Logo web page.
llvm devmtg
compiler_rt
- klee web page at klee.llvm.org
+ KLEE web page at klee.llvm.org
Many new papers added to /pubs/
Mention gcc plugin.
<div class="doc_text">
<p>
The LLVM 2.6 distribution currently consists of code from the core LLVM
-repository —which roughly includes the LLVM optimizers, code generators
-and supporting tools — and the llvm-gcc repository. In addition to this
-code, the LLVM Project includes other sub-projects that are in development. The
-two which are the most actively developed are the <a href="#clang">Clang
-Project</a> and the <a href="#vmkit">VMKit Project</a>.
+repository (which roughly includes the LLVM optimizers, code generators
+and supporting tools), the Clang repository and the llvm-gcc repository. In
+addition to this code, the LLVM Project includes other sub-projects that are in
+development. Here we include updates on these subprojects.
</p>
</div>
<p>The <a href="http://clang.llvm.org/">Clang project</a> is an effort to build
a set of new 'LLVM native' front-end technologies for the C family of languages.
LLVM 2.6 is the first release to officially include Clang, and it provides a
-production quality C and Objective-C compiler. If you are interested in fast
-compiles and good diagnostics, we encourage you to try it out.</p>
+production quality C and Objective-C compiler. If you are interested in <a
+href="http://clang.llvm.org/performance.html">fast compiles</a> and
+<a href="http://clang.llvm.org/diagnostics.html">good diagnostics</a>, we
+encourage you to try it out. Clang currently compiles typical Objective-C code
+3x faster than GCC and compiles C code about 30% faster than GCC at -O0 -g
+(which is when the most pressure is on the frontend).</p>
<p>In addition to supporting these languages, C++ support is also <a
href="http://clang.llvm.org/cxx_status.html">well under way</a>, and mainline
<ul>
<li>C and Objective-C support are now considered production quality.</li>
-<li>AuroraUX / FreeBSD & OpenBSD Toolchain support.</li>
+<li>AuroraUX, FreeBSD and OpenBSD are now supported.</li>
<li>Most of Objective-C 2.0 is now supported with the GNU runtime.</li>
-<li>Many many bugs are fixed and many features have been added.</li>
+<li>Many many bugs are fixed and lots of features have been added.</li>
</ul>
</div>
<div class="doc_text">
-<p><b>UPDATE!</b> Previously announced in the 2.4 and 2.5 LLVM releases, the Clang project also
+<p>Previously announced in the 2.4 and 2.5 LLVM releases, the Clang project also
includes an early stage static source code analysis tool for <a
href="http://clang.llvm.org/StaticAnalysis.html">automatically finding bugs</a>
-in C and Objective-C programs. The tool performs a growing set of checks to find
+in C and Objective-C programs. The tool performs checks to find
bugs that occur on a specific path within a program.</p>
-<p>In the LLVM 2.6 time-frame there have been many significant improvements to
-XYZ.</p>
+<p>In the LLVM 2.6 time-frame, the analyzer core has undergone several important
+improvements and cleanups and now includes a new <em>Checker</em> interface that
+is intended to eventually serve as a basis for domain-specific checks. Further,
+in addition to generating HTML files for reporting analysis results, the
+analyzer can now also emit bug reports in a structured XML format that is
+intended to be easily readable by other programs.</p>
<p>The set of checks performed by the static analyzer continues to expand, and
future plans for the tool include full source-level inter-procedural analysis
<div class="doc_text">
<p>
-<b>UPDATE!</b>
The <a href="http://vmkit.llvm.org/">VMKit project</a> is an implementation of
-a JVM and a CLI Virtual Machines (Microsoft .NET is an
-implementation of the CLI) using the Just-In-Time compiler of LLVM.</p>
+a JVM and a CLI Virtual Machine (Microsoft .NET is an
+implementation of the CLI) using LLVM for static and just-in-time
+compilation.</p>
-<p>Following LLVM 2.6, VMKit has its XYZ release that you can find on its
-<a href="http://vmkit.llvm.org/releases/">webpage</a>. The release includes
+<p>
+VMKit version 0.26 builds with LLVM 2.6 and you can find it on its
+<a href="http://vmkit.llvm.org/releases/">web page</a>. The release includes
bug fixes, cleanup and new features. The major changes are:</p>
<ul>
-<li>Something wonderful!</li>
+<li>A new llcj tool to generate shared libraries or executables of Java
+ files.</li>
+<li>Cooperative garbage collection. </li>
+<li>Fast subtype checking (paper from Click et al [JGI'02]). </li>
+<li>Implementation of a two-word header for Java objects instead of the original
+ three-word header. </li>
+<li>Better Java specification-compliance: division by zero checks, stack
+ overflow checks, finalization and references support. </li>
</ul>
</div>
is a simple library that provides an implementation of the low-level
target-specific hooks required by code generation and other runtime components.
