[RISCV][VP] Add fp test of widen and split for vp.setcc

[Jimerlife]

Summary

Add fp test of widen and split for vp.setcc

[frasercrmck]

LGTM, thanks

llvm/test/CodeGen/RISCV/rvv/fixed-vectors-setcc-fp-vp.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -mattr=+v,+zfh,+experimental-zvfh -target-abi=ilp32d -riscv-v-vector-bits-min=128 \
+; RUN: llc -mtriple=riscv32 -mattr=+v,+m,+zfh,+experimental-zvfh -target-abi=ilp32d -riscv-v-vector-bits-min=128 \
 ; RUN:   -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=riscv64 -mattr=+v,+zfh,+experimental-zvfh -target-abi=lp64d -riscv-v-vector-bits-min=128 \
+; RUN: llc -mtriple=riscv64 -mattr=+v,+m,+zfh,+experimental-zvfh -target-abi=lp64d -riscv-v-vector-bits-min=128 \
 ; RUN:   -verify-machineinstrs < %s | FileCheck %s
 
+declare <7 x i1> @llvm.vp.fcmp.v7f16(<7 x half>, <7 x half>, metadata, <7 x i1>, i32)
+
+define <7 x i1> @fcmp_oeq_vv_v7f16(<7 x half> %va, <7 x half> %vb, <7 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_v7f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9, v0.t
+; CHECK-NEXT:    ret
+  %v = call <7 x i1> @llvm.vp.fcmp.v7f16(<7 x half> %va, <7 x half> %vb, metadata !"oeq", <7 x i1> %m, i32 %evl)
+  ret <7 x i1> %v
+}
+
 declare <8 x i1> @llvm.vp.fcmp.v8f16(<8 x half>, <8 x half>, metadata, <8 x i1>, i32)
 
 define <8 x i1> @fcmp_oeq_vv_v8f16(<8 x half> %va, <8 x half> %vb, <8 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_v8f16:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v0, v8, v9, v0.t
 ; CHECK-NEXT:    ret

 ; CHECK-NEXT:    vmor.mm v0, v9, v8
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <8 x half> poison, half %b, i32 0
   %vb = shufflevector <8 x half> %elt.head, <8 x half> poison, <8 x i32> zeroinitializer
   %v = call <8 x i1> @llvm.vp.fcmp.v8f16(<8 x half> %vb, <8 x half> %va, metadata !"uno", <8 x i1> %m, i32 %evl)
   ret <8 x i1> %v
 }
 
+declare <128 x i1> @llvm.vp.fcmp.v128f16(<128 x half>, <128 x half>, metadata, <128 x i1>, i32)
+
+define <128 x i1> @fcmp_oeq_vv_v128f16(<128 x half> %va, <128 x half> %vb, <128 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_v128f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v1, v0
+; CHECK-NEXT:    addi a1, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a3, 0
+; CHECK-NEXT:    li a1, 64
+; CHECK-NEXT:    addi a4, a0, 128
+; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, mu
+; CHECK-NEXT:    vle16.v v24, (a4)
+; CHECK-NEXT:    vsetivli zero, 8, e8, m1, ta, mu
+; CHECK-NEXT:    addi a4, a2, -64
+; CHECK-NEXT:    vslidedown.vi v0, v0, 8
+; CHECK-NEXT:    bltu a2, a4, .LBB43_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a3, a4
+; CHECK-NEXT:  .LBB43_2:
+; CHECK-NEXT:    vsetvli zero, a1, e16, m8, ta, mu
+; CHECK-NEXT:    vle16.v v8, (a0)
+; CHECK-NEXT:    vsetvli zero, a3, e16, m8, ta, ma
+; CHECK-NEXT:    vmfeq.vv v2, v16, v24, v0.t
+; CHECK-NEXT:    bltu a2, a1, .LBB43_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    li a2, 64
+; CHECK-NEXT:  .LBB43_4:
+; CHECK-NEXT:    vsetvli zero, a2, e16, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v16, v24, v8, v0.t
+; CHECK-NEXT:    vsetivli zero, 16, e8, m1, tu, mu
+; CHECK-NEXT:    vslideup.vi v16, v2, 8
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <128 x i1> @llvm.vp.fcmp.v128f16(<128 x half> %va, <128 x half> %vb, metadata !"oeq", <128 x i1> %m, i32 %evl)
+  ret <128 x i1> %v
+}
+
+declare <7 x i1> @llvm.vp.fcmp.v7f64(<7 x double>, <7 x double>, metadata, <7 x i1>, i32)
+
+define <7 x i1> @fcmp_oeq_vv_v7f64(<7 x double> %va, <7 x double> %vb, <7 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_v7f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vmfeq.vv v16, v8, v12, v0.t
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    ret
+  %v = call <7 x i1> @llvm.vp.fcmp.v7f64(<7 x double> %va, <7 x double> %vb, metadata !"oeq", <7 x i1> %m, i32 %evl)
+  ret <7 x i1> %v
+}
+
 declare <8 x i1> @llvm.vp.fcmp.v8f64(<8 x double>, <8 x double>, metadata, <8 x i1>, i32)
 
