Dale Johannesen e4a4245906 Mark negative-zero-to-int conversion as Inexact, ...

since -0.0 can't be represented as an int.
This prevents llvm from reducing -0.0 to a
load-and-convert of int 0.  Fixes
ieee.exp/mzero[2356].c in gcc testsuite.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57224 91177308-0d34-0410-b5e6-96231b3b80d8

2008-10-07 00:40:01 +00:00

..

Allocator.cpp

…

Annotation.cpp

…

APFloat.cpp

Mark negative-zero-to-int conversion as Inexact,

2008-10-07 00:40:01 +00:00

APInt.cpp

…

APSInt.cpp

…

CMakeLists.txt

…

CommandLine.cpp

…

ConstantRange.cpp

…

Debug.cpp

…

Dwarf.cpp

…

FileUtilities.cpp

…

FoldingSet.cpp

…

GraphWriter.cpp

…

IsInf.cpp

…

IsNAN.cpp

…

Makefile

…

ManagedStatic.cpp

…

MemoryBuffer.cpp

…

PluginLoader.cpp

…

raw_ostream.cpp

…

SlowOperationInformer.cpp

…

SmallPtrSet.cpp

…

Statistic.cpp

…

Streams.cpp

…

StringExtras.cpp

…

StringMap.cpp

…

StringPool.cpp

…

SystemUtils.cpp

…

Timer.cpp

…