llvm/test/CodeGen/X86/avx-intrinsics-fast-isel.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -fast-isel -mtriple=i386-unknown-unknown -mattr=avx,aes,pclmul | FileCheck %s --check-prefix=ALL --check-prefix=X32
; RUN: llc < %s -fast-isel -mtriple=x86_64-unknown-unknown -mattr=avx,aes,pclmul | FileCheck %s --check-prefix=ALL --check-prefix=X64

; NOTE: This should use IR equivalent to what is generated by clang/test/CodeGen/avx-builtins.c

define <4 x double> @test_mm256_add_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_add_pd:
; X32:       # BB#0:
; X32-NEXT:    vaddpd %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_add_pd:
; X64:       # BB#0:
; X64-NEXT:    vaddpd %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fadd <4 x double> %a0, %a1
  ret <4 x double> %res
}

define <8 x float> @test_mm256_add_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_add_ps:
; X32:       # BB#0:
; X32-NEXT:    vaddps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_add_ps:
; X64:       # BB#0:
; X64-NEXT:    vaddps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fadd <8 x float> %a0, %a1
  ret <8 x float> %res
}

define <4 x double> @test_mm256_addsub_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_addsub_pd:
; X32:       # BB#0:
; X32-NEXT:    vaddsubpd %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_addsub_pd:
; X64:       # BB#0:
; X64-NEXT:    vaddsubpd %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = call <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double> %a0, <4 x double> %a1)
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double>, <4 x double>) nounwind readnone

define <8 x float> @test_mm256_addsub_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_addsub_ps:
; X32:       # BB#0:
; X32-NEXT:    vaddsubps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_addsub_ps:
; X64:       # BB#0:
; X64-NEXT:    vaddsubps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = call <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float> %a0, <8 x float> %a1)
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float>, <8 x float>) nounwind readnone

define <4 x double> @test_mm256_and_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_and_pd:
; X32:       # BB#0:
; X32-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_and_pd:
; X64:       # BB#0:
; X64-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <4 x double> %a0 to <4 x i64>
  %2 = bitcast <4 x double> %a1 to <4 x i64>
  %res = and <4 x i64> %1, %2
  %bc = bitcast <4 x i64> %res to <4 x double>
  ret <4 x double> %bc
}

define <8 x float> @test_mm256_and_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_and_ps:
; X32:       # BB#0:
; X32-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_and_ps:
; X64:       # BB#0:
; X64-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <8 x float> %a0 to <8 x i32>
  %2 = bitcast <8 x float> %a1 to <8 x i32>
  %res = and <8 x i32> %1, %2
  %bc = bitcast <8 x i32> %res to <8 x float>
  ret <8 x float> %bc
}

define <4 x double> @test_mm256_andnot_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_andnot_pd:
; X32:       # BB#0:
; X32-NEXT:    vpcmpeqd %xmm2, %xmm2, %xmm2
; X32-NEXT:    vinsertf128 $1, %xmm2, %ymm2, %ymm2
; X32-NEXT:    vxorps %ymm2, %ymm0, %ymm0
; X32-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_andnot_pd:
; X64:       # BB#0:
; X64-NEXT:    vpcmpeqd %xmm2, %xmm2, %xmm2
; X64-NEXT:    vinsertf128 $1, %xmm2, %ymm2, %ymm2
; X64-NEXT:    vxorps %ymm2, %ymm0, %ymm0
; X64-NEXT:    vandps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <4 x double> %a0 to <4 x i64>
  %2 = bitcast <4 x double> %a1 to <4 x i64>
  %3 = xor <4 x i64> %1, <i64 -1, i64 -1, i64 -1, i64 -1>
  %res = and <4 x i64> %3, %2
  %bc = bitcast <4 x i64> %res to <4 x double>
  ret <4 x double> %bc
}

define <8 x float> @test_mm256_andnot_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_andnot_ps:
; X32:       # BB#0:
; X32-NEXT:    vandnps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_andnot_ps:
; X64:       # BB#0:
; X64-NEXT:    vandnps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <8 x float> %a0 to <8 x i32>
  %2 = bitcast <8 x float> %a1 to <8 x i32>
  %3 = xor <8 x i32> %1, <i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1, i32 -1>
  %res = and <8 x i32> %3, %2
  %bc = bitcast <8 x i32> %res to <8 x float>
  ret <8 x float> %bc
}

