengthening vector add and subtract) have
 479   been removed.  They are replaced by vector add and vector subtract operations
 480   where one (vaddw, vsubw) or both (vaddl, vsubl) of the operands are either
 481   sign-extended or zero-extended.
 482 </li>
 483 <li>
 484   The llvm.arm.neon.vmulls, llvm.arm.neon.vmullu, llvm.arm.neon.vmlal*, and
 485   llvm.arm.neon.vmlsl* intrinsics (lengthening vector multiply with and without
 486   accumulation and subtraction) have been removed.  These operations are now
 487   represented as vector multiplications where the operands are either
 488   sign-extended or zero-extended, followed by a vector add for vmlal or a
 489   vector subtract for vmlsl.  Note that the polynomial vector multiply
 490   intrinsic, llvm.arm.neon.vmullp, remains unchanged.
 491 </li>
 492
 493 </ul>
 494 </div>
 495
 496 <!--=========================================================================-->
 497 <div class="doc_subsection">
 498 <a name="newapis">New Useful APIs</a>
 499 </div>
 500
 501 <div class="doc_text">
 502
 503 <p>This release includes a number of new APIs that are used internally, which
 504    may also be useful for external clients.
 505 </p>
 506
 507 <ul>
 508 <li></li>
 509 </ul>
 510
 511
 512 </div>
 513
 514 <!--=========================================================================-->
 515 <div class="doc_subsection">
 516 <a name="otherimprovements">Other Improvements and New Features</a>
 517 </div>
 518
 519 <div class="doc_text">
 520 <p>Other miscellaneous features include:</p>
 521
 522 <ul>
 523 <li></li>
 524 </ul>
 525
 526 </div>
 527
 528
 529 <!--=========================================================================-->
 530 <div class="doc_subsection">
 531 <a name="changes">Major Changes and Removed Features</a>
 532 </div>
 533
 534 <div class="doc_text">
 535
 536 <p>If you're already an LLVM user or developer with out-of-tree changes based
 537 on LLVM 2.7, this section lists some "gotchas" that you may run into upgrading
 538 from the previous release.</p>
 539
 540 <ul>
 541 <li>.ll file doesn't produce #uses comments anymore, to get them, run a .bc file
 542    through "llvm-dis --show-annotations".</li>
 543 <li>MSIL Backend removed.</li>
 544 <li>ABCD and SSI passes removed.</li>
 545 <li>'Union' LLVM IR feature removed.</li>
 546 </ul>
 547
 548 <p>In addition, many APIs have changed in this release.  Some of the major LLVM
 549 API changes are:</p>
 550
 551 <ul>
 552 </ul>
 553
 554 </div>
 555
 556
 557
 558 <!-- *********************************************************************** -->
 559 <div class="doc_section">
 560   <a name="portability">Portability and Supported Platforms</a>
 561 </div>
 562 <!-- *********************************************************************** -->
 563
 564 <div class="doc_text">
 565
 566 <p>LLVM is known to work on the following platforms:</p>
 567
 568 <ul>
 569 <li>Intel and AMD machines (IA32, X86-64, AMD64, EMT-64) running Red Hat
 570     Linux, Fedora Core, FreeBSD and AuroraUX (and probably other unix-like
 571     systems).</li>
 572 <li>PowerPC and X86-based Mac OS X systems, running 10.4 and above in 32-bit
 573     and 64-bit modes.</li>
 574 <li>Intel and AMD machines running on Win32 using MinGW libraries (native).</li>
 575 <li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
 576     support is available for native builds with Visual C++).</li>
 577 <li>Sun x86 and AMD64 machines running Solaris 10, OpenSolaris 0906.</li>
 578 <li>Alpha-based machines running Debian GNU/Linux.</li>
 579 </ul>
 580
 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>
 585
 586 </div>
 587
 588 <!-- *********************************************************************** -->
 589 <div class="doc_section">
 590   <a name="knownproblems">Known Problems</a>
 591 </div>
 592 <!-- *********************************************************************** -->
