<p>This document contains the release notes for the LLVM Compiler
Infrastructure, release 3.0. Here we describe the status of LLVM, including
- major improvements from the previous release and significant known problems.
+ major improvements from the previous release, improvements in various
+ subprojects of LLVM, and some of the current users of the code.
All LLVM releases may be downloaded from
the <a href="http://llvm.org/releases/">LLVM releases web site</a>.</p>
</div>
-<!-- Features that need text if they're finished for 3.1:
- ARM EHABI
- combiner-aa?
- strong phi elim
- loop dependence analysis
- CorrelatedValuePropagation
- lib/Transforms/IPO/MergeFunctions.cpp => consider for 3.1.
- -->
<!-- *********************************************************************** -->
<h2>
provides a modular, library-based architecture that makes it suitable for
creating or integrating with other development tools. Clang is considered a
production-quality compiler for C, Objective-C, C++ and Objective-C++ on x86
- (32- and 64-bit), and for darwin/arm targets.</p>
+ (32- and 64-bit), and for Darwin/ARM targets.</p>
<p>In the LLVM 3.0 time-frame, the Clang team has made many improvements:</p>
<li><a href="http://clang.llvm.org/cxx_status.html">Improved support</a> for
the <a href="http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=50372">C++
- 2011</a> standard, including implementations of non-static data member
- initializers, alias templates, delegating constructors, the range-based
- for loop, and implicitly-generated move constructors and move assignment
+ 2011</a> standard (aka "C++'0x"), including implementations of non-static data member
+ initializers, alias templates, delegating constructors, range-based
+ for loops, and implicitly-generated move constructors and move assignment
operators, among others.</li>
<li>Implemented support for some features of the upcoming C1x standard,
Automatic Reference Counting</a> (ARC) and an improved memory model
cleanly separating object and C memory.</li>
- <li>A migration tool for moving legacy code to ARC</li>
+ <li>A migration tool for moving manual retain/release code to ARC</li>
<li>Better support for data hiding, allowing instance variables to be
declared in implementation contexts or class extensions</li>
- <li>Weak linking for classes</li>
+ <li>Weak linking support for Objective-C classes</li>
<li>Improved static type checking by inferring the return type of methods
such as +alloc and -init.</li>
</ul>
- Some new features require either the Mac OS X 10.7 / iOS 5 Objective-C
- runtime, or version 1.6 or later of the GNUstep Objective-C runtime
- version.</li>
+ Some new Objective-C features require either the Mac OS X 10.7 / iOS 5
+ Objective-C runtime, or version 1.6 or later of the GNUstep Objective-C
+ runtime version.</li>
<li>Implemented a number of optimizations in <tt>libclang</tt>, the Clang C
interface, to improve the performance of code completion and the mapping
<p>The 3.0 release has the following notable changes:</p>
+ <ul>
<li>GCC version 4.6 is now fully supported.</li>
<li>Patching and building GCC is no longer required: the plugin should work
<li>The <tt>-fplugin-arg-dragonegg-enable-gcc-optzns</tt> option, which runs
GCC's optimizers as well as LLVM's, now works much better. This is the
- option to use if you want ultimate performance! It not yet completely
- stable: it may cause the plugin to crash.</li>
+ option to use if you want ultimate performance! It is still experimental
+ though: it may cause the plugin to crash.</li>
<li>The type and constant conversion logic has been almost entirely rewritten,
fixing a multitude of obscure bugs.</li>
-<ul>
-<!--
-<li></li>
--->
</ul>
</div>
implementations of this and other low-level routines (some are 3x faster than
the equivalent libgcc routines).</p>
-<p>In the LLVM 3.0 timeframe,</p>
+<p>In the LLVM 3.0 timeframe, the target specific ARM code has converted to
+ "unified" assembly syntax, and several new functions have been added to the
+ library.</p>
</div>
</h3>
<div>
+
+<p>LLDB is a ground-up implementation of a command line debugger, as well as a
+ debugger API that can be used from other applications. LLDB makes use of the
+ Clang parser to provide high-fidelity expression parsing (particularly for
+ C++) and uses the LLVM JIT for target support.</p>
<p>LLDB has advanced by leaps and bounds in the 3.0 timeframe. It is
dramatically more stable and useful, and includes both a
</div>
-
-<!--=========================================================================-->
-<h3>
-<a name="LLBrowse">LLBrowse: IR Browser</a>
-</h3>
-
-<div>
-
-<p><a href="http://llvm.org/svn/llvm-project/llbrowse/trunk/doc/LLBrowse.html">
- LLBrowse</a> is an interactive viewer for LLVM modules. It can load any LLVM
- module and displays its contents as an expandable tree view, facilitating an
- easy way to inspect types, functions, global variables, or metadata nodes. It
- is fully cross-platform, being based on the popular wxWidgets GUI
- toolkit.</p>
-
-</div>
-
<!--=========================================================================-->
<h3>
<a name="vmkit">VMKit</a>
</div>
-
+
+<!--=========================================================================-->
+<h3>
+<a name="LLBrowse">LLBrowse: IR Browser</a>
+</h3>
+
+<div>
+
+<p><a href="http://llvm.org/svn/llvm-project/llbrowse/trunk/doc/LLBrowse.html">
+ LLBrowse</a> is an interactive viewer for LLVM modules. It can load any LLVM
+ module and displays its contents as an expandable tree view, facilitating an
+ easy way to inspect types, functions, global variables, or metadata nodes. It
+ is fully cross-platform, being based on the popular wxWidgets GUI
+ toolkit.</p>
+
+</div>
+
+
<!--=========================================================================-->
<!--
<h3>
<p>Since version 0.96 it
has <a href="http://vrt-sourcefire.blogspot.com/2010/09/introduction-to-clamavs-low-level.html">bytecode
- signatures</a> that allow writing detections for complex malware.</p>
-
-<p>It uses LLVM's JIT to speed up the execution of bytecode on X86, X86-64,
+ signatures</a> that allow writing detections for complex malware.
