; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt %s -S -passes=msan 2>&1 | FileCheck %s

target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

define <4 x double> @test_x86_avx_addsub_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_addsub_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.addsub.pd.256(<4 x double> %a0, <4 x double> %a1) ; <<4 x double>> [#uses=1]
  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_x86_avx_addsub_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_addsub_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.addsub.ps.256(<8 x float> %a0, <8 x float> %a1) ; <<8 x float>> [#uses=1]
  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_x86_avx_blendv_pd_256(<4 x double> %a0, <4 x double> %a1, <4 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_blendv_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[_MSPROP1:%.*]] = or <4 x i64> [[_MSPROP]], [[TMP3]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.blendv.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]], <4 x double> [[A2:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP1]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %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_x86_avx_blendv_ps_256(<8 x float> %a0, <8 x float> %a1, <8 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_blendv_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 64) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[_MSPROP1:%.*]] = or <8 x i32> [[_MSPROP]], [[TMP3]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.blendv.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]], <8 x float> [[A2:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP1]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %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_x86_avx_cmp_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0:![0-9]+]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8:[0-9]+]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.cmp.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]], i8 7)
; CHECK-NEXT:    store <4 x i64> zeroinitializer, <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.cmp.pd.256(<4 x double> %a0, <4 x double> %a1, i8 7) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.cmp.pd.256(<4 x double>, <4 x double>, i8) nounwind readnone


define <8 x float> @test_x86_avx_cmp_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]], i8 7)
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 7) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}