define <4 x double> @test_mm256_blend_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_blend_pd:
; X32:       # BB#0:
; X32-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_blend_pd:
; X64:       # BB#0:
; X64-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3]
; X64-NEXT:    retq
  %res = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 0, i32 5, i32 6, i32 3>
  ret <4 x double> %res
}

define <8 x float> @test_mm256_blend_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_blend_ps:
; X32:       # BB#0:
; X32-NEXT:    vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4,5,6],ymm1[7]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_blend_ps:
; X64:       # BB#0:
; X64-NEXT:    vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3],ymm0[4,5,6],ymm1[7]
; X64-NEXT:    retq
  %res = shufflevector <8 x float> %a0, <8 x float> %a1, <8 x i32> <i32 0, i32 9, i32 10, i32 11, i32 4, i32 5, i32 6, i32 15>
  ret <8 x float> %res
}

define <4 x double> @test_mm256_blendv_pd(<4 x double> %a0, <4 x double> %a1, <4 x double> %a2) {
; X32-LABEL: test_mm256_blendv_pd:
; X32:       # BB#0:
; X32-NEXT:    vblendvpd %ymm2, %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_blendv_pd:
; X64:       # BB#0:
; X64-NEXT:    vblendvpd %ymm2, %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = call <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double> %a0, <4 x double> %a1, <4 x double> %a2) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double>, <4 x double>, <4 x double>) nounwind readnone

define <8 x float> @test_mm256_blendv_ps(<8 x float> %a0, <8 x float> %a1, <8 x float> %a2) {
; X32-LABEL: test_mm256_blendv_ps:
; X32:       # BB#0:
; X32-NEXT:    vblendvps %ymm2, %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_blendv_ps:
; X64:       # BB#0:
; X64-NEXT:    vblendvps %ymm2, %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = call <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float> %a0, <8 x float> %a1, <8 x float> %a2) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float>, <8 x float>, <8 x float>) nounwind readnone

define <4 x double> @test_mm256_broadcast_pd(i8* %a0) {
; X32-LABEL: test_mm256_broadcast_pd:
; X32:       # BB#0:
; X32-NEXT:    movl {{[0-9]+}}(%esp), %eax
; X32-NEXT:    vbroadcastf128 (%eax), %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_broadcast_pd:
; X64:       # BB#0:
; X64-NEXT:    vbroadcastf128 (%rdi), %ymm0
; X64-NEXT:    retq
  %res = call <4 x double> @llvm.x86.avx.vbroadcastf128.pd.256(i8* %a0) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.vbroadcastf128.pd.256(i8*) nounwind readonly

define <8 x float> @test_mm256_broadcast_ps(i8* %a0) {
; X32-LABEL: test_mm256_broadcast_ps:
; X32:       # BB#0:
; X32-NEXT:    movl {{[0-9]+}}(%esp), %eax
; X32-NEXT:    vbroadcastf128 (%eax), %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_broadcast_ps:
; X64:       # BB#0:
; X64-NEXT:    vbroadcastf128 (%rdi), %ymm0
; X64-NEXT:    retq
  %res = call <8 x float> @llvm.x86.avx.vbroadcastf128.ps.256(i8* %a0) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.vbroadcastf128.ps.256(i8*) nounwind readonly

; TODO test_mm256_broadcast_sd
; TODO test_mm_broadcast_ss
; TODO test_mm256_broadcast_sd
; TODO test_mm256_castpd_ps
; TODO test_mm256_castpd_si256
; TODO test_mm256_castpd128_pd256
; TODO test_mm256_castpd256_pd128
; TODO test_mm256_castps_pd
; TODO test_mm256_castps_si256
; TODO test_mm256_castps128_ps256
; TODO test_mm256_castps256_ps128
; TODO test_mm256_castsi128_si256
; TODO test_mm256_castsi256_pd
; TODO test_mm256_castsi256_ps
; TODO test_mm256_castsi256_si128
; TODO test_mm256_ceil_pd
; TODO test_mm256_ceil_ps
; TODO test_mm_cmp_pd
; TODO test_mm256_cmp_pd
; TODO test _mm_cmp_ps
; TODO test _mm256_cmp_ps
; TODO test_mm_cmp_sd
; TODO test_mm_cmp_ss
; TODO test_mm256_cvtepi32_pd
; TODO test_mm256_cvtepi32_ps
; TODO test_mm256_cvtpd_epi32
; TODO test_mm256_cvtpd_ps
; TODO test_mm256_cvtps_epi32
; TODO test_mm256_cvtps_pd
; TODO test_mm256_cvttpd_epi32
; TODO test_mm256_cvttps_epi32

define <4 x double> @test_mm256_div_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_div_pd:
; X32:       # BB#0:
; X32-NEXT:    vdivpd %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_div_pd:
; X64:       # BB#0:
; X64-NEXT:    vdivpd %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fdiv <4 x double> %a0, %a1
  ret <4 x double> %res
}

define <8 x float> @test_mm256_div_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_div_ps:
; X32:       # BB#0:
; X32-NEXT:    vdivps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_div_ps:
; X64:       # BB#0:
; X64-NEXT:    vdivps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fdiv <8 x float> %a0, %a1
  ret <8 x float> %res
}

