Michael Zolotukhin c9da41492f Use estimated number of optimized insns in unroll-threshold computation. ... If complete-unroll could help us to optimize away N% of instructions, we might want to do this even if the final size would exceed loop-unroll threshold. However, we don't want to unroll huge loop, and we are add AbsoluteThreshold to avoid that - this threshold will never be crossed, even if we expect to optimize 99% instructions after that. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@228434 91177308-0d34-0410-b5e6-96231b3b80d8		2015-02-06 20:20:40 +00:00
..
Analysis	[InstSimplify] Add SimplifyFPBinOp function.	2015-02-06 20:02:51 +00:00
AsmParser	AsmParser: Split out LineField, NFC	2015-02-04 22:59:18 +00:00
Bitcode	IR: Initialize MDNode abbreviations en masse, NFC	2015-02-04 21:54:12 +00:00
CodeGen	[LiveIntervalAnalysis] Speed up creation of live ranges for physical registers	2015-02-06 18:42:41 +00:00
DebugInfo	Revert "Create lib/DebugInfo/PDB."	2015-02-06 20:00:18 +00:00
ExecutionEngine	[Orc] Move SectionMemoryManager's implementation from MCJIT to ExecutionEngine.	2015-02-06 19:36:40 +00:00
Fuzzer	[fuzzer] move default sanitizer options to a separate file	2015-02-06 19:52:07 +00:00
IR	Teach isDereferenceablePointer() to look through bitcast constant expressions.	2015-02-05 09:15:37 +00:00
IRReader
LineEditor
Linker
LTO
MC	Try to fix the build in MCValue.cpp	2015-02-05 01:23:14 +00:00
Object
Option
ProfileData	InstrProf: Use a stable sort when reading coverage regions	2015-02-04 00:12:18 +00:00
Support	Make helper functions/classes/globals static. NFC.	2015-02-06 17:51:54 +00:00
TableGen
Target	Don't dllexport declarations	2015-02-06 17:59:49 +00:00
Transforms	Use estimated number of optimized insns in unroll-threshold computation.	2015-02-06 20:20:40 +00:00
CMakeLists.txt
LLVMBuild.txt
Makefile