+ It uses LLVM's JIT to speed up the execution of bytecode on X86, X86-64,
PPC32/64, falling back to its own interpreter otherwise. The git version was
updated to work with LLVM 3.0.</p>
<div>
<p><a href="http://cern.ch/cling">Cling</a> is an interactive compiler interface
- (aka C++ interpreter). It uses LLVM's JIT and clang; it currently supports
- C++ and C. It has a prompt interface, runs source files, calls into shared
+ (aka C++ interpreter). It supports C++ and C, and uses LLVM's JIT and the
+ Clang parser. It has a prompt interface, runs source files, calls into shared
libraries, prints the value of expressions, even does runtime lookup of
identifiers (dynamic scopes). And it just behaves like one would expect from
an interpreter.</p>
binary compatible with Microsoft.NET. Has an optional, dynamically-loaded
LLVM code generation backend in Mini, the JIT compiler.</p>
-<p>Note that we use a Git mirror of LLVM with some patches. See:
- https://github.com/mono/llvm</p>
+<p>Note that we use a Git mirror of LLVM <a
+ href="https://github.com/mono/llvm">with some patches</a>.</p>
</div>
<div>
-<p><b>llvm-gcc is gone</b></p>
+ <!-- Features that need text if they're finished for 3.1:
+ ARM EHABI
+ combiner-aa?
+ strong phi elim
+ loop dependence analysis
+ CorrelatedValuePropagation
+ lib/Transforms/IPO/MergeFunctions.cpp => consider for 3.1.
+ Integrated assembler on by default for arm/thumb?
-<p>LLVM 3.0 includes several major new capabilities:</p>
-
-<!-- Near dead:
+ -->
+
+ <!-- Near dead:
Analysis/RegionInfo.h + Dom Frontiers
SparseBitVector: used in LiveVar.
-
- -->
-
-<!--
- Type system rewrite: http://blog.llvm.org/2011/11/llvm-30-type-system-rewrite.html
- Better performance for Neon code in clang due to SRoA improvements.
- New regalloc on by default. Lin scan going away in 3.1
- PGO / builtin_expect improvements (summary needed)
- Big EH rewrite.
- AVX support, assembler, compiler and disassembler.
- IndVar improvements: andy
- PTX backend improvements: Justin
- llvm-rtdyld & MC JIT: JimG
- InstAliases now automatically used in the asmprinter where they are shorter.
- Integrated assembler on by default for arm/thumb?
- PostOrder Dominator frontiers were removed.
- Line Profiling / gcov support
- EH and debug information produced with CFI directives, yielding smaller executables: http://blog.mozilla.com/respindola/2011/05/12/cfi-directives/
- X86-64 generates smaller and faster code at -O0 (fast isel improvements)
- Better code generation for Cortex-A9
- Many APIs take ArrayRef's now.
- Pass manager extension API.
- ARM inline asm constraints implemented.
- LangRef.html#fnattrs uwtable attribute for asynch unwind tables.
- better performance for indirect gotos.
- llvm.prefetch now takes a 4th argument that specifies whether the prefetch happens to the icache or dcache.
- New PackedVector, TinyPtrVector class (see Programmer's Manual)
- New nonlazybind function attribute.
- ARC language specific optimizer (Transforms/ObjCARC) a decent example of language-specific transformation.
- LLVM 3.0 removes support for reading LLVM 2.8 and earlier files.
-
- New llvm.expect intrinsic.
- Table generated MC expansion logic for pseudo instructions that expand to multiple MC instructions through the PseudoInstExpansion class. (JimG)
- New llvm.fma intrinsic.
-
- Euro dev meeting and main one too.
- New atomics instructions, "#i_fence" instruction, cmpxchg, atomicrmw too. What target support?
- X86: inline assembler supports .code32 and .code64.
- Exception handling rewrite: new landingpad and resume instruction. Unwind gone.
- LowerSetJmp pass removed, unused.
- llvm-objdump (d0k) and object file parsing stuff (mspencer)
- Old arm disassembler replaced with a new one based on autogenerated encoding information from ARM .td files.
- -->
-
-<ul>
+ llvm/lib/Archive - replace with lib object?
