llvm-project/llvm/test/CodeGen/X86/sse2-intrinsics-x86.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -disable-peephole -mtriple=i386-apple-darwin -mattr=+sse2 -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X86,SSE,X86-SSE
; RUN: llc < %s -disable-peephole -mtriple=i386-apple-darwin -mattr=+avx -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X86,AVX,X86-AVX,AVX1,X86-AVX1
; RUN: llc < %s -disable-peephole -mtriple=i386-apple-darwin -mattr=+avx512f,+avx512bw,+avx512dq,+avx512vl -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X86,AVX,X86-AVX,AVX512,X86-AVX512
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+sse2 -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X64,SSE,X64-SSE
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+avx -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X64,AVX,X64-AVX,AVX1,X64-AVX1
; RUN: llc < %s -disable-peephole -mtriple=x86_64-apple-darwin -mattr=+avx512f,+avx512bw,+avx512dq,+avx512vl -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,X64,AVX,X64-AVX,AVX512,X64-AVX512

define <2 x double> @test_x86_sse2_cmp_pd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_cmp_pd:
; SSE:       ## %bb.0:
; SSE-NEXT:    cmpordpd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xc2,0xc1,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX-LABEL: test_x86_sse2_cmp_pd:
; AVX:       ## %bb.0:
; AVX-NEXT:    vcmpordpd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xc2,0xc1,0x07]
; AVX-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.cmp.pd(<2 x double>, <2 x double>, i8) nounwind readnone


define <2 x double> @test_x86_sse2_cmp_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_cmp_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    cmpordsd %xmm1, %xmm0 ## encoding: [0xf2,0x0f,0xc2,0xc1,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX-LABEL: test_x86_sse2_cmp_sd:
; AVX:       ## %bb.0:
; AVX-NEXT:    vcmpordsd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0xc2,0xc1,0x07]
; AVX-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double> %a0, <2 x double> %a1, i8 7) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounwind readnone


define i32 @test_x86_sse2_comieq_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comieq_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    comisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2f,0xc1]
; SSE-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; SSE-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; SSE-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; SSE-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comieq_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX1-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; AVX1-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; AVX1-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; AVX1-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comieq_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX512-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; AVX512-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; AVX512-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; AVX512-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_comige_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comige_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    comisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2f,0xc1]
; SSE-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comige_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vcomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX1-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comige_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vcomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX512-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comige.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_comigt_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comigt_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    comisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2f,0xc1]
; SSE-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comigt_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vcomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX1-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comigt_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vcomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX512-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comigt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comigt.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_comile_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comile_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    comisd %xmm0, %xmm1 ## encoding: [0x66,0x0f,0x2f,0xc8]
; SSE-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comile_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vcomisd %xmm0, %xmm1 ## encoding: [0xc5,0xf9,0x2f,0xc8]
; AVX1-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comile_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vcomisd %xmm0, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc8]
; AVX512-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comile.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comile.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_comilt_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comilt_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    comisd %xmm0, %xmm1 ## encoding: [0x66,0x0f,0x2f,0xc8]
; SSE-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comilt_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vcomisd %xmm0, %xmm1 ## encoding: [0xc5,0xf9,0x2f,0xc8]
; AVX1-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comilt_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vcomisd %xmm0, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc8]
; AVX512-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comilt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comilt.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_comineq_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_comineq_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    comisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2f,0xc1]
; SSE-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; SSE-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; SSE-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; SSE-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_comineq_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX1-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; AVX1-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; AVX1-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; AVX1-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_comineq_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2f,0xc1]
; AVX512-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; AVX512-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; AVX512-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; AVX512-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.comineq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.comineq.sd(<2 x double>, <2 x double>) nounwind readnone


define <4 x i32> @test_x86_sse2_cvtpd2dq(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_cvtpd2dq:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtpd2dq %xmm0, %xmm0 ## encoding: [0xf2,0x0f,0xe6,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvtpd2dq:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtpd2dq %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0xe6,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvtpd2dq:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtpd2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0xe6,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double>) nounwind readnone


define <2 x i64> @test_mm_cvtpd_epi32_zext(<2 x double> %a0) nounwind {
; SSE-LABEL: test_mm_cvtpd_epi32_zext:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtpd2dq %xmm0, %xmm0 ## encoding: [0xf2,0x0f,0xe6,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_mm_cvtpd_epi32_zext:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtpd2dq %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0xe6,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_mm_cvtpd_epi32_zext:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtpd2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0xe6,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %cvt = call <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double> %a0)
  %res = shufflevector <4 x i32> %cvt, <4 x i32> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  %bc = bitcast <4 x i32> %res to <2 x i64>
  ret <2 x i64> %bc
}