For example, when compiling for a 32-bit target, converting a double to a 64-bit
-unsigned integer is compiling into a runtime call to the "__fixunsdfdi"
-function. The compiler-rt library provides optimized implementations of this and
-other low-level routines.</p>
+unsigned integer is compiled into a runtime call to the "__fixunsdfdi"
+function. The compiler-rt library provides highly optimized implementations of
+this and other low-level routines (some are 3x faster than the equivalent
+libgcc routines).</p>
<p>
All of the code in the compiler-rt project is available under the standard LLVM
<!--=========================================================================-->
<div class="doc_subsection">
-<a name="klee">klee: Symbolic Execution and Automatic Test Case Generator</a>
+<a name="klee">KLEE: Symbolic Execution and Automatic Test Case Generator</a>
</div>
<div class="doc_text">
<p>
-The new LLVM <a href="http://klee.llvm.org/">klee project</a> is a symbolic
-execution framework for programs in LLVM bitcode form. Klee tries to
+The new LLVM <a href="http://klee.llvm.org/">KLEE project</a> is a symbolic
+execution framework for programs in LLVM bitcode form. KLEE tries to
symbolically evaluate "all" paths through the application and records state
transitions that lead to fault states. This allows it to construct testcases
that lead to faults and can even be used to verify algorithms. For more
details, please see the <a
href="http://llvm.org/pubs/2008-12-OSDI-KLEE.html">OSDI 2008 paper</a> about
-Klee.</p>
+KLEE.</p>
</div>
<!--=========================================================================-->
<div class="doc_subsection">
-<a name="dragonegg">Dragon Egg: An LLVM backend plugin for GCC</a>
+<a name="dragonegg">DragonEgg: GCC-4.5 as an LLVM frontend</a>
</div>
<div class="doc_text">
<p>
-<b>Duncan needs to write me</b>.
+The goal of <a href="http://dragonegg.llvm.org/">DragonEgg</a> is to make
+gcc-4.5 act like llvm-gcc without requiring any gcc modifications whatsoever.
+DragonEgg is a shared library (llvm.so) that is loaded by gcc at runtime. It
+uses the new gcc plugin architecture to disable the GCC optimizers and code
+generators, and schedule the LLVM optimizers and code generators (or direct
+output of LLVM IR) instead. Currently only Linux and Darwin are supported,
+and only on x86-32 and x86-64. It should be easy to add additional unix-like
+architectures and other processor families. Since gcc-4.5 has not yet been
+released, neither has <a href="http://dragonegg.llvm.org/">DragonEgg</a>.
+To build <a href="http://dragonegg.llvm.org/">DragonEgg</a> you will need to
+check out the development versions of <a href="http://gcc.gnu.org/svn.html/">
+gcc</a>, <a href="http://llvm.org/docs/GettingStarted.html#checkout">llvm</a>
+and <a href="http://dragonegg.llvm.org/">DragonEgg</a> from their respective
+subversion repositories.
</p>
</div>
</div>
<!-- *********************************************************************** -->
+<div class="doc_text">
+
+<p>An exciting aspect of LLVM is that it is used as an enabling technology for
+ a lot of other language and tools projects. This section lists some of the
+ projects that have already been updated to work with LLVM 2.6.</p>
+</div>
+
+
<!--=========================================================================-->
<div class="doc_subsection">
<a name="Rubinius">Rubinius</a>
remove dynamism from ruby execution and increase performance.</p>
<p>Since LLVM 2.5, Rubinius has made several major leaps forward, implementing
-a counter based JIT, type feedback, and speculative method inlining.
+a counter based JIT, type feedback and speculative method inlining.
</p>
</div>
<p>
<a href="http://macruby.org">MacRuby</a> is an implementation of Ruby on top of
core Mac OS X technologies, such as the Objective-C common runtime and garbage
-collector, and the CoreFoundation framework. It is principally developed by
+collector and the CoreFoundation framework. It is principally developed by
Apple and aims at enabling the creation of full-fledged Mac OS X applications.
</p>
The LDC project works great with the LLVM 2.6 release. General improvements in
this
cycle have included new inline asm constraint handling, better debug info
-support, general bugfixes, and better x86-64 support. This has allowed
-some major improvements in LDC, getting us much closer to being as
+support, general bug fixes and better x86-64 support. This has allowed
+some major improvements in LDC, getting it much closer to being as
fully featured as the original DMD compiler from DigitalMars.
</p>
</div>
<p>
<a href="http://code.roadsend.com/rphp">Roadsend PHP</a> (rphp) is an open
source implementation of the PHP programming
-language that uses LLVM for its optimizer, JIT, and static compiler. This is a
+language that uses LLVM for its optimizer, JIT and static compiler. This is a
reimplementation of an earlier project that is now based on LLVM.</p>
</div>
<div class="doc_text">
<p>
<a href="http://code.google.com/p/llvm-lua/">LLVM-Lua</a> uses LLVM to add JIT
-& static compiling support to the Lua VM. Lua bytecode is analyzed to
-remove type checks, then LLVM is used to compile those bytecodes down to machine
+and static compiling support to the Lua VM. Lua bytecode is analyzed to
+remove type checks, then LLVM is used to compile the bytecode down to machine
code.</p>
</div>
<div class="doc_text">
-<p>This release includes a huge number of bug fixes, performance tweaks, and
+<p>This release includes a huge number of bug fixes, performance tweaks and
minor improvements. Some of the major improvements and new features are listed
in this section.
