llvm/test/CodeGen/PowerPC/masked-urem.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
; RUN: llc -mtriple=powerpc64le < %s | FileCheck %s

; Legal
define <4 x i32> @urem_v4i32(<4 x i32> %x, <4 x i32> %y, <4 x i1> %m) {
; CHECK-LABEL: urem_v4i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    xxleqv 32, 32, 32
; CHECK-NEXT:    vspltisw 5, 1
; CHECK-NEXT:    xxsldwi 1, 34, 34, 1
; CHECK-NEXT:    vslw 4, 4, 0
; CHECK-NEXT:    xxswapd 4, 34
; CHECK-NEXT:    xxsldwi 6, 34, 34, 3
; CHECK-NEXT:    mffprwz 4, 1
; CHECK-NEXT:    vsraw 4, 4, 0
; CHECK-NEXT:    xxsel 0, 37, 35, 36
; CHECK-NEXT:    xxsldwi 2, 0, 0, 1
; CHECK-NEXT:    xxswapd 3, 0
; CHECK-NEXT:    xxsldwi 5, 0, 0, 3
; CHECK-NEXT:    mffprwz 3, 2
; CHECK-NEXT:    mffprwz 5, 3
; CHECK-NEXT:    divwu 6, 4, 3
; CHECK-NEXT:    mullw 3, 6, 3
; CHECK-NEXT:    mffprwz 6, 4
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    divwu 4, 6, 5
; CHECK-NEXT:    mullw 4, 4, 5
; CHECK-NEXT:    sub 4, 6, 4
; CHECK-NEXT:    rldimi 4, 3, 32, 0
; CHECK-NEXT:    mffprwz 3, 5
; CHECK-NEXT:    mtfprd 1, 4
; CHECK-NEXT:    mffprwz 4, 6
; CHECK-NEXT:    divwu 5, 4, 3
; CHECK-NEXT:    mullw 3, 5, 3
; CHECK-NEXT:    mfvsrwz 5, 34
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprwz 4, 0
; CHECK-NEXT:    divwu 6, 5, 4
; CHECK-NEXT:    mullw 4, 6, 4
; CHECK-NEXT:    sub 4, 5, 4
; CHECK-NEXT:    rldimi 4, 3, 32, 0
; CHECK-NEXT:    mtfprd 0, 4
; CHECK-NEXT:    xxmrghd 34, 0, 1
; CHECK-NEXT:    blr
  %res = call <4 x i32> @llvm.masked.urem(<4 x i32> %x, <4 x i32> %y, <4 x i1> %m)
  ret <4 x i32> %res
}

define <2 x i64> @urem_v2i64(<2 x i64> %x, <2 x i64> %y, <2 x i1> %m) {
; CHECK-LABEL: urem_v2i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    xxleqv 32, 32, 32
; CHECK-NEXT:    vspltisw 5, 1
; CHECK-NEXT:    mfvsrd 4, 34
; CHECK-NEXT:    xxswapd 2, 34
; CHECK-NEXT:    vsld 4, 4, 0
; CHECK-NEXT:    vsrad 4, 4, 0
; CHECK-NEXT:    vupklsw 5, 5
; CHECK-NEXT:    xxsel 0, 37, 35, 36
; CHECK-NEXT:    mffprd 3, 0
; CHECK-NEXT:    divdu 5, 4, 3
; CHECK-NEXT:    mulld 3, 5, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprd 4, 2
; CHECK-NEXT:    xxswapd 1, 0
; CHECK-NEXT:    mtfprd 0, 3
; CHECK-NEXT:    mffprd 3, 1
; CHECK-NEXT:    divdu 5, 4, 3
; CHECK-NEXT:    mulld 3, 5, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mtfprd 1, 3
; CHECK-NEXT:    xxmrghd 34, 0, 1
; CHECK-NEXT:    blr
  %res = call <2 x i64> @llvm.masked.urem(<2 x i64> %x, <2 x i64> %y, <2 x i1> %m)
  ret <2 x i64> %res
