; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple riscv64 -mattr=+experimental-v -verify-machineinstrs < %s | FileCheck %s

define <vscale x 4 x i32> @extract_nxv8i32_nxv4i32_0(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv4i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8m2 killed $v8m2 killed $v8m4
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> %vec, i64 0)
  ret <vscale x 4 x i32> %c
}

define <vscale x 4 x i32> @extract_nxv8i32_nxv4i32_4(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv4i32_4:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> %vec, i64 4)
  ret <vscale x 4 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv8i32_nxv2i32_0(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv2i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8 killed $v8 killed $v8m4
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %vec, i64 0)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv8i32_nxv2i32_2(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv2i32_2:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %vec, i64 2)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv8i32_nxv2i32_4(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv2i32_4:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v10
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %vec, i64 4)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv8i32_nxv2i32_6(<vscale x 8 x i32> %vec) {
; CHECK-LABEL: extract_nxv8i32_nxv2i32_6:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v11
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %vec, i64 6)
  ret <vscale x 2 x i32> %c
}

define <vscale x 8 x i32> @extract_nxv16i32_nxv8i32_0(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv8i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8m4 killed $v8m4 killed $v8m8
; CHECK-NEXT:    ret
  %c = call <vscale x 8 x i32> @llvm.experimental.vector.extract.nxv8i32.nxv16i32(<vscale x 16 x i32> %vec, i64 0)
  ret <vscale x 8 x i32> %c
}

define <vscale x 8 x i32> @extract_nxv16i32_nxv8i32_8(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv8i32_8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv4r.v v8, v12
; CHECK-NEXT:    ret
  %c = call <vscale x 8 x i32> @llvm.experimental.vector.extract.nxv8i32.nxv16i32(<vscale x 16 x i32> %vec, i64 8)
  ret <vscale x 8 x i32> %c
}

define <vscale x 4 x i32> @extract_nxv16i32_nxv4i32_0(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv4i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8m2 killed $v8m2 killed $v8m8
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv16i32(<vscale x 16 x i32> %vec, i64 0)
  ret <vscale x 4 x i32> %c
}

define <vscale x 4 x i32> @extract_nxv16i32_nxv4i32_4(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv4i32_4:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv2r.v v8, v10
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv16i32(<vscale x 16 x i32> %vec, i64 4)
  ret <vscale x 4 x i32> %c
}

define <vscale x 4 x i32> @extract_nxv16i32_nxv4i32_8(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv4i32_8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv2r.v v8, v12
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv16i32(<vscale x 16 x i32> %vec, i64 8)
  ret <vscale x 4 x i32> %c
}

define <vscale x 4 x i32> @extract_nxv16i32_nxv4i32_12(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv4i32_12:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv2r.v v8, v14
; CHECK-NEXT:    ret
  %c = call <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv16i32(<vscale x 16 x i32> %vec, i64 12)
  ret <vscale x 4 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_0(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8 killed $v8 killed $v8m8
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 0)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_2(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_2:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 2)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_4(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_4:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v10
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 4)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_6(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_6:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v11
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 6)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_8(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_8:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v12
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 8)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_10(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_10:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v13
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 10)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_12(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_12:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v14
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 12)
  ret <vscale x 2 x i32> %c
}

define <vscale x 2 x i32> @extract_nxv16i32_nxv2i32_14(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv2i32_14:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v15
; CHECK-NEXT:    ret
  %c = call <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 14)
  ret <vscale x 2 x i32> %c
}

define <vscale x 1 x i32> @extract_nxv16i32_nxv1i32_0(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv1i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    # kill: def $v8 killed $v8 killed $v8m8
; CHECK-NEXT:    ret
  %c = call <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv16i32(<vscale x 16 x i32> %vec, i64 0)
  ret <vscale x 1 x i32> %c
}

; TODO: Extracts that don't align to a vector register are not yet supported.
; In this case we want to extract the upper half of the lowest VR subregister
; in the LMUL group.
; define <vscale x 1 x i32> @extract_nxv16i32_nxv1i32_1(<vscale x 16 x i32> %vec) {
;   %c = call <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv16i32(<vscale x 16 x i32> %vec, i64 1)
;   ret <vscale x 1 x i32> %c
; }

define <vscale x 1 x i32> @extract_nxv16i32_nxv1i32_2(<vscale x 16 x i32> %vec) {
; CHECK-LABEL: extract_nxv16i32_nxv1i32_2:
; CHECK:       # %bb.0:
; CHECK-NEXT:    vmv1r.v v8, v9
; CHECK-NEXT:    ret
  %c = call <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv16i32(<vscale x 16 x i32> %vec, i64 2)
  ret <vscale x 1 x i32> %c
}

define <vscale x 1 x i32> @extract_nxv2i32_nxv1i32_0(<vscale x 2 x i32> %vec) {
; CHECK-LABEL: extract_nxv2i32_nxv1i32_0:
; CHECK:       # %bb.0:
; CHECK-NEXT:    ret
  %c = call <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv2i32(<vscale x 2 x i32> %vec, i64 0)
  ret <vscale x 1 x i32> %c
}

declare <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv2i32(<vscale x 2 x i32> %vec, i64 %idx)

declare <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv8i32(<vscale x 8 x i32> %vec, i64 %idx)
declare <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv8i32(<vscale x 8 x i32> %vec, i64 %idx)

declare <vscale x 1 x i32> @llvm.experimental.vector.extract.nxv1i32.nxv16i32(<vscale x 16 x i32> %vec, i64 %idx)
declare <vscale x 2 x i32> @llvm.experimental.vector.extract.nxv2i32.nxv16i32(<vscale x 16 x i32> %vec, i64 %idx)
declare <vscale x 4 x i32> @llvm.experimental.vector.extract.nxv4i32.nxv16i32(<vscale x 16 x i32> %vec, i64 %idx)
declare <vscale x 8 x i32> @llvm.experimental.vector.extract.nxv8i32.nxv16i32(<vscale x 16 x i32> %vec, i64 %idx)