The Nest
Explore
0
Compiler projects using llvm 
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+m,+v -riscv-v-vector-bits-min=128 < %s \
; RUN:   | FileCheck %s
; RUN: llc -mtriple=riscv64 -mattr=+m,+v -riscv-v-vector-bits-min=128 < %s \
; RUN:   | FileCheck %s

declare <4 x i16> @llvm.vp.sext.v4i16.v4i8(<4 x i8>, <4 x i1>, i32)

define <4 x i16> @vsext_v4i16_v4i8(<4 x i8> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i16_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
; CHECK-NEXT:    vsext.vf2 v9, v8, v0.t
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i16> @llvm.vp.sext.v4i16.v4i8(<4 x i8> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i16> %v
}

define <4 x i16> @vsext_v4i16_v4i8_unmasked(<4 x i8> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i16_v4i8_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
; CHECK-NEXT:    vsext.vf2 v9, v8
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i16> @llvm.vp.sext.v4i16.v4i8(<4 x i8> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i16> %v
}

declare <4 x i32> @llvm.vp.sext.v4i32.v4i8(<4 x i8>, <4 x i1>, i32)

define <4 x i32> @vsext_v4i32_v4i8(<4 x i8> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i32_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT:    vsext.vf4 v9, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i32> @llvm.vp.sext.v4i32.v4i8(<4 x i8> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i32> %v
}

define <4 x i32> @vsext_v4i32_v4i8_unmasked(<4 x i8> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i32_v4i8_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT:    vsext.vf4 v9, v8
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i32> @llvm.vp.sext.v4i32.v4i8(<4 x i8> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i32> %v
}

declare <4 x i64> @llvm.vp.sext.v4i64.v4i8(<4 x i8>, <4 x i1>, i32)

define <4 x i64> @vsext_v4i64_v4i8(<4 x i8> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf8 v10, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i8(<4 x i8> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i64> %v
}

define <4 x i64> @vsext_v4i64_v4i8_unmasked(<4 x i8> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i8_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf8 v10, v8
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i8(<4 x i8> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i64> %v
}

declare <4 x i32> @llvm.vp.sext.v4i32.v4i16(<4 x i16>, <4 x i1>, i32)

define <4 x i32> @vsext_v4i32_v4i16(<4 x i16> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i32_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT:    vsext.vf2 v9, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i32> @llvm.vp.sext.v4i32.v4i16(<4 x i16> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i32> %v
}

define <4 x i32> @vsext_v4i32_v4i16_unmasked(<4 x i16> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i32_v4i16_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
; CHECK-NEXT:    vsext.vf2 v9, v8
; CHECK-NEXT:    vmv.v.v v8, v9
; CHECK-NEXT:    ret
  %v = call <4 x i32> @llvm.vp.sext.v4i32.v4i16(<4 x i16> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i32> %v
}

declare <4 x i64> @llvm.vp.sext.v4i64.v4i16(<4 x i16>, <4 x i1>, i32)

define <4 x i64> @vsext_v4i64_v4i16(<4 x i16> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf4 v10, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i16(<4 x i16> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i64> %v
}

define <4 x i64> @vsext_v4i64_v4i16_unmasked(<4 x i16> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i16_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf4 v10, v8
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i16(<4 x i16> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i64> %v
}

declare <4 x i64> @llvm.vp.sext.v4i64.v4i32(<4 x i32>, <4 x i1>, i32)

define <4 x i64> @vsext_v4i64_v4i32(<4 x i32> %va, <4 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf2 v10, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i32(<4 x i32> %va, <4 x i1> %m, i32 %evl)
  ret <4 x i64> %v
}

define <4 x i64> @vsext_v4i64_v4i32_unmasked(<4 x i32> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v4i64_v4i32_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
; CHECK-NEXT:    vsext.vf2 v10, v8
; CHECK-NEXT:    vmv.v.v v8, v10
; CHECK-NEXT:    ret
  %v = call <4 x i64> @llvm.vp.sext.v4i64.v4i32(<4 x i32> %va, <4 x i1> shufflevector (<4 x i1> insertelement (<4 x i1> undef, i1 true, i32 0), <4 x i1> undef, <4 x i32> zeroinitializer), i32 %evl)
  ret <4 x i64> %v
}

declare <32 x i64> @llvm.vp.sext.v32i64.v32i32(<32 x i32>, <32 x i1>, i32)

define <32 x i64> @vsext_v32i64_v32i32(<32 x i32> %va, <32 x i1> %m, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v32i64_v32i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v1, v0
; CHECK-NEXT:    li a1, 0
; CHECK-NEXT:    vsetivli zero, 2, e8, mf4, ta, mu
; CHECK-NEXT:    addi a2, a0, -16
; CHECK-NEXT:    vslidedown.vi v0, v0, 2
; CHECK-NEXT:    bltu a0, a2, .LBB12_2
; CHECK-NEXT:  # %bb.1:
; CHECK-NEXT:    mv a1, a2
; CHECK-NEXT:  .LBB12_2:
; CHECK-NEXT:    vsetivli zero, 16, e32, m8, ta, mu
; CHECK-NEXT:    vslidedown.vi v24, v8, 16
; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
; CHECK-NEXT:    li a1, 16
; CHECK-NEXT:    vsext.vf2 v16, v24, v0.t
; CHECK-NEXT:    bltu a0, a1, .LBB12_4
; CHECK-NEXT:  # %bb.3:
; CHECK-NEXT:    li a0, 16
; CHECK-NEXT:  .LBB12_4:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
; CHECK-NEXT:    vmv1r.v v0, v1
; CHECK-NEXT:    vsext.vf2 v24, v8, v0.t
; CHECK-NEXT:    vmv.v.v v8, v24
; CHECK-NEXT:    ret
  %v = call <32 x i64> @llvm.vp.sext.v32i64.v32i32(<32 x i32> %va, <32 x i1> %m, i32 %evl)
  ret <32 x i64> %v
}

define <32 x i64> @vsext_v32i64_v32i32_unmasked(<32 x i32> %va, i32 zeroext %evl) {
; CHECK-LABEL: vsext_v32i64_v32i32_unmasked:
; CHECK:       # %bb.0:
; CHECK-NEXT:    addi a2, a0, -16
; CHECK-NEXT:    li a1, 0
; CHECK-NEXT:    bltu a0, a2, .LBB13_2
; CHECK-NEXT:  # %bb.1:
; CHECK-NEXT:    mv a1, a2
; CHECK-NEXT:  .LBB13_2:
; CHECK-NEXT:    vsetivli zero, 16, e32, m8, ta, mu
; CHECK-NEXT:    vslidedown.vi v24, v8, 16
; CHECK-NEXT:    vsetvli zero, a1, e64, m8, ta, mu
; CHECK-NEXT:    li a1, 16
; CHECK-NEXT:    vsext.vf2 v16, v24
; CHECK-NEXT:    bltu a0, a1, .LBB13_4
; CHECK-NEXT:  # %bb.3:
; CHECK-NEXT:    li a0, 16
; CHECK-NEXT:  .LBB13_4:
; CHECK-NEXT:    vsetvli zero, a0, e64, m8, ta, mu
; CHECK-NEXT:    vsext.vf2 v24, v8
; CHECK-NEXT:    vmv.v.v v8, v24
; CHECK-NEXT:    ret
  %v = call <32 x i64> @llvm.vp.sext.v32i64.v32i32(<32 x i32> %va, <32 x i1> shufflevector (<32 x i1> insertelement (<32 x i1> undef, i1 true, i32 0), <32 x i1> undef, <32 x i32> zeroinitializer), i32 %evl)
  ret <32 x i64> %v
}