</p>
<p>LLVM 2.6 includes several major new capabilities:</p>
<ul>
-<li>New <a href="#compiler-rt">compiler-rt</a>, <A href="#klee">klee</a>,
+<li>New <a href="#compiler-rt">compiler-rt</a>, <A href="#klee">KLEE</a>
and <a href="#mc">machine code toolkit</a> sub-projects.</li>
<li>Debug information now includes line numbers when optimizations are enabled.
- This allows statistical sampling tools like oprofile and Shark to map
+ This allows statistical sampling tools like OProfile and Shark to map
samples back to source lines.</li>
-<li>LLVM now includes new experimental backends to support the MSP430, SystemZ,
+<li>LLVM now includes new experimental backends to support the MSP430, SystemZ
and BlackFin architectures.</li>
<li>LLVM supports a new <a href="GoldPlugin.html">Gold Linker Plugin</a> which
enables support for <a href="LinkTimeOptimization.html">transparent
link-time optimization</a> on ELF targets when used with the Gold binutils
linker.</li>
-<li>LLVM now supports doing optimization and code generation on multiple threads
- by allowing multiple "LLVMContext" objects to exist. Please see the <a
- href="ProgrammersManual.html#threading">threading entry in the Programmer's
- Manual</a> for more information.</li>
+<li>LLVM now supports doing optimization and code generation on multiple
+ threads. Please see the <a href="ProgrammersManual.html#threading">LLVM
+ Programmer's Manual</a> for more information.</li>
<li>LLVM now has experimental support for <a
href="http://nondot.org/~sabre/LLVMNotes/EmbeddedMetadata.txt">embedded
metadata</a> in LLVM IR, though the implementation is not guaranteed to be
final and the .bc file format may change in future releases. Debug info
- does not yet use this format in LLVM 2.6.</p>
-
+ does not yet use this format in LLVM 2.6.</li>
</ul>
</div>
</div>
<div class="doc_text">
-<p>LLVM IR has several new features that are used by our existing front-ends and
-can be useful if you are writing a front-end for LLVM:</p>
+<p>LLVM IR has several new features for better support of new targets and that
+expose new optimization opportunities:</p>
<ul>
-<li>Getelementpr instruction now allows any integer type for array/pointer indexes.</li>
-<li>Inbounds for GEP</li>
-<li>NSW/NUW/exact div</li>
-<li>LSR promotes int induction variables to 64-bit on 64-bit targets, major perf boost for numerical code.</li>
-<li>LSR now analyzes pointer expressions (e.g. getelementptrs), not just integers.</li>
-<li>new linkage types linkonce_odr, weak_odr, linker_private, and available_externally.</li>
-<li>New fadd, fsub, fmul instructions and classes. </li>
-<li>Target intrinsics can now return multiple results.</li>
+<li>The <a href="LangRef.html#i_add">add</a>, <a
+ href="LangRef.html#i_sub">sub</a> and <a href="LangRef.html#i_mul">mul</a>
+ instructions have been split into integer and floating point versions (like
+ divide and remainder), introducing new <a
+ href="LangRef.html#i_fadd">fadd</a>, <a href="LangRef.html#i_fsub">fsub</a>,
+ and <a href="LangRef.html#i_fmul">fmul</a> instructions.</li>
+<li>The <a href="LangRef.html#i_add">add</a>, <a
+ href="LangRef.html#i_sub">sub</a> and <a href="LangRef.html#i_mul">mul</a>
+ instructions now support optional "nsw" and "nuw" bits which indicate that
+ the operation is guaranteed to not overflow (in the signed or
+ unsigned case, respectively). This gives the optimizer more information and
+ can be used for things like C signed integer values, which are undefined on
+ overflow.</li>
+<li>The <a href="LangRef.html#i_sdiv">sdiv</a> instruction now supports an
+ optional "exact" flag which indicates that the result of the division is
+ guaranteed to have a remainder of zero. This is useful for optimizing pointer
+ subtraction in C.</li>
+<li>The <a href="LangRef.html#i_getelementptr">getelementptr</a> instruction now
+ supports arbitrary integer index values for array/pointer indices. This
+ allows for better code generation on 16-bit targets like PIC16.</li>
+<li>The <a href="LangRef.html#i_getelementptr">getelementptr</a> instruction now
+ supports an "inbounds" optimization hint that tells the optimizer that the
+ pointer is guaranteed to be within its allocated object.</li>
+<li>LLVM now support a series of new linkage types for global values which allow
+ for better optimization and new capabilities:
+ <ul>
+ <li><a href="LangRef.html#linkage_linkonce">linkonce_odr</a> and
+ <a href="LangRef.html#linkage_weak">weak_odr</a> have the same linkage
+ semantics as the non-"odr" linkage types. The difference is that these
+ linkage types indicate that all definitions of the specified function
+ are guaranteed to have the same semantics. This allows inlining
+ templates functions in C++ but not inlining weak functions in C,
+ which previously both got the same linkage type.</li>
+ <li><a href="LangRef.html#linkage_available_externally">available_externally
+ </a> is a new linkage type that gives the optimizer visibility into the
+ definition of a function (allowing inlining and side effect analysis)
