From 94c495295135dd45a0b67a5b6d6c3c318f4f8628 Mon Sep 17 00:00:00 2001
From: Jingu Kang <jingu.kang@arm.com>
Date: Wed, 14 Jul 2021 11:43:29 +0100
Subject: [PATCH] [AArch64] Enable Upper bound unrolling universally

Differential Revision: https://reviews.llvm.org/D105996
---
 .../AArch64/AArch64TargetTransformInfo.cpp    |  3 +-
 .../LoopUnroll/AArch64/unroll-upperbound.ll   | 65 +++++++++++++++++++
 2 files changed, 67 insertions(+), 1 deletion(-)
 create mode 100644 llvm/test/Transforms/LoopUnroll/AArch64/unroll-upperbound.ll

diff --git a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
index 2fd1229f1826..b479b4808ec2 100644
--- a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
@@ -1754,6 +1754,8 @@ void AArch64TTIImpl::getUnrollingPreferences(Loop *L, ScalarEvolution &SE,
   // Enable partial unrolling and runtime unrolling.
   BaseT::getUnrollingPreferences(L, SE, UP, ORE);
 
+  UP.UpperBound = true;
+
   // For inner loop, it is more likely to be a hot one, and the runtime check
   // can be promoted out from LICM pass, so the overhead is less, let's try
   // a larger threshold to unroll more loops.
@@ -1794,7 +1796,6 @@ void AArch64TTIImpl::getUnrollingPreferences(Loop *L, ScalarEvolution &SE,
       !ST->getSchedModel().isOutOfOrder()) {
     UP.Runtime = true;
     UP.Partial = true;
-    UP.UpperBound = true;
     UP.UnrollRemainder = true;
     UP.DefaultUnrollRuntimeCount = 4;
 
diff --git a/llvm/test/Transforms/LoopUnroll/AArch64/unroll-upperbound.ll b/llvm/test/Transforms/LoopUnroll/AArch64/unroll-upperbound.ll
new file mode 100644
index 000000000000..3b82365d1a6e
--- /dev/null
+++ b/llvm/test/Transforms/LoopUnroll/AArch64/unroll-upperbound.ll
@@ -0,0 +1,65 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt < %s -S -loop-unroll -mtriple aarch64 | FileCheck %s
+
+; Below loop's trip count is not constant and it blocks to unroll the loop.
+; After setting up `UP.UpperBound = true` in `getUnrollingPreferences`,
+; the loop should be unrolled.
+
+define void @test(i1 %cond) {
+; CHECK-LABEL: @test(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_BODY:%.*]]
+; CHECK:       for.body:
+; CHECK-NEXT:    switch i32 0, label [[SW_DEFAULT:%.*]] [
+; CHECK-NEXT:    i32 2, label [[LATCH:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       sw.default:
+; CHECK-NEXT:    tail call void @foo()
+; CHECK-NEXT:    br label [[LATCH]]
+; CHECK:       latch:
+; CHECK-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[FOR_BODY_1:%.*]]
+; CHECK:       for.end:
+; CHECK-NEXT:    ret void
+; CHECK:       for.body.1:
+; CHECK-NEXT:    switch i32 1, label [[SW_DEFAULT_1:%.*]] [
+; CHECK-NEXT:    i32 2, label [[LATCH_1:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       sw.default.1:
+; CHECK-NEXT:    tail call void @foo()
+; CHECK-NEXT:    br label [[LATCH_1]]
+; CHECK:       latch.1:
+; CHECK-NEXT:    br i1 [[COND:%.*]], label [[FOR_END]], label [[FOR_BODY_2:%.*]]
+; CHECK:       for.body.2:
+; CHECK-NEXT:    switch i32 2, label [[SW_DEFAULT_2:%.*]] [
+; CHECK-NEXT:    i32 2, label [[LATCH_2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       sw.default.2:
+; CHECK-NEXT:    tail call void @foo()
+; CHECK-NEXT:    br label [[LATCH_2]]
+; CHECK:       latch.2:
+; CHECK-NEXT:    br label [[FOR_END]]
+;
+entry:
+  %0 = select i1 %cond, i32 2, i32 3
+  br label %for.body
+
+for.body:
+  %i.017 = phi i32 [ 0, %entry ], [ %inc, %latch ]
+  switch i32 %i.017, label %sw.default [
+  i32 2, label %latch
+  ]
+
+sw.default:
+  tail call void @foo()
+  br label %latch
+
+latch:
+  %inc = add nuw nsw i32 %i.017, 1
+  %exitcond.not = icmp eq i32 %inc, %0
+  br i1 %exitcond.not, label %for.end, label %for.body
+
+for.end:
+  ret void
+}
+
+declare void @foo()