}

; Splitting
define <4 x i64> @urem_v4i64(<4 x i64> %x, <4 x i64> %y, <4 x i1> %m) {
; CHECK-LABEL: urem_v4i64:
; CHECK:       # %bb.0:
; CHECK-NEXT:    xxmrglw 32, 38, 38
; CHECK-NEXT:    xxleqv 39, 39, 39
; CHECK-NEXT:    xxmrghw 33, 38, 38
; CHECK-NEXT:    mfvsrd 3, 34
; CHECK-NEXT:    vspltisw 6, 1
; CHECK-NEXT:    mfvsrd 4, 35
; CHECK-NEXT:    xxswapd 2, 34
; CHECK-NEXT:    xxswapd 4, 35
; CHECK-NEXT:    vsld 0, 0, 7
; CHECK-NEXT:    mffprd 5, 2
; CHECK-NEXT:    vsrad 0, 0, 7
; CHECK-NEXT:    vupklsw 6, 6
; CHECK-NEXT:    xxsel 0, 38, 36, 32
; CHECK-NEXT:    vsld 4, 1, 7
; CHECK-NEXT:    mffprd 6, 0
; CHECK-NEXT:    vsrad 4, 4, 7
; CHECK-NEXT:    divdu 9, 3, 6
; CHECK-NEXT:    mulld 6, 9, 6
; CHECK-NEXT:    sub 3, 3, 6
; CHECK-NEXT:    xxswapd 3, 0
; CHECK-NEXT:    mffprd 8, 3
; CHECK-NEXT:    mtfprd 0, 3
; CHECK-NEXT:    xxsel 1, 38, 37, 36
; CHECK-NEXT:    mffprd 7, 1
; CHECK-NEXT:    divdu 9, 4, 7
; CHECK-NEXT:    mulld 7, 9, 7
; CHECK-NEXT:    divdu 9, 5, 8
; CHECK-NEXT:    sub 4, 4, 7
; CHECK-NEXT:    mulld 8, 9, 8
; CHECK-NEXT:    sub 3, 5, 8
; CHECK-NEXT:    xxswapd 5, 1
; CHECK-NEXT:    mffprd 5, 4
; CHECK-NEXT:    mtfprd 1, 4
; CHECK-NEXT:    mffprd 4, 5
; CHECK-NEXT:    mtfprd 2, 3
; CHECK-NEXT:    divdu 3, 5, 4
; CHECK-NEXT:    mulld 3, 3, 4
; CHECK-NEXT:    sub 3, 5, 3
; CHECK-NEXT:    xxmrghd 34, 0, 2
; CHECK-NEXT:    mtfprd 0, 3
; CHECK-NEXT:    xxmrghd 35, 1, 0
; CHECK-NEXT:    blr
  %res = call <4 x i64> @llvm.masked.urem(<4 x i64> %x, <4 x i64> %y, <4 x i1> %m)
  ret <4 x i64> %res
}

; Widening
define <2 x i32> @urem_v2i32(<2 x i32> %x, <2 x i32> %y, <2 x i1> %m) {
; CHECK-LABEL: urem_v2i32:
; CHECK:       # %bb.0:
; CHECK-NEXT:    addis 3, 2, .LCPI3_0@toc@ha
; CHECK-NEXT:    xxlxor 32, 32, 32
; CHECK-NEXT:    xxsldwi 1, 34, 34, 1
; CHECK-NEXT:    addi 3, 3, .LCPI3_0@toc@l
; CHECK-NEXT:    mffprwz 4, 1
; CHECK-NEXT:    xxswapd 4, 34
; CHECK-NEXT:    xxsldwi 6, 34, 34, 3
; CHECK-NEXT:    lxvd2x 0, 0, 3
; CHECK-NEXT:    xxswapd 37, 0
; CHECK-NEXT:    vperm 4, 0, 4, 5
; CHECK-NEXT:    xxleqv 32, 32, 32
; CHECK-NEXT:    vspltisw 5, 1
; CHECK-NEXT:    vslw 4, 4, 0
; CHECK-NEXT:    vsraw 4, 4, 0
; CHECK-NEXT:    xxsel 0, 37, 35, 36
; CHECK-NEXT:    xxsldwi 2, 0, 0, 1
; CHECK-NEXT:    xxswapd 3, 0
; CHECK-NEXT:    xxsldwi 5, 0, 0, 3
; CHECK-NEXT:    mffprwz 3, 2
; CHECK-NEXT:    mffprwz 5, 3
; CHECK-NEXT:    divwu 6, 4, 3
; CHECK-NEXT:    mullw 3, 6, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprwz 4, 4