 593
 594 <div class="doc_text">
 595
 596 <p>This section contains significant known problems with the LLVM system,
 597 listed by component.  If you run into a problem, please check the <a
 598 href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
 599 there isn't already one.</p>
 600
 601 <ul>
 602 <li>LLVM will not correctly compile on Solaris and/or OpenSolaris
 603 using the stock GCC 3.x.x series 'out the box',
 604 See: <a href="GettingStarted.html#brokengcc">Broken versions of GCC and other tools</a>.
 605 However, A <a href="http://pkg.auroraux.org/GCC">Modern GCC Build</a>
 606 for x86/x86-64 has been made available from the third party AuroraUX Project
 607 that has been meticulously tested for bootstrapping LLVM &amp; Clang.</li>
 608 <li>There have been reports of Solaris and/or OpenSolaris build failures due
 609 to an incompatibility in the nm program as well. The nm from binutils does seem
 610 to work.</li>
 611 </ul>
 612
 613 </div>
 614
 615 <!-- ======================================================================= -->
 616 <div class="doc_subsection">
 617   <a name="experimental">Experimental features included with this release</a>
 618 </div>
 619
 620 <div class="doc_text">
 621
 622 <p>The following components of this LLVM release are either untested, known to
 623 be broken or unreliable, or are in early development.  These components should
 624 not be relied on, and bugs should not be filed against them, but they may be
 625 useful to some people.  In particular, if you would like to work on one of these
 626 components, please contact us on the <a
 627 href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
 628
 629 <ul>
 630 <li>The Alpha, SPU, MIPS, PIC16, Blackfin, MSP430, SystemZ and MicroBlaze
 631     backends are experimental.</li>
 632 <li><tt>llc</tt> "<tt>-filetype=asm</tt>" (the default) is the only
 633     supported value for this option.  XXX Update me</li>
 634 </ul>
 635
 636 </div>
 637
 638 <!-- ======================================================================= -->
 639 <div class="doc_subsection">
 640   <a name="x86-be">Known problems with the X86 back-end</a>
 641 </div>
 642
 643 <div class="doc_text">
 644
 645 <ul>
 646   <li>The X86 backend does not yet support
 647     all <a href="http://llvm.org/PR879">inline assembly that uses the X86
 648     floating point stack</a>.  It supports the 'f' and 't' constraints, but not
 649     'u'.</li>
 650   <li>Win64 code generation wasn't widely tested. Everything should work, but we
 651     expect small issues to happen. Also, llvm-gcc cannot build the mingw64
 652     runtime currently due to lack of support for the 'u' inline assembly
 653     constraint and for X87 floating point inline assembly.</li>
 654   <li>The X86-64 backend does not yet support the LLVM IR instruction
 655       <tt>va_arg</tt>. Currently, front-ends support variadic
 656       argument constructs on X86-64 by lowering them manually.</li>
 657 </ul>
 658
 659 </div>
 660
 661 <!-- ======================================================================= -->
 662 <div class="doc_subsection">
 663   <a name="ppc-be">Known problems with the PowerPC back-end</a>
 664 </div>
 665
 666 <div class="doc_text">
 667
 668 <ul>
 669 <li>The Linux PPC32/ABI support needs testing for the interpreter and static
 670 compilation, and lacks support for debug information.</li>
 671 </ul>
 672
 673 </div>
 674
 675 <!-- ======================================================================= -->
 676 <div class="doc_subsection">
 677   <a name="arm-be">Known problems with the ARM back-end</a>
 678 </div>
 679
 680 <div class="doc_text">
 681
 682 <ul>
 683 <li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
 684 processors, thumb programs can crash or produce wrong
 685 results (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
 686 <li>Compilation for ARM Linux OABI (old ABI) is supported but not fully tested.
 687 </li>
 688 </ul>
 689
 690 </div>
 691
 692 <!-- ======================================================================= -->
 693 <div class="doc_subsection">
 694   <a name="sparc-be">Known problems with the SPARC back-end</a>
 695 </div>
 696
 697 <div class="doc_text">
 698
 699 <ul>
 700 <li>The SPARC backend only supports the 32-bit SPARC ABI (-m32); it does not
 701     support the 64-bit SPARC ABI (-m64).</li>
 702 </ul>
 703
 704 </div>
 705
 706 <!-- ======================================================================= -->
 707 <div class="doc_subsection">
 708   <a name="mips-be">Known problems with the MIPS back-end</a>
 709 </div>
 710
 711 <div class="doc_text">
 712
 713 <ul>
 714 <li>64-bit MIPS targets are not supported yet.</li>
 715 </ul>
 716
 717 </div>
 718
 719 <!-- ======================================================================= -->
 720 <div class="doc_subsection">
 721   <a name="alpha-be">Known problems with the Alpha back-end</a>
 722 </div>
 723
 724 <div class="doc_text">
 725
 726 <ul>
 727
 728 <li>On 21164s, some rare FP arithmetic sequences which may trap do not have the
 729 appropriate nops inserted to ensure restartability.</li>
 730
 731 </ul>
 732 </div>
 733
 734 <!-- ======================================================================= -->
 735 <div class="doc_subsection">
 736   <a name="c-be">Known problems with the C back-end</a>
 737 </div>
 738
 739 <div class="doc_text">
 740
 741 <ul>
 742 <li><a href="http://llvm.org/PR802">The C backend has only basic support for
 743     inline assembly code</a>.</li>
 744 <li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
 745     C++ programs</a>, preventing intermixing between C++ compiled by the CBE and
 746     C++ code compiled with <tt>llc</tt> or native compilers.</li>
 747 <li>The C backend does not support all exception handling constructs.</li>
 748 <li>The C backend does not support arbitrary precision integers.</li>
 749 </ul>
 750
 751 </div>
 752
 753
 754 <!-- ======================================================================= -->
 755 <div class="doc_subsection">
 756   <a name="c-fe">Known problems with the llvm-gcc C and C++ front-end</a>
 757 </div>
 758
 759 <div class="doc_text">
 760
 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>
 766
 767 </div>
 768
 769 <!-- ======================================================================= -->
 770 <div class="doc_subsection">
 771   <a name="fortran-fe">Known problems with the llvm-gcc Fortran front-end</a>
 772 </div>
 773
 774 <div class="doc_text">
 775 <ul>
 776 <li>Fortran support generally works, but there are still several unresolved bugs
 777     in <a href="http://llvm.org/bugs/">Bugzilla</a>.  Please see the
 778     tools/gfortran component for details.</li>
 779 </ul>
 780 </div>
 781
 782 <!-- ======================================================================= -->
 783 <div class="doc_subsection">
 784   <a name="ada-fe">Known problems with the llvm-gcc Ada front-end</a>
 785 </div>
 786
 787 <div class="doc_text">
 788 The llvm-gcc 4.2 Ada compiler works fairly well; however, this is not a mature
 789 technology, and problems should be expected.