define <2 x i64> @test_mm_cvtpd_epi32_zext_load(<2 x double>* %p0) nounwind {
; X86-SSE-LABEL: test_mm_cvtpd_epi32_zext_load:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    cvtpd2dq (%eax), %xmm0 ## encoding: [0xf2,0x0f,0xe6,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_mm_cvtpd_epi32_zext_load:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX1-NEXT:    vcvtpd2dqx (%eax), %xmm0 ## encoding: [0xc5,0xfb,0xe6,0x00]
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_mm_cvtpd_epi32_zext_load:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX512-NEXT:    vcvtpd2dqx (%eax), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0xe6,0x00]
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_mm_cvtpd_epi32_zext_load:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    cvtpd2dq (%rdi), %xmm0 ## encoding: [0xf2,0x0f,0xe6,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_mm_cvtpd_epi32_zext_load:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vcvtpd2dqx (%rdi), %xmm0 ## encoding: [0xc5,0xfb,0xe6,0x07]
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_mm_cvtpd_epi32_zext_load:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vcvtpd2dqx (%rdi), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0xe6,0x07]
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %a0 = load <2 x double>, <2 x double>* %p0
  %cvt = call <4 x i32> @llvm.x86.sse2.cvtpd2dq(<2 x double> %a0)
  %res = shufflevector <4 x i32> %cvt, <4 x i32> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  %bc = bitcast <4 x i32> %res to <2 x i64>
  ret <2 x i64> %bc
}


define <4 x float> @test_x86_sse2_cvtpd2ps(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_cvtpd2ps:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtpd2ps %xmm0, %xmm0 ## encoding: [0x66,0x0f,0x5a,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvtpd2ps:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtpd2ps %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x5a,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvtpd2ps:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtpd2ps %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5a,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x float> @llvm.x86.sse2.cvtpd2ps(<2 x double> %a0) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
declare <4 x float> @llvm.x86.sse2.cvtpd2ps(<2 x double>) nounwind readnone

define <4 x float> @test_x86_sse2_cvtpd2ps_zext(<2 x double> %a0) nounwind {
; SSE-LABEL: test_x86_sse2_cvtpd2ps_zext:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtpd2ps %xmm0, %xmm0 ## encoding: [0x66,0x0f,0x5a,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvtpd2ps_zext:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtpd2ps %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x5a,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvtpd2ps_zext:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtpd2ps %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5a,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %cvt = call <4 x float> @llvm.x86.sse2.cvtpd2ps(<2 x double> %a0)
  %res = shufflevector <4 x float> %cvt, <4 x float> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  ret <4 x float> %res
}

define <4 x float> @test_x86_sse2_cvtpd2ps_zext_load(<2 x double>* %p0) nounwind {
; X86-SSE-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    cvtpd2ps (%eax), %xmm0 ## encoding: [0x66,0x0f,0x5a,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX1-NEXT:    vcvtpd2psx (%eax), %xmm0 ## encoding: [0xc5,0xf9,0x5a,0x00]
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX512-NEXT:    vcvtpd2psx (%eax), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5a,0x00]
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    cvtpd2ps (%rdi), %xmm0 ## encoding: [0x66,0x0f,0x5a,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vcvtpd2psx (%rdi), %xmm0 ## encoding: [0xc5,0xf9,0x5a,0x07]
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_x86_sse2_cvtpd2ps_zext_load:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vcvtpd2psx (%rdi), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5a,0x07]
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %a0 = load <2 x double>, <2 x double>* %p0
  %cvt = call <4 x float> @llvm.x86.sse2.cvtpd2ps(<2 x double> %a0)
  %res = shufflevector <4 x float> %cvt, <4 x float> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  ret <4 x float> %res
}

define <4 x i32> @test_x86_sse2_cvtps2dq(<4 x float> %a0) {
; SSE-LABEL: test_x86_sse2_cvtps2dq:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtps2dq %xmm0, %xmm0 ## encoding: [0x66,0x0f,0x5b,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvtps2dq:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtps2dq %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x5b,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvtps2dq:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtps2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5b,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.cvtps2dq(<4 x float> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.cvtps2dq(<4 x float>) nounwind readnone


define i32 @test_x86_sse2_cvtsd2si(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_cvtsd2si:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtsd2si %xmm0, %eax ## encoding: [0xf2,0x0f,0x2d,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvtsd2si:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvtsd2si %xmm0, %eax ## encoding: [0xc5,0xfb,0x2d,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvtsd2si:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvtsd2si %xmm0, %eax ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0x2d,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.cvtsd2si(<2 x double> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>) nounwind readnone


define <4 x float> @test_x86_sse2_cvtsd2ss(<4 x float> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_cvtsd2ss:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvtsd2ss %xmm1, %xmm0 ## encoding: [0xf2,0x0f,0x5a,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX-LABEL: test_x86_sse2_cvtsd2ss:
; AVX:       ## %bb.0:
; AVX-NEXT:    vcvtsd2ss %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5a,0xc1]
; AVX-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float> %a0, <2 x double> %a1) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}
declare <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float>, <2 x double>) nounwind readnone


define <4 x float> @test_x86_sse2_cvtsd2ss_load(<4 x float> %a0, <2 x double>* %p1) {
; X86-SSE-LABEL: test_x86_sse2_cvtsd2ss_load:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    cvtsd2ss (%eax), %xmm0 ## encoding: [0xf2,0x0f,0x5a,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX-LABEL: test_x86_sse2_cvtsd2ss_load:
; X86-AVX:       ## %bb.0:
; X86-AVX-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX-NEXT:    vcvtsd2ss (%eax), %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5a,0x00]
; X86-AVX-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_cvtsd2ss_load:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    cvtsd2ss (%rdi), %xmm0 ## encoding: [0xf2,0x0f,0x5a,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX-LABEL: test_x86_sse2_cvtsd2ss_load:
; X64-AVX:       ## %bb.0:
; X64-AVX-NEXT:    vcvtsd2ss (%rdi), %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5a,0x07]
; X64-AVX-NEXT:    retq ## encoding: [0xc3]
  %a1 = load <2 x double>, <2 x double>* %p1
  %res = call <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float> %a0, <2 x double> %a1) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}