define <8 x float> @test_x86_avx_cmp_ps_256_pseudo_op(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_cmp_ps_256_pseudo_op(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[A2:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]], i8 0)
; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i256 [[TMP9]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP10:%.*]], label [[TMP11:%.*]], !prof [[PROF0]]
; CHECK:       10:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       11:
; CHECK-NEXT:    [[A3:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A2]], i8 1)
; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP3:%.*]] = icmp ne i256 [[TMP12]], 0
; CHECK-NEXT:    br i1 [[_MSCMP3]], label [[TMP13:%.*]], label [[TMP14:%.*]], !prof [[PROF0]]
; CHECK:       13:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       14:
; CHECK-NEXT:    [[A4:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A3]], i8 2)
; CHECK-NEXT:    [[TMP15:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP4:%.*]] = icmp ne i256 [[TMP15]], 0
; CHECK-NEXT:    br i1 [[_MSCMP4]], label [[TMP16:%.*]], label [[TMP17:%.*]], !prof [[PROF0]]
; CHECK:       16:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       17:
; CHECK-NEXT:    [[A5:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A4]], i8 3)
; CHECK-NEXT:    [[TMP18:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP5:%.*]] = icmp ne i256 [[TMP18]], 0
; CHECK-NEXT:    br i1 [[_MSCMP5]], label [[TMP19:%.*]], label [[TMP20:%.*]], !prof [[PROF0]]
; CHECK:       19:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       20:
; CHECK-NEXT:    [[A6:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A5]], i8 4)
; CHECK-NEXT:    [[TMP21:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP6:%.*]] = icmp ne i256 [[TMP21]], 0
; CHECK-NEXT:    br i1 [[_MSCMP6]], label [[TMP22:%.*]], label [[TMP23:%.*]], !prof [[PROF0]]
; CHECK:       22:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       23:
; CHECK-NEXT:    [[A7:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A6]], i8 5)
; CHECK-NEXT:    [[TMP24:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP7:%.*]] = icmp ne i256 [[TMP24]], 0
; CHECK-NEXT:    br i1 [[_MSCMP7]], label [[TMP25:%.*]], label [[TMP26:%.*]], !prof [[PROF0]]
; CHECK:       25:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       26:
; CHECK-NEXT:    [[A8:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A7]], i8 6)
; CHECK-NEXT:    [[TMP27:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP8:%.*]] = icmp ne i256 [[TMP27]], 0
; CHECK-NEXT:    br i1 [[_MSCMP8]], label [[TMP28:%.*]], label [[TMP29:%.*]], !prof [[PROF0]]
; CHECK:       28:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       29:
; CHECK-NEXT:    [[A9:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A8]], i8 7)
; CHECK-NEXT:    [[TMP30:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP9:%.*]] = icmp ne i256 [[TMP30]], 0
; CHECK-NEXT:    br i1 [[_MSCMP9]], label [[TMP31:%.*]], label [[TMP32:%.*]], !prof [[PROF0]]
; CHECK:       31:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       32:
; CHECK-NEXT:    [[A10:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A9]], i8 8)
; CHECK-NEXT:    [[TMP33:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP10:%.*]] = icmp ne i256 [[TMP33]], 0
; CHECK-NEXT:    br i1 [[_MSCMP10]], label [[TMP34:%.*]], label [[TMP35:%.*]], !prof [[PROF0]]
; CHECK:       34:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       35:
; CHECK-NEXT:    [[A11:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A10]], i8 9)
; CHECK-NEXT:    [[TMP36:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP11:%.*]] = icmp ne i256 [[TMP36]], 0
; CHECK-NEXT:    br i1 [[_MSCMP11]], label [[TMP37:%.*]], label [[TMP38:%.*]], !prof [[PROF0]]
; CHECK:       37:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       38:
; CHECK-NEXT:    [[A12:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A11]], i8 10)
; CHECK-NEXT:    [[TMP39:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP12:%.*]] = icmp ne i256 [[TMP39]], 0
; CHECK-NEXT:    br i1 [[_MSCMP12]], label [[TMP40:%.*]], label [[TMP41:%.*]], !prof [[PROF0]]
; CHECK:       40:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       41:
; CHECK-NEXT:    [[A13:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A12]], i8 11)
; CHECK-NEXT:    [[TMP42:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP13:%.*]] = icmp ne i256 [[TMP42]], 0
; CHECK-NEXT:    br i1 [[_MSCMP13]], label [[TMP43:%.*]], label [[TMP44:%.*]], !prof [[PROF0]]
; CHECK:       43:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       44:
; CHECK-NEXT:    [[A14:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A13]], i8 12)
; CHECK-NEXT:    [[TMP45:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP14:%.*]] = icmp ne i256 [[TMP45]], 0
; CHECK-NEXT:    br i1 [[_MSCMP14]], label [[TMP46:%.*]], label [[TMP47:%.*]], !prof [[PROF0]]
; CHECK:       46:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       47:
; CHECK-NEXT:    [[A15:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A14]], i8 13)
; CHECK-NEXT:    [[TMP48:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP15:%.*]] = icmp ne i256 [[TMP48]], 0
; CHECK-NEXT:    br i1 [[_MSCMP15]], label [[TMP49:%.*]], label [[TMP50:%.*]], !prof [[PROF0]]
; CHECK:       49:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       50:
; CHECK-NEXT:    [[A16:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A15]], i8 14)
; CHECK-NEXT:    [[TMP51:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP16:%.*]] = icmp ne i256 [[TMP51]], 0
; CHECK-NEXT:    br i1 [[_MSCMP16]], label [[TMP52:%.*]], label [[TMP53:%.*]], !prof [[PROF0]]
; CHECK:       52:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       53:
; CHECK-NEXT:    [[A17:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A16]], i8 15)
; CHECK-NEXT:    [[TMP54:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP17:%.*]] = icmp ne i256 [[TMP54]], 0
; CHECK-NEXT:    br i1 [[_MSCMP17]], label [[TMP55:%.*]], label [[TMP56:%.*]], !prof [[PROF0]]
; CHECK:       55:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       56:
; CHECK-NEXT:    [[A18:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A17]], i8 16)
; CHECK-NEXT:    [[TMP57:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP18:%.*]] = icmp ne i256 [[TMP57]], 0
; CHECK-NEXT:    br i1 [[_MSCMP18]], label [[TMP58:%.*]], label [[TMP59:%.*]], !prof [[PROF0]]
; CHECK:       58:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       59:
; CHECK-NEXT:    [[A19:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A18]], i8 17)
; CHECK-NEXT:    [[TMP60:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP19:%.*]] = icmp ne i256 [[TMP60]], 0
; CHECK-NEXT:    br i1 [[_MSCMP19]], label [[TMP61:%.*]], label [[TMP62:%.*]], !prof [[PROF0]]
; CHECK:       61:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       62:
; CHECK-NEXT:    [[A20:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A19]], i8 18)
; CHECK-NEXT:    [[TMP63:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP20:%.*]] = icmp ne i256 [[TMP63]], 0
; CHECK-NEXT:    br i1 [[_MSCMP20]], label [[TMP64:%.*]], label [[TMP65:%.*]], !prof [[PROF0]]
; CHECK:       64:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       65:
; CHECK-NEXT:    [[A21:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A20]], i8 19)
; CHECK-NEXT:    [[TMP66:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP21:%.*]] = icmp ne i256 [[TMP66]], 0
; CHECK-NEXT:    br i1 [[_MSCMP21]], label [[TMP67:%.*]], label [[TMP68:%.*]], !prof [[PROF0]]
