The patch lowers vector rint/nearbyint like vp.rint/nearbyint.
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D148619
setOperationAction({ISD::FMINNUM, ISD::FMAXNUM}, VT, Legal);
- setOperationAction(
- {ISD::FTRUNC, ISD::FCEIL, ISD::FFLOOR, ISD::FROUND, ISD::FROUNDEVEN},
- VT, Custom);
+ setOperationAction({ISD::FTRUNC, ISD::FCEIL, ISD::FFLOOR, ISD::FROUND,
+ ISD::FROUNDEVEN, ISD::FRINT, ISD::FNEARBYINT},
+ VT, Custom);
setOperationAction(FloatingPointVecReduceOps, VT, Custom);
setOperationAction(ISD::FLOG, VT, Expand);
setOperationAction(ISD::FLOG2, VT, Expand);
setOperationAction(ISD::FLOG10, VT, Expand);
- setOperationAction(ISD::FRINT, VT, Expand);
- setOperationAction(ISD::FNEARBYINT, VT, Expand);
setOperationAction(ISD::FCOPYSIGN, VT, Legal);
setOperationAction({ISD::FP_ROUND, ISD::FP_EXTEND}, VT, Custom);
setOperationAction({ISD::FTRUNC, ISD::FCEIL, ISD::FFLOOR, ISD::FROUND,
- ISD::FROUNDEVEN},
+ ISD::FROUNDEVEN, ISD::FRINT, ISD::FNEARBYINT},
VT, Custom);
setCondCodeAction(VFPCCToExpand, VT, Expand);
Truncated = DAG.getNode(RISCVISD::VFCVT_RTZ_X_F_VL, DL, IntVT, Src,
Mask, VL);
break;
+ case ISD::FRINT:
case ISD::VP_FRINT:
Truncated = DAG.getNode(RISCVISD::VFCVT_X_F_VL, DL, IntVT, Src, Mask, VL);
break;
+ case ISD::FNEARBYINT:
case ISD::VP_FNEARBYINT:
Truncated = DAG.getNode(RISCVISD::VFROUND_NOEXCEPT_VL, DL, ContainerVT, Src,
Mask, VL);
}
// VFROUND_NOEXCEPT_VL includes SINT_TO_FP_VL.
- if (Op.getOpcode() != ISD::VP_FNEARBYINT)
+ if (Truncated.getOpcode() != RISCVISD::VFROUND_NOEXCEPT_VL)
Truncated = DAG.getNode(RISCVISD::SINT_TO_FP_VL, DL, ContainerVT, Truncated,
Mask, VL);
case ISD::FTRUNC:
case ISD::FCEIL:
case ISD::FFLOOR:
+ case ISD::FNEARBYINT:
case ISD::FRINT:
case ISD::FROUND:
case ISD::FROUNDEVEN:
{Intrinsic::roundeven, MVT::nxv2f64, 9},
{Intrinsic::roundeven, MVT::nxv4f64, 9},
{Intrinsic::roundeven, MVT::nxv8f64, 9},
+ {Intrinsic::rint, MVT::v2f32, 7},
+ {Intrinsic::rint, MVT::v4f32, 7},
+ {Intrinsic::rint, MVT::v8f32, 7},
+ {Intrinsic::rint, MVT::v16f32, 7},
+ {Intrinsic::rint, MVT::nxv1f32, 7},
+ {Intrinsic::rint, MVT::nxv2f32, 7},
+ {Intrinsic::rint, MVT::nxv4f32, 7},
+ {Intrinsic::rint, MVT::nxv8f32, 7},
+ {Intrinsic::rint, MVT::nxv16f32, 7},
+ {Intrinsic::rint, MVT::v2f64, 7},
+ {Intrinsic::rint, MVT::v4f64, 7},
+ {Intrinsic::rint, MVT::v8f64, 7},
+ {Intrinsic::rint, MVT::v16f64, 7},
+ {Intrinsic::rint, MVT::nxv1f64, 7},
+ {Intrinsic::rint, MVT::nxv2f64, 7},
+ {Intrinsic::rint, MVT::nxv4f64, 7},
+ {Intrinsic::rint, MVT::nxv8f64, 7},
+ {Intrinsic::nearbyint, MVT::v2f32, 9},
+ {Intrinsic::nearbyint, MVT::v4f32, 9},
+ {Intrinsic::nearbyint, MVT::v8f32, 9},
+ {Intrinsic::nearbyint, MVT::v16f32, 9},
+ {Intrinsic::nearbyint, MVT::nxv1f32, 9},
+ {Intrinsic::nearbyint, MVT::nxv2f32, 9},
+ {Intrinsic::nearbyint, MVT::nxv4f32, 9},
+ {Intrinsic::nearbyint, MVT::nxv8f32, 9},
+ {Intrinsic::nearbyint, MVT::nxv16f32, 9},
+ {Intrinsic::nearbyint, MVT::v2f64, 9},
+ {Intrinsic::nearbyint, MVT::v4f64, 9},
+ {Intrinsic::nearbyint, MVT::v8f64, 9},
+ {Intrinsic::nearbyint, MVT::v16f64, 9},
+ {Intrinsic::nearbyint, MVT::nxv1f64, 9},
+ {Intrinsic::nearbyint, MVT::nxv2f64, 9},
+ {Intrinsic::nearbyint, MVT::nxv4f64, 9},
+ {Intrinsic::nearbyint, MVT::nxv8f64, 9},
{Intrinsic::bswap, MVT::v2i16, 3},
{Intrinsic::bswap, MVT::v4i16, 3},
{Intrinsic::bswap, MVT::v8i16, 3},
define void @rint() {
; CHECK-LABEL: 'rint'
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.rint.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %2 = call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %3 = call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %4 = call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %5 = call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.rint.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 19 for instruction: %12 = call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 39 for instruction: %13 = call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 79 for instruction: %14 = call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 159 for instruction: %15 = call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %19 = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call float @llvm.rint.f32(float undef)
define void @nearbyint() {
; CHECK-LABEL: 'nearbyint'
