llvm/test/Transforms/VectorCombine/X86/shuffle.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=vector-combine -S -mtriple=x86_64-- -mattr=SSE2 | FileCheck %s --check-prefixes=CHECK,SSE
; RUN: opt < %s -passes=vector-combine -S -mtriple=x86_64-- -mattr=AVX2 | FileCheck %s --check-prefixes=CHECK,AVX

; x86 does not have a cheap v16i8 shuffle until SSSE3 (pshufb)

define <16 x i8> @bitcast_shuf_narrow_element(<4 x i32> %v) {
; SSE-LABEL: @bitcast_shuf_narrow_element(
; SSE-NEXT:    [[SHUF:%.*]] = shufflevector <4 x i32> [[V:%.*]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; SSE-NEXT:    [[R:%.*]] = bitcast <4 x i32> [[SHUF]] to <16 x i8>
; SSE-NEXT:    ret <16 x i8> [[R]]
;
; AVX-LABEL: @bitcast_shuf_narrow_element(
; AVX-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[V:%.*]] to <16 x i8>
; AVX-NEXT:    [[R:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> poison, <16 x i32> <i32 12, i32 13, i32 14, i32 15, i32 8, i32 9, i32 10, i32 11, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX-NEXT:    ret <16 x i8> [[R]]
;
  %shuf = shufflevector <4 x i32> %v, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %r = bitcast <4 x i32> %shuf to <16 x i8>
  ret <16 x i8> %r
}

; v4f32 is the same cost as v4i32, so this always works

define <4 x float> @bitcast_shuf_same_size(<4 x i32> %v) {
; CHECK-LABEL: @bitcast_shuf_same_size(
; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[V:%.*]] to <4 x float>
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x float> [[TMP1]], <4 x float> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    ret <4 x float> [[R]]
;
  %shuf = shufflevector <4 x i32> %v, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %r = bitcast <4 x i32> %shuf to <4 x float>
  ret <4 x float> %r
}

; Negative test - length-changing shuffle

define <16 x i8> @bitcast_shuf_narrow_element_wrong_size(<2 x i32> %v) {
; CHECK-LABEL: @bitcast_shuf_narrow_element_wrong_size(
; CHECK-NEXT:    [[SHUF:%.*]] = shufflevector <2 x i32> [[V:%.*]], <2 x i32> poison, <4 x i32> <i32 1, i32 0, i32 1, i32 0>
; CHECK-NEXT:    [[R:%.*]] = bitcast <4 x i32> [[SHUF]] to <16 x i8>
; CHECK-NEXT:    ret <16 x i8> [[R]]
;
  %shuf = shufflevector <2 x i32> %v, <2 x i32> poison, <4 x i32> <i32 1, i32 0, i32 1, i32 0>
  %r = bitcast <4 x i32> %shuf to <16 x i8>
  ret <16 x i8> %r
}

; Negative test - must cast to vector type

define i128 @bitcast_shuf_narrow_element_wrong_type(<4 x i32> %v) {
; CHECK-LABEL: @bitcast_shuf_narrow_element_wrong_type(
; CHECK-NEXT:    [[SHUF:%.*]] = shufflevector <4 x i32> [[V:%.*]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    [[R:%.*]] = bitcast <4 x i32> [[SHUF]] to i128
; CHECK-NEXT:    ret i128 [[R]]
;
  %shuf = shufflevector <4 x i32> %v, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %r = bitcast <4 x i32> %shuf to i128
  ret i128 %r
}

; Widen shuffle elements

define <4 x i32> @bitcast_shuf_wide_element(<8 x i16> %v) {
; CHECK-LABEL: @bitcast_shuf_wide_element(
; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i16> [[V:%.*]] to <4 x i32>
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x i32> [[TMP1]], <4 x i32> poison, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %shuf = shufflevector <8 x i16> %v, <8 x i16> poison, <8 x i32> <i32 0, i32 1, i32 0, i32 1, i32 2, i32 3, i32 2, i32 3>
  %r = bitcast <8 x i16> %shuf to <4 x i32>
  ret <4 x i32> %r
}

declare void @use(<4 x i32>)

