David Majnemer b1b738e036 InstCombine: (icmp eq B, 0) | (icmp ult A, B) -> (icmp ule A, B-1) ... This transform allows us to turn IR that looks like: %1 = icmp eq i64 %b, 0 %2 = icmp ult i64 %a, %b %3 = or i1 %1, %2 ret i1 %3 into: %0 = add i64 %b, -1 %1 = icmp uge i64 %0, %a ret i1 %1 which means we go from lowering: cmpq %rsi, %rdi setb %cl testq %rsi, %rsi sete %al orb %cl, %al ret to lowering: decq %rsi cmpq %rdi, %rsi setae %al ret git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@185677 91177308-0d34-0410-b5e6-96231b3b80d8		2013-07-05 00:31:17 +00:00
..
CMakeLists.txt
InstCombine.h	Reapply with r182909 with a fix to the calculation of the new indices for	2013-05-31 00:59:42 +00:00
InstCombineAddSub.cpp	In InstCombine{AddSub,MulDivRem} convert APFloat.isFiniteNonZero() && !APFloat.isDenormal => APFloat.isNormal.	2013-06-26 23:17:31 +00:00
InstCombineAndOrXor.cpp	InstCombine: (icmp eq B, 0) | (icmp ult A, B) -> (icmp ule A, B-1)	2013-07-05 00:31:17 +00:00
InstCombineCalls.cpp	Fix using arg_end() - arg_begin() instead of arg_size()	2013-06-28 00:25:40 +00:00
InstCombineCasts.cpp
InstCombineCompares.cpp	InstCombine: FoldGEPICmp shouldn't change sign of base pointer comparison	2013-06-29 10:28:04 +00:00
InstCombineLoadStoreAlloca.cpp
InstCombineMulDivRem.cpp	InstCombine: Reimplementation of visitUDivOperand	2013-07-04 21:17:49 +00:00
InstCombinePHI.cpp
InstCombineSelect.cpp	InstCombine: Also turn selects fed by an and into arithmetic when the types don't match.	2013-06-29 21:17:04 +00:00
InstCombineShifts.cpp
InstCombineSimplifyDemanded.cpp
InstCombineVectorOps.cpp	Delete dead safety check.	2013-06-03 23:15:20 +00:00
InstCombineWorklist.h
InstructionCombining.cpp	Use SmallVectorImpl::iterator/const_iterator instead of SmallVector to avoid specifying the vector size.	2013-07-04 01:31:24 +00:00
LLVMBuild.txt
Makefile