+ -->
-<!--
-<li></li>
--->
+<p>LLVM 3.0 includes several major changes and big features:</p>
+
+<ul>
+<li>llvm-gcc is no longer supported, and not included in the release. We
+ recommend switching to <a
+ href="http://clang.llvm.org/">Clang</a> or <a
+ href="http://dragonegg.llvm.org/">DragonEgg</a>.</li>
+
+<li>The linear scan register allocator has been replaced with a new "greedy"
+ register allocator, enabling live range splitting and many other
+ optimizations that lead to better code quality. Please see its <a
+ href="http://blog.llvm.org/2011/09/greedy-register-allocation-in-llvm-30.html">blog post</a> or its talk at the <a
+ href="http://llvm.org/devmtg/2011-11/">Developer Meeting</a>
+ for more information.</li>
+<li>LLVM IR now includes full support for <a href="Atomics.html">atomics
+ memory operations</a> intended to support the C++'11 and C'1x memory models.
+ This includes <a href="LangRef.html#memoryops">atomic load and store,
+ compare and exchange, and read/modify/write instructions</a> as well as a
+ full set of <a href="LangRef.html#ordering">memory ordering constraints</a>.
+ Please see the <a href="Atomics.html">Atomics Guide</a> for more
+ information.
+</li>
+<li>The LLVM IR exception handling representation has been redesigned and
+ reimplemented, making it more elegant, fixing a huge number of bugs, and
+ enabling inlining and other optimizations. Please see its blog post (XXX
+ not yet) and the <a href="ExceptionHandling.html">Exception Handling
+ documentation</a> for more information.</li>
+<li>The LLVM IR Type system has been redesigned and reimplemented, making it
+ faster and solving some long-standing problems.
+ Please see its <a
+ href="http://blog.llvm.org/2011/11/llvm-30-type-system-rewrite.html">blog
+ post</a> for more information.</li>
-</ul>
+<li>The MIPS backend has made major leaps in this release, going from an
+ experimental target to being virtually production quality and supporting a
+ wide variety of MIPS subtargets. See the <a href="#MIPS">MIPS section</a>
+ below for more information.</li>
+<li>The optimizer and code generator now supports gprof and gcov-style coverage
+ and profiling information, and includes a new llvm-cov tool (but also works
+ with gcov). Clang exposes coverage and profiling through GCC-compatible
+ command line options.</li>
+</ul>
+
</div>
+
<!--=========================================================================-->
<h3>
<a name="coreimprovements">LLVM IR and Core Improvements</a>
<p>LLVM IR has several new features for better support of new targets and that
expose new optimization opportunities:</p>
-<p>One of the biggest changes is that 3.0 has a new exception handling
- system. The old system used LLVM intrinsics to convey the exception handling
- information to the code generator. It worked in most cases, but not
- all. Inlining was especially difficult to get right. Also, the intrinsics
- could be moved away from the <code>invoke</code> instruction, making it hard
- to recover that information.</p>
-
-<p>The new EH system makes exception handling a first-class member of the IR. It
- adds two new instructions:</p>
-
-<ul>
- <li><a href="LangRef.html#i_landingpad"><code>landingpad</code></a> —
- this instruction defines a landing pad basic block. It contains all of the
- information that's needed by the code generator. It's also required to be
- the first non-PHI instruction in the landing pad. In addition, a landing
- pad may be jumped to only by the unwind edge of an <code>invoke</code>
- instruction.</li>
-
- <li><a href="LangRef.html#i_resume"><code>resume</code></a> — this
- instruction causes the current exception to resume traveling up the
- stack. It replaces the <code>@llvm.eh.resume</code> intrinsic.</li>
-</ul>
-
-<p>Converting from the old EH API to the new EH API is rather simple, because a
- lot of complexity has been removed. The two intrinsics,
- <code>@llvm.eh.exception</code> and <code>@llvm.eh.selector</code> have been
- superseded by the <code>landingpad</code> instruction. Instead of generating
- a call to <code>@llvm.eh.exception</code> and <code>@llvm.eh.selector</code>:
-
-<div class="doc_code">
-<pre>
-Function *ExcIntr = Intrinsic::getDeclaration(TheModule,
- Intrinsic::eh_exception);
-Function *SlctrIntr = Intrinsic::getDeclaration(TheModule,
- Intrinsic::eh_selector);
-
-// The exception pointer.
-Value *ExnPtr = Builder.CreateCall(ExcIntr, "exc_ptr");
-
-std::vector<Value*> Args;
-Args.push_back(ExnPtr);
-Args.push_back(Builder.CreateBitCast(Personality,
- Type::getInt8PtrTy(Context)));
-
-<i>// Add selector clauses to Args.</i>
-
-// The selector call.