; CHECK-NEXT:    divwu 6, 4, 5
; CHECK-NEXT:    mullw 5, 6, 5
; CHECK-NEXT:    sub 4, 4, 5
; CHECK-NEXT:    rldimi 4, 3, 32, 0
; CHECK-NEXT:    mffprwz 3, 5
; CHECK-NEXT:    mtfprd 1, 4
; CHECK-NEXT:    mffprwz 4, 6
; CHECK-NEXT:    divwu 5, 4, 3
; CHECK-NEXT:    mullw 3, 5, 3
; CHECK-NEXT:    mfvsrwz 5, 34
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprwz 4, 0
; CHECK-NEXT:    divwu 6, 5, 4
; CHECK-NEXT:    mullw 4, 6, 4
; CHECK-NEXT:    sub 4, 5, 4
; CHECK-NEXT:    rldimi 4, 3, 32, 0
; CHECK-NEXT:    mtfprd 0, 4
; CHECK-NEXT:    xxmrghd 34, 0, 1
; CHECK-NEXT:    blr
  %res = call <2 x i32> @llvm.masked.urem(<2 x i32> %x, <2 x i32> %y, <2 x i1> %m)
  ret <2 x i32> %res
}

; Promotion
define <4 x i16> @urem_v4i16(<4 x i16> %x, <4 x i16> %y, <4 x i1> %m) {
; CHECK-LABEL: urem_v4i16:
; CHECK:       # %bb.0:
; CHECK-NEXT:    xxswapd 0, 36
; CHECK-NEXT:    xxsldwi 1, 36, 36, 1
; CHECK-NEXT:    mfvsrwz 3, 36
; CHECK-NEXT:    li 7, 0
; CHECK-NEXT:    xxsldwi 2, 36, 36, 3
; CHECK-NEXT:    std 25, -56(1) # 8-byte Folded Spill
; CHECK-NEXT:    std 30, -16(1) # 8-byte Folded Spill
; CHECK-NEXT:    std 29, -24(1) # 8-byte Folded Spill
; CHECK-NEXT:    mffprwz 4, 0
; CHECK-NEXT:    mffprwz 5, 1
; CHECK-NEXT:    std 28, -32(1) # 8-byte Folded Spill
; CHECK-NEXT:    std 27, -40(1) # 8-byte Folded Spill
; CHECK-NEXT:    mffprwz 6, 2
; CHECK-NEXT:    std 26, -48(1) # 8-byte Folded Spill
; CHECK-NEXT:    mtvsrd 36, 3
; CHECK-NEXT:    mfvsrd 3, 34
; CHECK-NEXT:    mtvsrd 37, 4
; CHECK-NEXT:    mtvsrd 32, 5
; CHECK-NEXT:    rldicl 8, 3, 48, 48
; CHECK-NEXT:    rldicl 9, 3, 32, 48
; CHECK-NEXT:    clrlwi 8, 8, 16
; CHECK-NEXT:    clrlwi 9, 9, 16
; CHECK-NEXT:    vmrghh 5, 0, 5
; CHECK-NEXT:    mtvsrd 32, 6
; CHECK-NEXT:    vmrghh 4, 0, 4
; CHECK-NEXT:    mtvsrd 32, 7
; CHECK-NEXT:    clrldi 7, 3, 48
; CHECK-NEXT:    rldicl 3, 3, 16, 48
; CHECK-NEXT:    clrlwi 7, 7, 16
; CHECK-NEXT:    clrlwi 3, 3, 16
; CHECK-NEXT:    xxmrglw 1, 36, 37
; CHECK-NEXT:    vspltish 4, 15
; CHECK-NEXT:    vsplth 0, 0, 3
; CHECK-NEXT:    xxspltw 0, 32, 3
; CHECK-NEXT:    vspltish 0, 1
; CHECK-NEXT:    xxmrgld 37, 0, 1
; CHECK-NEXT:    xxswapd 1, 34
; CHECK-NEXT:    vslh 5, 5, 4
; CHECK-NEXT:    mffprd 4, 1
; CHECK-NEXT:    vsrah 4, 5, 4
; CHECK-NEXT:    clrldi 10, 4, 48
; CHECK-NEXT:    rldicl 11, 4, 48, 48
; CHECK-NEXT:    rldicl 12, 4, 32, 48
; CHECK-NEXT:    rldicl 4, 4, 16, 48
; CHECK-NEXT:    clrlwi 10, 10, 16
; CHECK-NEXT:    clrlwi 11, 11, 16
; CHECK-NEXT:    clrlwi 12, 12, 16
