[TTI/CostModel] improve TTI::getGEPCost and use it in CostModel::getInstructionCost

Summary:
This patch updates TargetTransformInfoImplCRTPBase::getGEPCost to consider
addressing modes. It now returns TCC_Free when the GEP can be completely folded
to an addresing mode.

I started this patch as I refactored SLSR. Function isGEPFoldable looks common
and is indeed used by some WIP of mine. So I extracted that logic to getGEPCost.

Furthermore, I noticed getGEPCost wasn't directly tested anywhere. The best
testing bed seems CostModel, but its getInstructionCost method invokes
getAddressComputationCost for GEPs which provides very coarse estimation. So
this patch also makes getInstructionCost call the updated getGEPCost for GEPs.
This change inevitably breaks some tests because the cost model changes, but
nothing looks seriously wrong -- if we believe the new cost model is the right
way to go, these tests should be updated.

This patch is not perfect yet -- the comments in some tests need to be updated.
I want to know whether this is a right approach before fixing those details.

Reviewers: chandlerc, hfinkel

Subscribers: aschwaighofer, llvm-commits, aemerson

Differential Revision: http://reviews.llvm.org/D9819

llvm-svn: 243250

1 files changed, 6 insertions, 3 deletions

diff --git a/llvm/test/Transforms/LoopUnroll/X86/partial.ll b/llvm/test/Transforms/LoopUnroll/X86/partial.ll
index 4566f792deb..104a38779e5 100644
--- a/llvm/test/Transforms/LoopUnroll/X86/partial.ll
+++ b/llvm/test/Transforms/LoopUnroll/X86/partial.ll
@@ -86,17 +86,20 @@ for.body:                                         ; preds = %entry, %for.body
   %reduction.026 = phi i16 [ %add14, %for.body ], [ 0, %entry ]
   %arrayidx = getelementptr inbounds i16, i16* %arr, i64 %indvars.iv
   %0 = load i16, i16* %arrayidx, align 2
-  %add = add i16 %0, %reduction.026
+  %mul = shl i16 %0, 1
+  %add = add i16 %mul, %reduction.026
   %sext = mul i64 %indvars.iv, 12884901888
   %idxprom3 = ashr exact i64 %sext, 32
   %arrayidx4 = getelementptr inbounds i16, i16* %arr, i64 %idxprom3
   %1 = load i16, i16* %arrayidx4, align 2
-  %add7 = add i16 %add, %1
+  %mul2 = shl i16 %1, 1
+  %add7 = add i16 %add, %mul2
   %sext28 = mul i64 %indvars.iv, 21474836480
   %idxprom10 = ashr exact i64 %sext28, 32
   %arrayidx11 = getelementptr inbounds i16, i16* %arr, i64 %idxprom10
   %2 = load i16, i16* %arrayidx11, align 2
-  %add14 = add i16 %add7, %2
+  %mul3 = shl i16 %2, 1
+  %add14 = add i16 %add7, %mul3
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, %n