[PATCH] D30719: AMDGPU/SI: Disable unrolling in the loop vectorizer if the loop is not vectorized.

[Changpeng Fang via Phabricator via llvm-commits]

cfang updated this revision to Diff 91088.
cfang added a comment.

add lit.local.cfg file in the newly created test/Transforms/LoopVectorize/AMDGPU directory.


https://reviews.llvm.org/D30719

Files:
  lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
  test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
  test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll


Index: test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll
===================================================================
--- /dev/null
+++ test/Transforms/LoopVectorize/AMDGPU/unroll-in-loop-vectorizer.ll
@@ -0,0 +1,29 @@
+; 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
+}
+
Index: test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
===================================================================
--- /dev/null
+++ test/Transforms/LoopVectorize/AMDGPU/lit.local.cfg
@@ -0,0 +1,3 @@
+if not 'AMDGPU' in config.root.targets:
+    config.unsupported = True
+
Index: lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
===================================================================
--- lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -155,6 +155,10 @@
 }
 
 unsigned AMDGPUTTIImpl::getMaxInterleaveFactor(unsigned VF) {
+  // Disable unrolling if the loop is not vectorized.
+  if (VF == 1)
+    return 1;
+
   // Semi-arbitrary large amount.
   return 64;
 }


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D30719.91088.patch
Type: text/x-patch
Size: 1908 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170308/e7d20fe3/attachment.bin>