; CHECK:       67:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       68:
; CHECK-NEXT:    [[A22:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A21]], i8 20)
; CHECK-NEXT:    [[TMP69:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP22:%.*]] = icmp ne i256 [[TMP69]], 0
; CHECK-NEXT:    br i1 [[_MSCMP22]], label [[TMP70:%.*]], label [[TMP71:%.*]], !prof [[PROF0]]
; CHECK:       70:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       71:
; CHECK-NEXT:    [[A23:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A22]], i8 21)
; CHECK-NEXT:    [[TMP72:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP23:%.*]] = icmp ne i256 [[TMP72]], 0
; CHECK-NEXT:    br i1 [[_MSCMP23]], label [[TMP73:%.*]], label [[TMP74:%.*]], !prof [[PROF0]]
; CHECK:       73:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       74:
; CHECK-NEXT:    [[A24:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A23]], i8 22)
; CHECK-NEXT:    [[TMP75:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP24:%.*]] = icmp ne i256 [[TMP75]], 0
; CHECK-NEXT:    br i1 [[_MSCMP24]], label [[TMP76:%.*]], label [[TMP77:%.*]], !prof [[PROF0]]
; CHECK:       76:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       77:
; CHECK-NEXT:    [[A25:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A24]], i8 23)
; CHECK-NEXT:    [[TMP78:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP25:%.*]] = icmp ne i256 [[TMP78]], 0
; CHECK-NEXT:    br i1 [[_MSCMP25]], label [[TMP79:%.*]], label [[TMP80:%.*]], !prof [[PROF0]]
; CHECK:       79:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       80:
; CHECK-NEXT:    [[A26:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A25]], i8 24)
; CHECK-NEXT:    [[TMP81:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP26:%.*]] = icmp ne i256 [[TMP81]], 0
; CHECK-NEXT:    br i1 [[_MSCMP26]], label [[TMP82:%.*]], label [[TMP83:%.*]], !prof [[PROF0]]
; CHECK:       82:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       83:
; CHECK-NEXT:    [[A27:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A26]], i8 25)
; CHECK-NEXT:    [[TMP84:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP27:%.*]] = icmp ne i256 [[TMP84]], 0
; CHECK-NEXT:    br i1 [[_MSCMP27]], label [[TMP85:%.*]], label [[TMP86:%.*]], !prof [[PROF0]]
; CHECK:       85:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       86:
; CHECK-NEXT:    [[A28:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A27]], i8 26)
; CHECK-NEXT:    [[TMP87:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP28:%.*]] = icmp ne i256 [[TMP87]], 0
; CHECK-NEXT:    br i1 [[_MSCMP28]], label [[TMP88:%.*]], label [[TMP89:%.*]], !prof [[PROF0]]
; CHECK:       88:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       89:
; CHECK-NEXT:    [[A29:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A28]], i8 27)
; CHECK-NEXT:    [[TMP90:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP29:%.*]] = icmp ne i256 [[TMP90]], 0
; CHECK-NEXT:    br i1 [[_MSCMP29]], label [[TMP91:%.*]], label [[TMP92:%.*]], !prof [[PROF0]]
; CHECK:       91:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       92:
; CHECK-NEXT:    [[A30:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A29]], i8 28)
; CHECK-NEXT:    [[TMP93:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP30:%.*]] = icmp ne i256 [[TMP93]], 0
; CHECK-NEXT:    br i1 [[_MSCMP30]], label [[TMP94:%.*]], label [[TMP95:%.*]], !prof [[PROF0]]
; CHECK:       94:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       95:
; CHECK-NEXT:    [[A31:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A30]], i8 29)
; CHECK-NEXT:    [[TMP96:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP31:%.*]] = icmp ne i256 [[TMP96]], 0
; CHECK-NEXT:    br i1 [[_MSCMP31]], label [[TMP97:%.*]], label [[TMP98:%.*]], !prof [[PROF0]]
; CHECK:       97:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       98:
; CHECK-NEXT:    [[A32:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A31]], i8 30)
; CHECK-NEXT:    [[TMP99:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP32:%.*]] = icmp ne i256 [[TMP99]], 0
; CHECK-NEXT:    br i1 [[_MSCMP32]], label [[TMP100:%.*]], label [[TMP101:%.*]], !prof [[PROF0]]
; CHECK:       100:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       101:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> [[A0]], <8 x float> [[A32]], i8 31)
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %a2 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 0) ; <<8 x float>> [#uses=1]
  %a3 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a2, i8 1) ; <<8 x float>> [#uses=1]
  %a4 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a3, i8 2) ; <<8 x float>> [#uses=1]
  %a5 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a4, i8 3) ; <<8 x float>> [#uses=1]
  %a6 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a5, i8 4) ; <<8 x float>> [#uses=1]
  %a7 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a6, i8 5) ; <<8 x float>> [#uses=1]
  %a8 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a7, i8 6) ; <<8 x float>> [#uses=1]
  %a9 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a8, i8 7) ; <<8 x float>> [#uses=1]
  %a10 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a9, i8 8) ; <<8 x float>> [#uses=1]
  %a11 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a10, i8 9) ; <<8 x float>> [#uses=1]
  %a12 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a11, i8 10) ; <<8 x float>> [#uses=1]
  %a13 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a12, i8 11) ; <<8 x float>> [#uses=1]
  %a14 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a13, i8 12) ; <<8 x float>> [#uses=1]
  %a15 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a14, i8 13) ; <<8 x float>> [#uses=1]
  %a16 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a15, i8 14) ; <<8 x float>> [#uses=1]
  %a17 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a16, i8 15) ; <<8 x float>> [#uses=1]
  %a18 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a17, i8 16) ; <<8 x float>> [#uses=1]
  %a19 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a18, i8 17) ; <<8 x float>> [#uses=1]
  %a20 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a19, i8 18) ; <<8 x float>> [#uses=1]
  %a21 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a20, i8 19) ; <<8 x float>> [#uses=1]
  %a22 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a21, i8 20) ; <<8 x float>> [#uses=1]
  %a23 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a22, i8 21) ; <<8 x float>> [#uses=1]
  %a24 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a23, i8 22) ; <<8 x float>> [#uses=1]
  %a25 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a24, i8 23) ; <<8 x float>> [#uses=1]
  %a26 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a25, i8 24) ; <<8 x float>> [#uses=1]
  %a27 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a26, i8 25) ; <<8 x float>> [#uses=1]
  %a28 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a27, i8 26) ; <<8 x float>> [#uses=1]
  %a29 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a28, i8 27) ; <<8 x float>> [#uses=1]
  %a30 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a29, i8 28) ; <<8 x float>> [#uses=1]
  %a31 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a30, i8 29) ; <<8 x float>> [#uses=1]
  %a32 = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a31, i8 30) ; <<8 x float>> [#uses=1]
  %res = call <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float> %a0, <8 x float> %a32, i8 31) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.cmp.ps.256(<8 x float>, <8 x float>, i8) nounwind readnone