-Builder.CreateCall(SlctrIntr, Args, "exc_sel");
-</pre>
-</div>
-
-<p>You should instead generate a <code>landingpad</code> instruction, that
- returns an exception object and selector value:</p>
-
-<div class="doc_code">
-<pre>
-LandingPadInst *LPadInst =
- Builder.CreateLandingPad(StructType::get(Int8PtrTy, Int32Ty, NULL),
- Personality, 0);
-
-Value *LPadExn = Builder.CreateExtractValue(LPadInst, 0);
-Builder.CreateStore(LPadExn, getExceptionSlot());
-
-Value *LPadSel = Builder.CreateExtractValue(LPadInst, 1);
-Builder.CreateStore(LPadSel, getEHSelectorSlot());
-</pre>
-</div>
-
-<p>It's now trivial to add the individual clauses to the <code>landingpad</code>
- instruction.</p>
-
-<div class="doc_code">
-<pre>
-<i><b>// Adding a catch clause</b></i>
-Constant *TypeInfo = getTypeInfo();
-LPadInst->addClause(TypeInfo);
-
-<i><b>// Adding a C++ catch-all</b></i>
-LPadInst->addClause(Constant::getNullValue(Builder.getInt8PtrTy()));
-
-<i><b>// Adding a cleanup</b></i>
-LPadInst->setCleanup(true);
-
-<i><b>// Adding a filter clause</b></i>
-std::vector<Constant*> TypeInfos;
-Constant *TypeInfo = getFilterTypeInfo();
-TypeInfos.push_back(Builder.CreateBitCast(TypeInfo, Builder.getInt8PtrTy()));
-
-ArrayType *FilterTy = ArrayType::get(Int8PtrTy, TypeInfos.size());
-LPadInst->addClause(ConstantArray::get(FilterTy, TypeInfos));
-</pre>
-</div>
-
-<p>Converting from using the <code>@llvm.eh.resume</code> intrinsic to
- the <code>resume</code> instruction is trivial. It takes the exception
- pointer and exception selector values returned by
- the <code>landingpad</code> instruction:</p>
-
-<div class="doc_code">
-<pre>
-Type *UnwindDataTy = StructType::get(Builder.getInt8PtrTy(),
- Builder.getInt32Ty(), NULL);
-Value *UnwindData = UndefValue::get(UnwindDataTy);
-Value *ExcPtr = Builder.CreateLoad(getExceptionObjSlot());
-Value *ExcSel = Builder.CreateLoad(getExceptionSelSlot());
-UnwindData = Builder.CreateInsertValue(UnwindData, ExcPtr, 0, "exc_ptr");
-UnwindData = Builder.CreateInsertValue(UnwindData, ExcSel, 1, "exc_sel");
-Builder.CreateResume(UnwindData);
-</pre>
-</div>
-
-</div>
-
-<!--=========================================================================-->
-<h3>
-<a name="loopoptimization">Loop Optimization Improvements</a>
-</h3>
-
-<div>
-<p>The induction variable simplification pass in 3.0 only modifies
- induction variables when profitable. Sign and zero extension
- elimination, linear function test replacement, loop unrolling, and
- other simplifications that require induction variable analysis have
- been generalized so they no longer require loops to be rewritten in a
- typically suboptimal form prior to optimization. This new design
- preserves more IR level information, avoids undoing earlier loop
- optimizations (particularly hand-optimized loops), and no longer
- strongly depends on the code generator rewriting loops a second time
- in a now optimal form--an intractable problem.</p>
-
-<p>The original behavior can be restored with -mllvm -enable-iv-rewrite;
- however, support for this mode will be short lived. As such, bug
- reports should be filed for any significant performance regressions
- when moving from -mllvm -enable-iv-rewrite to the 3.0 default mode.</p>
+ <ul>
+ <li><a href="Atomics.html">Atomic memory accesses and memory ordering</a> are
+ now directly expressible in the IR.</li>
+ <li>A new <a href="LangRef.html#int_fma">llvm.fma intrinsic</a> directly
+ represents floating point multiply accumulate operations without an
+ intermediate rounding stage.</li>
+ <li>A new llvm.expect intrinsic (XXX not documented in langref) allows a
+ frontend to express expected control flow (and the __builtin_expect builtin
+ from GNU C).</li>
+ <li>The <a href="LangRef.html#int_prefetch">llvm.prefetch intrinsic</a> now
+ takes a 4th argument that specifies whether the prefetch happens from the
+ icache or dcache.</li>
+ <li>The new <a href="LangRef.html#uwtable">uwtable function attribute</a>
+ allows a frontend to control emission of unwind tables.</li>
+ <li>The new <a href="LangRef.html#fnattrs">nonlazybind function
+ attribute</a> allow optimization of Global Offset Table (GOT) accesses.</li>
+ <li>The new <a href="LangRef.html#returns_twice">returns_twice attribute</a>
+ allows better modeling of functions like setjmp.</li>
+ <li>The <a href="LangRef.html#datalayout">target datalayout</a> string can now
+ encode the natural alignment of the target's stack for better optimization.