+ but that does not cause code to be generated. This allows better
+ optimization of "GNU inline" functions, extern templates, etc.</li>
+ <li><a href="LangRef.html#linkage_linker_private">linker_private</a> is a
+ new linkage type (which is only useful on Mac OS X) that is used for
+ some metadata generation and other obscure things.</li>
+ </ul></li>
+<li>Finally, target-specific intrinsics can now return multiple values, which
+ is useful for modeling target operations with multiple results.</li>
</ul>
</div>
<div class="doc_text">
-<p>In addition to a large array of bug fixes and minor performance tweaks, this
+<p>In addition to a large array of minor performance tweaks and bug fixes, this
release includes a few major enhancements and additions to the optimizers:</p>
<ul>
-<li>SRoA improvements for vector unions, memset, arbitrary weird bitfield accesses etc. It now produces "strange" sized integers.</li>
-<li>Inliner reuse stack space when inlining arrays?</li>
-<li>Enabled GVN Load PRE.</li>
-<li>New Static Single Information (SSI) construction pass (not used by anything yet, experimental).</li>
+<li>The <a href="Passes.html#scalarrepl">Scalar Replacement of Aggregates</a>
+ pass has many improvements that allow it to better promote vector unions,
+ variables which are memset, and much more strange code that can happen to
+ do bitfield accesses to register operations. An interesting change is that
+ it now produces "unusual" integer sizes (like i1704) in some cases and lets
+ other optimizers clean things up.</li>
+<li>The <a href="Passes.html#loop-reduce">Loop Strength Reduction</a> pass now
+ promotes small integer induction variables to 64-bit on 64-bit targets,
+ which provides a major performance boost for much numerical code. It also
+ promotes shorts to int on 32-bit hosts, etc. LSR now also analyzes pointer
+ expressions (e.g. getelementptrs), as well as integers.</li>
+<li>The <a href="Passes.html#gvn">GVN</a> pass now eliminates partial
+ redundancies of loads in simple cases.</li>
+<li>The <a href="Passes.html#inline">Inliner</a> now reuses stack space when
+ inlining similar arrays from multiple callees into one caller.</li>
+<li>LLVM includes a new experimental Static Single Information (SSI)
+ construction pass.</li>
</li>
</ul>
</div>
+
+<!--=========================================================================-->
+<div class="doc_subsection">
+<a name="executionengine">Interpreter and JIT Improvements</a>
+</div>
+
+<div class="doc_text">
+
+<ul>
+<li>LLVM has a new "EngineBuilder" class which makes it more obvious how to
+ set up and configure an ExecutionEngine (a JIT or interpreter).</li>
+<li>The JIT now supports generating more than 16M of code.</li>
+<li>When configured with --with-oprofile, the JIT can now inform OProfile about
+ JIT'd code, allowing OProfile to get line number and function name
+ information for JIT'd functions.</li>
+<li>When "libffi" is available, the LLVM interpreter now uses it, which supports
+ calling almost arbitrary external (natively compiled) functions.</li>
+<li>Clients of the JIT can now register a 'JITEventListener' object to receive
+ callbacks when the JIT emits or frees machine code. The OProfile support
+ uses this mechanism.</li>
+</ul>
+
+</div>
+
<!--=========================================================================-->
<div class="doc_subsection">
<a name="codegen">Target Independent Code Generator Improvements</a>
<ul>
-<li> -asm-verbose now prints location info (with -g) and loop nest info.</li>
+<li>The <tt>llc -asm-verbose</tt> option (exposed from llvm-gcc and clang as
+ <tt>-fverbose-asm</tt>) now adds a lot of useful information in comments to
+ the generated .s file. This information includes location information (if
+ built with <tt>-g</tt>) and loop nest information.</li>
+<li>The code generator now supports a new MachineVerifier pass which is useful
+ for finding bugs in targets and codegen passes.</li>
+<li>The Machine LICM is now enabled by default. It hoists instructions out of
+ loops (such as constant pool loads, loads from read-only stubs, vector
+ constant synthesization code, etc.) and is currently configured to only do
+ so when the hoisted operation can be rematerialized.</li>
+<li>The Machine Sinking pass is now enabled by default. This pass moves
+ side-effect free operations down the CFG so that they are executed on fewer
+ paths through a function.</li>
+<li>The code generator now performs "stack slot coloring" of register spills,
+ which allows spill slots to be reused. This leads to smaller stack frames
+ in cases where there are lots of register spills.</li>
+<li>The register allocator has many improvements to take better advantage of
+ commutable operations, various spiller peephole optimizations, and can now
+ coalesce cross-register-class copies.</li>
<li>Tblgen now supports multiclass inheritance and a number of new string and
list operations like !(subst), !(foreach), !car, !cdr, !null, !if, !cast.