 define <8 x i1> @fcmp_oeq_vv_v8f64(<8 x double> %va, <8 x double> %vb, <8 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_v8f64:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v16, v8, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v16

 ; CHECK-NEXT:    vsetvli zero, zero, e8, mf2, ta, mu
 ; CHECK-NEXT:    vmor.mm v0, v16, v12
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <8 x double> poison, double %b, i32 0
   %vb = shufflevector <8 x double> %elt.head, <8 x double> poison, <8 x i32> zeroinitializer
   %v = call <8 x i1> @llvm.vp.fcmp.v8f64(<8 x double> %vb, <8 x double> %va, metadata !"uno", <8 x i1> %m, i32 %evl)
   ret <8 x i1> %v
 }
+
+declare <32 x i1> @llvm.vp.fcmp.v32f64(<32 x double>, <32 x double>, metadata, <32 x i1>, i32)
+
+define <32 x i1> @fcmp_oeq_vv_v32f64(<32 x double> %va, <32 x double> %vb, <32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_v32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v2, v0
+; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
+; CHECK-NEXT:    vslidedown.vi v0, v0, 2
+; CHECK-NEXT:    addi a1, a0, 128
+; CHECK-NEXT:    vsetivli zero, 16, e64, m8, ta, mu
+; CHECK-NEXT:    vle64.v v24, (a1)
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    addi a3, a2, -16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 4
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a1, 0
+; CHECK-NEXT:    bltu a2, a3, .LBB87_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a1, a3
+; CHECK-NEXT:  .LBB87_2:
+; CHECK-NEXT:    vle64.v v8, (a0)
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, ma
+; CHECK-NEXT:    li a0, 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8re8.v v8, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v1, v16, v8, v0.t
+; CHECK-NEXT:    bltu a2, a0, .LBB87_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    li a2, 16
+; CHECK-NEXT:  .LBB87_4:
+; CHECK-NEXT:    vsetvli zero, a2, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v2
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v16, v24, v8, v0.t
+; CHECK-NEXT:    vsetivli zero, 4, e8, mf4, tu, mu
+; CHECK-NEXT:    vslideup.vi v16, v1, 2
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a1, 24
+; CHECK-NEXT:    mul a0, a0, a1
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <32 x i1> @llvm.vp.fcmp.v32f64(<32 x double> %va, <32 x double> %vb, metadata !"oeq", <32 x i1> %m, i32 %evl)
+  ret <32 x i1> %v
+}

llvm/test/CodeGen/RISCV/rvv/setcc-fp-vp.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=riscv32 -mattr=+v,+zfh,+experimental-zvfh -target-abi=ilp32d \
+; RUN: llc -mtriple=riscv32 -mattr=+v,+m,+zfh,+experimental-zvfh -target-abi=ilp32d \
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=riscv64 -mattr=+v,+zfh,+experimental-zvfh -target-abi=lp64d \
+; RUN: llc -mtriple=riscv64 -mattr=+v,+m,+zfh,+experimental-zvfh -target-abi=lp64d \
 ; RUN:     -verify-machineinstrs < %s | FileCheck %s
 
 declare <vscale x 1 x i1> @llvm.vp.fcmp.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, metadata, <vscale x 1 x i1>, i32)
 
 define <vscale x 1 x i1> @fcmp_oeq_vv_nxv1f16(<vscale x 1 x half> %va, <vscale x 1 x half> %vb, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_nxv1f16:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, ma