+ </li>
+ </ul>
</div>
<!--=========================================================================-->
<div>
-<p>In addition to a large array of minor performance tweaks and bug fixes, this
+<p>In addition to many minor performance tweaks and bug fixes, this
release includes a few major enhancements and additions to the
optimizers:</p>
<ul>
+<li>The pass manager now has an extension API that allows front-ends and plugins
+ to insert their own optimizations in the well-known places in the standard
+ pass optimization pipeline.</li>
+
<li>Information about <a href="BranchWeightMetadata.html">branch probability</a>
and basic block frequency is now available within LLVM, based on a
combination of static branch prediction heuristics and
register spill placement and if-conversion, with additional optimizations
planned for future releases. The same framework is intended for eventual
use with profile-guided optimization.</li>
-</li>
+<li>The "-indvars" induction variable simplification pass only modifies
+ induction variables when profitable. Sign and zero extension
+ elimination, linear function test replacement, loop unrolling, and
+ other simplifications that require induction variable analysis have
+ been generalized so they no longer require loops to be rewritten into
+ canonical form prior to optimization. This new design
+ preserves more IR level information, avoids undoing earlier loop
+ optimizations (particularly hand-optimized loops), and no longer
+ requires the code generator to reconstruct loops into an optimal form -
+ an intractable problem.</li>
+
+<li>LLVM now includes a pass to optimize retain/release calls for the
+ <a href="http://clang.llvm.org/docs/AutomaticReferenceCounting.html">Automatic
+ Reference Counting</a> (ARC) Objective-C language feature (in
+ lib/Transforms/Scalar/ObjCARC.cpp). It is a decent example of implementing
+ a source-language-specific optimization in LLVM.</li>
+
</ul>
</div>
<p>The LLVM Machine Code (aka MC) subsystem was created to solve a number of
problems in the realm of assembly, disassembly, object file format handling,
and a number of other related areas that CPU instruction-set level tools work
- in.</p>
+ in. For more information, please see
+ the <a href="http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html">Intro
+ to the LLVM MC Project Blog Post</a>.</p>
<ul>
- <li>The ELF object streamers are much more full featured.</li>
- <li>Target dependent relocation handling has been refactored into the Targets.</li>
- <li>Early stage MC-JIT infrastructure has been implemented.</li>
+ <li>The MC layer has undergone significant refactoring to eliminate layering
+ violations that caused it to pull in the LLVM compiler backend code.</li>
+ <li>The ELF object file writers are much more full featured.</li>
+ <li>The integrated assembler now supports #line directives.</li>
+ <li>An early implementation of a JIT built on top of the MC framework (known
+ as MC-JIT) has been implemented and will eventually replace the old JIT.
+ It emits object files direct to memory and uses a runtime dynamic linker to
+ resolve references and drive lazy compilation. The MC-JIT enables much
+ greater code reuse between the JIT and the static compiler and provides
+ better integration with the platform ABI as a result.
+ </li>
+ <li>The assembly printer now makes uses of assemblers instruction aliases
+ (InstAliases) to print simplified mneumonics when possible.</li>
+ <li>TableGen can now autogenerate MC expansion logic for pseudo
+ instructions that expand to multiple MC instructions (through the
+ PseudoInstExpansion class).</li>
+
+ <li>XXX: llvm-objdump / dwarf parser library / llvm-dwarfdump (d0k)
+ object file parsing stuff and llvm-size (mspencer). Status?</li>
</ul>
-<p>The MC-JIT is a major new feature for MC, and will eventually grow to replace
-the current JIT implementation. It emits object files direct to memory and
-uses a runtime dynamic linker to resolve references and drive lazy compilation.
-The MC-JIT enables much greater code reuse between the JIT and the static
-compiler and provides better integration with the platform ABI as a result.</p>
-
-<p>For more information, please see
- the <a href="http://blog.llvm.org/2010/04/intro-to-llvm-mc-project.html">Intro
- to the LLVM MC Project Blog Post</a>.</p>
-
</div>
<!--=========================================================================-->
make it run faster:</p>
<ul>
-<!--
-<li></li>
--->
+<li>XXX: Segmented stacks.</li>
+<li>LLVM generates substantially better code for indirect gotos due to a new
+ tail duplication pass, which can be a substantial performance win for
+ interpreter loops that use them.</li>
+<li>Exception handling and debug information is now emitted with CFI directives,
+ yielding <a href="http://blog.mozilla.com/respindola/2011/05/12/cfi-directives/">much smaller executables</a> for some C++ applications.
+</li>
+
+<li>The code generator now supports vector "select" operations on vector
+ comparisons, turning them into various optimized code sequences (e.g.
+ using the SSE4/AVX "blend" instructions).</li>
+<li>XXX: Domain fixing pass is now target independent (ExecutionDepsFix pass). (Jakob)</li>
</ul>
</div>
<p>New features and major changes in the X86 target include:</p>
<ul>
- <li>The X86 backend, assembler and disassembler now completely support AVX.
- To enable it pass <code>-mavx</code> to the compiler.</li>
-
- <li>The X86 backend now supports
- all <a href="http://llvm.org/PR879">inline assembly that uses the X86
- floating point stack</a>.</li>
-
- <li>The CRC32 intrinsics have been renamed. The intrinsics were previously
- <code>@llvm.x86.sse42.crc32.[8|16|32]</code>
- and <code>@llvm.x86.sse42.crc64.[8|64]</code>. They have been renamed to
- <code>@llvm.x86.sse42.crc32.32.[8|16|32]</code> and
- <code>@llvm.x86.sse42.crc32.64.[8|64]</code>.</li>
-
+<li>The X86 backend, assembler and disassembler now have full support for AVX 1.
