[llvm] r216140 - [LoopVectorizer] Limit unroll factor in the presence of nested reductions.

[James Molloy]

Author: jamesm
Date: Wed Aug 20 18:53:52 2014
New Revision: 216140

URL: http://llvm.org/viewvc/llvm-project?rev=216140&view=rev
Log:
[LoopVectorizer] Limit unroll factor in the presence of nested reductions.

If we have a scalar reduction, we can increase the critical path length if the loop we're unrolling is inside another loop. Limit, by default to 2, so the critical path only gets increased by one reduction operation.

Modified:
    llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp

Modified: llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp?rev=216140&r1=216139&r2=216140&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp (original)
+++ llvm/trunk/lib/Transforms/Vectorize/LoopVectorize.cpp Wed Aug 20 18:53:52 2014
@@ -204,6 +204,11 @@ static cl::opt<bool> EnableCondStoresVec
     "enable-cond-stores-vec", cl::init(false), cl::Hidden,
     cl::desc("Enable if predication of stores during vectorization."));
 
+static cl::opt<unsigned> MaxNestedScalarReductionUF(
+    "max-nested-scalar-reduction-unroll", cl::init(2), cl::Hidden,
+    cl::desc("The maximum unroll factor to use when unrolling a scalar "
+             "reduction in a nested loop."));
+
 namespace {
 
 // Forward declarations.
@@ -5546,6 +5551,18 @@ LoopVectorizationCostModel::selectUnroll
     unsigned StoresUF = UF / (Legal->NumStores ? Legal->NumStores : 1);
     unsigned LoadsUF = UF /  (Legal->NumLoads ? Legal->NumLoads : 1);
 
+    // If we have a scalar reduction (vector reductions are already dealt with
+    // by this point), we can increase the critical path length if the loop
+    // we're unrolling is inside another loop. Limit, by default to 2, so the
+    // critical path only gets increased by one reduction operation.
+    if (Legal->getReductionVars()->size() &&
+        TheLoop->getLoopDepth() > 1) {
+      unsigned F = static_cast<unsigned>(MaxNestedScalarReductionUF);
+      SmallUF = std::min(SmallUF, F);
+      StoresUF = std::min(StoresUF, F);
+      LoadsUF = std::min(LoadsUF, F);
+    }
+
     if (EnableLoadStoreRuntimeUnroll && std::max(StoresUF, LoadsUF) > SmallUF) {
       DEBUG(dbgs() << "LV: Unrolling to saturate store or load ports.\n");
       return std::max(StoresUF, LoadsUF);