; Negative test - don't create an extra shuffle

define <16 x i8> @bitcast_shuf_uses(<4 x i32> %v) {
; CHECK-LABEL: @bitcast_shuf_uses(
; CHECK-NEXT:    [[SHUF:%.*]] = shufflevector <4 x i32> [[V:%.*]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; CHECK-NEXT:    call void @use(<4 x i32> [[SHUF]])
; CHECK-NEXT:    [[R:%.*]] = bitcast <4 x i32> [[SHUF]] to <16 x i8>
; CHECK-NEXT:    ret <16 x i8> [[R]]
;
  %shuf = shufflevector <4 x i32> %v, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  call void @use(<4 x i32> %shuf)
  %r = bitcast <4 x i32> %shuf to <16 x i8>
  ret <16 x i8> %r
}

define <2 x i64> @PR35454_1(<2 x i64> %v) {
; SSE-LABEL: @PR35454_1(
; SSE-NEXT:    [[BC:%.*]] = bitcast <2 x i64> [[V:%.*]] to <4 x i32>
; SSE-NEXT:    [[PERMIL:%.*]] = shufflevector <4 x i32> [[BC]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; SSE-NEXT:    [[BC1:%.*]] = bitcast <4 x i32> [[PERMIL]] to <16 x i8>
; SSE-NEXT:    [[ADD:%.*]] = shl <16 x i8> [[BC1]], <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
; SSE-NEXT:    [[BC2:%.*]] = bitcast <16 x i8> [[ADD]] to <4 x i32>
; SSE-NEXT:    [[PERMIL1:%.*]] = shufflevector <4 x i32> [[BC2]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; SSE-NEXT:    [[BC3:%.*]] = bitcast <4 x i32> [[PERMIL1]] to <2 x i64>
; SSE-NEXT:    ret <2 x i64> [[BC3]]
;
; AVX-LABEL: @PR35454_1(
; AVX-NEXT:    [[BC:%.*]] = bitcast <2 x i64> [[V:%.*]] to <4 x i32>
; AVX-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[BC]] to <16 x i8>
; AVX-NEXT:    [[BC1:%.*]] = shufflevector <16 x i8> [[TMP1]], <16 x i8> poison, <16 x i32> <i32 12, i32 13, i32 14, i32 15, i32 8, i32 9, i32 10, i32 11, i32 4, i32 5, i32 6, i32 7, i32 0, i32 1, i32 2, i32 3>
; AVX-NEXT:    [[ADD:%.*]] = shl <16 x i8> [[BC1]], <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
; AVX-NEXT:    [[BC2:%.*]] = bitcast <16 x i8> [[ADD]] to <4 x i32>
; AVX-NEXT:    [[PERMIL1:%.*]] = shufflevector <4 x i32> [[BC2]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; AVX-NEXT:    [[BC3:%.*]] = bitcast <4 x i32> [[PERMIL1]] to <2 x i64>
; AVX-NEXT:    ret <2 x i64> [[BC3]]
;
  %bc = bitcast <2 x i64> %v to <4 x i32>
  %permil = shufflevector <4 x i32> %bc, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %bc1 = bitcast <4 x i32> %permil to <16 x i8>
  %add = shl <16 x i8> %bc1, <i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1, i8 1>
  %bc2 = bitcast <16 x i8> %add to <4 x i32>
  %permil1 = shufflevector <4 x i32> %bc2, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %bc3 = bitcast <4 x i32> %permil1 to <2 x i64>
  ret <2 x i64> %bc3
}