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call float @llvm.nearbyint.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %2 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %3 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 95 for instruction: %4 = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 191 for instruction: %5 = call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %6 = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %7 = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %8 = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %9 = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %10 = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %11 = call double @llvm.nearbyint.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 23 for instruction: %12 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %13 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 95 for instruction: %14 = call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 191 for instruction: %15 = call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %16 = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %17 = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %18 = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Invalid cost for instruction: %19 = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: ret void
;
call float @llvm.nearbyint.f32(float undef)
}
declare <2 x double> @llvm.round.v2f64(<2 x double>)
+define void @rint_v8f16(ptr %x) {
+; CHECK-LABEL: rint_v8f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma
+; CHECK-NEXT: vle16.v v8, (a0)
+; CHECK-NEXT: lui a1, %hi(.LCPI103_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI103_0)(a1)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse16.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <8 x half>, ptr %x
+ %b = call <8 x half> @llvm.rint.v8f16(<8 x half> %a)
+ store <8 x half> %b, ptr %x
+ ret void
+}
+declare <8 x half> @llvm.rint.v8f16(<8 x half>)
+
+define void @rint_v4f32(ptr %x) {
+; CHECK-LABEL: rint_v4f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma
+; CHECK-NEXT: vle32.v v8, (a0)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a1, 307200
+; CHECK-NEXT: fmv.w.x fa5, a1
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse32.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <4 x float>, ptr %x
+ %b = call <4 x float> @llvm.rint.v4f32(<4 x float> %a)
+ store <4 x float> %b, ptr %x
+ ret void
+}
+declare <4 x float> @llvm.rint.v4f32(<4 x float>)
+
+define void @rint_v2f64(ptr %x) {
+; CHECK-LABEL: rint_v2f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 2, e64, m1, ta, ma
+; CHECK-NEXT: vle64.v v8, (a0)
+; CHECK-NEXT: lui a1, %hi(.LCPI105_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI105_0)(a1)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse64.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <2 x double>, ptr %x
+ %b = call <2 x double> @llvm.rint.v2f64(<2 x double> %a)
+ store <2 x double> %b, ptr %x
+ ret void
+}
+declare <2 x double> @llvm.rint.v2f64(<2 x double>)
+
+define void @nearbyint_v8f16(ptr %x) {
+; CHECK-LABEL: nearbyint_v8f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 8, e16, m1, ta, ma
+; CHECK-NEXT: vle16.v v8, (a0)
+; CHECK-NEXT: lui a1, %hi(.LCPI106_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI106_0)(a1)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a1
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a1
+; CHECK-NEXT: vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse16.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <8 x half>, ptr %x
+ %b = call <8 x half> @llvm.nearbyint.v8f16(<8 x half> %a)
+ store <8 x half> %b, ptr %x
+ ret void
+}
+declare <8 x half> @llvm.nearbyint.v8f16(<8 x half>)
+
+define void @nearbyint_v4f32(ptr %x) {
+; CHECK-LABEL: nearbyint_v4f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 4, e32, m1, ta, ma
+; CHECK-NEXT: vle32.v v8, (a0)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a1, 307200
+; CHECK-NEXT: fmv.w.x fa5, a1
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a1
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a1
+; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse32.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <4 x float>, ptr %x
+ %b = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %a)
+ store <4 x float> %b, ptr %x
+ ret void
+}
+declare <4 x float> @llvm.nearbyint.v4f32(<4 x float>)
+
+define void @nearbyint_v2f64(ptr %x) {
+; CHECK-LABEL: nearbyint_v2f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetivli zero, 2, e64, m1, ta, ma
+; CHECK-NEXT: vle64.v v8, (a0)
+; CHECK-NEXT: lui a1, %hi(.LCPI108_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI108_0)(a1)
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a1
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a1
+; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: vse64.v v8, (a0)
+; CHECK-NEXT: ret
+ %a = load <2 x double>, ptr %x
+ %b = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %a)
