Boost the effective chain depth of loads and stores.

By default, boost the chain depth contribution of loads and stores. This will allow a load/store pair to vectorize even when it would not otherwise be long enough to satisfy the chain depth requirement.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@149761 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Transforms/Vectorize/BBVectorize.cpp b/lib/Transforms/Vectorize/BBVectorize.cpp
index ce6e994c463d70422b523c00a2ebf3d4968b93c0..0780b4489ff1e32839c49ff17441dc70040c27a3 100644 (file)
--- a/lib/Transforms/Vectorize/BBVectorize.cpp
+++ b/lib/Transforms/Vectorize/BBVectorize.cpp
@@ -103,6 +103,11 @@ static cl::opt<bool>
 AlignedOnly("bb-vectorize-aligned-only", cl::init(false), cl::Hidden,
   cl::desc("Only generate aligned loads and stores"));
 
+static cl::opt<bool>
+NoMemOpBoost("bb-vectorize-no-mem-op-boost",
+  cl::init(false), cl::Hidden,
+  cl::desc("Don't boost the chain-depth contribution of loads and stores"));
+
 static cl::opt<bool>
 FastDep("bb-vectorize-fast-dep", cl::init(false), cl::Hidden,
   cl::desc("Use a fast instruction dependency analysis"));
@@ -340,6 +345,11 @@ namespace {
       if (isa<InsertElementInst>(V) || isa<ExtractElementInst>(V))
         return 0;
 
+      // Give a load or store half of the required depth so that load/store
+      // pairs will vectorize.
+      if (!NoMemOpBoost && (isa<LoadInst>(V) || isa<StoreInst>(V)))
+        return ReqChainDepth/2;
+
       return 1;
     }
 

diff --git a/test/Transforms/BBVectorize/mem-op-depth.ll b/test/Transforms/BBVectorize/mem-op-depth.ll
new file mode 100644 (file)
index 0000000..84f16bd
--- /dev/null
+++ b/test/Transforms/BBVectorize/mem-op-depth.ll
@@ -0,0 +1,22 @@
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+target triple = "x86_64-unknown-linux-gnu"
+; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=6 -instcombine -gvn -S | FileCheck %s
+
+@A = common global [1024 x float] zeroinitializer, align 16
+@B = common global [1024 x float] zeroinitializer, align 16
+
+define i32 @test1() nounwind {
+; CHECK: @test1
+  %V1 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 0), align 16
+  %V2 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 1), align 4
+  %V3= load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 2), align 8
+  %V4 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 3), align 4
+; CHECK:   %V1 = load <4 x float>* bitcast ([1024 x float]* @A to <4 x float>*), align 16
+  store float %V1, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 0), align 16
+  store float %V2, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 1), align 4
+  store float %V3, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 2), align 8
+  store float %V4, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 3), align 4
+; CHECK-NEXT: store <4 x float> %V1, <4 x float>* bitcast ([1024 x float]* @B to <4 x float>*), align 16
+  ret i32 0
+; CHECK-NEXT: ret i32 0
+}