 790 <ul>
 791 <li>The Ada front-end currently only builds on X86-32.  This is mainly due
 792 to lack of trampoline support (pointers to nested functions) on other platforms.
 793 However, it <a href="http://llvm.org/PR2006">also fails to build on X86-64</a>
 794 which does support trampolines.</li>
 795 <li>The Ada front-end <a href="http://llvm.org/PR2007">fails to bootstrap</a>.
 796 This is due to lack of LLVM support for <tt>setjmp</tt>/<tt>longjmp</tt> style
 797 exception handling, which is used internally by the compiler.
 798 Workaround: configure with <tt>--disable-bootstrap</tt>.</li>
 799 <li>The c380004, <a href="http://llvm.org/PR2010">c393010</a>
 800 and <a href="http://llvm.org/PR2421">cxg2021</a> ACATS tests fail
 801 (c380004 also fails with gcc-4.2 mainline).
 802 If the compiler is built with checks disabled then <a href="http://llvm.org/PR2010">c393010</a>
 803 causes the compiler to go into an infinite loop, using up all system memory.</li>
 804 <li>Some GCC specific Ada tests continue to crash the compiler.</li>
 805 <li>The <tt>-E</tt> binder option (exception backtraces)
 806 <a href="http://llvm.org/PR1982">does not work</a> and will result in programs
 807 crashing if an exception is raised.  Workaround: do not use <tt>-E</tt>.</li>
 808 <li>Only discrete types <a href="http://llvm.org/PR1981">are allowed to start
 809 or finish at a non-byte offset</a> in a record.  Workaround: do not pack records
 810 or use representation clauses that result in a field of a non-discrete type
 811 starting or finishing in the middle of a byte.</li>
 812 <li>The <tt>lli</tt> interpreter <a href="http://llvm.org/PR2009">considers
 813 'main' as generated by the Ada binder to be invalid</a>.
 814 Workaround: hand edit the file to use pointers for <tt>argv</tt> and
 815 <tt>envp</tt> rather than integers.</li>
 816 <li>The <tt>-fstack-check</tt> option <a href="http://llvm.org/PR2008">is
 817 ignored</a>.</li>
 818 </ul>
 819 </div>
 820
 821 <!-- *********************************************************************** -->
 822 <div class="doc_section">
 823   <a name="additionalinfo">Additional Information</a>
 824 </div>
 825 <!-- *********************************************************************** -->
 826
 827 <div class="doc_text">
 828
 829 <p>A wide variety of additional information is available on the <a
 830 href="http://llvm.org">LLVM web page</a>, in particular in the <a
 831 href="http://llvm.org/docs/">documentation</a> section.  The web page also
 832 contains versions of the API documentation which is up-to-date with the
 833 Subversion version of the source code.
 834 You can access versions of these documents specific to this release by going
 835 into the "<tt>llvm/doc/</tt>" directory in the LLVM tree.</p>
 836
 837 <p>If you have any questions or comments about LLVM, please feel free to contact
 838 us via the <a href="http://llvm.org/docs/#maillist"> mailing
 839 lists</a>.</p>
 840
 841 </div>
 842
 843 <!-- *********************************************************************** -->
 844
 845 <hr>
 846 <address>
 847   <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
 848   src="http://jigsaw.w3.org/css-validator/images/vcss-blue" alt="Valid CSS"></a>
 849   <a href="http://validator.w3.org/check/referer"><img
 850   src="http://www.w3.org/Icons/valid-html401-blue" alt="Valid HTML 4.01"></a>
 851
 852   <a href="http://llvm.org/">LLVM Compiler Infrastructure</a><br>
 853   Last modified: $Date$
 854 </address>
 855
 856 </body>
 857 </html>