 ; CHECK-NEXT:    vmor.mm v0, v9, v8
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <vscale x 1 x half> poison, half %b, i32 0
   %vb = shufflevector <vscale x 1 x half> %elt.head, <vscale x 1 x half> poison, <vscale x 1 x i32> zeroinitializer
   %v = call <vscale x 1 x i1> @llvm.vp.fcmp.nxv1f16(<vscale x 1 x half> %vb, <vscale x 1 x half> %va, metadata !"uno", <vscale x 1 x i1> %m, i32 %evl)
   ret <vscale x 1 x i1> %v
 }
 
+declare <vscale x 3 x i1> @llvm.vp.fcmp.nxv3f16(<vscale x 3 x half>, <vscale x 3 x half>, metadata, <vscale x 3 x i1>, i32)
+
+define <vscale x 3 x i1> @fcmp_oeq_vv_nxv3f16(<vscale x 3 x half> %va, <vscale x 3 x half> %vb, <vscale x 3 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_nxv3f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, ma
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9, v0.t
+; CHECK-NEXT:    ret
+  %v = call <vscale x 3 x i1> @llvm.vp.fcmp.nxv3f16(<vscale x 3 x half> %va, <vscale x 3 x half> %vb, metadata !"oeq", <vscale x 3 x i1> %m, i32 %evl)
+  ret <vscale x 3 x i1> %v
+}
+
 declare <vscale x 8 x i1> @llvm.vp.fcmp.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, metadata, <vscale x 8 x i1>, i32)
 
 define <vscale x 8 x i1> @fcmp_oeq_vv_nxv8f16(<vscale x 8 x half> %va, <vscale x 8 x half> %vb, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_nxv8f16:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v12, v8, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v12

 ; CHECK-NEXT:    vmor.mm v0, v12, v10
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <vscale x 8 x half> poison, half %b, i32 0
   %vb = shufflevector <vscale x 8 x half> %elt.head, <vscale x 8 x half> poison, <vscale x 8 x i32> zeroinitializer
   %v = call <vscale x 8 x i1> @llvm.vp.fcmp.nxv8f16(<vscale x 8 x half> %vb, <vscale x 8 x half> %va, metadata !"uno", <vscale x 8 x i1> %m, i32 %evl)
   ret <vscale x 8 x i1> %v
 }
 
+declare <vscale x 64 x i1> @llvm.vp.fcmp.nxv64f16(<vscale x 64 x half>, <vscale x 64 x half>, metadata, <vscale x 64 x i1>, i32)
+
+define <vscale x 64 x i1> @fcmp_oeq_vv_nxv64f16(<vscale x 64 x half> %va, <vscale x 64 x half> %vb, <vscale x 64 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_nxv64f16:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v1, v0
+; CHECK-NEXT:    addi a1, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    li a4, 0
+; CHECK-NEXT:    csrr a3, vlenb
+; CHECK-NEXT:    srli a1, a3, 1
+; CHECK-NEXT:    vsetvli a5, zero, e8, m1, ta, mu
+; CHECK-NEXT:    slli a5, a3, 3
+; CHECK-NEXT:    add a5, a0, a5
+; CHECK-NEXT:    vl8re16.v v24, (a5)
+; CHECK-NEXT:    slli a3, a3, 2
+; CHECK-NEXT:    sub a5, a2, a3
+; CHECK-NEXT:    vslidedown.vx v0, v0, a1
+; CHECK-NEXT:    bltu a2, a5, .LBB85_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a4, a5
+; CHECK-NEXT:  .LBB85_2:
+; CHECK-NEXT:    vl8re16.v v8, (a0)
+; CHECK-NEXT:    vsetvli zero, a4, e16, m8, ta, ma
+; CHECK-NEXT:    vmfeq.vv v2, v16, v24, v0.t
+; CHECK-NEXT:    bltu a2, a3, .LBB85_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    mv a2, a3
+; CHECK-NEXT:  .LBB85_4:
+; CHECK-NEXT:    vsetvli zero, a2, e16, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v16, v24, v8, v0.t
+; CHECK-NEXT:    add a0, a1, a1
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, tu, mu
+; CHECK-NEXT:    vslideup.vx v16, v2, a1
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 64 x i1> @llvm.vp.fcmp.nxv64f16(<vscale x 64 x half> %va, <vscale x 64 x half> %vb, metadata !"oeq", <vscale x 64 x i1> %m, i32 %evl)
+  ret <vscale x 64 x i1> %v
+}
+
 declare <vscale x 1 x i1> @llvm.vp.fcmp.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, metadata, <vscale x 1 x i1>, i32)
 