+ store <2 x double> %b, ptr %x
+ ret void
+}
+declare <2 x double> @llvm.nearbyint.v2f64(<2 x double>)
+
define void @fmuladd_v8f16(ptr %x, ptr %y, ptr %z) {
; CHECK-LABEL: fmuladd_v8f16:
; CHECK: # %bb.0:
--- /dev/null
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=ilp32d \
+; RUN: -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=lp64d \
+; RUN: -verify-machineinstrs < %s | FileCheck %s
+
+define <vscale x 1 x half> @nearbyint_nxv1f16(<vscale x 1 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv1f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI0_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI0_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x half> @llvm.nearbyint.nxv1f16(<vscale x 1 x half> %x)
+ ret <vscale x 1 x half> %a
+}
+declare <vscale x 1 x half> @llvm.nearbyint.nxv1f16(<vscale x 1 x half>)
+
+define <vscale x 2 x half> @nearbyint_nxv2f16(<vscale x 2 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv2f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI1_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI1_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x half> @llvm.nearbyint.nxv2f16(<vscale x 2 x half> %x)
+ ret <vscale x 2 x half> %a
+}
+declare <vscale x 2 x half> @llvm.nearbyint.nxv2f16(<vscale x 2 x half>)
+
+define <vscale x 4 x half> @nearbyint_nxv4f16(<vscale x 4 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv4f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI2_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI2_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x half> @llvm.nearbyint.nxv4f16(<vscale x 4 x half> %x)
+ ret <vscale x 4 x half> %a
+}
+declare <vscale x 4 x half> @llvm.nearbyint.nxv4f16(<vscale x 4 x half>)
+
+define <vscale x 8 x half> @nearbyint_nxv8f16(<vscale x 8 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv8f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI3_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI3_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x half> @llvm.nearbyint.nxv8f16(<vscale x 8 x half> %x)
+ ret <vscale x 8 x half> %a
+}
+declare <vscale x 8 x half> @llvm.nearbyint.nxv8f16(<vscale x 8 x half>)
+
+define <vscale x 16 x half> @nearbyint_nxv16f16(<vscale x 16 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv16f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI4_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI4_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 16 x half> @llvm.nearbyint.nxv16f16(<vscale x 16 x half> %x)
+ ret <vscale x 16 x half> %a
+}
+declare <vscale x 16 x half> @llvm.nearbyint.nxv16f16(<vscale x 16 x half>)
+
+define <vscale x 32 x half> @nearbyint_nxv32f16(<vscale x 32 x half> %x) {
+; CHECK-LABEL: nearbyint_nxv32f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI5_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI5_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e16, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 32 x half> @llvm.nearbyint.nxv32f16(<vscale x 32 x half> %x)
+ ret <vscale x 32 x half> %a
+}
+declare <vscale x 32 x half> @llvm.nearbyint.nxv32f16(<vscale x 32 x half>)
+
+define <vscale x 1 x float> @nearbyint_nxv1f32(<vscale x 1 x float> %x) {
+; CHECK-LABEL: nearbyint_nxv1f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> %x)
+ ret <vscale x 1 x float> %a
+}
+declare <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float>)
+
+define <vscale x 2 x float> @nearbyint_nxv2f32(<vscale x 2 x float> %x) {
+; CHECK-LABEL: nearbyint_nxv2f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> %x)
+ ret <vscale x 2 x float> %a
+}
+declare <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float>)
+
+define <vscale x 4 x float> @nearbyint_nxv4f32(<vscale x 4 x float> %x) {
+; CHECK-LABEL: nearbyint_nxv4f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> %x)
+ ret <vscale x 4 x float> %a
+}
+declare <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float>)
+
+define <vscale x 8 x float> @nearbyint_nxv8f32(<vscale x 8 x float> %x) {
+; CHECK-LABEL: nearbyint_nxv8f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> %x)
+ ret <vscale x 8 x float> %a
+}
+declare <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float>)
+
+define <vscale x 16 x float> @nearbyint_nxv16f32(<vscale x 16 x float> %x) {
+; CHECK-LABEL: nearbyint_nxv16f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e32, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> %x)
+ ret <vscale x 16 x float> %a
+}
+declare <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float>)
+
+define <vscale x 1 x double> @nearbyint_nxv1f64(<vscale x 1 x double> %x) {