These make the .td files more expressive and allow more aggressive factoring
of duplication across instruction patterns.</li>
-<li>New MachineVerifier pass.</li>
-<li>Machine LICM, hoists things like constant pool loads, loads from readonly stubs, vector constant synthesization code, etc.</li>
-<li>Machine Sinking</li>
-<li>target-specific intrinsics (r63765)</li>
-<li>Regalloc improvements for commuting, various spiller peephole optimizations, cross-class coalescing.</li>
-<li><tt>llc -enable-value-prop</tt>, propagation of value info (sign/zero ext info) from one MBB to another</li>
-<li>Regalloc hints for allocation stuff: Evan r73381/r73671. Finished/enabled?</li>
-<li>Stack slot coloring for register spills (denser stack frames)</li>
-<li>SelectionDAGS: New BuildVectorSDNode (r65296), and ISD::VECTOR_SHUFFLE (r69952 / PR2957)</li>
-<li>Experimental support for shrink wrapping support in PEI.</li>
-<li>Experimental support for writing ELF .o files directly from the compiler,
- it works well for many simple C testcases, but doesn't support exception
- handling, debug info, inline assembly, etc.</li>
+<li>Target-specific intrinsics can now be added without having to hack VMCore to
+ add them. This makes it easier to maintain out-of-tree targets.</li>
+<li>The instruction selector is better at propagating information about values
+ (such as whether they are sign/zero extended etc.) across basic block
+ boundaries.</li>
+<li>The SelectionDAG datastructure has new nodes for representing buildvector
+ and <a href="http://llvm.org/PR2957">vector shuffle</a> operations. This
+ makes operations and pattern matching more efficient and easier to get
+ right.</li>
+<li>The Prolog/Epilog Insertion Pass now has experimental support for performing
+ the "shrink wrapping" optimization, which moves spills and reloads around in
+ the CFG to avoid doing saves on paths that don't need them.</li>
+<li>LLVM includes new experimental support for writing ELF .o files directly
+ from the compiler. It works well for many simple C testcases, but doesn't
+ support exception handling, debug info, inline assembly, etc.</li>
<li>Targets can now specify register allocation hints through
- MachineRegisterInfo:: setRegAllocationHint. A regalloc hint consists 1) hint
- type, 2) physical register number. A hint type of zero specifies a register
- allocation preference. Other hint type values are target specific which are
- resolved by TargetRegisterInfo::ResolveRegAllocHint. An example of which is
- the ARM target can uses register hint to request that the register allocator
- provide an even / odd register pair to two virtual registers. It is
- important to note the register allocation hints are just hints. There is no
- guarantee the register allocators will be able to satisfy the hints.</li>
-
+ MachineRegisterInfo::setRegAllocationHint. A regalloc hint consists of hint
+ type and physical register number. A hint type of zero specifies a register
+ allocation preference. Other hint type values are target specific which are
+ resolved by TargetRegisterInfo::ResolveRegAllocHint. An example is the ARM
+ target which uses register hints to request that the register allocator
+ provide an even / odd register pair to two virtual registers.</li>
</ul>
</div>
<ul>
-<li>Preliminary support for addrspace 256 -> GS, 257 -> FS, known problems: CodeGenerator.html#x86_memory</li>
-<li>Support for softfloat modes, typically used by OS kernels.</li>
-
-<li>X86-64: better modeling of implicit zero extensions, eliminates a lot of redundant zexts</li>
-<li>X86-64 TLS support for local exec and initial exec.</li>
-<li>Better modeling of H registerts as subregs.</li>
-<li>Vector icmp/fcmp now work with SSE codegen.</li>
-<li>SSE 4.2 support.</li>
-<li>all global variable reference logic is now in ClassifyGlobalReference.</li>
+<li>SSE 4.2 builtins are now supported.</li>
+<li>GCC-compatible soft float modes are now supported, which are typically used
+ by OS kernels.</li>
+<li>X86-64 now models implicit zero extensions better, which allows the code
+ generator to remove a lot of redundant zexts. It also models the 8-bit "H"
+ registers as subregs, which allows them to be used in some tricky
+ situations.</li>
+<li>X86-64 now supports the "local exec" and "initial exec" thread local storage
+ model.</li>
+<li>The vector forms of the <a href="LangRef.html#i_icmp">icmp</a> and <a
+ href="LangRef.html#i_fcmp">fcmp</a> instructions now select to efficient
+ SSE operations.</li>
+<li>Support for the win64 calling conventions have improved. The primary
+ missing feature is support for varargs function definitions. It seems to
+ work well for many win64 JIT purposes.</li>
+<li>The X86 backend has preliminary support for <a
+ href="CodeGenerator.html#x86_memory">mapping address spaces to segment
+ register references</a>. This allows you to write GS or FS relative memory
+ accesses directly in LLVM IR for cases where you know exactly what you're
+ doing (such as in an OS kernel). There are some known problems with this
+ support, but it works in simple cases.</li>
+<li>The X86 code generator has been refactored to move all global variable
+ reference logic to one place
+ (<tt>X86Subtarget::ClassifyGlobalReference</tt>) which
+ makes it easier to reason about.</li>
</li>
</ul>
<li>Preliminary support for processors, such as the Cortex-A8 and Cortex-A9,
that implement version v7-A of the ARM architecture. The ARM backend now
-supports both the Thumb2 and Advanced SIMD (Neon) instruction sets. The
-AAPCS-VFP "hard float" calling conventions are also supported with the
-<tt>-float-abi=hard</tt> flag. These features are still somewhat experimental
-and subject to change. The Neon intrinsics, in particular, may change in future
-releases of LLVM.