 define <vscale x 1 x i1> @fcmp_oeq_vv_nxv1f64(<vscale x 1 x double> %va, <vscale x 1 x double> %vb, <vscale x 1 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_nxv1f64:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v0, v8, v9, v0.t
 ; CHECK-NEXT:    ret

 ; CHECK-NEXT:    vmor.mm v0, v9, v8
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <vscale x 1 x double> poison, double %b, i32 0
   %vb = shufflevector <vscale x 1 x double> %elt.head, <vscale x 1 x double> poison, <vscale x 1 x i32> zeroinitializer
   %v = call <vscale x 1 x i1> @llvm.vp.fcmp.nxv1f64(<vscale x 1 x double> %vb, <vscale x 1 x double> %va, metadata !"uno", <vscale x 1 x i1> %m, i32 %evl)
   ret <vscale x 1 x i1> %v
 }
 
+declare <vscale x 3 x i1> @llvm.vp.fcmp.nxv3f64(<vscale x 3 x double>, <vscale x 3 x double>, metadata, <vscale x 3 x i1>, i32)
+
+define <vscale x 3 x i1> @fcmp_oeq_vv_nxv3f64(<vscale x 3 x double> %va, <vscale x 3 x double> %vb, <vscale x 3 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_nxv3f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, ma
+; CHECK-NEXT:    vmfeq.vv v16, v8, v12, v0.t
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 3 x i1> @llvm.vp.fcmp.nxv3f64(<vscale x 3 x double> %va, <vscale x 3 x double> %vb, metadata !"oeq", <vscale x 3 x i1> %m, i32 %evl)
+  ret <vscale x 3 x i1> %v
+}
+
 declare <vscale x 8 x i1> @llvm.vp.fcmp.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, metadata, <vscale x 8 x i1>, i32)
 
 define <vscale x 8 x i1> @fcmp_oeq_vv_nxv8f64(<vscale x 8 x double> %va, <vscale x 8 x double> %vb, <vscale x 8 x i1> %m, i32 zeroext %evl) {
 ; CHECK-LABEL: fcmp_oeq_vv_nxv8f64:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, ma
 ; CHECK-NEXT:    vmfeq.vv v24, v8, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v24

