From 961d666be4f3714452ff9f15470f9255d85f0506 Mon Sep 17 00:00:00 2001
From: Duncan Sands <baldrick@free.fr>
Date: Sat, 7 Apr 2012 20:04:00 +0000
Subject: [PATCH] Convert floating point division by a constant into
 multiplication by the reciprocal if converting to the reciprocal is exact. 
 Do it even if inexact if -ffast-math.  This substantially speeds up ac.f90
 from the polyhedron benchmarks.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@154265 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/CodeGen/SelectionDAG/DAGCombiner.cpp | 13 ++++++++++
 test/CodeGen/ARM/vdiv_combine.ll         | 12 ++++-----
 test/CodeGen/X86/fdiv.ll                 | 32 ++++++++++++++++++++++++
 3 files changed, 51 insertions(+), 6 deletions(-)
 create mode 100644 test/CodeGen/X86/fdiv.ll

diff --git a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index e03668777d8..e0fd3abfb45 100644
--- a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -5725,6 +5725,19 @@ SDValue DAGCombiner::visitFDIV(SDNode *N) {
   if (N0CFP && N1CFP && VT != MVT::ppcf128)
     return DAG.getNode(ISD::FDIV, N->getDebugLoc(), VT, N0, N1);
 
+  // fold (fdiv X, c2) -> fmul X, 1/c2 if there is no precision loss or if
+  // losing precision is acceptable.
+  if (N1CFP && VT != MVT::ppcf128) {
+    // Compute the reciprocal 1.0 / c2.
+    APFloat N1APF = N1CFP->getValueAPF();
+    APFloat Recip(N1APF.getSemantics(), 1); // 1.0
+    APFloat::opStatus st = Recip.divide(N1APF, APFloat::rmNearestTiesToEven);
+    // Only do the transform if the reciprocal is not too horrible (eg not NaN).
+    if (st == APFloat::opOK || (st == APFloat::opInexact &&
+                                DAG.getTarget().Options.UnsafeFPMath))
+      return DAG.getNode(ISD::FMUL, N->getDebugLoc(), VT, N0,
+                         DAG.getConstantFP(Recip, VT));
+  }
 
   // (fdiv (fneg X), (fneg Y)) -> (fdiv X, Y)
   if (char LHSNeg = isNegatibleForFree(N0, LegalOperations, TLI,
diff --git a/test/CodeGen/ARM/vdiv_combine.ll b/test/CodeGen/ARM/vdiv_combine.ll
index 13873932abd..7fddbed1ed5 100644
--- a/test/CodeGen/ARM/vdiv_combine.ll
+++ b/test/CodeGen/ARM/vdiv_combine.ll
@@ -8,7 +8,7 @@ declare void @foo_int32x4_t(<4 x i32>)
 
 ; Test signed conversion.
 ; CHECK: t1
-; CHECK-NOT: vdiv
+; CHECK-NOT: {{vdiv|vmul}}
 define void @t1() nounwind {
 entry:
   %tmp = load i32* @iin, align 4, !tbaa !3
@@ -24,7 +24,7 @@ declare void @foo_float32x2_t(<2 x float>)
 
 ; Test unsigned conversion.
 ; CHECK: t2
-; CHECK-NOT: vdiv
+; CHECK-NOT: {{vdiv|vmul}}
 define void @t2() nounwind {
 entry:
   %tmp = load i32* @uin, align 4, !tbaa !3
@@ -38,7 +38,7 @@ entry:
 
 ; Test which should not fold due to non-power of 2.
 ; CHECK: t3
-; CHECK: vdiv
+; CHECK: {{vdiv|vmul}}
 define void @t3() nounwind {
 entry:
   %tmp = load i32* @iin, align 4, !tbaa !3
@@ -52,7 +52,7 @@ entry:
 
 ; Test which should not fold due to power of 2 out of range.
 ; CHECK: t4
-; CHECK: vdiv
+; CHECK: {{vdiv|vmul}}
 define void @t4() nounwind {
 entry:
   %tmp = load i32* @iin, align 4, !tbaa !3
@@ -66,7 +66,7 @@ entry:
 
 ; Test case where const is max power of 2 (i.e., 2^32).
 ; CHECK: t5
-; CHECK-NOT: vdiv
+; CHECK-NOT: {{vdiv|vmul}}
 define void @t5() nounwind {
 entry:
   %tmp = load i32* @iin, align 4, !tbaa !3
@@ -80,7 +80,7 @@ entry:
 
 ; Test quadword.
 ; CHECK: t6
-; CHECK-NOT: vdiv
+; CHECK-NOT: {{vdiv|vmul}}
 define void @t6() nounwind {
 entry:
   %tmp = load i32* @iin, align 4, !tbaa !3
diff --git a/test/CodeGen/X86/fdiv.ll b/test/CodeGen/X86/fdiv.ll
new file mode 100644
index 00000000000..553f14efa13
--- /dev/null
+++ b/test/CodeGen/X86/fdiv.ll
@@ -0,0 +1,32 @@
+; RUN: llc < %s -march=x86-64 | FileCheck %s
+; RUN: llc < %s -march=x86-64 -enable-unsafe-fp-math | FileCheck -check-prefix=UNSAFE %s
+
+define double @exact(double %x) {
+; Exact division by a constant always converted to multiplication.
+; CHECK: @exact
+; CHECK: mulsd
+; UNSAFE: @exact
+; UNSAFE: mulsd
+  %div = fdiv double %x, 2.0
+  ret double %div
+}
+
+define double @inexact(double %x) {
+; Inexact division by a constant converted to multiplication if unsafe-math.
+; CHECK: @inexact
+; CHECK: divsd
+; UNSAFE: @inexact
+; UNSAFE: mulsd
+  %div = fdiv double %x, 0x41DFFFFFFFC00000 
+  ret double %div
+}
+
+define double @funky(double %x) {
+; No conversion to multiplication if too funky.
+; CHECK: @funky
+; CHECK: divsd
+; UNSAFE: @funky
+; UNSAFE: divsd
+  %div = fdiv double %x, 0.0
+  ret double %div
+}