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

Changpeng Fang · Changpeng Fang · commit 1be9b9f81694 · 2017-03-09T00:07:00.000Z
Reviewers: arsenm Differential Revision: http://reviews.llvm.org/D30719 llvm-svn: 297328
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/llvm/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/llvm/test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg b/llvm/test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'AMDGPU' in config.root.targets:
+    config.unsupported = True
diff --git a/llvm/test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll b/llvm/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
+}

Original file line number	Diff line number	Diff line change
`@@ -155,6 +155,10 @@ bool AMDGPUTTIImpl::isLegalToVectorizeStoreChain(unsigned ChainSizeInBytes,`
`155`	`155`	`}`
`156`	`156`
`157`	`157`	`unsigned AMDGPUTTIImpl::getMaxInterleaveFactor(unsigned VF) {`
	`158`	`+ // Disable unrolling if the loop is not vectorized.`
	`159`	`+ if (VF == 1)`
	`160`	`+ return 1;`
	`161`	`+`
`158`	`162`	`// Semi-arbitrary large amount.`
`159`	`163`	`return 64;`
`160`	`164`	`}`
Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	`+if not 'AMDGPU' in config.root.targets:`
	`2`	`+ config.unsupported = True`