 ; CHECK-NEXT:    vsetvli zero, zero, e8, m1, ta, mu
 ; CHECK-NEXT:    vmor.mm v0, v24, v16
 ; CHECK-NEXT:    ret
   %elt.head = insertelement <vscale x 8 x double> poison, double %b, i32 0
   %vb = shufflevector <vscale x 8 x double> %elt.head, <vscale x 8 x double> poison, <vscale x 8 x i32> zeroinitializer
   %v = call <vscale x 8 x i1> @llvm.vp.fcmp.nxv8f64(<vscale x 8 x double> %vb, <vscale x 8 x double> %va, metadata !"uno", <vscale x 8 x i1> %m, i32 %evl)
   ret <vscale x 8 x i1> %v
 }
+
+declare <vscale x 32 x i1> @llvm.vp.fcmp.nxv32f64(<vscale x 32 x double>, <vscale x 32 x double>, metadata, <vscale x 32 x i1>, i32)
+
+define <vscale x 32 x i1> @fcmp_oeq_vv_nxv32f64(<vscale x 32 x double> %va, <vscale x 32 x double> %vb, <vscale x 32 x i1> %m, i32 zeroext %evl) {
+; CHECK-LABEL: fcmp_oeq_vv_nxv32f64:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi sp, sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 5
+; CHECK-NEXT:    sub sp, sp, a1
+; CHECK-NEXT:    vmv1r.v v24, v0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 4
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    li a3, 24
+; CHECK-NEXT:    mul a1, a1, a3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a3, vlenb
+; CHECK-NEXT:    srli a1, a3, 3
+; CHECK-NEXT:    slli a5, a3, 3
+; CHECK-NEXT:    slli a7, a3, 1
+; CHECK-NEXT:    add a4, a2, a5
+; CHECK-NEXT:    mv t0, a6
+; CHECK-NEXT:    bltu a6, a7, .LBB171_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv t0, a7
+; CHECK-NEXT:  .LBB171_2:
+; CHECK-NEXT:    li t1, 0
+; CHECK-NEXT:    vsetvli t2, zero, e8, mf4, ta, mu
+; CHECK-NEXT:    vl8re64.v v16, (a4)
+; CHECK-NEXT:    srli a4, a3, 2
+; CHECK-NEXT:    sub t2, t0, a3
+; CHECK-NEXT:    vslidedown.vx v0, v24, a1
+; CHECK-NEXT:    bltu t0, t2, .LBB171_4
+; CHECK-NEXT:  # %bb.3:
+; CHECK-NEXT:    mv t1, t2
+; CHECK-NEXT:  .LBB171_4:
+; CHECK-NEXT:    li t2, 24
+; CHECK-NEXT:    vsetvli t3, zero, e8, mf2, ta, mu
+; CHECK-NEXT:    vslidedown.vx v1, v24, a4
+; CHECK-NEXT:    vl8re64.v v8, (a2)
+; CHECK-NEXT:    csrr t3, vlenb
+; CHECK-NEXT:    slli t3, t3, 3
+; CHECK-NEXT:    add t3, sp, t3
+; CHECK-NEXT:    addi t3, t3, 16
+; CHECK-NEXT:    vs8r.v v8, (t3) # Unknown-size Folded Spill
+; CHECK-NEXT:    slli t3, a3, 4
+; CHECK-NEXT:    vsetvli zero, t1, e64, m8, ta, ma
+; CHECK-NEXT:    csrr t1, vlenb
+; CHECK-NEXT:    slli t1, t1, 4
+; CHECK-NEXT:    add t1, sp, t1
+; CHECK-NEXT:    addi t1, t1, 16
+; CHECK-NEXT:    vl8re8.v v8, (t1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v2, v8, v16, v0.t
+; CHECK-NEXT:    bltu t0, a3, .LBB171_6
+; CHECK-NEXT:  # %bb.5:
+; CHECK-NEXT:    mv t0, a3
+; CHECK-NEXT:  .LBB171_6:
+; CHECK-NEXT:    li t1, 0
+; CHECK-NEXT:    mul t4, a3, t2
+; CHECK-NEXT:    add t2, a2, t3
+; CHECK-NEXT:    vsetvli zero, t0, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v24
+; CHECK-NEXT:    csrr t0, vlenb
+; CHECK-NEXT:    li t3, 24
+; CHECK-NEXT:    mul t0, t0, t3
+; CHECK-NEXT:    add t0, sp, t0
+; CHECK-NEXT:    addi t0, t0, 16
+; CHECK-NEXT:    vl8re8.v v24, (t0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr t0, vlenb
+; CHECK-NEXT:    slli t0, t0, 3
