[SLP] Enable -slp-vectorize-hor by default.

Measurements primarily on AArch64 have shown this feature does not
significantly effect compile-time. The are no significant perf changes in LNT,
but for AArch64 at least, there are wins in third party benchmarks.

As discussed on llvm-dev, we're going to try turning this on by default and see
how other targets react to the change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@252733 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Vectorize/SLPVectorizer.cpp b/lib/Transforms/Vectorize/SLPVectorizer.cpp
index ab81430a584136e291aac414bcf283c9d5b1c9ff..ad28a72deee3239adfb7d9cfbebb4a340d66025f 100644 (file)
--- a/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -61,7 +61,7 @@ static cl::opt<int>
                               "number "));
 
 static cl::opt<bool>
-ShouldVectorizeHor("slp-vectorize-hor", cl::init(false), cl::Hidden,
+ShouldVectorizeHor("slp-vectorize-hor", cl::init(true), cl::Hidden,
                    cl::desc("Attempt to vectorize horizontal reductions"));
 
 static cl::opt<bool> ShouldStartVectorizeHorAtStore(