ScalarEvolution: Compute exit counts for loops with a power-of-2 step.

If we have a loop of the form
for (unsigned n = 0; n != (k & -32); n += 32) {}
then we know that n is always divisible by 32 and the loop must
terminate. Even if we have a condition where the loop counter will
overflow it'll always hold this invariant.

PR19183. Our loop vectorizer creates this pattern and it's also
occasionally formed by loop counters derived from pointers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204728 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Analysis/ScalarEvolution.cpp b/lib/Analysis/ScalarEvolution.cpp
index 39bf95b92d9d8f42a55e72e85b539772af113217..08de6213e2242a56d78f3a381185550c65165f3a 100644 (file)
--- a/lib/Analysis/ScalarEvolution.cpp
+++ b/lib/Analysis/ScalarEvolution.cpp
@@ -5744,6 +5744,16 @@ ScalarEvolution::HowFarToZero(const SCEV *V, const Loop *L, bool IsSubExpr) {
       getUDivExpr(Distance, CountDown ? getNegativeSCEV(Step) : Step);
     return ExitLimit(Exact, Exact, /*MustExit=*/false);
   }
+
+  // If Step is a power of two that evenly divides Start we know that the loop
+  // will always terminate.  Start may not be a constant so we just have the
+  // number of trailing zeros available.  This is safe even in presence of
+  // overflow as the recurrence will overflow to exactly 0.
+  const APInt &StepV = StepC->getValue()->getValue();
+  if (StepV.isPowerOf2() &&
+      GetMinTrailingZeros(getNegativeSCEV(Start)) >= StepV.countTrailingZeros())
+    return getUDivExactExpr(Distance, CountDown ? getNegativeSCEV(Step) : Step);
+
   // Then, try to solve the above equation provided that Start is constant.
   if (const SCEVConstant *StartC = dyn_cast<SCEVConstant>(Start))
     return SolveLinEquationWithOverflow(StepC->getValue()->getValue(),

diff --git a/test/Analysis/ScalarEvolution/trip-count-pow2.ll b/test/Analysis/ScalarEvolution/trip-count-pow2.ll
new file mode 100644 (file)
index 0000000..2c5b72e
--- /dev/null
+++ b/test/Analysis/ScalarEvolution/trip-count-pow2.ll
@@ -0,0 +1,53 @@
+; RUN: opt < %s -scalar-evolution -analyze | FileCheck %s
+
+define void @test1(i32 %n) {
+entry:
+  %s = mul i32 %n, 96
+  br label %loop
+loop:
+  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
+  %i.next = add i32 %i, 32
+  %t = icmp ne i32 %i.next, %s
+  br i1 %t, label %loop, label %exit
+exit:
+  ret void
+
+; CHECK-LABEL: @test1
+; CHECK: Loop %loop: backedge-taken count is ((-32 + (96 * %n)) /u 32)
+; CHECK: Loop %loop: max backedge-taken count is ((-32 + (96 * %n)) /u 32)
+}
+
+; PR19183
+define i32 @test2(i32 %n) {
+entry:
+  %s = and i32 %n, -32
+  br label %loop
+loop:
+  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
+  %i.next = add i32 %i, 32
+  %t = icmp ne i32 %i.next, %s
+  br i1 %t, label %loop, label %exit
+exit:
+  ret i32 %i
+
+; CHECK-LABEL: @test2
+; CHECK: Loop %loop: backedge-taken count is ((-32 + (32 * (%n /u 32))) /u 32)
+; CHECK: Loop %loop: max backedge-taken count is ((-32 + (32 * (%n /u 32))) /u 32)
+}
+
+define void @test3(i32 %n) {
+entry:
+  %s = mul i32 %n, 96
+  br label %loop
+loop:
+  %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
+  %i.next = add i32 %i, 96
+  %t = icmp ne i32 %i.next, %s
+  br i1 %t, label %loop, label %exit
+exit:
+  ret void
+
+; CHECK-LABEL: @test3
+; CHECK: Loop %loop: Unpredictable backedge-taken count.
+; CHECK: Loop %loop: Unpredictable max backedge-taken count.
+}