+supports both the Thumb2 and Advanced SIMD (Neon) instruction sets.</li>
+
+<li>The AAPCS-VFP "hard float" calling conventions are also supported with the
+<tt>-float-abi=hard</tt> flag.</li>
+
+<li>The ARM calling convention code is now tblgen generated instead of resorting
+ to C++ code.</li>
</li>
- ARM AAPCS-VFP hard float ABI is supported.
- ARM calling convention code is now tblgen generated instead of manual.
- ARM: NEON support. neonfp for doing single precision fp with neon instead of VFP.
+
+<p>These features are still somewhat experimental
+and subject to change. The Neon intrinsics, in particular, may change in future
+releases of LLVM. ARMv7 support has progressed a lot on top of tree since 2.6
+branched.</p>
</ul>
</div>
-<!--=========================================================================-->
-<div class="doc_subsection">
-<a name="executionengine">Interpreter and JIT Improvements</a>
-</div>
-
-<div class="doc_text">
-
-<ul>
-<li>The JIT now supports generating more than 16M of code.</li>
-<li>When configured with --with-oprofile, the JIT can now inform oprofile about
- JIT'd code, allowing oprofile to get line number and function name
- information for JIT'd functions.</li>
-<li>When "libffi" is available, the LLVM interpreter now uses it, which supports
- calling almost arbitrary external (natively compiled) functions.</li>
-<li>Clients of the JIT can now register a 'JITEventListener' object to receive
- callbacks when the JIT emits or frees machine code. The OProfile support
- uses this mechanism.</li>
-</ul>
-
-</div>
-
-
<!--=========================================================================-->
<div class="doc_subsection">
<a name="newapis">New Useful APIs</a>
<div class="doc_text">
-<ul>
-<li>New EngineBuilder class for creating JITs: r76276</li>
- New PrettyStackTrace, crashes of llvm tools should give some indication of what the compiler was doing at the time of the crash (e.g. running a pass), and print out command line arguments.
- StringRef class, Twine class.
- New WeakVH and AssertingVH and CallbackVH classes.
- New llvm/ADT/Triple class.
- llvm_report_error() error handling API (llvm/Support/ErrorHandling.h)
- New llvm/System/Atomic.h, llvm/System/RWMutex.h for portable atomic ops, rw locks.
- New SourceMgr, SMLoc classes for simple parsers with caret diagnostics and #include support, (used by
- tablegen, llvm-mc, the .ll parser, FileCheck, etc)
+<p>This release includes a number of new APIs that are used internally, which
+ may also be useful for external clients.
+</p>
+<ul>
+<li>New <a href="http://llvm.org/doxygen/PrettyStackTrace_8h-source.html">
+ PrettyStackTrace classes</a> allows crashes of llvm tools (and applications
+ that integrate them) to provide more detailed indication of what the
+ compiler was doing at the time of the crash (e.g. running a pass).
+ At the top level for each LLVM tool, it includes the command line arguments.
+ </li>
+<li>New <a href="http://llvm.org/doxygen/StringRef_8h-source.html">StringRef</a>
+ and <a href="http://llvm.org/doxygen/Twine_8h-source.html">Twine</a> classes
+ make operations on character ranges and
+ string concatenation to be more efficient. StringRef is just a <tt>const
+ char*</tt> with a length, Twine is a light-weight rope.</li>
+<li>LLVM has new WeakVH, AssertingVH and CallbackVH classes, which make it
+ easier to write LLVM IR transformations. WeakVH is automatically drops to
+ null when the referenced Value is deleted, and is updated across a
+ replaceAllUsesWith operation. AssertingVH aborts the program if the
+ referenced value is destroyed while it is being referenced. CallbackVH is
+ a customizable class for handling value references. See <a
+ href="http://llvm.org/doxygen/ValueHandle_8h-source.html">ValueHandle.h</a>
+ for more information.</li>
+<li>The new '<a href="http://llvm.org/doxygen/Triple_8h-source.html">Triple
+ </a>' class centralizes a lot of logic that reasons about target
+ triples.</li>
+<li>The new '<a href="http://llvm.org/doxygen/ErrorHandling_8h-source.html">
+ llvm_report_error()</a>' set of APIs allows tools to embed the LLVM
+ optimizer and backend and recover from previously unrecoverable errors.</li>
+<li>LLVM has new abstractions for <a
+ href="http://llvm.org/doxygen/Atomic_8h-source.html">atomic operations</a>
+ and <a href="http://llvm.org/doxygen/RWMutex_8h-source.html">reader/writer
+ locks</a>.</li>
+<li>LLVM has new <a href="http://llvm.org/doxygen/SourceMgr_8h-source.html">
+ SourceMgr and SMLoc classes</a> which implement caret diagnostics and basic
+ include stack processing for simple parsers. It is used by tablegen,
+ llvm-mc, the .ll parser and FileCheck.</li>
</ul>
overhead. Combined, the generated profile information is both more correct and
imposes about half as much overhead (2.6. from 12% to 6% overhead on SPEC
CPU2000).</li>
-<li>Many extensions to the C APIs.</li>
+<li>The C bindings (in the llvm/include/llvm-c directory) include many newly
+ supported APIs.</li>
<li>LLVM 2.6 includes a brand new experimental LLVM bindings to the Ada2005
-programming language.</li>
-
-<li>LLVMC:
+ programming language.</li>
-* Dynamic plugins now work on Windows.