define <4 x float> @test_x86_sse2_cvtsd2ss_load_optsize(<4 x float> %a0, <2 x double>* %p1) optsize {
; X86-SSE-LABEL: test_x86_sse2_cvtsd2ss_load_optsize:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    cvtsd2ss (%eax), %xmm0 ## encoding: [0xf2,0x0f,0x5a,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX-LABEL: test_x86_sse2_cvtsd2ss_load_optsize:
; X86-AVX:       ## %bb.0:
; X86-AVX-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX-NEXT:    vcvtsd2ss (%eax), %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5a,0x00]
; X86-AVX-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_cvtsd2ss_load_optsize:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    cvtsd2ss (%rdi), %xmm0 ## encoding: [0xf2,0x0f,0x5a,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX-LABEL: test_x86_sse2_cvtsd2ss_load_optsize:
; X64-AVX:       ## %bb.0:
; X64-AVX-NEXT:    vcvtsd2ss (%rdi), %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5a,0x07]
; X64-AVX-NEXT:    retq ## encoding: [0xc3]
  %a1 = load <2 x double>, <2 x double>* %p1
  %res = call <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float> %a0, <2 x double> %a1) ; <<4 x float>> [#uses=1]
  ret <4 x float> %res
}


define <4 x i32> @test_x86_sse2_cvttpd2dq(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_cvttpd2dq:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvttpd2dq %xmm0, %xmm0 ## encoding: [0x66,0x0f,0xe6,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvttpd2dq:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvttpd2dq %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe6,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvttpd2dq:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvttpd2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe6,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double>) nounwind readnone


define <2 x i64> @test_mm_cvttpd_epi32_zext(<2 x double> %a0) nounwind {
; SSE-LABEL: test_mm_cvttpd_epi32_zext:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvttpd2dq %xmm0, %xmm0 ## encoding: [0x66,0x0f,0xe6,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_mm_cvttpd_epi32_zext:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvttpd2dq %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe6,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_mm_cvttpd_epi32_zext:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvttpd2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe6,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %cvt = call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> %a0)
  %res = shufflevector <4 x i32> %cvt, <4 x i32> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  %bc = bitcast <4 x i32> %res to <2 x i64>
  ret <2 x i64> %bc
}


define <2 x i64> @test_mm_cvttpd_epi32_zext_load(<2 x double>* %p0) nounwind {
; X86-SSE-LABEL: test_mm_cvttpd_epi32_zext_load:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    cvttpd2dq (%eax), %xmm0 ## encoding: [0x66,0x0f,0xe6,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_mm_cvttpd_epi32_zext_load:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX1-NEXT:    vcvttpd2dqx (%eax), %xmm0 ## encoding: [0xc5,0xf9,0xe6,0x00]
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_mm_cvttpd_epi32_zext_load:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX512-NEXT:    vcvttpd2dqx (%eax), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe6,0x00]
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_mm_cvttpd_epi32_zext_load:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    cvttpd2dq (%rdi), %xmm0 ## encoding: [0x66,0x0f,0xe6,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_mm_cvttpd_epi32_zext_load:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vcvttpd2dqx (%rdi), %xmm0 ## encoding: [0xc5,0xf9,0xe6,0x07]
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_mm_cvttpd_epi32_zext_load:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vcvttpd2dqx (%rdi), %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe6,0x07]
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %a0 = load <2 x double>, <2 x double>* %p0
  %cvt = call <4 x i32> @llvm.x86.sse2.cvttpd2dq(<2 x double> %a0)
  %res = shufflevector <4 x i32> %cvt, <4 x i32> zeroinitializer, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
  %bc = bitcast <4 x i32> %res to <2 x i64>
  ret <2 x i64> %bc
}


define <4 x i32> @test_x86_sse2_cvttps2dq(<4 x float> %a0) {
; SSE-LABEL: test_x86_sse2_cvttps2dq:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvttps2dq %xmm0, %xmm0 ## encoding: [0xf3,0x0f,0x5b,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvttps2dq:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvttps2dq %xmm0, %xmm0 ## encoding: [0xc5,0xfa,0x5b,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvttps2dq:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvttps2dq %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfa,0x5b,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float> %a0) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.cvttps2dq(<4 x float>) nounwind readnone


define i32 @test_x86_sse2_cvttsd2si(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_cvttsd2si:
; SSE:       ## %bb.0:
; SSE-NEXT:    cvttsd2si %xmm0, %eax ## encoding: [0xf2,0x0f,0x2c,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_cvttsd2si:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vcvttsd2si %xmm0, %eax ## encoding: [0xc5,0xfb,0x2c,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_cvttsd2si:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vcvttsd2si %xmm0, %eax ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0x2c,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.cvttsd2si(<2 x double> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.cvttsd2si(<2 x double>) nounwind readnone


