AMDGPU/SI: Disable unrolling in the loop vectorizer if the loop is not vectorized.

Reviewers:
  arsenm

Differential Revision:
  http://reviews.llvm.org/D30719

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@297328 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
index a780a76c86a604f3be0938f689c3a035a067b7e5..4a6d12bd883e7c3ae7abd1ffef01becb18491304 100644 (file)
--- a/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ b/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -155,6 +155,10 @@ bool AMDGPUTTIImpl::isLegalToVectorizeStoreChain(unsigned ChainSizeInBytes,
 }
 
 unsigned AMDGPUTTIImpl::getMaxInterleaveFactor(unsigned VF) {
+  // Disable unrolling if the loop is not vectorized.
+  if (VF == 1)
+    return 1;
+
   // Semi-arbitrary large amount.
   return 64;
 }

diff --git a/test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg b/test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
new file mode 100644 (file)
index 0000000..2a665f0
--- /dev/null
+++ b/test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'AMDGPU' in config.root.targets:
+    config.unsupported = True

diff --git a/test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll b/test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll
new file mode 100644 (file)
index 0000000..85ba95c
--- /dev/null
+++ b/test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll
@@ -0,0 +1,28 @@
+; RUN: opt -S -mtriple=amdgcn-unknown-amdhsa -mcpu=fiji -loop-vectorize < %s | FileCheck %s
+
+
+; For AMDGPU, loop unroll in loop vectorizer is disabled when VF==1.
+;
+; CHECK-LABEL: @small_loop(
+; CHECK: store i32
+; CHECK-NOT: store i32
+; CHECK: ret
+define void @small_loop(i32* nocapture %inArray, i32 %size) nounwind {
+entry:
+  %0 = icmp sgt i32 %size, 0
+  br i1 %0, label %loop, label %exit
+
+loop:                                          ; preds = %entry, %loop
+  %iv = phi i32 [ %iv1, %loop ], [ 0, %entry ]
+  %1 = getelementptr inbounds i32, i32* %inArray, i32 %iv
+  %2 = load i32, i32* %1, align 4
+  %3 = add nsw i32 %2, 6
+  store i32 %3, i32* %1, align 4
+  %iv1 = add i32 %iv, 1
+;  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %cond = icmp eq i32 %iv1, %size
+  br i1 %cond, label %exit, label %loop
+
+exit:                                         ; preds = %loop, %entry
+  ret void
+}