-* New option property: init. Makes possible to provide default values for
- options defined in plugins (interface to cl::init).
-* New example: Skeleton, shows how to create a standalone LLVMC-based driver.
-* New example: mcc16, a driver for the PIC16 toolchain.</li>
+<li>The LLVMC driver has several new features:
+ <ul>
+ <li>Dynamic plugins now work on Windows.</li>
+ <li>New option property: init. Makes possible to provide default values for
+ options defined in plugins (interface to cl::init).</li>
+ <li>New example: Skeleton, shows how to create a standalone LLVMC-based
+ driver.</li>
+ <li>New example: mcc16, a driver for the PIC16 toolchain.</li>
+ </ul>
+</li>
</ul>
from the previous release.</p>
<ul>
-
-<li>The Itanium (IA64) backend has been removed. It was not supported and
- bitrotted.</li>
-<li>The BigBlock register allocator has been removed, it also bitrotted.</li>
+<li>The Itanium (IA64) backend has been removed. It was not actively supported
+ and had bitrotted.</li>
+<li>The BigBlock register allocator has been removed, it had also bitrotted.</li>
<li>The C Backend (-march=c) is no longer considered part of the LLVM release
criteria. We still want it to work, but no one is maintaining it and it lacks
support for arbitrary precision integers and other important IR features.</li>
-</ul>
- LLVM build now builds all libraries as .a files instead of some
+<li>All LLVM tools now default to overwriting their output file, behaving more
+ like standard unix tools. Previously, this only happened with the '-f'
+ option.</li>
+<li>LLVM build now builds all libraries as .a files instead of some
libraries as relinked .o files. This requires some APIs like
- InitializeAllTargets.h. TargetRegistry!
-
-
+ InitializeAllTargets.h.
+ </li>
+</ul>
<p>In addition, many APIs have changed in this release. Some of the major LLVM
API changes are:</p>
-
- API Cleanup:
- no use of hash_set/hash_map, no more llvm::OStream
- Use raw_ostream for everything, killed off llvm/Streams.h and DOUT
-
<ul>
+<li>All uses of hash_set and hash_map have been removed from the LLVM tree and
+ the wrapper headers have been removed.</li>
+<li>The llvm/Streams.h and DOUT member of Debug.h have been removed. The
+ llvm::Ostream class has been completely removed and replaced with uses of
+ raw_ostream.</li>
<li>LLVM's global uniquing tables for <tt>Type</tt>s and <tt>Constant</tt>s have
been privatized into members of an <tt>LLVMContext</tt>. A number of APIs
now take an <tt>LLVMContext</tt> as a parameter. To smooth the transition
context which can be passed in any and all cases where a context is
required.
<li>The <tt>getABITypeSize</tt> methods are now called <tt>getAllocSize</tt>.</li>
-<li>The <tt>Add</tt>, <tt>Sub</tt>, and <tt>Mul</tt> operators are no longer
- overloaded for floating-point types. Floating-point addition, subtraction,
+<li>The <tt>Add</tt>, <tt>Sub</tt> and <tt>Mul</tt> operators are no longer
+ overloaded for floating-point types. Floating-point addition, subtraction
and multiplication are now represented with new operators <tt>FAdd</tt>,
- <tt>FSub</tt>, and <tt>FMul</tt>. In the <tt>IRBuilder</tt> API,
- <tt>CreateAdd</tt>, <tt>CreateSub</tt>, <tt>CreateMul</tt>, and
+ <tt>FSub</tt> and <tt>FMul</tt>. In the <tt>IRBuilder</tt> API,
+ <tt>CreateAdd</tt>, <tt>CreateSub</tt>, <tt>CreateMul</tt> and
<tt>CreateNeg</tt> should only be used for integer arithmetic now;
- <tt>CreateFAdd</tt>, <tt>CreateFSub</tt>, <tt>CreateFMul</tt>, and
+ <tt>CreateFAdd</tt>, <tt>CreateFSub</tt>, <tt>CreateFMul</tt> and
<tt>CreateFNeg</tt> should now be used for floating-point arithmetic.</li>
<li>The DynamicLibrary class can no longer be constructed, its functionality has
moved to static member functions.</li>
and <tt>Twine</tt> classes instead of passing <tt>const char*</tt>
or <tt>std::string</tt>, as described in
the <a href="ProgrammersManual.html#string_apis">Programmer's Manual</a>. Most
-clients should be unaffected by this transition, unless they are used to <tt>Value::getName()</tt> returning a string. Here are some tips on updating to 2.6:
+clients should be unaffected by this transition, unless they are used to
+<tt>Value::getName()</tt> returning a string. Here are some tips on updating to
+2.6:
<ul>
<li><tt>getNameStr()</tt> is still available, and matches the old
behavior. Replacing <tt>getName()</tt> calls with this is an safe option,
<li>If you were using <tt>getName().c_str()</tt> to get a <tt>const
char*</tt> pointer to the name, you can use <tt>getName().data()</tt>.
