[LoopUnroll] By default disable unrolling when optimizing for size.

Summary:
In clang commit r268509 we started to invoke loop-unroll pass from the
driver even under -Os. However, we happen to not initialize optsize
thresholds properly, which si fixed with this change.

r268509 led to some big compile time regressions, because we started to
unroll some loops that we didn't unroll before. With this change I hope
to recover most of the regressions. We still are slightly slower than
before, because we do some checks here and there in loop-unrolling
before we bail out, but at least the slowdown is not that huge now.

Reviewers: hfinkel, chandlerc

Subscribers: mzolotukhin, llvm-commits

Differential Revision: https://reviews.llvm.org/D23388

llvm-svn: 279585

diff --git a/llvm/include/llvm/CodeGen/BasicTTIImpl.h b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
index c4480ff..b1a29d5 100644 (file)
--- a/llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -285,7 +285,11 @@ public:
 
     // Enable runtime and partial unrolling up to the specified size.
     UP.Partial = UP.Runtime = true;
-    UP.PartialThreshold = UP.PartialOptSizeThreshold = MaxOps;
+    UP.PartialThreshold = MaxOps;
+
+    // Avoid unrolling when optimizing for size.
+    UP.OptSizeThreshold = 0;
+    UP.PartialOptSizeThreshold = 0;
   }
 
   /// @}

diff --git a/llvm/lib/Transforms/Scalar/LoopUnrollPass.cpp b/llvm/lib/Transforms/Scalar/LoopUnrollPass.cpp
index f53ac86..1018b61 100644 (file)
--- a/llvm/lib/Transforms/Scalar/LoopUnrollPass.cpp
+++ b/llvm/lib/Transforms/Scalar/LoopUnrollPass.cpp
@@ -948,6 +948,10 @@ static bool tryToUnrollLoop(Loop *L, DominatorTree &DT, LoopInfo *LI,
       L, TTI, ProvidedThreshold, ProvidedCount, ProvidedAllowPartial,
       ProvidedRuntime);
 
+  // Exit early if unrolling is disabled.
+  if (UP.Threshold == 0 && (!UP.Partial || UP.PartialThreshold == 0))
+    return false;
+
   // If the loop contains a convergent operation, the prelude we'd add
   // to do the first few instructions before we hit the unrolled loop
   // is unsafe -- it adds a control-flow dependency to the convergent