+ To enable it pass <code>-mavx</code> to the compiler. AVX2 implementation is
+ underway on mainline.</li>
+<li>The integrated assembler and disassembler now support a broad range of new
+ instructions including Atom, Ivy Bridge, <a
+ href="http://en.wikipedia.org/wiki/SSE4a">SSE4a/BMI</a> instructions, <a
+ href="http://en.wikipedia.org/wiki/RdRand">rdrand</a> and many others.</li>
+<li>The X86 backend now fully supports the <a href="http://llvm.org/PR879">X87
+ floating point stack inline assembly constraints</a>.</li>
+<li>The integrated assembler now supports the <tt>.code32</tt> and
+ <tt>.code64</tt> directives to switch between 32-bit and 64-bit
+ instructions.</li>
+<li>The X86 backend now synthesizes horizontal add/sub instructions from generic
+ vector code when the appropriate instructions are enabled.</li>
+<li>The X86-64 backend generates smaller and faster code at -O0 due to
+ improvements in fast instruction selection.</li>
+<li><a href="http://code.google.com/p/nativeclient/">Native Client</a>
+ subtarget support has been added.</li>
+
+<li>The CRC32 intrinsics have been renamed. The intrinsics were previously
+ <code>@llvm.x86.sse42.crc32.[8|16|32]</code>
+ and <code>@llvm.x86.sse42.crc64.[8|64]</code>. They have been renamed to
+ <code>@llvm.x86.sse42.crc32.32.[8|16|32]</code> and
+ <code>@llvm.x86.sse42.crc32.64.[8|64]</code>.</li>
</ul>
</div>
<p>New features of the ARM target include:</p>
<ul>
- <li>Reworked Set Jump Long Jump EH Lowering,</li>
- <li>improved support for Cortex-M series processors, and</li>
- <li>beta quality integrated assembler support.</li>
+<li>The ARM backend generates much faster code for Cortex-A9 chips.</li>
+<li>The ARM backend has improved support for Cortex-M series processors.</li>
+<li>The ARM inline assembly constraints have been implemented and are now fully
+ supported.</li>
+<li>NEON code produced by Clang often runs much faster due to improvements in
+ the Scalar Replacement of Aggregates pass.</li>
+<li>The old ARM disassembler is replaced with a new one based on autogenerated
+ encoding information from ARM .td files.</li>
+<li>The integrated assembler has made major leaps forward, but is still beta quality in LLVM 3.0.</li>
</ul>
</div>
<div>
-<p>New features and major changes in the MIPS target include:</p>
+<p>This release has seen major new work on just about every aspect of the MIPS
+ backend. Some of the major new features include:</p>
<ul>
<li>Most MIPS32r1 and r2 instructions are now supported.</li>
from the previous release.</p>
<ul>
- <li>The <code>LLVMC</code> front end code was removed while separating
- out language independence.</li>
+ <li>LLVM's configure script doesn't depend on llvm-gcc anymore, clean layering.</p>
+ LLVM 3.0 removes support for reading LLVM 2.8 and earlier files. Aim to maintain compatibility all the way back to 3.0 "forever".
+ Frontend tests removed from llvm/test/Frontend*
+ Tablegen has been split into a library, clang tblgen pieces now live in clang.
+ The llvm version is now named llvm-tblgen instead of tblgen.
+ PostOrder Dominator frontiers were removed.
+ LowerSetJmp pass removed, unused.
+
+
+ <li>The <code>LLVMC</code> meta compiler driver was removed.</li>
<li>The <code>LowerSetJmp</code> pass wasn't used effectively by any
target and has been removed.</li>
<li>The old <code>TailDup</code> pass was not used in the standard pipeline
<div>
-<p>This section contains significant known problems with the LLVM system, listed
- by component. If you run into a problem, please check
- the <a href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
- there isn't already one.</p>
-
-<!-- ======================================================================= -->
-<h3>
- <a name="experimental">Experimental features included with this release</a>
-</h3>
-
-<div>
-
-<p>The following components of this LLVM release are either untested, known to
- be broken or unreliable, or are in early development. These components
- should not be relied on, and bugs should not be filed against them, but they
- may be useful to some people. In particular, if you would like to work on
- one of these components, please contact us on
- the <a href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev
- list</a>.</p>
-
+<p>LLVM is generally a production quality compiler, and is used by a broad range
+ of applications and shipping in many products. That said, not every
+ subsystem is as mature as the aggregate, particularly the more obscure
+ targets. If you run into a problem, please check the <a
+ href="http://llvm.org/bugs/">LLVM bug database</a> and submit a bug if
+ there isn't already one or ask on the <a
+ href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev
+ list</a>.</p>
+
+ <p>Known problem areas include:</p>
+
<ul>
- <li>The Alpha, Blackfin, CellSPU, MicroBlaze, MSP430, MIPS, PTX, SystemZ and
- XCore backends are experimental.</li>
-
- <li><tt>llc</tt> "<tt>-filetype=obj</tt>" is experimental on all targets other
- than darwin and ELF X86 systems.</li>
+ <li>The Alpha, Blackfin, CellSPU, MSP430, PTX, SystemZ and
+ XCore backends are experimental, and the Alpha, Blackfin and SystemZ
+ targets have already been removed from mainline.</li>
+
+ <li>The integrated assembler, disassembler, and JIT is not supported by
+ several targets. If an integrated assembler is not supported, then a
+ system assembler is required. For more details, see the <a
+ href="CodeGenerator.html#targetfeatures">Target Features Matrix</a>.