; TODO test_mm256_dp_ps
; TODO test_mm256_extract_epi16
; TODO test_mm256_extract_epi32
; TODO test_mm256_extract_epi64
; TODO test_mm256_extract_epi8
; TODO test_mm256_extractf128_pd
; TODO test_mm256_extractf128_ps
; TODO test_mm256_extractf128_si256
; TODO test_mm256_floor_pd
; TODO test_mm256_floor_ps
; TODO test_mm256_hadd_pd
; TODO test_mm256_hadd_ps
; TODO test_mm256_hsub_pd
; TODO test_mm256_hsub_ps
; TODO test_mm256_insert_epi16
; TODO test_mm256_insert_epi32
; TODO test_mm256_insert_epi64
; TODO test_mm256_insert_epi8
; TODO test_mm256_insertf128_pd
; TODO test_mm256_insertf128_ps
; TODO test_mm256_insertf128_si256
; TODO test_mm256_lddqu_si256
; TODO test_mm256_load_pd
; TODO test_mm256_load_ps
; TODO test_mm256_load_si256
; TODO test_mm256_loadu_pd
; TODO test_mm256_loadu_ps
; TODO test_mm256_loadu_si256
; TODO test_mm256_loadu2_m128
; TODO test_mm256_loadu2_m128d
; TODO test_mm256_loadu2_m128i
; TODO test_mm_maskload_pd
; TODO test_mm256_maskload_pd
; TODO test_mm_maskload_ps
; TODO test_mm256_maskload_ps
; TODO test_mm_maskstore_pd
; TODO test_mm256_maskstore_pd
; TODO test_mm_maskstore_ps
; TODO test_mm256_maskstore_ps
; TODO test_mm256_max_pd
; TODO test_mm256_max_ps
; TODO test_mm256_min_pd
; TODO test_mm256_min_ps
; TODO test_mm256_movedup_pd
; TODO test_mm256_movehdup_ps
; TODO test_mm256_moveldup_ps
; TODO test_mm256_movemask_pd
; TODO test_mm256_movemask_ps

define <4 x double> @test_mm256_mul_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_mul_pd:
; X32:       # BB#0:
; X32-NEXT:    vmulpd %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_mul_pd:
; X64:       # BB#0:
; X64-NEXT:    vmulpd %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fmul <4 x double> %a0, %a1
  ret <4 x double> %res
}

define <8 x float> @test_mm256_mul_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_mul_ps:
; X32:       # BB#0:
; X32-NEXT:    vmulps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_mul_ps:
; X64:       # BB#0:
; X64-NEXT:    vmulps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fmul <8 x float> %a0, %a1
  ret <8 x float> %res
}

; TODO test_mm256_or_pd
; TODO test_mm256_or_ps
; TODO test_mm_permute_pd
; TODO test_mm256_permute_pd
; TODO test_mm_permute_ps
; TODO test_mm256_permute_ps

define <4 x double> @test_mm256_permute2f128_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_permute2f128_pd:
; X32:       # BB#0:
; X32-NEXT:    vperm2f128 {{.*#+}} ymm0 = zero,zero,ymm1[0,1]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_permute2f128_pd:
; X64:       # BB#0:
; X64-NEXT:    vperm2f128 {{.*#+}} ymm0 = zero,zero,ymm1[0,1]
; X64-NEXT:    retq
  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 44)
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double>, <4 x double>, i8) nounwind readnone

