llvm-project/clang/lib/CodeGen
John McCall 999110f903 When building the alloca for a local variable, set its name
separately from building the instruction so that it's
preserved even in -Asserts builds.

Employ C++'s mystical "comment" feature to discourage
breaking this in the future.

llvm-svn: 246991
2015-09-08 09:18:30 +00:00
..
ABIInfo.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
Address.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
BackendUtil.cpp Convert SampleProfile pass into a Module pass. 2015-08-25 15:25:13 +00:00
CGAtomic.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGBlocks.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGBlocks.h Move BlockByrefHelpers back to CodeGenModule.h to placate MSVC. 2015-09-08 08:21:11 +00:00
CGBuilder.h Remove unnecessary braces; this resolves against a 2015-09-08 08:57:00 +00:00
CGBuiltin.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCUDANV.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCUDARuntime.cpp Pass expressions instead of argument ranges to EmitCall/EmitCXXConstructorCall. 2014-08-21 20:26:47 +00:00
CGCUDARuntime.h Revert r240270 ("Fixed/added namespace ending comments using clang-tidy"). 2015-06-22 23:07:51 +00:00
CGCXX.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCXXABI.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCXXABI.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCall.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCall.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGClass.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCleanup.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGCleanup.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGDebugInfo.cpp CGDebugInfo: Instead of uniquing RetainedTypes, just refrain from retaining 2015-08-27 22:56:46 +00:00
CGDebugInfo.h CGDebugInfo: Factor out a getOrCreateStandaloneType() method. 2015-08-27 21:21:19 +00:00
CGDecl.cpp When building the alloca for a local variable, set its name 2015-09-08 09:18:30 +00:00
CGDeclCXX.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGException.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExpr.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExprAgg.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExprCXX.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExprComplex.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExprConstant.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGExprScalar.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGLoopInfo.cpp Add new llvm.loop.unroll.enable metadata for use with "#pragma unroll". 2015-08-10 17:29:39 +00:00
CGLoopInfo.h Add new llvm.loop.unroll.enable metadata for use with "#pragma unroll". 2015-08-10 17:29:39 +00:00
CGObjC.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGObjCGNU.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGObjCMac.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGObjCRuntime.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGObjCRuntime.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGOpenCLRuntime.cpp [C++11] Use 'nullptr'. CodeGen edition. 2014-05-21 05:09:00 +00:00
CGOpenCLRuntime.h Revert r240270 ("Fixed/added namespace ending comments using clang-tidy"). 2015-06-22 23:07:51 +00:00
CGOpenMPRuntime.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGOpenMPRuntime.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGRecordLayout.h Respect alignment of nested bitfields 2015-07-10 17:30:00 +00:00
CGRecordLayoutBuilder.cpp Respect alignment of nested bitfields 2015-07-10 17:30:00 +00:00
CGStmt.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGStmtOpenMP.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGVTT.cpp Remove and forbid raw_svector_ostream::flush() calls. 2015-08-13 18:12:56 +00:00
CGVTables.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CGVTables.h Header guard canonicalization, clang part. 2014-08-13 16:25:19 +00:00
CGValue.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CMakeLists.txt [CMake] Fill up required libs, corresponding to r241653. 2015-07-08 02:06:21 +00:00
CodeGenABITypes.cpp LLVM API Change: the Module always owns the DataLayout 2015-07-24 16:04:29 +00:00
CodeGenAction.cpp Initialize the AST consumer as soon as we have both an ASTConsumer and an 2015-08-18 20:39:29 +00:00
CodeGenFunction.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CodeGenFunction.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CodeGenModule.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CodeGenModule.h Move BlockByrefHelpers back to CodeGenModule.h to placate MSVC. 2015-09-08 08:21:11 +00:00
CodeGenPGO.cpp Switch users of the 'for (StmtRange range = stmt->children(); range; ++range)‘ pattern to range for loops. 2015-07-02 21:03:14 +00:00
CodeGenPGO.h InstrProf: Cede ownership of createProfileWeights to CGF 2015-05-02 05:00:55 +00:00
CodeGenTBAA.cpp Remove and forbid raw_svector_ostream::flush() calls. 2015-08-13 18:12:56 +00:00
CodeGenTBAA.h Header guard canonicalization, clang part. 2014-08-13 16:25:19 +00:00
CodeGenTypeCache.h Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
CodeGenTypes.cpp LLVM API Change: the Module always owns the DataLayout 2015-07-24 16:04:29 +00:00
CodeGenTypes.h Remove superfluous private:, TypeCache is private by default. 2015-08-13 07:12:03 +00:00
CoverageMappingGen.cpp Use llvm::reverse to make a bunch of loops use foreach. NFC. 2015-07-30 17:22:52 +00:00
CoverageMappingGen.h [cleanup] Re-sort *all* #include lines with llvm/utils/sort_includes.py 2015-01-14 11:29:14 +00:00
EHScopeStack.h Devirtualize EHScopeStack::Cleanup's dtor because it's never destroyed polymorphically 2015-08-18 22:40:54 +00:00
ItaniumCXXABI.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
Makefile
MicrosoftCXXABI.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
ModuleBuilder.cpp Rename DescriptionString -> DataLayoutString as it matches the actual 2015-08-05 23:48:05 +00:00
ObjectFilePCHContainerOperations.cpp Add a -gmodules option to the driver and a -dwarf-ext-refs to cc1 2015-08-27 19:46:20 +00:00
README.txt
SanitizerMetadata.cpp [ASan] Initial support for Kernel AddressSanitizer 2015-06-19 12:19:07 +00:00
SanitizerMetadata.h Removing LLVM_DELETED_FUNCTION, as MSVC 2012 was the last reason for requiring the macro. NFC; Clang edition. 2015-02-15 22:54:08 +00:00
TargetInfo.cpp Compute and preserve alignment more faithfully in IR-generation. 2015-09-08 08:05:57 +00:00
TargetInfo.h [OPENMP] Introduced type trait "__builtin_omp_required_simd_align" for default simd alignment. 2015-07-02 03:40:19 +00:00

README.txt

IRgen optimization opportunities.

//===---------------------------------------------------------------------===//

The common pattern of
--
short x; // or char, etc
(x == 10)
--
generates an zext/sext of x which can easily be avoided.

//===---------------------------------------------------------------------===//

Bitfields accesses can be shifted to simplify masking and sign
extension. For example, if the bitfield width is 8 and it is
appropriately aligned then is is a lot shorter to just load the char
directly.

//===---------------------------------------------------------------------===//

It may be worth avoiding creation of alloca's for formal arguments
for the common situation where the argument is never written to or has
its address taken. The idea would be to begin generating code by using
the argument directly and if its address is taken or it is stored to
then generate the alloca and patch up the existing code.

In theory, the same optimization could be a win for block local
variables as long as the declaration dominates all statements in the
block.

NOTE: The main case we care about this for is for -O0 -g compile time
performance, and in that scenario we will need to emit the alloca
anyway currently to emit proper debug info. So this is blocked by
being able to emit debug information which refers to an LLVM
temporary, not an alloca.

//===---------------------------------------------------------------------===//

We should try and avoid generating basic blocks which only contain
jumps. At -O0, this penalizes us all the way from IRgen (malloc &
instruction overhead), all the way down through code generation and
assembly time.

On 176.gcc:expr.ll, it looks like over 12% of basic blocks are just
direct branches!

//===---------------------------------------------------------------------===//