Reapply 184685 after the SetVector iteration order fix.

This should hopefully have fixed the stage2/stage3 miscompare on the dragonegg
testers.

"LoopVectorize: Use the dependence test utility class

We now no longer need alias analysis - the cases that alias analysis would
handle are now handled as accesses with a large dependence distance.

We can now vectorize loops with simple constant dependence distances.

  for (i = 8; i < 256; ++i) {
    a[i] = a[i+4] * a[i+8];
  }

  for (i = 8; i < 256; ++i) {
    a[i] = a[i-4] * a[i-8];
  }

We would be able to vectorize about 200 more loops (in many cases the cost model
instructs us no to) in the test suite now. Results on x86-64 are a wash.

I have seen one degradation in ammp. Interestingly, the function in which we
now vectorize a loop is never executed so we probably see some instruction
cache effects. There is a 2% improvement in h264ref. There is one or the other
TSCV loop kernel that speeds up.

radar://13681598"

llvm-svn: 184724

1 files changed, 1 insertions, 1 deletions

diff --git a/llvm/test/Transforms/LoopVectorize/12-12-11-if-conv.ll b/llvm/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
index bab6300f2e7..6ef101074de 100644
--- a/llvm/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
+++ b/llvm/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
@@ -30,7 +30,7 @@ if.then:                                          ; preds = %for.body
 if.end:                                           ; preds = %for.body, %if.then
   %z.0 = phi i32 [ %add1, %if.then ], [ 9, %for.body ]
   store i32 %z.0, i32* %arrayidx, align 4
-  %indvars.iv.next = add i64 %indvars.iv, 1
+  %indvars.iv.next = add nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, %x
   br i1 %exitcond, label %for.end, label %for.body