define <4 x float> @test_x86_avx_cvt_pd2_ps_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_pd2_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x float> @llvm.x86.avx.cvt.pd2.ps.256(<4 x double> [[A0:%.*]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x float> [[RES]]
;
  %res = call <4 x float> @llvm.x86.avx.cvt.pd2.ps.256(<4 x double> %a0) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
declare <4 x float> @llvm.x86.avx.cvt.pd2.ps.256(<4 x double>) nounwind readnone


define <4 x i32> @test_x86_avx_cvt_pd2dq_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_pd2dq_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x i32> @llvm.x86.avx.cvt.pd2dq.256(<4 x double> [[A0:%.*]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x i32> [[RES]]
;
  %res = call <4 x i32> @llvm.x86.avx.cvt.pd2dq.256(<4 x double> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.avx.cvt.pd2dq.256(<4 x double>) nounwind readnone


define <8 x i32> @test_x86_avx_cvt_ps2dq_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvt_ps2dq_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x i32> @llvm.x86.avx.cvt.ps2dq.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x i32> [[RES]]
;
  %res = call <8 x i32> @llvm.x86.avx.cvt.ps2dq.256(<8 x float> %a0) ; <<8 x i32>> [#uses=1]
  ret <8 x i32> %res
}
declare <8 x i32> @llvm.x86.avx.cvt.ps2dq.256(<8 x float>) nounwind readnone


define <4 x i32> @test_x86_avx_cvtt_pd2dq_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvtt_pd2dq_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double> [[A0:%.*]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x i32> [[RES]]
;
  %res = call <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.avx.cvtt.pd2dq.256(<4 x double>) nounwind readnone


define <8 x i32> @test_x86_avx_cvtt_ps2dq_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_cvtt_ps2dq_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x i32> [[RES]]
;
  %res = call <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float> %a0) ; <<8 x i32>> [#uses=1]
  ret <8 x i32> %res
}
declare <8 x i32> @llvm.x86.avx.cvtt.ps2dq.256(<8 x float>) nounwind readnone


define <8 x float> @test_x86_avx_dp_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_dp_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.dp.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]], i8 7)
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.dp.ps.256(<8 x float> %a0, <8 x float> %a1, i8 7) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.dp.ps.256(<8 x float>, <8 x float>, i8) nounwind readnone


define <4 x double> @test_x86_avx_hadd_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hadd_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double> %a0, <4 x double> %a1) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.hadd.pd.256(<4 x double>, <4 x double>) nounwind readnone


define <8 x float> @test_x86_avx_hadd_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hadd_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float> %a0, <8 x float> %a1) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float>, <8 x float>) nounwind readnone


define <4 x double> @test_x86_avx_hsub_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hsub_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double> %a0, <4 x double> %a1) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.hsub.pd.256(<4 x double>, <4 x double>) nounwind readnone


define <8 x float> @test_x86_avx_hsub_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_hsub_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float> %a0, <8 x float> %a1) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.hsub.ps.256(<8 x float>, <8 x float>) nounwind readnone


define <32 x i8> @test_x86_avx_ldu_dq_256(i8* %a0) #0 {
; CHECK-LABEL: @test_x86_avx_ldu_dq_256(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint i8* [[A0:%.*]] to i64
; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to <32 x i8>*
; CHECK-NEXT:    [[_MSLD:%.*]] = load <32 x i8>, <32 x i8>* [[TMP4]], align 1
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP5:%.*]], label [[TMP6:%.*]], !prof [[PROF0]]
; CHECK:       5:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       6:
; CHECK-NEXT:    [[RES:%.*]] = call <32 x i8> @llvm.x86.avx.ldu.dq.256(i8* [[A0]])
; CHECK-NEXT:    store <32 x i8> [[_MSLD]], <32 x i8>* bitcast ([100 x i64]* @__msan_retval_tls to <32 x i8>*), align 8
; CHECK-NEXT:    ret <32 x i8> [[RES]]
;
  %res = call <32 x i8> @llvm.x86.avx.ldu.dq.256(i8* %a0) ; <<32 x i8>> [#uses=1]
  ret <32 x i8> %res
}
declare <32 x i8> @llvm.x86.avx.ldu.dq.256(i8*) nounwind readonly


define <2 x double> @test_x86_avx_maskload_pd(i8* %a0, <2 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP5]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
; CHECK:       6:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       7:
; CHECK-NEXT:    [[RES:%.*]] = call <2 x double> @llvm.x86.avx.maskload.pd(i8* [[A0:%.*]], <2 x i64> [[MASK:%.*]])
; CHECK-NEXT:    store <2 x i64> zeroinitializer, <2 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <2 x i64>*), align 8
; CHECK-NEXT:    ret <2 x double> [[RES]]
;
  %res = call <2 x double> @llvm.x86.avx.maskload.pd(i8* %a0, <2 x i64> %mask) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.avx.maskload.pd(i8*, <2 x i64>) nounwind readonly


define <4 x double> @test_x86_avx_maskload_pd_256(i8* %a0, <4 x i64> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP5]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
; CHECK:       6:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       7:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.maskload.pd.256(i8* [[A0:%.*]], <4 x i64> [[MASK:%.*]])
; CHECK-NEXT:    store <4 x i64> zeroinitializer, <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.maskload.pd.256(i8* %a0, <4 x i64> %mask) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.maskload.pd.256(i8*, <4 x i64>) nounwind readonly