; CHECK-NEXT:    clrlwi 4, 4, 16
; CHECK-NEXT:    xxsel 0, 32, 35, 36
; CHECK-NEXT:    mffprd 5, 0
; CHECK-NEXT:    clrldi 0, 5, 48
; CHECK-NEXT:    rldicl 30, 5, 48, 48
; CHECK-NEXT:    rldicl 29, 5, 32, 48
; CHECK-NEXT:    rldicl 5, 5, 16, 48
; CHECK-NEXT:    clrlwi 0, 0, 16
; CHECK-NEXT:    clrlwi 30, 30, 16
; CHECK-NEXT:    clrlwi 29, 29, 16
; CHECK-NEXT:    clrlwi 5, 5, 16
; CHECK-NEXT:    xxswapd 2, 0
; CHECK-NEXT:    mffprd 6, 2
; CHECK-NEXT:    clrldi 28, 6, 48
; CHECK-NEXT:    rldicl 27, 6, 48, 48
; CHECK-NEXT:    rldicl 26, 6, 32, 48
; CHECK-NEXT:    rldicl 6, 6, 16, 48
; CHECK-NEXT:    divwu 25, 7, 0
; CHECK-NEXT:    clrlwi 28, 28, 16
; CHECK-NEXT:    clrlwi 27, 27, 16
; CHECK-NEXT:    clrlwi 26, 26, 16
; CHECK-NEXT:    clrlwi 6, 6, 16
; CHECK-NEXT:    mullw 0, 25, 0
; CHECK-NEXT:    divwu 25, 8, 30
; CHECK-NEXT:    sub 7, 7, 0
; CHECK-NEXT:    mtvsrd 34, 7
; CHECK-NEXT:    mullw 30, 25, 30
; CHECK-NEXT:    divwu 25, 9, 29
; CHECK-NEXT:    sub 8, 8, 30
; CHECK-NEXT:    ld 30, -16(1) # 8-byte Folded Reload
; CHECK-NEXT:    mtvsrd 35, 8
; CHECK-NEXT:    mullw 29, 25, 29
; CHECK-NEXT:    divwu 25, 3, 5
; CHECK-NEXT:    sub 9, 9, 29
; CHECK-NEXT:    ld 29, -24(1) # 8-byte Folded Reload
; CHECK-NEXT:    mtvsrd 36, 9
; CHECK-NEXT:    mullw 5, 25, 5
; CHECK-NEXT:    divwu 25, 10, 28
; CHECK-NEXT:    sub 3, 3, 5
; CHECK-NEXT:    mtvsrd 37, 3
; CHECK-NEXT:    mullw 28, 25, 28
; CHECK-NEXT:    divwu 25, 11, 27
; CHECK-NEXT:    sub 3, 10, 28
; CHECK-NEXT:    ld 28, -32(1) # 8-byte Folded Reload
; CHECK-NEXT:    mullw 27, 25, 27
; CHECK-NEXT:    divwu 25, 12, 26
; CHECK-NEXT:    sub 5, 11, 27
; CHECK-NEXT:    ld 27, -40(1) # 8-byte Folded Reload
; CHECK-NEXT:    mullw 26, 25, 26
; CHECK-NEXT:    divwu 25, 4, 6
; CHECK-NEXT:    sub 7, 12, 26
; CHECK-NEXT:    ld 26, -48(1) # 8-byte Folded Reload
; CHECK-NEXT:    mullw 6, 25, 6
; CHECK-NEXT:    ld 25, -56(1) # 8-byte Folded Reload
; CHECK-NEXT:    vmrghh 2, 3, 2
; CHECK-NEXT:    vmrghh 3, 5, 4
; CHECK-NEXT:    mtvsrd 36, 3
; CHECK-NEXT:    mtvsrd 37, 5
; CHECK-NEXT:    sub 3, 4, 6
; CHECK-NEXT:    mtvsrd 32, 3
; CHECK-NEXT:    xxmrglw 0, 35, 34
; CHECK-NEXT:    vmrghh 4, 5, 4
; CHECK-NEXT:    mtvsrd 37, 7
; CHECK-NEXT:    vmrghh 5, 0, 5
; CHECK-NEXT:    xxmrglw 1, 37, 36
; CHECK-NEXT:    xxmrgld 34, 0, 1
; CHECK-NEXT:    blr
  %res = call <4 x i16> @llvm.masked.urem(<4 x i16> %x, <4 x i16> %y, <4 x i1> %m)
  ret <4 x i16> %res
}

; Scalarization