define <2 x double> @test_x86_sse2_max_pd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_max_pd:
; SSE:       ## %bb.0:
; SSE-NEXT:    maxpd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x5f,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_max_pd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vmaxpd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x5f,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_max_pd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vmaxpd %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5f,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.max.pd(<2 x double> %a0, <2 x double> %a1) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.max.pd(<2 x double>, <2 x double>) nounwind readnone


define <2 x double> @test_x86_sse2_max_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_max_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    maxsd %xmm1, %xmm0 ## encoding: [0xf2,0x0f,0x5f,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_max_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5f,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_max_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vmaxsd %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0x5f,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.max.sd(<2 x double> %a0, <2 x double> %a1) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.max.sd(<2 x double>, <2 x double>) nounwind readnone


define <2 x double> @test_x86_sse2_min_pd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_min_pd:
; SSE:       ## %bb.0:
; SSE-NEXT:    minpd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x5d,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_min_pd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vminpd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x5d,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_min_pd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vminpd %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x5d,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.min.pd(<2 x double> %a0, <2 x double> %a1) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.min.pd(<2 x double>, <2 x double>) nounwind readnone


define <2 x double> @test_x86_sse2_min_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_min_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    minsd %xmm1, %xmm0 ## encoding: [0xf2,0x0f,0x5d,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_min_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vminsd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xfb,0x5d,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_min_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vminsd %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xfb,0x5d,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x double> @llvm.x86.sse2.min.sd(<2 x double> %a0, <2 x double> %a1) ; <<2 x double>> [#uses=1]
  ret <2 x double> %res
}
declare <2 x double> @llvm.x86.sse2.min.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_movmsk_pd(<2 x double> %a0) {
; SSE-LABEL: test_x86_sse2_movmsk_pd:
; SSE:       ## %bb.0:
; SSE-NEXT:    movmskpd %xmm0, %eax ## encoding: [0x66,0x0f,0x50,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX-LABEL: test_x86_sse2_movmsk_pd:
; AVX:       ## %bb.0:
; AVX-NEXT:    vmovmskpd %xmm0, %eax ## encoding: [0xc5,0xf9,0x50,0xc0]
; AVX-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.movmsk.pd(<2 x double> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.movmsk.pd(<2 x double>) nounwind readnone


define <8 x i16> @test_x86_sse2_packssdw_128(<4 x i32> %a0, <4 x i32> %a1) {
; SSE-LABEL: test_x86_sse2_packssdw_128:
; SSE:       ## %bb.0:
; SSE-NEXT:    packssdw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x6b,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_packssdw_128:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpackssdw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x6b,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_packssdw_128:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpackssdw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x6b,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32> %a0, <4 x i32> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32>, <4 x i32>) nounwind readnone


define <8 x i16> @test_x86_sse2_packssdw_128_fold() {
; X86-SSE-LABEL: test_x86_sse2_packssdw_128_fold:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X86-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X86-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI30_0, kind: FK_Data_4
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_x86_sse2_packssdw_128_fold:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X86-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI30_0, kind: FK_Data_4
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_x86_sse2_packssdw_128_fold:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    vmovaps LCPI30_0, %xmm0 ## EVEX TO VEX Compression xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X86-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI30_0, kind: FK_Data_4
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_packssdw_128_fold:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X64-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X64-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI30_0-4, kind: reloc_riprel_4byte
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_x86_sse2_packssdw_128_fold:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X64-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI30_0-4, kind: reloc_riprel_4byte
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_x86_sse2_packssdw_128_fold:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vmovaps {{.*}}(%rip), %xmm0 ## EVEX TO VEX Compression xmm0 = [0,0,0,0,32767,32767,65535,32768]
; X64-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI30_0-4, kind: reloc_riprel_4byte
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.packssdw.128(<4 x i32> zeroinitializer, <4 x i32> <i32 65535, i32 65536, i32 -1, i32 -131072>)
  ret <8 x i16> %res
}


define <16 x i8> @test_x86_sse2_packsswb_128(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_packsswb_128:
; SSE:       ## %bb.0:
; SSE-NEXT:    packsswb %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x63,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_packsswb_128:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpacksswb %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x63,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_packsswb_128:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpacksswb %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x63,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16> %a0, <8 x i16> %a1) ; <<16 x i8>> [#uses=1]
  ret <16 x i8> %res
}
declare <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16>, <8 x i16>) nounwind readnone