define <4 x float> @test_x86_avx_maskload_ps(i8* %a0, <4 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP5]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
; CHECK:       6:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       7:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x float> @llvm.x86.avx.maskload.ps(i8* [[A0:%.*]], <4 x i32> [[MASK:%.*]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x float> [[RES]]
;
  %res = call <4 x float> @llvm.x86.avx.maskload.ps(i8* %a0, <4 x i32> %mask) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
declare <4 x float> @llvm.x86.avx.maskload.ps(i8*, <4 x i32>) nounwind readonly


define <8 x float> @test_x86_avx_maskload_ps_256(i8* %a0, <8 x i32> %mask) #0 {
; CHECK-LABEL: @test_x86_avx_maskload_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP5]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
; CHECK:       6:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       7:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.maskload.ps.256(i8* [[A0:%.*]], <8 x i32> [[MASK:%.*]])
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.maskload.ps.256(i8* %a0, <8 x i32> %mask) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.maskload.ps.256(i8*, <8 x i32>) nounwind readonly


define void @test_x86_avx_maskstore_pd(i8* %a0, <2 x i64> %mask, <2 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i64> [[TMP3]] to i128
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i128 [[TMP9]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP10:%.*]], label [[TMP11:%.*]], !prof [[PROF0]]
; CHECK:       10:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       11:
; CHECK-NEXT:    call void @llvm.x86.avx.maskstore.pd(i8* [[A0:%.*]], <2 x i64> [[MASK:%.*]], <2 x double> [[A2:%.*]])
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.maskstore.pd(i8* %a0, <2 x i64> %mask, <2 x double> %a2)
  ret void
}
declare void @llvm.x86.avx.maskstore.pd(i8*, <2 x i64>, <2 x double>) nounwind


define void @test_x86_avx_maskstore_pd_256(i8* %a0, <4 x i64> %mask, <4 x double> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i64> [[TMP3]] to i256
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i256 [[TMP9]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP10:%.*]], label [[TMP11:%.*]], !prof [[PROF0]]
; CHECK:       10:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       11:
; CHECK-NEXT:    call void @llvm.x86.avx.maskstore.pd.256(i8* [[A0:%.*]], <4 x i64> [[MASK:%.*]], <4 x double> [[A2:%.*]])
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.maskstore.pd.256(i8* %a0, <4 x i64> %mask, <4 x double> %a2)
  ret void
}
declare void @llvm.x86.avx.maskstore.pd.256(i8*, <4 x i64>, <4 x double>) nounwind


define void @test_x86_avx_maskstore_ps(i8* %a0, <4 x i32> %mask, <4 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i32>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 24) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i32> [[TMP3]] to i128
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i128 [[TMP9]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP10:%.*]], label [[TMP11:%.*]], !prof [[PROF0]]
; CHECK:       10:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       11:
; CHECK-NEXT:    call void @llvm.x86.avx.maskstore.ps(i8* [[A0:%.*]], <4 x i32> [[MASK:%.*]], <4 x float> [[A2:%.*]])
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.maskstore.ps(i8* %a0, <4 x i32> %mask, <4 x float> %a2)
  ret void
}
declare void @llvm.x86.avx.maskstore.ps(i8*, <4 x i32>, <4 x float>) nounwind


define void @test_x86_avx_maskstore_ps_256(i8* %a0, <8 x i32> %mask, <8 x float> %a2) #0 {
; CHECK-LABEL: @test_x86_avx_maskstore_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP3:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 40) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <8 x i32> [[TMP3]] to i256
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i256 [[TMP9]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP10:%.*]], label [[TMP11:%.*]], !prof [[PROF0]]
; CHECK:       10:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       11:
; CHECK-NEXT:    call void @llvm.x86.avx.maskstore.ps.256(i8* [[A0:%.*]], <8 x i32> [[MASK:%.*]], <8 x float> [[A2:%.*]])
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.maskstore.ps.256(i8* %a0, <8 x i32> %mask, <8 x float> %a2)
  ret void
}
declare void @llvm.x86.avx.maskstore.ps.256(i8*, <8 x i32>, <8 x float>) nounwind


define <4 x double> @test_x86_avx_max_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_max_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.max.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.max.pd.256(<4 x double> %a0, <4 x double> %a1) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.max.pd.256(<4 x double>, <4 x double>) nounwind readnone


define <8 x float> @test_x86_avx_max_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_max_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.max.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.max.ps.256(<8 x float> %a0, <8 x float> %a1) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.max.ps.256(<8 x float>, <8 x float>) nounwind readnone


define <4 x double> @test_x86_avx_min_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_min_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.min.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.min.pd.256(<4 x double> %a0, <4 x double> %a1) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.min.pd.256(<4 x double>, <4 x double>) nounwind readnone


define <8 x float> @test_x86_avx_min_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_min_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <8 x i32> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.min.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> [[_MSPROP]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.min.ps.256(<8 x float> %a0, <8 x float> %a1) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.min.ps.256(<8 x float>, <8 x float>) nounwind readnone