Note that this string (as before), may not be the entire name if the
- name containts embedded null characters.</li>
+ name contains embedded null characters.</li>
<li>If you were using operator plus on the result of <tt>getName()</tt> and
treating the result as an <tt>std::string</tt>, you can either
- uses <tt>Twine::str</tt> to get the result as an <tt>std::string</tt>, or
+ use <tt>Twine::str</tt> to get the result as an <tt>std::string</tt>, or
could move to a <tt>Twine</tt> based design.</li>
<li><tt>isName()</tt> should be replaced with comparison
</li>
<li>The registration interfaces for backend Targets has changed (what was
-previously TargetMachineRegistry). For backend authors, see the <a href="WritingAnLLVMBackend.html#TargetRegistration">Writing An LLVM Backend</a> guide. For clients, the notable API changes are:
+previously TargetMachineRegistry). For backend authors, see the <a
+href="WritingAnLLVMBackend.html#TargetRegistration">Writing An LLVM Backend</a>
+guide. For clients, the notable API changes are:
<ul>
<li><tt>TargetMachineRegistry</tt> has been renamed
to <tt>TargetRegistry</tt>.</li>
function to find targets.</li>
</ul>
</li>
-
-<li>llvm-dis now fails if output file exists, instead of dumping to stdout.
-FIXME: describe any other tool changes due to the raw_fd_ostream change. FIXME:
-This is not an API change, maybe there should be a tool changes section?</li>
-<li>temporarely due to Context API change passes should call doInitialization()
-method of the pass they inherit from, otherwise Context is NULL.
-FIXME: remove this entry when this is no longer needed.<li>
</ul>
</div>
<ul>
<li>Intel and AMD machines (IA32, X86-64, AMD64, EMT-64) running Red Hat
-Linux, Fedora Core, FreeBSD and AuroraUX (and probably other unix-like systems).</li>
+ Linux, Fedora Core, FreeBSD and AuroraUX (and probably other unix-like
+ systems).</li>
<li>PowerPC and X86-based Mac OS X systems, running 10.3 and above in 32-bit
-and 64-bit modes.</li>
+ and 64-bit modes.</li>
<li>Intel and AMD machines running on Win32 using MinGW libraries (native).</li>
<li>Intel and AMD machines running on Win32 with the Cygwin libraries (limited
support is available for native builds with Visual C++).</li>
using the stock GCC 3.x.x series 'out the box',
See: <a href="#brokengcc">Broken versions of GCC and other tools</a>.
However, A <a href="http://pkg.auroraux.org/GCC">Modern GCC Build</a>
-for x86/x64 has been made available from the third party AuroraUX Project
+for x86/x86-64 has been made available from the third party AuroraUX Project
that has been meticulously tested for bootstrapping LLVM & Clang.</li>
</ul>
href="http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev">LLVMdev list</a>.</p>
<ul>
-<li>The MSIL, Alpha, SPU, MIPS, and PIC16 backends are experimental.</li>
+<li>The MSIL, Alpha, SPU, MIPS, PIC16, Blackfin, MSP430 and SystemZ backends are
+ experimental.</li>
<li>The <tt>llc</tt> "<tt>-filetype=asm</tt>" (the default) is the only
- supported value for this option.</li>
+ supported value for this option. The ELF writer is experimental.</li>
+<li>The implementation of Andersen's Alias Analysis has many known bugs.</li>
</ul>
</div>
<div class="doc_text">
<ul>
-<li>The O32 ABI is not fully supported.</li>
<li>64-bit MIPS targets are not supported yet.</li>
</ul>
<div class="doc_text">
-<p>llvm-gcc does not currently support <a href="http://llvm.org/PR869">Link-Time
-Optimization</a> on most platforms "out-of-the-box". Please inquire on the
-LLVMdev mailing list if you are interested.</p>
-
<p>The only major language feature of GCC not supported by llvm-gcc is
the <tt>__builtin_apply</tt> family of builtins. However, some extensions
are only supported on some targets. For example, trampolines are only
<div class="doc_text">
<ul>
<li>Fortran support generally works, but there are still several unresolved bugs
- in Bugzilla. Please see the tools/gfortran component for details.</li>
+ in <a href="http://llvm.org/bugs/">Bugzilla</a>. Please see the
+ tools/gfortran component for details.</li>
</ul>
</div>
projects / oota-llvm.git / blobdiff