+; CHECK-NEXT:    add t0, sp, t0
+; CHECK-NEXT:    addi t0, t0, 16
+; CHECK-NEXT:    vl8re8.v v8, (t0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v17, v24, v8, v0.t
+; CHECK-NEXT:    sub t0, a6, a7
+; CHECK-NEXT:    add a7, a1, a1
+; CHECK-NEXT:    bltu a6, t0, .LBB171_8
+; CHECK-NEXT:  # %bb.7:
+; CHECK-NEXT:    mv t1, t0
+; CHECK-NEXT:  .LBB171_8:
+; CHECK-NEXT:    add a2, a2, t4
+; CHECK-NEXT:    vl8re64.v v8, (t2)
+; CHECK-NEXT:    csrr a6, vlenb
+; CHECK-NEXT:    li t0, 24
+; CHECK-NEXT:    mul a6, a6, t0
+; CHECK-NEXT:    add a6, sp, a6
+; CHECK-NEXT:    addi a6, a6, 16
+; CHECK-NEXT:    vs8r.v v8, (a6) # Unknown-size Folded Spill
+; CHECK-NEXT:    vl8re64.v v8, (a0)
+; CHECK-NEXT:    csrr a6, vlenb
+; CHECK-NEXT:    slli a6, a6, 4
+; CHECK-NEXT:    add a6, sp, a6
+; CHECK-NEXT:    addi a6, a6, 16
+; CHECK-NEXT:    vs8r.v v8, (a6) # Unknown-size Folded Spill
+; CHECK-NEXT:    add a0, a0, a5
+; CHECK-NEXT:    vsetvli zero, a7, e8, mf2, tu, mu
+; CHECK-NEXT:    vslideup.vx v17, v2, a1
+; CHECK-NEXT:    mv a5, t1
+; CHECK-NEXT:    bltu t1, a3, .LBB171_10
+; CHECK-NEXT:  # %bb.9:
+; CHECK-NEXT:    mv a5, a3
+; CHECK-NEXT:  .LBB171_10:
+; CHECK-NEXT:    li a6, 0
+; CHECK-NEXT:    vsetvli a7, zero, e8, mf4, ta, mu
+; CHECK-NEXT:    vslidedown.vx v16, v1, a1
+; CHECK-NEXT:    vl8re64.v v8, (a2)
+; CHECK-NEXT:    csrr a2, vlenb
+; CHECK-NEXT:    slli a2, a2, 3
+; CHECK-NEXT:    add a2, sp, a2
+; CHECK-NEXT:    addi a2, a2, 16
+; CHECK-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
+; CHECK-NEXT:    vl8re64.v v8, (a0)
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vsetvli zero, a5, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v1
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    li a2, 24
+; CHECK-NEXT:    mul a0, a0, a2
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 4
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v18, v8, v24, v0.t
+; CHECK-NEXT:    add a0, a4, a1
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, tu, mu
+; CHECK-NEXT:    sub a0, t1, a3
+; CHECK-NEXT:    vslideup.vx v17, v18, a4
+; CHECK-NEXT:    bltu t1, a0, .LBB171_12
+; CHECK-NEXT:  # %bb.11:
+; CHECK-NEXT:    mv a6, a0
+; CHECK-NEXT:  .LBB171_12:
+; CHECK-NEXT:    vsetvli zero, a6, e64, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v16
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8re8.v v24, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    addi a0, sp, 16
+; CHECK-NEXT:    vl8re8.v v8, (a0) # Unknown-size Folded Reload
+; CHECK-NEXT:    vmfeq.vv v16, v8, v24, v0.t
+; CHECK-NEXT:    slli a0, a1, 1
+; CHECK-NEXT:    add a0, a0, a1
+; CHECK-NEXT:    add a1, a0, a1
+; CHECK-NEXT:    vsetvli zero, a1, e8, mf2, tu, mu
+; CHECK-NEXT:    vslideup.vx v17, v16, a0
+; CHECK-NEXT:    vmv1r.v v0, v17
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 5
+; CHECK-NEXT:    add sp, sp, a0
+; CHECK-NEXT:    addi sp, sp, 16
+; CHECK-NEXT:    ret
+  %v = call <vscale x 32 x i1> @llvm.vp.fcmp.nxv32f64(<vscale x 32 x double> %va, <vscale x 32 x double> %vb, metadata !"oeq", <vscale x 32 x i1> %m, i32 %evl)
+  ret <vscale x 32 x i1> %v
+}