+; CHECK-LABEL: nearbyint_nxv1f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI11_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI11_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> %x)
+ ret <vscale x 1 x double> %a
+}
+declare <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double>)
+
+define <vscale x 2 x double> @nearbyint_nxv2f64(<vscale x 2 x double> %x) {
+; CHECK-LABEL: nearbyint_nxv2f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI12_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI12_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> %x)
+ ret <vscale x 2 x double> %a
+}
+declare <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double>)
+
+define <vscale x 4 x double> @nearbyint_nxv4f64(<vscale x 4 x double> %x) {
+; CHECK-LABEL: nearbyint_nxv4f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI13_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI13_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> %x)
+ ret <vscale x 4 x double> %a
+}
+declare <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double>)
+
+define <vscale x 8 x double> @nearbyint_nxv8f64(<vscale x 8 x double> %x) {
+; CHECK-LABEL: nearbyint_nxv8f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI14_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI14_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: frflags a0
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: fsflags a0
+; CHECK-NEXT: vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> %x)
+ ret <vscale x 8 x double> %a
+}
+declare <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double>)
--- /dev/null
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=riscv32 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=ilp32d \
+; RUN: -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=riscv64 -mattr=+d,+zfh,+experimental-zvfh,+v -target-abi=lp64d \
+; RUN: -verify-machineinstrs < %s | FileCheck %s
+
+define <vscale x 1 x half> @rint_nxv1f16(<vscale x 1 x half> %x) {
+; CHECK-LABEL: rint_nxv1f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI0_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI0_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, mf4, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, mf4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x half> @llvm.rint.nxv1f16(<vscale x 1 x half> %x)
+ ret <vscale x 1 x half> %a
+}
+declare <vscale x 1 x half> @llvm.rint.nxv1f16(<vscale x 1 x half>)
+
+define <vscale x 2 x half> @rint_nxv2f16(<vscale x 2 x half> %x) {
+; CHECK-LABEL: rint_nxv2f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI1_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI1_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, mf2, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, mf2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x half> @llvm.rint.nxv2f16(<vscale x 2 x half> %x)
+ ret <vscale x 2 x half> %a
+}
+declare <vscale x 2 x half> @llvm.rint.nxv2f16(<vscale x 2 x half>)
+
+define <vscale x 4 x half> @rint_nxv4f16(<vscale x 4 x half> %x) {
+; CHECK-LABEL: rint_nxv4f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI2_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI2_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x half> @llvm.rint.nxv4f16(<vscale x 4 x half> %x)
+ ret <vscale x 4 x half> %a
+}
+declare <vscale x 4 x half> @llvm.rint.nxv4f16(<vscale x 4 x half>)
+
+define <vscale x 8 x half> @rint_nxv8f16(<vscale x 8 x half> %x) {
+; CHECK-LABEL: rint_nxv8f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI3_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI3_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x half> @llvm.rint.nxv8f16(<vscale x 8 x half> %x)
+ ret <vscale x 8 x half> %a
+}
+declare <vscale x 8 x half> @llvm.rint.nxv8f16(<vscale x 8 x half>)
+
+define <vscale x 16 x half> @rint_nxv16f16(<vscale x 16 x half> %x) {
+; CHECK-LABEL: rint_nxv16f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI4_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI4_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 16 x half> @llvm.rint.nxv16f16(<vscale x 16 x half> %x)
+ ret <vscale x 16 x half> %a
+}
+declare <vscale x 16 x half> @llvm.rint.nxv16f16(<vscale x 16 x half>)
+
+define <vscale x 32 x half> @rint_nxv32f16(<vscale x 32 x half> %x) {
+; CHECK-LABEL: rint_nxv32f16:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI5_0)
+; CHECK-NEXT: flh fa5, %lo(.LCPI5_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e16, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e16, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 32 x half> @llvm.rint.nxv32f16(<vscale x 32 x half> %x)