define i32 @test_x86_avx_movmsk_pd_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_movmsk_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.movmsk.pd.256(<4 x double> [[A0:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.movmsk.pd.256(<4 x double> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.movmsk.pd.256(<4 x double>) nounwind readnone


define i32 @test_x86_avx_movmsk_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_movmsk_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.movmsk.ps.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.movmsk.ps.256(<8 x float> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.movmsk.ps.256(<8 x float>) nounwind readnone


define i32 @test_x86_avx_ptestc_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestc_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.ptestc.256(<4 x i64> [[A0:%.*]], <4 x i64> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.ptestc.256(<4 x i64> %a0, <4 x i64> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.ptestc.256(<4 x i64>, <4 x i64>) nounwind readnone


define i32 @test_x86_avx_ptestnzc_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestnzc_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.ptestnzc.256(<4 x i64> [[A0:%.*]], <4 x i64> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.ptestnzc.256(<4 x i64> %a0, <4 x i64> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.ptestnzc.256(<4 x i64>, <4 x i64>) nounwind readnone


define i32 @test_x86_avx_ptestz_256(<4 x i64> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_ptestz_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.ptestz.256(<4 x i64> [[A0:%.*]], <4 x i64> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.ptestz.256(<4 x i64> %a0, <4 x i64> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.ptestz.256(<4 x i64>, <4 x i64>) nounwind readnone


define <8 x float> @test_x86_avx_rcp_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_rcp_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT:    store <8 x i32> [[TMP1]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float> %a0) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.rcp.ps.256(<8 x float>) nounwind readnone


define <4 x double> @test_x86_avx_round_pd_256(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_round_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.round.pd.256(<4 x double> [[A0:%.*]], i32 7)
; CHECK-NEXT:    store <4 x i64> zeroinitializer, <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.round.pd.256(<4 x double> %a0, i32 7) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.round.pd.256(<4 x double>, i32) nounwind readnone


define <8 x float> @test_x86_avx_round_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_round_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.round.ps.256(<8 x float> [[A0:%.*]], i32 7)
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.round.ps.256(<8 x float> %a0, i32 7) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.round.ps.256(<8 x float>, i32) nounwind readnone


define <8 x float> @test_x86_avx_rsqrt_ps_256(<8 x float> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_rsqrt_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float> [[A0:%.*]])
; CHECK-NEXT:    store <8 x i32> [[TMP1]], <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float> %a0) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.rsqrt.ps.256(<8 x float>) nounwind readnone

define <2 x double> @test_x86_avx_vpermilvar_pd(<2 x double> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> [[A0:%.*]], <2 x i64> [[A1:%.*]])
; CHECK-NEXT:    store <2 x i64> zeroinitializer, <2 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <2 x i64>*), align 8
; CHECK-NEXT:    ret <2 x double> [[RES]]
;
  %res = call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %a0, <2 x i64> %a1) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double>, <2 x i64>) nounwind readnone


define <4 x double> @test_x86_avx_vpermilvar_pd_256(<4 x double> %a0, <4 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> [[A0:%.*]], <4 x i64> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i64> zeroinitializer, <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %a0, <4 x i64> %a1) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}
declare <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double>, <4 x i64>) nounwind readnone

define <4 x double> @test_x86_avx_vpermilvar_pd_256_2(<4 x double> %a0) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_pd_256_2(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> [[A0:%.*]], <4 x i64> <i64 2, i64 0, i64 0, i64 2>)
; CHECK-NEXT:    store <4 x i64> zeroinitializer, <4 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i64>*), align 8
; CHECK-NEXT:    ret <4 x double> [[RES]]
;
  %res = call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %a0, <4 x i64> <i64 2, i64 0, i64 0, i64 2>) ; <<4 x double>> [#uses=1]
  ret <4 x double> %res
}

define <4 x float> @test_x86_avx_vpermilvar_ps(<4 x float> %a0, <4 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> [[A0:%.*]], <4 x i32> [[A1:%.*]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x float> [[RES]]
;
  %res = call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> %a0, <4 x i32> %a1) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
define <4 x float> @test_x86_avx_vpermilvar_ps_load(<4 x float> %a0, <4 x i32>* %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps_load(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to i64*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[A2:%.*]] = load <4 x i32>, <4 x i32>* [[A1:%.*]], align 16
; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint <4 x i32>* [[A1]] to i64
; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to <4 x i32>*
; CHECK-NEXT:    [[_MSLD:%.*]] = load <4 x i32>, <4 x i32>* [[TMP7]], align 16
; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP8]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP9:%.*]], label [[TMP10:%.*]], !prof [[PROF0]]
; CHECK:       9:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       10:
; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <4 x i32> [[_MSLD]] to i128
; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i128 [[TMP11]], 0
; CHECK-NEXT:    br i1 [[_MSCMP2]], label [[TMP12:%.*]], label [[TMP13:%.*]], !prof [[PROF0]]
; CHECK:       12:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       13:
; CHECK-NEXT:    [[RES:%.*]] = call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> [[A0:%.*]], <4 x i32> [[A2]])
; CHECK-NEXT:    store <4 x i32> zeroinitializer, <4 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <4 x i32>*), align 8
; CHECK-NEXT:    ret <4 x float> [[RES]]
;
  %a2 = load <4 x i32>, <4 x i32>* %a1
  %res = call <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float> %a0, <4 x i32> %a2) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
declare <4 x float> @llvm.x86.avx.vpermilvar.ps(<4 x float>, <4 x i32>) nounwind readnone


