Cameron Zwarich b2fd770136 Add support to scalar replacement for partial vector accesses of an alloca, e.g. ... a union of a float, <2 x float>, and <4 x float>. This mostly comes up with the use of vector intrinsics, especially in NEON when programmers know the layout of the register file. This enables codegen to eliminate a lot of the subregister traffic it would otherwise generate. This commit only enables this for a small number of floating-point cases, but a lot more integer cases. I assume this is okay for all ports, but I did not do extensive testing of the quality of code involving i512 vectors and the like. If there is a use case where this generates worse code than before, let me know and we can scale it back. This fixes <rdar://problem/9036264>. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127317 91177308-0d34-0410-b5e6-96231b3b80d8		2011-03-09 05:43:05 +00:00
..
ADCE
ArgumentPromotion	fix PR8932, a case where arg promotion could infinitely promote.	2011-01-16 08:09:24 +00:00
BlockPlacement
BranchFolding
CodeExtractor
CodeGenPrepare	temporarily revert r123526. While working on a follow-on patch I	2011-01-15 07:51:19 +00:00
ConstantMerge	Don't merge two constants if we care about the address of both.	2011-01-16 17:05:09 +00:00
ConstProp	When loading from a constant, fold inttoptr if the integer type and the resulting pointer type both have the same size.	2011-02-06 20:11:56 +00:00
CorrelatedValuePropagation	Test for lazy value info's ability to prove the absense of NULLs in pointers.	2011-01-16 21:57:20 +00:00
DeadArgElim	Add an obvious missing safety check to DAE::RemoveDeadArgumentsFromCallers.	2011-03-01 00:33:47 +00:00
DeadStoreElimination
EarlyCSE	Duncan deftly points out that readnone functions aren't	2011-01-03 23:38:13 +00:00
FunctionAttrs
GlobalDCE
GlobalOpt	Add unnamed_addr when we can show that address of a global is not used.	2011-01-19 16:32:21 +00:00
GVN
IndVarSimplify
Inline	relax testcase a bit.	2011-01-14 07:46:33 +00:00
InstCombine	PR9346: Prevent SimplifyDemandedBits from incorrectly introducing	2011-03-09 01:28:35 +00:00
InstSimplify	Thread comparisons over udiv/sdiv/ashr/lshr exact and lshr nuw/nsw whenever	2011-03-05 05:19:11 +00:00
Internalize	Add test for r127138.	2011-03-07 21:28:14 +00:00
IPConstantProp
JumpThreading	Fix PR9331. Simplified version of a patch by Jakub Staszak.	2011-03-08 12:39:03 +00:00
LCSSA	Rename 'loopsimplify' to 'loop-simplify'.	2011-02-10 23:38:10 +00:00
LICM	While sinking an instruction, do not lose llvm.dbg.value intrinsic.	2011-03-08 03:06:19 +00:00
LoopDeletion	Make LoopDeletion work on loops with multiple edges, as long as the incoming	2011-02-22 22:25:39 +00:00
LoopIdiom	Preserve line no. info.	2011-03-07 22:43:45 +00:00
LoopRotate	Do not hoist @llvm.dbg.value. Here, @llvm.dbg.value is "referring" a value that is modified inside loop.	2011-02-14 23:03:23 +00:00
LoopSimplify	Rename 'loopsimplify' to 'loop-simplify'.	2011-02-10 23:38:10 +00:00
LoopStrengthReduce	Teach RecursivelyDeleteDeadPHINodes to handle multiple self-references. Patch	2011-02-20 08:38:20 +00:00
LoopUnroll	add a testcase for r125827	2011-02-18 05:05:01 +00:00
LoopUnswitch
LowerAtomic
LowerInvoke
LowerSetJmp
LowerSwitch
Mem2Reg
MemCpyOpt	Actually check memcpy lengths, instead of just commenting about	2011-01-21 22:07:57 +00:00
MergeFunc	When removing a function from the function set and adding it to deferred, we	2011-02-09 06:32:02 +00:00
PruneEH
Reassociate	fix PR9215, preventing -reassociate from clearing nsw/nuw when	2011-02-17 01:29:24 +00:00
ScalarRepl	Add support to scalar replacement for partial vector accesses of an alloca, e.g.	2011-03-09 05:43:05 +00:00
SCCP
SimplifyCFG	PR9420; an instruction before an unreachable is guaranteed not to have any	2011-03-09 00:48:33 +00:00
SimplifyLibCalls	Optimize fprintf -> iprintf if there are no floating point arguments	2011-03-03 14:20:22 +00:00
Sink
SRETPromotion
StripSymbols
TailCallElim	Add a test for TCE return duplication.	2011-01-29 04:53:35 +00:00
TailDup