define <2 x i64> @PR35454_2(<2 x i64> %v) {
; SSE-LABEL: @PR35454_2(
; SSE-NEXT:    [[BC:%.*]] = bitcast <2 x i64> [[V:%.*]] to <4 x i32>
; SSE-NEXT:    [[PERMIL:%.*]] = shufflevector <4 x i32> [[BC]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; SSE-NEXT:    [[BC1:%.*]] = bitcast <4 x i32> [[PERMIL]] to <8 x i16>
; SSE-NEXT:    [[ADD:%.*]] = shl <8 x i16> [[BC1]], <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
; SSE-NEXT:    [[BC2:%.*]] = bitcast <8 x i16> [[ADD]] to <4 x i32>
; SSE-NEXT:    [[PERMIL1:%.*]] = shufflevector <4 x i32> [[BC2]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; SSE-NEXT:    [[BC3:%.*]] = bitcast <4 x i32> [[PERMIL1]] to <2 x i64>
; SSE-NEXT:    ret <2 x i64> [[BC3]]
;
; AVX-LABEL: @PR35454_2(
; AVX-NEXT:    [[BC:%.*]] = bitcast <2 x i64> [[V:%.*]] to <4 x i32>
; AVX-NEXT:    [[TMP1:%.*]] = bitcast <4 x i32> [[BC]] to <8 x i16>
; AVX-NEXT:    [[BC1:%.*]] = shufflevector <8 x i16> [[TMP1]], <8 x i16> poison, <8 x i32> <i32 6, i32 7, i32 4, i32 5, i32 2, i32 3, i32 0, i32 1>
; AVX-NEXT:    [[ADD:%.*]] = shl <8 x i16> [[BC1]], <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
; AVX-NEXT:    [[BC2:%.*]] = bitcast <8 x i16> [[ADD]] to <4 x i32>
; AVX-NEXT:    [[PERMIL1:%.*]] = shufflevector <4 x i32> [[BC2]], <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
; AVX-NEXT:    [[BC3:%.*]] = bitcast <4 x i32> [[PERMIL1]] to <2 x i64>
; AVX-NEXT:    ret <2 x i64> [[BC3]]
;
  %bc = bitcast <2 x i64> %v to <4 x i32>
  %permil = shufflevector <4 x i32> %bc, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %bc1 = bitcast <4 x i32> %permil to <8 x i16>
  %add = shl <8 x i16> %bc1, <i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1, i16 1>
  %bc2 = bitcast <8 x i16> %add to <4 x i32>
  %permil1 = shufflevector <4 x i32> %bc2, <4 x i32> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
  %bc3 = bitcast <4 x i32> %permil1 to <2 x i64>
  ret <2 x i64> %bc3
}

; Shuffle is much cheaper than fdiv. FMF are intersected.

define <4 x float> @shuf_fdiv_v4f32_yy(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
; CHECK-LABEL: @shuf_fdiv_v4f32_yy(
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[X:%.*]], <4 x float> [[Z:%.*]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[Y:%.*]], <4 x float> poison, <4 x i32> <i32 1, i32 3, i32 1, i32 3>
; CHECK-NEXT:    [[R:%.*]] = fdiv arcp <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    ret <4 x float> [[R]]
;
  %b0 = fdiv fast <4 x float> %x, %y
  %b1 = fdiv arcp <4 x float> %z, %y
  %r = shufflevector <4 x float> %b0, <4 x float> %b1, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
  ret <4 x float> %r
}

; Common operand is op0 of the binops.

define <4 x i32> @shuf_add_v4i32_xx(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
; CHECK-LABEL: @shuf_add_v4i32_xx(
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[X:%.*]], <4 x i32> poison, <4 x i32> <i32 undef, i32 undef, i32 2, i32 0>
; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]], <4 x i32> <i32 undef, i32 undef, i32 6, i32 0>
; CHECK-NEXT:    [[R:%.*]] = add <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %b0 = add <4 x i32> %x, %y
  %b1 = add <4 x i32> %x, %z
  %r = shufflevector <4 x i32> %b0, <4 x i32> %b1, <4 x i32> <i32 poison, i32 poison, i32 6, i32 0>
  ret <4 x i32> %r
}

; For commutative instructions, common operand may be swapped.

define <4 x float> @shuf_fmul_v4f32_xx_swap(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
; CHECK-LABEL: @shuf_fmul_v4f32_xx_swap(
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[Y:%.*]], <4 x float> [[Z:%.*]], <4 x i32> <i32 0, i32 3, i32 4, i32 7>
; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[X:%.*]], <4 x float> poison, <4 x i32> <i32 0, i32 3, i32 0, i32 3>
; CHECK-NEXT:    [[R:%.*]] = fmul <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    ret <4 x float> [[R]]
;
  %b0 = fmul <4 x float> %x, %y
  %b1 = fmul <4 x float> %z, %x
  %r = shufflevector <4 x float> %b0, <4 x float> %b1, <4 x i32> <i32 0, i32 3, i32 4, i32 7>
  ret <4 x float> %r
}