define <8 x float> @test_x86_avx_vpermilvar_ps_256(<8 x float> %a0, <8 x i32> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vpermilvar_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float> [[A0:%.*]], <8 x i32> [[A1:%.*]])
; CHECK-NEXT:    store <8 x i32> zeroinitializer, <8 x i32>* bitcast ([100 x i64]* @__msan_retval_tls to <8 x i32>*), align 8
; CHECK-NEXT:    ret <8 x float> [[RES]]
;
  %res = call <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float> %a0, <8 x i32> %a1) ; <<8 x float>> [#uses=1]
  ret <8 x float> %res
}
declare <8 x float> @llvm.x86.avx.vpermilvar.ps.256(<8 x float>, <8 x i32>) nounwind readnone


define i32 @test_x86_avx_vtestc_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestc.pd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestc.pd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestc.pd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_avx_vtestc_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestc.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestc.pd.256(<4 x double> %a0, <4 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestc.pd.256(<4 x double>, <4 x double>) nounwind readnone


define i32 @test_x86_avx_vtestc_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestc.ps(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestc.ps(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestc.ps(<4 x float>, <4 x float>) nounwind readnone


define i32 @test_x86_avx_vtestc_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestc_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestc.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestc.ps.256(<8 x float> %a0, <8 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestc.ps.256(<8 x float>, <8 x float>) nounwind readnone


define i32 @test_x86_avx_vtestnzc_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestnzc.pd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestnzc.pd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestnzc.pd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_avx_vtestnzc_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double> %a0, <4 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double>, <4 x double>) nounwind readnone


define i32 @test_x86_avx_vtestnzc_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestnzc.ps(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestnzc.ps(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestnzc.ps(<4 x float>, <4 x float>) nounwind readnone


define i32 @test_x86_avx_vtestnzc_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestnzc_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float> %a0, <8 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float>, <8 x float>) nounwind readnone


define i32 @test_x86_avx_vtestz_pd(<2 x double> %a0, <2 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i64> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i64> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestz.pd(<2 x double> [[A0:%.*]], <2 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestz.pd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestz.pd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_avx_vtestz_pd_256(<4 x double> %a0, <4 x double> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_pd_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <4 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i64> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i64> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestz.pd.256(<4 x double> [[A0:%.*]], <4 x double> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestz.pd.256(<4 x double> %a0, <4 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestz.pd.256(<4 x double>, <4 x double>) nounwind readnone


define i32 @test_x86_avx_vtestz_ps(<4 x float> %a0, <4 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <4 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <4 x i32>, <4 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <4 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i32> [[TMP2]] to i128
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i128 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestz.ps(<4 x float> [[A0:%.*]], <4 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestz.ps(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestz.ps(<4 x float>, <4 x float>) nounwind readnone


define i32 @test_x86_avx_vtestz_ps_256(<8 x float> %a0, <8 x float> %a1) #0 {
; CHECK-LABEL: @test_x86_avx_vtestz_ps_256(
; CHECK-DAG:    [[TMP1:%.*]] = load <8 x i32>, <8 x i32>* bitcast ([100 x i64]* @__msan_param_tls to <8 x i32>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 32) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i32> [[TMP1]] to i256
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i256 [[TMP3]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
; CHECK:       4:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       5:
; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i32> [[TMP2]] to i256
; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i256 [[TMP6]], 0
; CHECK-NEXT:    br i1 [[_MSCMP1]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0]]
; CHECK:       7:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       8:
; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.x86.avx.vtestz.ps.256(<8 x float> [[A0:%.*]], <8 x float> [[A1:%.*]])
; CHECK-NEXT:    store i32 0, i32* bitcast ([100 x i64]* @__msan_retval_tls to i32*), align 8
; CHECK-NEXT:    ret i32 [[RES]]
;
  %res = call i32 @llvm.x86.avx.vtestz.ps.256(<8 x float> %a0, <8 x float> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.avx.vtestz.ps.256(<8 x float>, <8 x float>) nounwind readnone


define void @test_x86_avx_vzeroall() #0 {
; CHECK-LABEL: @test_x86_avx_vzeroall(
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    call void @llvm.x86.avx.vzeroall()
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.vzeroall()
  ret void
}
declare void @llvm.x86.avx.vzeroall() nounwind


define void @test_x86_avx_vzeroupper() #0 {
; CHECK-LABEL: @test_x86_avx_vzeroupper(
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    call void @llvm.x86.avx.vzeroupper()
; CHECK-NEXT:    ret void
;
  call void @llvm.x86.avx.vzeroupper()
  ret void
}
declare void @llvm.x86.avx.vzeroupper() nounwind