+ ret <vscale x 32 x half> %a
+}
+declare <vscale x 32 x half> @llvm.rint.nxv32f16(<vscale x 32 x half>)
+
+define <vscale x 1 x float> @rint_nxv1f32(<vscale x 1 x float> %x) {
+; CHECK-LABEL: rint_nxv1f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, mf2, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, mf2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> %x)
+ ret <vscale x 1 x float> %a
+}
+declare <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float>)
+
+define <vscale x 2 x float> @rint_nxv2f32(<vscale x 2 x float> %x) {
+; CHECK-LABEL: rint_nxv2f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> %x)
+ ret <vscale x 2 x float> %a
+}
+declare <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float>)
+
+define <vscale x 4 x float> @rint_nxv4f32(<vscale x 4 x float> %x) {
+; CHECK-LABEL: rint_nxv4f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> %x)
+ ret <vscale x 4 x float> %a
+}
+declare <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float>)
+
+define <vscale x 8 x float> @rint_nxv8f32(<vscale x 8 x float> %x) {
+; CHECK-LABEL: rint_nxv8f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> %x)
+ ret <vscale x 8 x float> %a
+}
+declare <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float>)
+
+define <vscale x 16 x float> @rint_nxv16f32(<vscale x 16 x float> %x) {
+; CHECK-LABEL: rint_nxv16f32:
+; CHECK: # %bb.0:
+; CHECK-NEXT: vsetvli a0, zero, e32, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: lui a0, 307200
+; CHECK-NEXT: fmv.w.x fa5, a0
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e32, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> %x)
+ ret <vscale x 16 x float> %a
+}
+declare <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float>)
+
+define <vscale x 1 x double> @rint_nxv1f64(<vscale x 1 x double> %x) {
+; CHECK-LABEL: rint_nxv1f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI11_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI11_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT: vfabs.v v9, v8
+; CHECK-NEXT: vmflt.vf v0, v9, fa5
+; CHECK-NEXT: vfcvt.x.f.v v9, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v9, v9, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e64, m1, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v9, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> %x)
+ ret <vscale x 1 x double> %a
+}
+declare <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double>)
+
+define <vscale x 2 x double> @rint_nxv2f64(<vscale x 2 x double> %x) {
+; CHECK-LABEL: rint_nxv2f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI12_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI12_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m2, ta, ma
+; CHECK-NEXT: vfabs.v v10, v8
+; CHECK-NEXT: vmflt.vf v0, v10, fa5
+; CHECK-NEXT: vfcvt.x.f.v v10, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v10, v10, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e64, m2, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v10, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> %x)
+ ret <vscale x 2 x double> %a
+}
+declare <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double>)
+
+define <vscale x 4 x double> @rint_nxv4f64(<vscale x 4 x double> %x) {
+; CHECK-LABEL: rint_nxv4f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI13_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI13_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m4, ta, ma
+; CHECK-NEXT: vfabs.v v12, v8
+; CHECK-NEXT: vmflt.vf v0, v12, fa5
+; CHECK-NEXT: vfcvt.x.f.v v12, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v12, v12, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e64, m4, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v12, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> %x)
+ ret <vscale x 4 x double> %a
+}
+declare <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double>)
+
+define <vscale x 8 x double> @rint_nxv8f64(<vscale x 8 x double> %x) {
+; CHECK-LABEL: rint_nxv8f64:
+; CHECK: # %bb.0:
+; CHECK-NEXT: lui a0, %hi(.LCPI14_0)
+; CHECK-NEXT: fld fa5, %lo(.LCPI14_0)(a0)
+; CHECK-NEXT: vsetvli a0, zero, e64, m8, ta, ma
+; CHECK-NEXT: vfabs.v v16, v8
+; CHECK-NEXT: vmflt.vf v0, v16, fa5
+; CHECK-NEXT: vfcvt.x.f.v v16, v8, v0.t
+; CHECK-NEXT: vfcvt.f.x.v v16, v16, v0.t
+; CHECK-NEXT: vsetvli zero, zero, e64, m8, ta, mu
+; CHECK-NEXT: vfsgnj.vv v8, v16, v8, v0.t
+; CHECK-NEXT: ret
+ %a = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> %x)
+ ret <vscale x 8 x double> %a
+}
+declare <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double>)