define <16 x i8> @test_x86_sse2_packsswb_128_fold() {
; X86-SSE-LABEL: test_x86_sse2_packsswb_128_fold:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X86-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X86-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI32_0, kind: FK_Data_4
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_x86_sse2_packsswb_128_fold:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X86-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI32_0, kind: FK_Data_4
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_x86_sse2_packsswb_128_fold:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    vmovaps LCPI32_0, %xmm0 ## EVEX TO VEX Compression xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X86-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI32_0, kind: FK_Data_4
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_packsswb_128_fold:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X64-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X64-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI32_0-4, kind: reloc_riprel_4byte
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_x86_sse2_packsswb_128_fold:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X64-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI32_0-4, kind: reloc_riprel_4byte
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_x86_sse2_packsswb_128_fold:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vmovaps {{.*}}(%rip), %xmm0 ## EVEX TO VEX Compression xmm0 = [0,127,127,255,255,128,128,128,0,0,0,0,0,0,0,0]
; X64-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI32_0-4, kind: reloc_riprel_4byte
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.packsswb.128(<8 x i16> <i16 0, i16 255, i16 256, i16 65535, i16 -1, i16 -255, i16 -256, i16 -32678>, <8 x i16> zeroinitializer)
  ret <16 x i8> %res
}


define <16 x i8> @test_x86_sse2_packuswb_128(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_packuswb_128:
; SSE:       ## %bb.0:
; SSE-NEXT:    packuswb %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x67,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_packuswb_128:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x67,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_packuswb_128:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpackuswb %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x67,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16> %a0, <8 x i16> %a1) ; <<16 x i8>> [#uses=1]
  ret <16 x i8> %res
}
declare <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16>, <8 x i16>) nounwind readnone


define <16 x i8> @test_x86_sse2_packuswb_128_fold() {
; X86-SSE-LABEL: test_x86_sse2_packuswb_128_fold:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X86-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X86-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI34_0, kind: FK_Data_4
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_x86_sse2_packuswb_128_fold:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X86-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI34_0, kind: FK_Data_4
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_x86_sse2_packuswb_128_fold:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    vmovaps LCPI34_0, %xmm0 ## EVEX TO VEX Compression xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X86-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X86-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI34_0, kind: FK_Data_4
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_packuswb_128_fold:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    movaps {{.*#+}} xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X64-SSE-NEXT:    ## encoding: [0x0f,0x28,0x05,A,A,A,A]
; X64-SSE-NEXT:    ## fixup A - offset: 3, value: LCPI34_0-4, kind: reloc_riprel_4byte
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_x86_sse2_packuswb_128_fold:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vmovaps {{.*#+}} xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X64-AVX1-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX1-NEXT:    ## fixup A - offset: 4, value: LCPI34_0-4, kind: reloc_riprel_4byte
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_x86_sse2_packuswb_128_fold:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vmovaps {{.*}}(%rip), %xmm0 ## EVEX TO VEX Compression xmm0 = [0,255,255,0,0,0,0,0,0,0,0,0,0,0,0,0]
; X64-AVX512-NEXT:    ## encoding: [0xc5,0xf8,0x28,0x05,A,A,A,A]
; X64-AVX512-NEXT:    ## fixup A - offset: 4, value: LCPI34_0-4, kind: reloc_riprel_4byte
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.packuswb.128(<8 x i16> <i16 0, i16 255, i16 256, i16 65535, i16 -1, i16 -255, i16 -256, i16 -32678>, <8 x i16> zeroinitializer)
  ret <16 x i8> %res
}


define <4 x i32> @test_x86_sse2_pmadd_wd(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_pmadd_wd:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmaddwd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xf5,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmadd_wd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpmaddwd %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xf5,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmadd_wd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpmaddwd %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xf5,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16> %a0, <8 x i16> %a1) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.pmadd.wd(<8 x i16>, <8 x i16>) nounwind readnone


define <8 x i16> @test_x86_sse2_pmaxs_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_pmaxs_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmaxsw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xee,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmaxs_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpmaxsw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xee,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmaxs_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpmaxsw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xee,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.pmaxs.w(<8 x i16>, <8 x i16>) nounwind readnone


define <16 x i8> @test_x86_sse2_pmaxu_b(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: test_x86_sse2_pmaxu_b:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmaxub %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xde,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmaxu_b:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpmaxub %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xde,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmaxu_b:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpmaxub %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xde,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8> %a0, <16 x i8> %a1) ; <<16 x i8>> [#uses=1]
  ret <16 x i8> %res
}
declare <16 x i8> @llvm.x86.sse2.pmaxu.b(<16 x i8>, <16 x i8>) nounwind readnone


define <8 x i16> @test_x86_sse2_pmins_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_pmins_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    pminsw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xea,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmins_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpminsw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xea,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmins_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpminsw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xea,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.pmins.w(<8 x i16>, <8 x i16>) nounwind readnone


define <16 x i8> @test_x86_sse2_pminu_b(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: test_x86_sse2_pminu_b:
; SSE:       ## %bb.0:
; SSE-NEXT:    pminub %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xda,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pminu_b:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpminub %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xda,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pminu_b:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpminub %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xda,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8> %a0, <16 x i8> %a1) ; <<16 x i8>> [#uses=1]
  ret <16 x i8> %res
}
declare <16 x i8> @llvm.x86.sse2.pminu.b(<16 x i8>, <16 x i8>) nounwind readnone


define i32 @test_x86_sse2_pmovmskb_128(<16 x i8> %a0) {
; SSE-LABEL: test_x86_sse2_pmovmskb_128:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmovmskb %xmm0, %eax ## encoding: [0x66,0x0f,0xd7,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX-LABEL: test_x86_sse2_pmovmskb_128:
; AVX:       ## %bb.0:
; AVX-NEXT:    vpmovmskb %xmm0, %eax ## encoding: [0xc5,0xf9,0xd7,0xc0]
; AVX-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.pmovmskb.128(<16 x i8> %a0) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.pmovmskb.128(<16 x i8>) nounwind readnone