define <1 x i64> @urem_v1i164(<1 x i64> %x, <1 x i64> %y, <1 x i1> %m) {
; CHECK-LABEL: urem_v1i164:
; CHECK:       # %bb.0:
; CHECK-NEXT:    andi. 5, 5, 1
; CHECK-NEXT:    li 5, 1
; CHECK-NEXT:    iselgt 4, 4, 5
; CHECK-NEXT:    divdu 5, 3, 4
; CHECK-NEXT:    mulld 4, 5, 4
; CHECK-NEXT:    sub 3, 3, 4
; CHECK-NEXT:    blr
  %res = call <1 x i64> @llvm.masked.urem(<1 x i64> %x, <1 x i64> %y, <1 x i1> %m)
  ret <1 x i64> %res
}

; Expansion
define <2 x i128> @urem_v2i128(<2 x i128> %x, <2 x i128> %y, <2 x i1> %m) nounwind {
; CHECK-LABEL: urem_v2i128:
; CHECK:       # %bb.0:
; CHECK-NEXT:    mfocrf 12, 32
; CHECK-NEXT:    stw 12, 8(1)
; CHECK-NEXT:    mflr 0
; CHECK-NEXT:    stdu 1, -128(1)
; CHECK-NEXT:    li 3, 48
; CHECK-NEXT:    std 0, 144(1)
; CHECK-NEXT:    xxswapd 0, 38
; CHECK-NEXT:    xxswapd 1, 37
; CHECK-NEXT:    std 30, 112(1) # 8-byte Folded Spill
; CHECK-NEXT:    li 30, 1
; CHECK-NEXT:    std 29, 104(1) # 8-byte Folded Spill
; CHECK-NEXT:    li 29, 0
; CHECK-NEXT:    stxvd2x 61, 1, 3 # 16-byte Folded Spill
; CHECK-NEXT:    li 3, 64
; CHECK-NEXT:    mfvsrd 4, 35
; CHECK-NEXT:    stxvd2x 62, 1, 3 # 16-byte Folded Spill
; CHECK-NEXT:    li 3, 80
; CHECK-NEXT:    vmr 30, 2
; CHECK-NEXT:    stxvd2x 63, 1, 3 # 16-byte Folded Spill
; CHECK-NEXT:    mffprd 3, 0
; CHECK-NEXT:    vmr 31, 4
; CHECK-NEXT:    andi. 3, 3, 1
; CHECK-NEXT:    mfvsrd 3, 38
; CHECK-NEXT:    crmove 8, 1
; CHECK-NEXT:    andi. 3, 3, 1
; CHECK-NEXT:    mffprd 3, 1
; CHECK-NEXT:    iselgt 5, 3, 30
; CHECK-NEXT:    mfvsrd 3, 37
; CHECK-NEXT:    xxswapd 0, 35
; CHECK-NEXT:    iselgt 6, 3, 29
; CHECK-NEXT:    mffprd 3, 0
; CHECK-NEXT:    bl __umodti3
; CHECK-NEXT:    nop
; CHECK-NEXT:    xxswapd 0, 63
; CHECK-NEXT:    mtfprd 1, 3
; CHECK-NEXT:    mtfprd 2, 4
; CHECK-NEXT:    mfvsrd 4, 62
; CHECK-NEXT:    mffprd 3, 0
; CHECK-NEXT:    isel 5, 3, 30, 8
; CHECK-NEXT:    mfvsrd 3, 63
; CHECK-NEXT:    isel 6, 3, 29, 8
; CHECK-NEXT:    xxswapd 0, 62
; CHECK-NEXT:    mffprd 3, 0
; CHECK-NEXT:    xxmrghd 61, 2, 1
; CHECK-NEXT:    bl __umodti3
; CHECK-NEXT:    nop
; CHECK-NEXT:    mtfprd 0, 3
; CHECK-NEXT:    li 3, 80
; CHECK-NEXT:    mtfprd 1, 4
; CHECK-NEXT:    ld 30, 112(1) # 8-byte Folded Reload
; CHECK-NEXT:    vmr 3, 29
; CHECK-NEXT:    ld 29, 104(1) # 8-byte Folded Reload
; CHECK-NEXT:    lxvd2x 63, 1, 3 # 16-byte Folded Reload
; CHECK-NEXT:    li 3, 64
; CHECK-NEXT:    lxvd2x 62, 1, 3 # 16-byte Folded Reload
; CHECK-NEXT:    li 3, 48
; CHECK-NEXT:    lxvd2x 61, 1, 3 # 16-byte Folded Reload