; For commutative instructions, common operand may be swapped.

define <2 x i64> @shuf_and_v2i64_yy_swap(<2 x i64> %x, <2 x i64> %y, <2 x i64> %z) {
; CHECK-LABEL: @shuf_and_v2i64_yy_swap(
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <2 x i64> [[Y:%.*]], <2 x i64> poison, <2 x i32> <i32 1, i32 0>
; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <2 x i64> [[X:%.*]], <2 x i64> [[Z:%.*]], <2 x i32> <i32 3, i32 0>
; CHECK-NEXT:    [[R:%.*]] = and <2 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    ret <2 x i64> [[R]]
;
  %b0 = and <2 x i64> %x, %y
  %b1 = and <2 x i64> %y, %z
  %r = shufflevector <2 x i64> %b0, <2 x i64> %b1, <2 x i32> <i32 3, i32 0>
  ret <2 x i64> %r
}

; non-commutative binop, but common op0

define <4 x i32> @shuf_shl_v4i32_xx(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
; CHECK-LABEL: @shuf_shl_v4i32_xx(
; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[X:%.*]], <4 x i32> poison, <4 x i32> <i32 3, i32 1, i32 1, i32 2>
; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]], <4 x i32> <i32 3, i32 1, i32 1, i32 6>
; CHECK-NEXT:    [[R:%.*]] = shl <4 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %b0 = shl <4 x i32> %x, %y
  %b1 = shl <4 x i32> %x, %z
  %r = shufflevector <4 x i32> %b0, <4 x i32> %b1, <4 x i32> <i32 3, i32 1, i32 1, i32 6>
  ret <4 x i32> %r
}

; negative test - common operand, but not commutable

define <4 x i32> @shuf_shl_v4i32_xx_swap(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
; CHECK-LABEL: @shuf_shl_v4i32_xx_swap(
; CHECK-NEXT:    [[B0:%.*]] = shl <4 x i32> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = shl <4 x i32> [[Z:%.*]], [[X]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 3, i32 2, i32 2, i32 5>
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %b0 = shl <4 x i32> %x, %y
  %b1 = shl <4 x i32> %z, %x
  %r = shufflevector <4 x i32> %b0, <4 x i32> %b1, <4 x i32> <i32 3, i32 2, i32 2, i32 5>
  ret <4 x i32> %r
}

; negative test - mismatched opcodes

define <2 x i64> @shuf_sub_add_v2i64_yy(<2 x i64> %x, <2 x i64> %y, <2 x i64> %z) {
; CHECK-LABEL: @shuf_sub_add_v2i64_yy(
; CHECK-NEXT:    [[B0:%.*]] = sub <2 x i64> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = add <2 x i64> [[Z:%.*]], [[Y]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <2 x i64> [[B0]], <2 x i64> [[B1]], <2 x i32> <i32 3, i32 0>
; CHECK-NEXT:    ret <2 x i64> [[R]]
;
  %b0 = sub <2 x i64> %x, %y
  %b1 = add <2 x i64> %z, %y
  %r = shufflevector <2 x i64> %b0, <2 x i64> %b1, <2 x i32> <i32 3, i32 0>
  ret <2 x i64> %r
}

; negative test - type change via shuffle

define <8 x float> @shuf_fmul_v4f32_xx_type(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
; CHECK-LABEL: @shuf_fmul_v4f32_xx_type(
; CHECK-NEXT:    [[B0:%.*]] = fmul <4 x float> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = fmul <4 x float> [[Z:%.*]], [[X]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x float> [[B0]], <4 x float> [[B1]], <8 x i32> <i32 0, i32 3, i32 4, i32 7, i32 0, i32 1, i32 1, i32 6>
; CHECK-NEXT:    ret <8 x float> [[R]]
;
  %b0 = fmul <4 x float> %x, %y
  %b1 = fmul <4 x float> %z, %x
  %r = shufflevector <4 x float> %b0, <4 x float> %b1, <8 x i32> <i32 0, i32 3, i32 4, i32 7, i32 0, i32 1, i32 1, i32 6>
  ret <8 x float> %r
}