define void @movnt_dq(i8* %p, <2 x i64> %a1) nounwind #0 {
; CHECK-LABEL: @movnt_dq(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <2 x i64> [[TMP1]], zeroinitializer
; CHECK-NEXT:    [[A2:%.*]] = add <2 x i64> [[A1:%.*]], <i64 1, i64 1>
; CHECK-NEXT:    [[_MSPROP1:%.*]] = shufflevector <2 x i64> [[_MSPROP]], <2 x i64> <i64 -1, i64 -1>, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
; CHECK-NEXT:    [[A3:%.*]] = shufflevector <2 x i64> [[A2]], <2 x i64> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
; CHECK-NEXT:    [[CAST:%.*]] = bitcast i8* [[P:%.*]] to <4 x i64>*
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint <4 x i64>* [[CAST]] to i64
; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to <4 x i64>*
; CHECK-NEXT:    store <4 x i64> [[_MSPROP1]], <4 x i64>* [[TMP7]], align 32
; CHECK-NEXT:    store <4 x i64> [[A3]], <4 x i64>* [[CAST]], align 32, !nontemporal !1
; CHECK-NEXT:    ret void
;
  %a2 = add <2 x i64> %a1, <i64 1, i64 1>
  %a3 = shufflevector <2 x i64> %a2, <2 x i64> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
  tail call void @llvm.x86.avx.movnt.dq.256(i8* %p, <4 x i64> %a3) nounwind
  ret void
}
declare void @llvm.x86.avx.movnt.dq.256(i8*, <4 x i64>) nounwind

define void @movnt_ps(i8* %p, <8 x float> %a) nounwind #0 {
; CHECK-LABEL: @movnt_ps(
; CHECK-DAG:    [[TMP1:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <8 x i32>, <8 x i32>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <8 x i32>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[CAST:%.*]] = bitcast i8* [[P:%.*]] to <8 x float>*
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint <8 x float>* [[CAST]] to i64
; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to <8 x i32>*
; CHECK-NEXT:    store <8 x i32> [[TMP2]], <8 x i32>* [[TMP7]], align 32
; CHECK-NEXT:    store <8 x float> [[A:%.*]], <8 x float>* [[CAST]], align 32, !nontemporal !1
; CHECK-NEXT:    ret void
;
  tail call void @llvm.x86.avx.movnt.ps.256(i8* %p, <8 x float> %a) nounwind
  ret void
}
declare void @llvm.x86.avx.movnt.ps.256(i8*, <8 x float>) nounwind

define void @movnt_pd(i8* %p, <4 x double> %a1) nounwind #0 {
  ; add operation forces the execution domain.
; CHECK-LABEL: @movnt_pd(
; CHECK-DAG:    [[TMP1:%.*]] = load <4 x i64>, <4 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 8) to <4 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load i64, i64* getelementptr inbounds ([100 x i64], [100 x i64]* @__msan_param_tls, i32 0, i32 0), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <4 x i64> [[TMP1]], zeroinitializer
; CHECK-NEXT:    [[A2:%.*]] = fadd <4 x double> [[A1:%.*]], zeroinitializer
; CHECK-NEXT:    [[CAST:%.*]] = bitcast i8* [[P:%.*]] to <4 x double>*
; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP2]], 0
; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
; CHECK:       3:
; CHECK-NEXT:    call void @__msan_warning_with_origin_noreturn(i32 0) #[[ATTR8]]
; CHECK-NEXT:    unreachable
; CHECK:       4:
; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint <4 x double>* [[CAST]] to i64
; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to <4 x i64>*
; CHECK-NEXT:    store <4 x i64> [[_MSPROP]], <4 x i64>* [[TMP7]], align 32
; CHECK-NEXT:    store <4 x double> [[A2]], <4 x double>* [[CAST]], align 32, !nontemporal !1
; CHECK-NEXT:    ret void
;
  %a2 = fadd <4 x double> %a1, <double 0x0, double 0x0, double 0x0, double 0x0>
  tail call void @llvm.x86.avx.movnt.pd.256(i8* %p, <4 x double> %a2) nounwind
  ret void
}
declare void @llvm.x86.avx.movnt.pd.256(i8*, <4 x double>) nounwind


define <2 x i64> @test_x86_pclmulqdq(<2 x i64> %a0, <2 x i64> %a1) #0 {
; CHECK-LABEL: @test_x86_pclmulqdq(
; CHECK-DAG:    [[TMP1:%.*]] = load <2 x i64>, <2 x i64>* bitcast ([100 x i64]* @__msan_param_tls to <2 x i64>*), align 8
; CHECK-DAG:    [[TMP2:%.*]] = load <2 x i64>, <2 x i64>* inttoptr (i64 add (i64 ptrtoint ([100 x i64]* @__msan_param_tls to i64), i64 16) to <2 x i64>*), align 8
; CHECK-NEXT:    call void @llvm.donothing()
; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i64> [[TMP1]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <2 x i64> [[TMP2]], <2 x i64> poison, <2 x i32> zeroinitializer
; CHECK-NEXT:    [[_MSPROP:%.*]] = or <2 x i64> [[TMP3]], [[TMP4]]
; CHECK-NEXT:    [[RES:%.*]] = call <2 x i64> @llvm.x86.pclmulqdq(<2 x i64> [[A0:%.*]], <2 x i64> [[A1:%.*]], i8 0)
; CHECK-NEXT:    store <2 x i64> [[_MSPROP]], <2 x i64>* bitcast ([100 x i64]* @__msan_retval_tls to <2 x i64>*), align 8
; CHECK-NEXT:    ret <2 x i64> [[RES]]
;
  %res = call <2 x i64> @llvm.x86.pclmulqdq(<2 x i64> %a0, <2 x i64> %a1, i8 0) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.pclmulqdq(<2 x i64>, <2 x i64>, i8) nounwind readnone

attributes #0 = { sanitize_memory }