; CHECK-NEXT:    xxmrghd 34, 1, 0
; CHECK-NEXT:    addi 1, 1, 128
; CHECK-NEXT:    ld 0, 16(1)
; CHECK-NEXT:    lwz 12, 8(1)
; CHECK-NEXT:    mtlr 0
; CHECK-NEXT:    mtocrf 32, 12
; CHECK-NEXT:    blr
  %res = call <2 x i128> @llvm.masked.urem(<2 x i128> %x, <2 x i128> %y, <2 x i1> %m)
  ret <2 x i128> %res
}

; Promotion and widening
define <3 x i10> @urem_v3i10(<3 x i10> %x, <3 x i10> %y, <3 x i1> %m) {
; CHECK-LABEL: urem_v3i10:
; CHECK:       # %bb.0:
; CHECK-NEXT:    lbz 11, 96(1)
; CHECK-NEXT:    li 12, 0
; CHECK-NEXT:    rldimi 9, 10, 32, 0
; CHECK-NEXT:    addis 10, 2, .LCPI7_0@toc@ha
; CHECK-NEXT:    mtvsrwz 33, 8
; CHECK-NEXT:    vspltisw 4, -10
; CHECK-NEXT:    xxleqv 38, 38, 38
; CHECK-NEXT:    addi 10, 10, .LCPI7_0@toc@l
; CHECK-NEXT:    mtfprd 0, 9
; CHECK-NEXT:    vsrw 4, 4, 4
; CHECK-NEXT:    rldimi 11, 12, 32, 0
; CHECK-NEXT:    lxvd2x 1, 0, 10
; CHECK-NEXT:    mtfprd 2, 11
; CHECK-NEXT:    xxswapd 34, 1
; CHECK-NEXT:    mtfprwz 1, 7
; CHECK-NEXT:    xxmrghd 35, 2, 0
; CHECK-NEXT:    mtfprwz 0, 6
; CHECK-NEXT:    vslw 3, 3, 6
; CHECK-NEXT:    vsraw 3, 3, 6
; CHECK-NEXT:    xxmrghw 37, 1, 0
; CHECK-NEXT:    mtfprwz 0, 3
; CHECK-NEXT:    mtfprwz 1, 4
; CHECK-NEXT:    vperm 5, 1, 5, 2
; CHECK-NEXT:    mtvsrwz 33, 5
; CHECK-NEXT:    xxmrghw 32, 1, 0
; CHECK-NEXT:    xxland 0, 37, 36
; CHECK-NEXT:    vperm 0, 1, 0, 2
; CHECK-NEXT:    vspltisw 1, 1
; CHECK-NEXT:    xxland 1, 32, 36
; CHECK-NEXT:    xxsel 0, 33, 0, 35
; CHECK-NEXT:    xxswapd 2, 0
; CHECK-NEXT:    xxswapd 3, 1
; CHECK-NEXT:    xxsldwi 4, 0, 0, 1
; CHECK-NEXT:    xxsldwi 5, 1, 1, 1
; CHECK-NEXT:    mffprwz 3, 2
; CHECK-NEXT:    mffprwz 4, 3
; CHECK-NEXT:    divwu 5, 4, 3
; CHECK-NEXT:    mullw 3, 5, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprwz 4, 5
; CHECK-NEXT:    mtfprwz 2, 3
; CHECK-NEXT:    mffprwz 3, 4
; CHECK-NEXT:    divwu 5, 4, 3
; CHECK-NEXT:    mullw 3, 5, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mffprwz 4, 1
; CHECK-NEXT:    mtfprwz 3, 3
; CHECK-NEXT:    mffprwz 3, 0
; CHECK-NEXT:    divwu 5, 4, 3
; CHECK-NEXT:    mullw 3, 5, 3
; CHECK-NEXT:    sub 3, 4, 3
; CHECK-NEXT:    mtvsrwz 36, 3
; CHECK-NEXT:    xxmrghw 35, 3, 2
; CHECK-NEXT:    vperm 2, 4, 3, 2
; CHECK-NEXT:    mfvsrwz 5, 34
; CHECK-NEXT:    xxswapd 0, 34
; CHECK-NEXT:    xxsldwi 1, 34, 34, 1
; CHECK-NEXT:    mffprwz 3, 0
; CHECK-NEXT:    mffprwz 4, 1
; CHECK-NEXT:    blr
  %res = call <3 x i10> @llvm.masked.urem(<3 x i10> %x, <3 x i10> %y, <3 x i1> %m)
  ret <3 x i10> %res
}