define <8 x i16> @test_x86_sse2_pmulh_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_pmulh_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmulhw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xe5,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmulh_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpmulhw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe5,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmulh_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpmulhw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe5,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.pmulh.w(<8 x i16>, <8 x i16>) nounwind readnone


define <8 x i16> @test_x86_sse2_pmulhu_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_pmulhu_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    pmulhuw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xe4,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pmulhu_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpmulhuw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe4,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pmulhu_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpmulhuw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe4,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.pmulhu.w(<8 x i16>, <8 x i16>) nounwind readnone


define <2 x i64> @test_x86_sse2_psad_bw(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: test_x86_sse2_psad_bw:
; SSE:       ## %bb.0:
; SSE-NEXT:    psadbw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xf6,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psad_bw:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsadbw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xf6,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psad_bw:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsadbw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xf6,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8> %a0, <16 x i8> %a1) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.sse2.psad.bw(<16 x i8>, <16 x i8>) nounwind readnone


define <4 x i32> @test_x86_sse2_psll_d(<4 x i32> %a0, <4 x i32> %a1) {
; SSE-LABEL: test_x86_sse2_psll_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    pslld %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xf2,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psll_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpslld %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xf2,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psll_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpslld %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xf2,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone


define <2 x i64> @test_x86_sse2_psll_q(<2 x i64> %a0, <2 x i64> %a1) {
; SSE-LABEL: test_x86_sse2_psll_q:
; SSE:       ## %bb.0:
; SSE-NEXT:    psllq %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xf3,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psll_q:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsllq %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xf3,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psll_q:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsllq %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xf3,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64>, <2 x i64>) nounwind readnone


define <8 x i16> @test_x86_sse2_psll_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_psll_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psllw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xf1,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psll_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsllw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xf1,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psll_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsllw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xf1,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.psll.w(<8 x i16>, <8 x i16>) nounwind readnone


define <4 x i32> @test_x86_sse2_pslli_d(<4 x i32> %a0) {
; SSE-LABEL: test_x86_sse2_pslli_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    pslld $7, %xmm0 ## encoding: [0x66,0x0f,0x72,0xf0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pslli_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpslld $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x72,0xf0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pslli_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpslld $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x72,0xf0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32> %a0, i32 7) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.pslli.d(<4 x i32>, i32) nounwind readnone


define <2 x i64> @test_x86_sse2_pslli_q(<2 x i64> %a0) {
; SSE-LABEL: test_x86_sse2_pslli_q:
; SSE:       ## %bb.0:
; SSE-NEXT:    psllq $7, %xmm0 ## encoding: [0x66,0x0f,0x73,0xf0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pslli_q:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsllq $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x73,0xf0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pslli_q:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsllq $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x73,0xf0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.sse2.pslli.q(<2 x i64>, i32) nounwind readnone


define <8 x i16> @test_x86_sse2_pslli_w(<8 x i16> %a0) {
; SSE-LABEL: test_x86_sse2_pslli_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psllw $7, %xmm0 ## encoding: [0x66,0x0f,0x71,0xf0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_pslli_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsllw $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x71,0xf0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_pslli_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsllw $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x71,0xf0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16> %a0, i32 7) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.pslli.w(<8 x i16>, i32) nounwind readnone


define <4 x i32> @test_x86_sse2_psra_d(<4 x i32> %a0, <4 x i32> %a1) {
; SSE-LABEL: test_x86_sse2_psra_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrad %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xe2,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psra_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrad %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe2,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psra_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrad %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe2,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.psra.d(<4 x i32>, <4 x i32>) nounwind readnone


define <8 x i16> @test_x86_sse2_psra_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_psra_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psraw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xe1,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psra_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsraw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xe1,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psra_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsraw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xe1,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.psra.w(<8 x i16>, <8 x i16>) nounwind readnone


define <4 x i32> @test_x86_sse2_psrai_d(<4 x i32> %a0) {
; SSE-LABEL: test_x86_sse2_psrai_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrad $7, %xmm0 ## encoding: [0x66,0x0f,0x72,0xe0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrai_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrad $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x72,0xe0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrai_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrad $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x72,0xe0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32> %a0, i32 7) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.psrai.d(<4 x i32>, i32) nounwind readnone


define <8 x i16> @test_x86_sse2_psrai_w(<8 x i16> %a0) {
; SSE-LABEL: test_x86_sse2_psrai_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psraw $7, %xmm0 ## encoding: [0x66,0x0f,0x71,0xe0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrai_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsraw $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x71,0xe0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrai_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsraw $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x71,0xe0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16> %a0, i32 7) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.psrai.w(<8 x i16>, i32) nounwind readnone


define <4 x i32> @test_x86_sse2_psrl_d(<4 x i32> %a0, <4 x i32> %a1) {
; SSE-LABEL: test_x86_sse2_psrl_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrld %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xd2,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrl_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrld %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xd2,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrl_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrld %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xd2,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32> %a0, <4 x i32> %a1) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32>, <4 x i32>) nounwind readnone