; PR26667
define <8 x float> @test_mm256_permute2f128_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_permute2f128_ps:
; X32:       # BB#0:
; X32-NEXT:    vmovaps %ymm1, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_permute2f128_ps:
; X64:       # BB#0:
; X64-NEXT:    vmovaps %ymm1, %ymm0
; X64-NEXT:    retq
  %res = call <8 x float> @llvm.x86.avx.vperm2f128.ps.256(<8 x float> %a0, <8 x float> %a1, i8 50)
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.vperm2f128.ps.256(<8 x float>, <8 x float>, i8) nounwind readnone

define <4 x i64> @test_mm256_permute2f128_si256(<4 x i64> %a0, <4 x i64> %a1) {
; X32-LABEL: test_mm256_permute2f128_si256:
; X32:       # BB#0:
; X32-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm1[2,3,0,1]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_permute2f128_si256:
; X64:       # BB#0:
; X64-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm1[2,3,0,1]
; X64-NEXT:    retq
  %1 = bitcast <4 x i64> %a0 to <8 x i32>
  %2 = bitcast <4 x i64> %a1 to <8 x i32>
  %res = call <8 x i32> @llvm.x86.avx.vperm2f128.si.256(<8 x i32> %1, <8 x i32> %2, i8 35)
  %bc = bitcast <8 x i32> %res to <4 x i64>
  ret <4 x i64> %bc
}
declare <8 x i32> @llvm.x86.avx.vperm2f128.si.256(<8 x i32>, <8 x i32>, i8) nounwind readnone

; TODO test_mm_permutevar_pd
; TODO test_mm256_permutevar_pd
; TODO test_mm_permutevar_ps
; TODO test_mm256_permutevar_ps
; TODO test_mm256_rcp_ps
; TODO test_mm256_round_pd
; TODO test_mm256_round_ps
; TODO test_mm256_rsqrt_ps
; TODO test_mm256_set_epi16
; TODO test_mm256_set_epi32
; TODO test_mm256_set_epi64x
; TODO test_mm256_set_epi8
; TODO test_mm256_set_m128
; TODO test_mm256_set_m128d
; TODO test_mm256_set_m128i
; TODO test_mm256_set_pd
; TODO test_mm256_set_ps
; TODO test_mm256_set1_epi16
; TODO test_mm256_set1_epi32
; TODO test_mm256_set1_epi64x
; TODO test_mm256_set1_epi8
; TODO test_mm256_set1_pd
; TODO test_mm256_set1_ps
; TODO test_mm256_setr_epi16
; TODO test_mm256_setr_epi32
; TODO test_mm256_setr_epi64x
; TODO test_mm256_setr_epi8
; TODO test_mm256_setr_m128
; TODO test_mm256_setr_m128d
; TODO test_mm256_setr_m128i
; TODO test_mm256_setr_pd
; TODO test_mm256_setr_ps
; TODO test_mm256_setzero_pd
; TODO test_mm256_setzero_ps
; TODO test_mm256_setzero_si256
; TODO test_mm256_shuffle_pd
; TODO test_mm256_shuffle_ps
; TODO test_mm256_sqrt_pd
; TODO test_mm256_sqrt_ps
; TODO test_mm256_store_pd
; TODO test_mm256_store_ps
; TODO test_mm256_store_si256
; TODO test_mm256_storeu_pd
; TODO test_mm256_storeu_ps
; TODO test_mm256_storeu_si256
; TODO test_mm256_storeu2_m128
; TODO test_mm256_storeu2_m128d
; TODO test_mm256_storeu2_m128i
; TODO test_mm256_stream_pd
; TODO test_mm256_stream_ps
; TODO test_mm256_stream_si256

define <4 x double> @test_mm256_sub_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_sub_pd:
; X32:       # BB#0:
; X32-NEXT:    vsubpd %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_sub_pd:
; X64:       # BB#0:
; X64-NEXT:    vsubpd %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fsub <4 x double> %a0, %a1
  ret <4 x double> %res
}

define <8 x float> @test_mm256_sub_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_sub_ps:
; X32:       # BB#0:
; X32-NEXT:    vsubps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_sub_ps:
; X64:       # BB#0:
; X64-NEXT:    vsubps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %res = fsub <8 x float> %a0, %a1
  ret <8 x float> %res
}

; TODO test_mm_testc_pd
; TODO test_mm256_testc_pd
; TODO test_mm_testc_ps
; TODO test_mm256_testc_ps
; TODO test_mm256_testc_si256
; TODO test_mm_testnzc_pd
; TODO test_mm256_testnzc_pd
; TODO test_mm_testnzc_ps
; TODO test_mm256_testnzc_ps
; TODO test_mm256_testnzc_si256
; TODO test_mm_testz_pd
; TODO test_mm256_testz_pd
; TODO test_mm_testz_ps
; TODO test_mm256_testz_ps
; TODO test_mm256_testz_si256

define <2 x double> @test_mm_undefined_pd() {
; X32-LABEL: test_mm_undefined_pd:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm_undefined_pd:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <2 x double> undef
}

