[llvm] 7704c50 - [RISCV] Use positive 0.0 for the neutral element in fadd reductions if nsz is present.

[Craig Topper via llvm-commits]

Author: Craig Topper
Date: 2021-12-23T10:38:00-06:00
New Revision: 7704c503ecb8b20574fa8a8dc45dd8545216c450

URL: https://github.com/llvm/llvm-project/commit/7704c503ecb8b20574fa8a8dc45dd8545216c450
DIFF: https://github.com/llvm/llvm-project/commit/7704c503ecb8b20574fa8a8dc45dd8545216c450.diff

LOG: [RISCV] Use positive 0.0 for the neutral element in fadd reductions if nsz is present.

-0.0 requires a constant pool. +0.0 can be made with vmv.v.x x0.

Not doing this in getNeutralElement for fear of changing other targets.

Differential Revision: https://reviews.llvm.org/D115978

Added: 
    

Modified: 
    llvm/lib/Target/RISCV/RISCVISelLowering.cpp
    llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll
    llvm/test/CodeGen/RISCV/rvv/vreductions-fp-sdnode.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index 3919aae25630a..4322f9bc1ce7a 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -4538,9 +4538,12 @@ getRVVFPReductionOpAndOperands(SDValue Op, SelectionDAG &DAG, EVT EltVT) {
   switch (Opcode) {
   default:
     llvm_unreachable("Unhandled reduction");
-  case ISD::VECREDUCE_FADD:
-    return std::make_tuple(RISCVISD::VECREDUCE_FADD_VL, Op.getOperand(0),
-                           DAG.getNeutralElement(BaseOpcode, DL, EltVT, Flags));
+  case ISD::VECREDUCE_FADD: {
+    // Use positive zero if we can. It is cheaper to materialize.
+    SDValue Zero =
+        DAG.getConstantFP(Flags.hasNoSignedZeros() ? 0.0 : -0.0, DL, EltVT);
+    return std::make_tuple(RISCVISD::VECREDUCE_FADD_VL, Op.getOperand(0), Zero);
+  }
   case ISD::VECREDUCE_SEQ_FADD:
     return std::make_tuple(RISCVISD::VECREDUCE_SEQ_FADD_VL, Op.getOperand(1),
                            Op.getOperand(0));

diff  --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll
index 2d3c3a2e18e48..9217101c946c4 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-fp.ll
@@ -1408,3 +1408,20 @@ define double @vreduce_fmax_v32f64(<32 x double>* %x) {
   %red = call double @llvm.vector.reduce.fmax.v32f64(<32 x double> %v)
   ret double %red
 }
+
+define float @vreduce_nsz_fadd_v4f32(<4 x float>* %x, float %s) {
+; CHECK-LABEL: vreduce_nsz_fadd_v4f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, mu
+; CHECK-NEXT:    vle32.v v8, (a0)
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, mu
+; CHECK-NEXT:    vmv.v.i v9, 0
+; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, mu
+; CHECK-NEXT:    vfredusum.vs v8, v8, v9
+; CHECK-NEXT:    vfmv.f.s ft0, v8
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %v = load <4 x float>, <4 x float>* %x
+  %red = call reassoc nsz float @llvm.vector.reduce.fadd.v4f32(float %s, <4 x float> %v)
+  ret float %red
+}

diff  --git a/llvm/test/CodeGen/RISCV/rvv/vreductions-fp-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vreductions-fp-sdnode.ll
index ce802a00163ab..8404398fdc855 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vreductions-fp-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vreductions-fp-sdnode.ll
@@ -882,3 +882,17 @@ define double @vreduce_fmax_nxv16f64(<vscale x 16 x double> %v) {
   %red = call double @llvm.vector.reduce.fmax.nxv16f64(<vscale x 16 x double> %v)
   ret double %red
 }
+
+define float @vreduce_nsz_fadd_nxv1f32(<vscale x 1 x float> %v, float %s) {
+; CHECK-LABEL: vreduce_nsz_fadd_nxv1f32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, mu
+; CHECK-NEXT:    vmv.v.i v9, 0
+; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, mu
+; CHECK-NEXT:    vfredusum.vs v8, v8, v9
+; CHECK-NEXT:    vfmv.f.s ft0, v8
+; CHECK-NEXT:    fadd.s fa0, fa0, ft0
+; CHECK-NEXT:    ret
+  %red = call reassoc nsz float @llvm.vector.reduce.fadd.nxv1f32(float %s, <vscale x 1 x float> %v)
+  ret float %red
+}