+ </li>
+
+ <li>The C backend has numerous problems and is not being actively maintained.
+ Depending on it for anything serious is not advised.</li>
</ul>
</div>
-<!-- ======================================================================= -->
-<h3>
- <a name="x86-be">Known problems with the X86 back-end</a>
-</h3>
-
-<div>
-
-<ul>
- <li>The X86-64 backend <a href="http://llvm.org/PR1740">does not yet support
- the <tt>va_arg</tt> LLVM IR instruction</a>. Currently, front-ends support
- variadic argument constructs on X86-64 by lowering them manually.</li>
-</ul>
-
</div>
-<!-- ======================================================================= -->
-<h3>
- <a name="ppc-be">Known problems with the PowerPC back-end</a>
-</h3>
+<!-- *********************************************************************** -->
+<h2>
+ <a name="additionalinfo">Additional Information</a>
+</h2>
+<!-- *********************************************************************** -->
<div>
-<ul>
- <li>The PPC32/ELF support lacks PIC support.</li>
-</ul>
-
-</div>
-
-<!-- ======================================================================= -->
-<h3>
- <a name="arm-be">Known problems with the ARM back-end</a>
-</h3>
+<p>A wide variety of additional information is available on
+ the <a href="http://llvm.org/">LLVM web page</a>, in particular in
+ the <a href="http://llvm.org/docs/">documentation</a> section. The web page
+ also contains versions of the API documentation which is up-to-date with the
+ Subversion version of the source code. You can access versions of these
+ documents specific to this release by going into the "<tt>llvm/doc/</tt>"
+ directory in the LLVM tree.</p>
-<div>
+<p>If you have any questions or comments about LLVM, please feel free to contact
+ us via the <a href="http://llvm.org/docs/#maillist"> mailing lists</a>.</p>
-<ul>
- <li>Thumb mode works only on ARMv6 or higher processors. On sub-ARMv6
- processors, thumb programs can crash or produce wrong results
- (<a href="http://llvm.org/PR1388">PR1388</a>).</li>
+</div>
- <li>Compilation for ARM Linux OABI (old ABI) is supported but not fully
- tested.</li>
-</ul>
+<!--=========================================================================-->
-</div>
+<!-- EH details: to be moved to a blog post:
+
-<!-- ======================================================================= -->
-<h3>
- <a name="sparc-be">Known problems with the SPARC back-end</a>
-</h3>
+
+
+<p>One of the biggest changes is that 3.0 has a new exception handling
+ system. The old system used LLVM intrinsics to convey the exception handling
+ information to the code generator. It worked in most cases, but not
+ all. Inlining was especially difficult to get right. Also, the intrinsics
+ could be moved away from the <code>invoke</code> instruction, making it hard
+ to recover that information.</p>
-<div>
+<p>The new EH system makes exception handling a first-class member of the IR. It
+ adds two new instructions:</p>
<ul>
- <li>The SPARC backend only supports the 32-bit SPARC ABI (-m32); it does not
- support the 64-bit SPARC ABI (-m64).</li>
+ <li><a href="LangRef.html#i_landingpad"><code>landingpad</code></a> —
+ this instruction defines a landing pad basic block. It contains all of the
+ information that's needed by the code generator. It's also required to be
+ the first non-PHI instruction in the landing pad. In addition, a landing
+ pad may be jumped to only by the unwind edge of an <code>invoke</code>
+ instruction.</li>
+
+ <li><a href="LangRef.html#i_resume"><code>resume</code></a> — this
+ instruction causes the current exception to resume traveling up the
+ stack. It replaces the <code>@llvm.eh.resume</code> intrinsic.</li>
</ul>
-</div>
+<p>Converting from the old EH API to the new EH API is rather simple, because a
+ lot of complexity has been removed. The two intrinsics,
+ <code>@llvm.eh.exception</code> and <code>@llvm.eh.selector</code> have been
+ superseded by the <code>landingpad</code> instruction. Instead of generating
+ a call to <code>@llvm.eh.exception</code> and <code>@llvm.eh.selector</code>:
-<!-- ======================================================================= -->
-<h3>
- <a name="mips-be">Known problems with the MIPS back-end</a>
-</h3>
+<div class="doc_code">
+<pre>
+Function *ExcIntr = Intrinsic::getDeclaration(TheModule,
+ Intrinsic::eh_exception);
+Function *SlctrIntr = Intrinsic::getDeclaration(TheModule,
+ Intrinsic::eh_selector);
-<div>
+// The exception pointer.