define <2 x i64> @test_x86_sse2_psrl_q(<2 x i64> %a0, <2 x i64> %a1) {
; SSE-LABEL: test_x86_sse2_psrl_q:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrlq %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xd3,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrl_q:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrlq %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xd3,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrl_q:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrlq %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xd3,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64>, <2 x i64>) nounwind readnone


define <8 x i16> @test_x86_sse2_psrl_w(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: test_x86_sse2_psrl_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrlw %xmm1, %xmm0 ## encoding: [0x66,0x0f,0xd1,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrl_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrlw %xmm1, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xd1,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrl_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrlw %xmm1, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xd1,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16>, <8 x i16>) nounwind readnone


define <8 x i16> @test_x86_sse2_psrl_w_load(<8 x i16> %a0, <8 x i16>* %p) {
; X86-SSE-LABEL: test_x86_sse2_psrl_w_load:
; X86-SSE:       ## %bb.0:
; X86-SSE-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-SSE-NEXT:    psrlw (%eax), %xmm0 ## encoding: [0x66,0x0f,0xd1,0x00]
; X86-SSE-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX1-LABEL: test_x86_sse2_psrl_w_load:
; X86-AVX1:       ## %bb.0:
; X86-AVX1-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX1-NEXT:    vpsrlw (%eax), %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xd1,0x00]
; X86-AVX1-NEXT:    retl ## encoding: [0xc3]
;
; X86-AVX512-LABEL: test_x86_sse2_psrl_w_load:
; X86-AVX512:       ## %bb.0:
; X86-AVX512-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-AVX512-NEXT:    vpsrlw (%eax), %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xd1,0x00]
; X86-AVX512-NEXT:    retl ## encoding: [0xc3]
;
; X64-SSE-LABEL: test_x86_sse2_psrl_w_load:
; X64-SSE:       ## %bb.0:
; X64-SSE-NEXT:    psrlw (%rdi), %xmm0 ## encoding: [0x66,0x0f,0xd1,0x07]
; X64-SSE-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX1-LABEL: test_x86_sse2_psrl_w_load:
; X64-AVX1:       ## %bb.0:
; X64-AVX1-NEXT:    vpsrlw (%rdi), %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0xd1,0x07]
; X64-AVX1-NEXT:    retq ## encoding: [0xc3]
;
; X64-AVX512-LABEL: test_x86_sse2_psrl_w_load:
; X64-AVX512:       ## %bb.0:
; X64-AVX512-NEXT:    vpsrlw (%rdi), %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0xd1,0x07]
; X64-AVX512-NEXT:    retq ## encoding: [0xc3]
  %a1 = load <8 x i16>, <8 x i16>* %p
  %res = call <8 x i16> @llvm.x86.sse2.psrl.w(<8 x i16> %a0, <8 x i16> %a1) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}


define <4 x i32> @test_x86_sse2_psrli_d(<4 x i32> %a0) {
; SSE-LABEL: test_x86_sse2_psrli_d:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrld $7, %xmm0 ## encoding: [0x66,0x0f,0x72,0xd0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrli_d:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrld $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x72,0xd0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrli_d:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrld $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x72,0xd0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32> %a0, i32 7) ; <<4 x i32>> [#uses=1]
  ret <4 x i32> %res
}
declare <4 x i32> @llvm.x86.sse2.psrli.d(<4 x i32>, i32) nounwind readnone


define <2 x i64> @test_x86_sse2_psrli_q(<2 x i64> %a0) {
; SSE-LABEL: test_x86_sse2_psrli_q:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrlq $7, %xmm0 ## encoding: [0x66,0x0f,0x73,0xd0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrli_q:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrlq $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x73,0xd0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrli_q:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrlq $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x73,0xd0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
  ret <2 x i64> %res
}
declare <2 x i64> @llvm.x86.sse2.psrli.q(<2 x i64>, i32) nounwind readnone


define <8 x i16> @test_x86_sse2_psrli_w(<8 x i16> %a0) {
; SSE-LABEL: test_x86_sse2_psrli_w:
; SSE:       ## %bb.0:
; SSE-NEXT:    psrlw $7, %xmm0 ## encoding: [0x66,0x0f,0x71,0xd0,0x07]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_psrli_w:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vpsrlw $7, %xmm0, %xmm0 ## encoding: [0xc5,0xf9,0x71,0xd0,0x07]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_psrli_w:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vpsrlw $7, %xmm0, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x71,0xd0,0x07]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16> %a0, i32 7) ; <<8 x i16>> [#uses=1]
  ret <8 x i16> %res
}
declare <8 x i16> @llvm.x86.sse2.psrli.w(<8 x i16>, i32) nounwind readnone