; negative test - uses

define <4 x i32> @shuf_lshr_v4i32_yy_use1(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
; CHECK-LABEL: @shuf_lshr_v4i32_yy_use1(
; CHECK-NEXT:    [[B0:%.*]] = lshr <4 x i32> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    call void @use(<4 x i32> [[B0]])
; CHECK-NEXT:    [[B1:%.*]] = lshr <4 x i32> [[Z:%.*]], [[Y]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 0, i32 2, i32 4, i32 6>
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %b0 = lshr <4 x i32> %x, %y
  call void @use(<4 x i32> %b0)
  %b1 = lshr <4 x i32> %z, %y
  %r = shufflevector <4 x i32> %b0, <4 x i32> %b1, <4 x i32> <i32 0, i32 2, i32 4, i32 6>
  ret <4 x i32> %r
}

; negative test - uses

define <4 x i32> @shuf_mul_v4i32_yy_use2(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
; CHECK-LABEL: @shuf_mul_v4i32_yy_use2(
; CHECK-NEXT:    [[B0:%.*]] = mul <4 x i32> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = mul <4 x i32> [[Z:%.*]], [[Y]]
; CHECK-NEXT:    call void @use(<4 x i32> [[B1]])
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; CHECK-NEXT:    ret <4 x i32> [[R]]
;
  %b0 = mul <4 x i32> %x, %y
  %b1 = mul <4 x i32> %z, %y
  call void @use(<4 x i32> %b1)
  %r = shufflevector <4 x i32> %b0, <4 x i32> %b1, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
  ret <4 x i32> %r
}

; negative test - must have matching operand

define <4 x float> @shuf_fadd_v4f32_no_common_op(<4 x float> %x, <4 x float> %y, <4 x float> %z, <4 x float> %w) {
; CHECK-LABEL: @shuf_fadd_v4f32_no_common_op(
; CHECK-NEXT:    [[B0:%.*]] = fadd <4 x float> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = fadd <4 x float> [[Z:%.*]], [[W:%.*]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x float> [[B0]], <4 x float> [[B1]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
; CHECK-NEXT:    ret <4 x float> [[R]]
;
  %b0 = fadd <4 x float> %x, %y
  %b1 = fadd <4 x float> %z, %w
  %r = shufflevector <4 x float> %b0, <4 x float> %b1, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
  ret <4 x float> %r
}

; negative test - binops may be relatively cheap

define <16 x i16> @shuf_and_v16i16_yy_expensive_shuf(<16 x i16> %x, <16 x i16> %y, <16 x i16> %z) {
; CHECK-LABEL: @shuf_and_v16i16_yy_expensive_shuf(
; CHECK-NEXT:    [[B0:%.*]] = and <16 x i16> [[X:%.*]], [[Y:%.*]]
; CHECK-NEXT:    [[B1:%.*]] = and <16 x i16> [[Y]], [[Z:%.*]]
; CHECK-NEXT:    [[R:%.*]] = shufflevector <16 x i16> [[B0]], <16 x i16> [[B1]], <16 x i32> <i32 15, i32 22, i32 25, i32 13, i32 28, i32 0, i32 undef, i32 3, i32 0, i32 30, i32 3, i32 7, i32 9, i32 19, i32 2, i32 22>
; CHECK-NEXT:    ret <16 x i16> [[R]]
;
  %b0 = and <16 x i16> %x, %y
  %b1 = and <16 x i16> %y, %z
  %r = shufflevector <16 x i16> %b0, <16 x i16> %b1, <16 x i32> <i32 15, i32 22, i32 25, i32 13, i32 28, i32 0, i32 poison, i32 3, i32 0, i32 30, i32 3, i32 7, i32 9, i32 19, i32 2, i32 22>
  ret <16 x i16> %r
}