+Value *ExnPtr = Builder.CreateCall(ExcIntr, "exc_ptr");
-<ul>
- <li>64-bit MIPS targets are not supported yet.</li>
-</ul>
+std::vector<Value*> Args;
+Args.push_back(ExnPtr);
+Args.push_back(Builder.CreateBitCast(Personality,
+ Type::getInt8PtrTy(Context)));
+<i>// Add selector clauses to Args.</i>
+
+// The selector call.
+Builder.CreateCall(SlctrIntr, Args, "exc_sel");
+</pre>
</div>
-<!-- ======================================================================= -->
-<h3>
- <a name="alpha-be">Known problems with the Alpha back-end</a>
-</h3>
+<p>You should instead generate a <code>landingpad</code> instruction, that
+ returns an exception object and selector value:</p>
-<div>
+<div class="doc_code">
+<pre>
+LandingPadInst *LPadInst =
+ Builder.CreateLandingPad(StructType::get(Int8PtrTy, Int32Ty, NULL),
+ Personality, 0);
-<ul>
- <li>On 21164s, some rare FP arithmetic sequences which may trap do not have
- the appropriate nops inserted to ensure restartability.</li>
-</ul>
+Value *LPadExn = Builder.CreateExtractValue(LPadInst, 0);
+Builder.CreateStore(LPadExn, getExceptionSlot());
+Value *LPadSel = Builder.CreateExtractValue(LPadInst, 1);
+Builder.CreateStore(LPadSel, getEHSelectorSlot());
+</pre>
</div>
-<!-- ======================================================================= -->
-<h3>
- <a name="c-be">Known problems with the C back-end</a>
-</h3>
-
-<div>
-
-<p>The C backend has numerous problems and is not being actively maintained.
- Depending on it for anything serious is not advised.</p>
-
-<ul>
- <li><a href="http://llvm.org/PR802">The C backend has only basic support for
- inline assembly code</a>.</li>
+<p>It's now trivial to add the individual clauses to the <code>landingpad</code>
+ instruction.</p>
- <li><a href="http://llvm.org/PR1658">The C backend violates the ABI of common
- C++ programs</a>, preventing intermixing between C++ compiled by the CBE
- and C++ code compiled with <tt>llc</tt> or native compilers.</li>
+<div class="doc_code">
+<pre>
+<i><b>// Adding a catch clause</b></i>
+Constant *TypeInfo = getTypeInfo();
+LPadInst->addClause(TypeInfo);
- <li>The C backend does not support all exception handling constructs.</li>
+<i><b>// Adding a C++ catch-all</b></i>
+LPadInst->addClause(Constant::getNullValue(Builder.getInt8PtrTy()));
- <li>The C backend does not support arbitrary precision integers.</li>
-</ul>
+<i><b>// Adding a cleanup</b></i>
+LPadInst->setCleanup(true);
-</div>
+<i><b>// Adding a filter clause</b></i>
+std::vector<Constant*> TypeInfos;
+Constant *TypeInfo = getFilterTypeInfo();
+TypeInfos.push_back(Builder.CreateBitCast(TypeInfo, Builder.getInt8PtrTy()));
+ArrayType *FilterTy = ArrayType::get(Int8PtrTy, TypeInfos.size());
+LPadInst->addClause(ConstantArray::get(FilterTy, TypeInfos));
+</pre>
</div>
-<!-- *********************************************************************** -->
-<h2>
- <a name="additionalinfo">Additional Information</a>
-</h2>
-<!-- *********************************************************************** -->
-
-<div>
+<p>Converting from using the <code>@llvm.eh.resume</code> intrinsic to
+ the <code>resume</code> instruction is trivial. It takes the exception
+ pointer and exception selector values returned by
+ the <code>landingpad</code> instruction:</p>
-<p>A wide variety of additional information is available on
- the <a href="http://llvm.org/">LLVM web page</a>, in particular in
- the <a href="http://llvm.org/docs/">documentation</a> section. The web page
- also contains versions of the API documentation which is up-to-date with the
- Subversion version of the source code. You can access versions of these
- documents specific to this release by going into the "<tt>llvm/doc/</tt>"
- directory in the LLVM tree.</p>
+<div class="doc_code">
+<pre>
+Type *UnwindDataTy = StructType::get(Builder.getInt8PtrTy(),
+ Builder.getInt32Ty(), NULL);
+Value *UnwindData = UndefValue::get(UnwindDataTy);
+Value *ExcPtr = Builder.CreateLoad(getExceptionObjSlot());
+Value *ExcSel = Builder.CreateLoad(getExceptionSelSlot());
+UnwindData = Builder.CreateInsertValue(UnwindData, ExcPtr, 0, "exc_ptr");
+UnwindData = Builder.CreateInsertValue(UnwindData, ExcSel, 1, "exc_sel");
+Builder.CreateResume(UnwindData);
+</pre>
+</div>
-<p>If you have any questions or comments about LLVM, please feel free to contact
- us via the <a href="http://llvm.org/docs/#maillist"> mailing lists</a>.</p>
-</div>
+
+
+ -->
+
<!-- *********************************************************************** -->
<hr>
projects / oota-llvm.git / blobdiff