define i32 @test_x86_sse2_ucomieq_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomieq_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    ucomisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2e,0xc1]
; SSE-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; SSE-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; SSE-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; SSE-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomieq_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vucomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX1-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; AVX1-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; AVX1-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; AVX1-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomieq_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vucomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX512-NEXT:    setnp %al ## encoding: [0x0f,0x9b,0xc0]
; AVX512-NEXT:    sete %cl ## encoding: [0x0f,0x94,0xc1]
; AVX512-NEXT:    andb %al, %cl ## encoding: [0x20,0xc1]
; AVX512-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomieq.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_ucomige_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomige_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    ucomisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2e,0xc1]
; SSE-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomige_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vucomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX1-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomige_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vucomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX512-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomige.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_ucomigt_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomigt_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    ucomisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2e,0xc1]
; SSE-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomigt_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vucomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX1-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomigt_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vucomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX512-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomigt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomigt.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_ucomile_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomile_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    ucomisd %xmm0, %xmm1 ## encoding: [0x66,0x0f,0x2e,0xc8]
; SSE-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomile_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vucomisd %xmm0, %xmm1 ## encoding: [0xc5,0xf9,0x2e,0xc8]
; AVX1-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomile_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vucomisd %xmm0, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc8]
; AVX512-NEXT:    setae %al ## encoding: [0x0f,0x93,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomile.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomile.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_ucomilt_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomilt_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; SSE-NEXT:    ucomisd %xmm0, %xmm1 ## encoding: [0x66,0x0f,0x2e,0xc8]
; SSE-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomilt_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX1-NEXT:    vucomisd %xmm0, %xmm1 ## encoding: [0xc5,0xf9,0x2e,0xc8]
; AVX1-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomilt_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    xorl %eax, %eax ## encoding: [0x31,0xc0]
; AVX512-NEXT:    vucomisd %xmm0, %xmm1 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc8]
; AVX512-NEXT:    seta %al ## encoding: [0x0f,0x97,0xc0]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomilt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomilt.sd(<2 x double>, <2 x double>) nounwind readnone


define i32 @test_x86_sse2_ucomineq_sd(<2 x double> %a0, <2 x double> %a1) {
; SSE-LABEL: test_x86_sse2_ucomineq_sd:
; SSE:       ## %bb.0:
; SSE-NEXT:    ucomisd %xmm1, %xmm0 ## encoding: [0x66,0x0f,0x2e,0xc1]
; SSE-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; SSE-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; SSE-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; SSE-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; SSE-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX1-LABEL: test_x86_sse2_ucomineq_sd:
; AVX1:       ## %bb.0:
; AVX1-NEXT:    vucomisd %xmm1, %xmm0 ## encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX1-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; AVX1-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; AVX1-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; AVX1-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX1-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
;
; AVX512-LABEL: test_x86_sse2_ucomineq_sd:
; AVX512:       ## %bb.0:
; AVX512-NEXT:    vucomisd %xmm1, %xmm0 ## EVEX TO VEX Compression encoding: [0xc5,0xf9,0x2e,0xc1]
; AVX512-NEXT:    setp %al ## encoding: [0x0f,0x9a,0xc0]
; AVX512-NEXT:    setne %cl ## encoding: [0x0f,0x95,0xc1]
; AVX512-NEXT:    orb %al, %cl ## encoding: [0x08,0xc1]
; AVX512-NEXT:    movzbl %cl, %eax ## encoding: [0x0f,0xb6,0xc1]
; AVX512-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  %res = call i32 @llvm.x86.sse2.ucomineq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
  ret i32 %res
}
declare i32 @llvm.x86.sse2.ucomineq.sd(<2 x double>, <2 x double>) nounwind readnone

define void @test_x86_sse2_pause() {
; CHECK-LABEL: test_x86_sse2_pause:
; CHECK:       ## %bb.0:
; CHECK-NEXT:    pause ## encoding: [0xf3,0x90]
; CHECK-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  tail call void @llvm.x86.sse2.pause()
  ret void
}
declare void @llvm.x86.sse2.pause() nounwind

define void @lfence() nounwind {
; CHECK-LABEL: lfence:
; CHECK:       ## %bb.0:
; CHECK-NEXT:    lfence ## encoding: [0x0f,0xae,0xe8]
; CHECK-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  tail call void @llvm.x86.sse2.lfence()
  ret void
}
declare void @llvm.x86.sse2.lfence() nounwind

define void @mfence() nounwind {
; CHECK-LABEL: mfence:
; CHECK:       ## %bb.0:
; CHECK-NEXT:    mfence ## encoding: [0x0f,0xae,0xf0]
; CHECK-NEXT:    ret{{[l|q]}} ## encoding: [0xc3]
  tail call void @llvm.x86.sse2.mfence()
  ret void
}
declare void @llvm.x86.sse2.mfence() nounwind

define void @clflush(i8* %p) nounwind {
; X86-LABEL: clflush:
; X86:       ## %bb.0:
; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax ## encoding: [0x8b,0x44,0x24,0x04]
; X86-NEXT:    clflush (%eax) ## encoding: [0x0f,0xae,0x38]
; X86-NEXT:    retl ## encoding: [0xc3]
;
; X64-LABEL: clflush:
; X64:       ## %bb.0:
; X64-NEXT:    clflush (%rdi) ## encoding: [0x0f,0xae,0x3f]
; X64-NEXT:    retq ## encoding: [0xc3]
  tail call void @llvm.x86.sse2.clflush(i8* %p)
  ret void
}
declare void @llvm.x86.sse2.clflush(i8*) nounwind