llvm/llvm/test/Transforms/PhaseOrdering/X86/vector-math.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -O2 -S -mattr=sse < %s | FileCheck %s

; This file should represent the nearly raw (mem2reg was run to make it more direct)
; IR for code written using x86 SSE intrinsics to compute integer abs/max functions.
;
; https://llvm.org/PR34047

define available_externally <2 x i64> @cmpgt_i32_sel_m128i(<2 x i64> %a, <2 x i64> %b, <2 x i64> %c, <2 x i64> %d) {
  %call = call <2 x i64> @_mm_cmpgt_epi32(<2 x i64> %a, <2 x i64> %b)
  %call1 = call <2 x i64> @_mm_andnot_si128(<2 x i64> %call, <2 x i64> %c)
  %call2 = call <2 x i64> @_mm_and_si128(<2 x i64> %call, <2 x i64> %d)
  %call3 = call <2 x i64> @_mm_or_si128(<2 x i64> %call1, <2 x i64> %call2)
  ret <2 x i64> %call3
}

define internal <2 x i64> @_mm_set1_epi32(i32 %__i) {
  %call = call <2 x i64> @_mm_set_epi32(i32 %__i, i32 %__i, i32 %__i, i32 %__i)
  ret <2 x i64> %call
}

define internal <2 x i64> @_mm_sub_epi32(<2 x i64> %__a, <2 x i64> %__b) {
  %t0 = bitcast <2 x i64> %__a to <4 x i32>
  %t1 = bitcast <2 x i64> %__b to <4 x i32>
  %sub = sub <4 x i32> %t0, %t1
  %t2 = bitcast <4 x i32> %sub to <2 x i64>
  ret <2 x i64> %t2
}

define internal <2 x i64> @_mm_setzero_si128() {
  ret <2 x i64> zeroinitializer
}

define internal <2 x i64> @_mm_cmpgt_epi32(<2 x i64> %__a, <2 x i64> %__b) {
  %t0 = bitcast <2 x i64> %__a to <4 x i32>
  %t1 = bitcast <2 x i64> %__b to <4 x i32>
  %cmp = icmp sgt <4 x i32> %t0, %t1
  %sext = sext <4 x i1> %cmp to <4 x i32>
  %t2 = bitcast <4 x i32> %sext to <2 x i64>
  ret <2 x i64> %t2
}

define internal <2 x i64> @_mm_or_si128(<2 x i64> %__a, <2 x i64> %__b) {
  %or = or <2 x i64> %__a, %__b
  ret <2 x i64> %or
}

define internal <2 x i64> @_mm_andnot_si128(<2 x i64> %__a, <2 x i64> %__b) {
  %neg = xor <2 x i64> %__a, <i64 -1, i64 -1>
  %and = and <2 x i64> %neg, %__b
  ret <2 x i64> %and
}

define internal <2 x i64> @_mm_and_si128(<2 x i64> %__a, <2 x i64> %__b) {
  %and = and <2 x i64> %__a, %__b
  ret <2 x i64> %and
}

define internal <2 x i64> @_mm_set_epi32(i32 %__i3, i32 %__i2, i32 %__i1, i32 %__i0) {
  %vecinit = insertelement <4 x i32> undef, i32 %__i0, i32 0
  %vecinit1 = insertelement <4 x i32> %vecinit, i32 %__i1, i32 1
  %vecinit2 = insertelement <4 x i32> %vecinit1, i32 %__i2, i32 2
  %vecinit3 = insertelement <4 x i32> %vecinit2, i32 %__i3, i32 3
  %t0 = bitcast <4 x i32> %vecinit3 to <2 x i64>
  ret <2 x i64> %t0
}

define <2 x i64> @abs_v4i32(<2 x i64> %x) {
; CHECK-LABEL: @abs_v4i32(
; CHECK-NEXT:    [[T1_I:%.*]] = bitcast <2 x i64> [[X:%.*]] to <4 x i32>
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x i32> @llvm.abs.v4i32(<4 x i32> [[T1_I]], i1 false)
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to <2 x i64>
; CHECK-NEXT:    ret <2 x i64> [[TMP2]]
;
  %call = call <2 x i64> @_mm_set1_epi32(i32 -1)
  %call1 = call <2 x i64> @_mm_setzero_si128()
  %call2 = call <2 x i64> @_mm_sub_epi32(<2 x i64> %call1, <2 x i64> %x)
  %call3 = call <2 x i64> @cmpgt_i32_sel_m128i(<2 x i64> %x, <2 x i64> %call, <2 x i64> %call2, <2 x i64> %x)
  ret <2 x i64> %call3
}

define <2 x i64> @max_v4i32(<2 x i64> %x, <2 x i64> %y) {
; CHECK-LABEL: @max_v4i32(
; CHECK-NEXT:    [[T0_I_I:%.*]] = bitcast <2 x i64> [[X:%.*]] to <4 x i32>
; CHECK-NEXT:    [[T1_I_I:%.*]] = bitcast <2 x i64> [[Y:%.*]] to <4 x i32>
; CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x i32> @llvm.smax.v4i32(<4 x i32> [[T0_I_I]], <4 x i32> [[T1_I_I]])
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i32> [[TMP1]] to <2 x i64>
; CHECK-NEXT:    ret <2 x i64> [[TMP2]]
;
  %call = call <2 x i64> @cmpgt_i32_sel_m128i(<2 x i64> %x, <2 x i64> %y, <2 x i64> %y, <2 x i64> %x)
  ret <2 x i64> %call
}