define <4 x double> @test_mm256_undefined_pd() {
; X32-LABEL: test_mm256_undefined_pd:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_undefined_pd:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <4 x double> undef
}

define <4 x float> @test_mm_undefined_ps() {
; X32-LABEL: test_mm_undefined_ps:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm_undefined_ps:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <4 x float> undef
}

define <8 x float> @test_mm256_undefined_ps() {
; X32-LABEL: test_mm256_undefined_ps:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_undefined_ps:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <8 x float> undef
}

define <2 x i64> @test_mm_undefined_si256() {
; X32-LABEL: test_mm_undefined_si256:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm_undefined_si256:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <2 x i64> undef
}

define <4 x i64> @test_mm256_undefined_si256() {
; X32-LABEL: test_mm256_undefined_si256:
; X32:       # BB#0:
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_undefined_si256:
; X64:       # BB#0:
; X64-NEXT:    retq
  ret <4 x i64> undef
}

define <4 x double> @test_mm256_unpackhi_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_unpackhi_pd:
; X32:       # BB#0:
; X32-NEXT:    vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_unpackhi_pd:
; X64:       # BB#0:
; X64-NEXT:    vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
; X64-NEXT:    retq
  %res = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
  ret <4 x double> %res
}

define <8 x float> @test_mm256_unpackhi_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_unpackhi_ps:
; X32:       # BB#0:
; X32-NEXT:    vunpckhps {{.*#+}} ymm0 = ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[6],ymm1[6],ymm0[7],ymm1[7]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_unpackhi_ps:
; X64:       # BB#0:
; X64-NEXT:    vunpckhps {{.*#+}} ymm0 = ymm0[2],ymm1[2],ymm0[3],ymm1[3],ymm0[6],ymm1[6],ymm0[7],ymm1[7]
; X64-NEXT:    retq
  %res = shufflevector <8 x float> %a0, <8 x float> %a1, <8 x i32> <i32 2, i32 10, i32 3, i32 11, i32 6, i32 14, i32 7, i32 15>
  ret <8 x float> %res
}

define <4 x double> @test_mm256_unpacklo_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_unpacklo_pd:
; X32:       # BB#0:
; X32-NEXT:    vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_unpacklo_pd:
; X64:       # BB#0:
; X64-NEXT:    vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
; X64-NEXT:    retq
  %res = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
  ret <4 x double> %res
}

define <8 x float> @test_mm256_unpacklo_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_unpacklo_ps:
; X32:       # BB#0:
; X32-NEXT:    vunpcklps {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[4],ymm1[4],ymm0[5],ymm1[5]
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_unpacklo_ps:
; X64:       # BB#0:
; X64-NEXT:    vunpcklps {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[1],ymm1[1],ymm0[4],ymm1[4],ymm0[5],ymm1[5]
; X64-NEXT:    retq
  %res = shufflevector <8 x float> %a0, <8 x float> %a1, <8 x i32> <i32 0, i32 8, i32 1, i32 9, i32 4, i32 12, i32 5, i32 13>
  ret <8 x float> %res
}

define <4 x double> @test_mm256_xor_pd(<4 x double> %a0, <4 x double> %a1) {
; X32-LABEL: test_mm256_xor_pd:
; X32:       # BB#0:
; X32-NEXT:    vxorps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_xor_pd:
; X64:       # BB#0:
; X64-NEXT:    vxorps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <4 x double> %a0 to <4 x i64>
  %2 = bitcast <4 x double> %a1 to <4 x i64>
  %res = xor <4 x i64> %1, %2
  %bc = bitcast <4 x i64> %res to <4 x double>
  ret <4 x double> %bc
}

define <8 x float> @test_mm256_xor_ps(<8 x float> %a0, <8 x float> %a1) {
; X32-LABEL: test_mm256_xor_ps:
; X32:       # BB#0:
; X32-NEXT:    vxorps %ymm1, %ymm0, %ymm0
; X32-NEXT:    retl
;
; X64-LABEL: test_mm256_xor_ps:
; X64:       # BB#0:
; X64-NEXT:    vxorps %ymm1, %ymm0, %ymm0
; X64-NEXT:    retq
  %1 = bitcast <8 x float> %a0 to <8 x i32>
  %2 = bitcast <8 x float> %a1 to <8 x i32>
  %res = xor <8 x i32> %1, %2
  %bc = bitcast <8 x i32> %res to <8 x float>
  ret <8 x float> %bc
}

; TODO test_mm256_zeroall
; TODO test_mm256_zeroupper