; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i386-pc-win32       -mattr=+avx512f -mattr=+avx512vl -mattr=+avx512bw -mattr=+avx512dq -verify-machineinstrs  | FileCheck %s --check-prefix=X32
; RUN: llc < %s -mtriple=x86_64-win32        -mattr=+avx512f -mattr=+avx512vl -mattr=+avx512bw -mattr=+avx512dq -verify-machineinstrs  | FileCheck %s --check-prefix=WIN64
; RUN: llc < %s -mtriple=x86_64-linux-gnu    -mattr=+avx512f -mattr=+avx512vl -mattr=+avx512bw -mattr=+avx512dq -verify-machineinstrs  | FileCheck %s --check-prefix=LINUXOSX64

; Test regcall when receiving/returning i1
define dso_local x86_regcallcc i1 @test_argReti1(i1 %a)  {
; X32-LABEL: test_argReti1:
; X32:       # %bb.0:
; X32-NEXT:    incb %al
; X32-NEXT:    # kill: def $al killed $al killed $eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argReti1:
; WIN64:       # %bb.0:
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    # kill: def $al killed $al killed $eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argReti1:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    # kill: def $al killed $al killed $eax
; LINUXOSX64-NEXT:    retq
  %add = add i1 %a, 1
  ret i1 %add
}

; Test regcall when passing/retrieving i1
define dso_local x86_regcallcc i1 @test_CallargReti1(i1 %a)  {
; X32-LABEL: test_CallargReti1:
; X32:       # %bb.0:
; X32-NEXT:    incb %al
; X32-NEXT:    movzbl %al, %eax
; X32-NEXT:    calll _test_argReti1
; X32-NEXT:    incb %al
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargReti1:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    movzbl %al, %eax
; WIN64-NEXT:    callq test_argReti1
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargReti1:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    movzbl %al, %eax
; LINUXOSX64-NEXT:    callq test_argReti1
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = add i1 %a, 1
  %c = call x86_regcallcc i1 @test_argReti1(i1 %b)
  %d = add i1 %c, 1
  ret i1 %d
}

; Test regcall when receiving/returning i8
define dso_local x86_regcallcc i8 @test_argReti8(i8 %a)  {
; X32-LABEL: test_argReti8:
; X32:       # %bb.0:
; X32-NEXT:    incb %al
; X32-NEXT:    # kill: def $al killed $al killed $eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argReti8:
; WIN64:       # %bb.0:
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    # kill: def $al killed $al killed $eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argReti8:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    # kill: def $al killed $al killed $eax
; LINUXOSX64-NEXT:    retq
  %add = add i8 %a, 1
  ret i8 %add
}

; Test regcall when passing/retrieving i8
define dso_local x86_regcallcc i8 @test_CallargReti8(i8 %a)  {
; X32-LABEL: test_CallargReti8:
; X32:       # %bb.0:
; X32-NEXT:    incb %al
; X32-NEXT:    movzbl %al, %eax
; X32-NEXT:    calll _test_argReti8
; X32-NEXT:    incb %al
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargReti8:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    movzbl %al, %eax
; WIN64-NEXT:    callq test_argReti8
; WIN64-NEXT:    incb %al
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargReti8:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    movzbl %al, %eax
; LINUXOSX64-NEXT:    callq test_argReti8
; LINUXOSX64-NEXT:    incb %al
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = add i8 %a, 1
  %c = call x86_regcallcc i8 @test_argReti8(i8 %b)
  %d = add i8 %c, 1
  ret i8 %d
}

; Test regcall when receiving/returning i16
define dso_local x86_regcallcc i16 @test_argReti16(i16 %a)  {
; X32-LABEL: test_argReti16:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    # kill: def $ax killed $ax killed $eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argReti16:
; WIN64:       # %bb.0:
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    # kill: def $ax killed $ax killed $eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argReti16:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    # kill: def $ax killed $ax killed $eax
; LINUXOSX64-NEXT:    retq
  %add = add i16 %a, 1
  ret i16 %add
}

; Test regcall when passing/retrieving i16
define dso_local x86_regcallcc i16 @test_CallargReti16(i16 %a)  {
; X32-LABEL: test_CallargReti16:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    calll _test_argReti16
; X32-NEXT:    # kill: def $ax killed $ax def $eax
; X32-NEXT:    incl %eax
; X32-NEXT:    # kill: def $ax killed $ax killed $eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargReti16:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    callq test_argReti16
; WIN64-NEXT:    # kill: def $ax killed $ax def $eax
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    # kill: def $ax killed $ax killed $eax
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargReti16:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    callq test_argReti16
; LINUXOSX64-NEXT:    # kill: def $ax killed $ax def $eax
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    # kill: def $ax killed $ax killed $eax
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = add i16 %a, 1
  %c = call x86_regcallcc i16 @test_argReti16(i16 %b)
  %d = add i16 %c, 1
  ret i16 %d
}

; Test regcall when receiving/returning i32
define dso_local x86_regcallcc i32 @test_argReti32(i32 %a)  {
; X32-LABEL: test_argReti32:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argReti32:
; WIN64:       # %bb.0:
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argReti32:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    retq
  %add = add i32 %a, 1
  ret i32 %add
}

; Test regcall when passing/retrieving i32
define dso_local x86_regcallcc i32 @test_CallargReti32(i32 %a)  {
; X32-LABEL: test_CallargReti32:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    calll _test_argReti32
; X32-NEXT:    incl %eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargReti32:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    callq test_argReti32
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargReti32:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    callq test_argReti32
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = add i32 %a, 1
  %c = call x86_regcallcc i32 @test_argReti32(i32 %b)
  %d = add i32 %c, 1
  ret i32 %d
}

; Test regcall when receiving/returning i64
define dso_local x86_regcallcc i64 @test_argReti64(i64 %a)  {
; X32-LABEL: test_argReti64:
; X32:       # %bb.0:
; X32-NEXT:    addl $3, %eax
; X32-NEXT:    adcl $1, %ecx
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argReti64:
; WIN64:       # %bb.0:
; WIN64-NEXT:    movabsq $4294967299, %rcx # imm = 0x100000003
; WIN64-NEXT:    addq %rcx, %rax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argReti64:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    movabsq $4294967299, %rcx # imm = 0x100000003
; LINUXOSX64-NEXT:    addq %rcx, %rax
; LINUXOSX64-NEXT:    retq
  %add = add i64 %a, 4294967299
  ret i64 %add
}

; Test regcall when passing/retrieving i64
define dso_local x86_regcallcc i64 @test_CallargReti64(i64 %a)  {
; X32-LABEL: test_CallargReti64:
; X32:       # %bb.0:
; X32-NEXT:    addl $1, %eax
; X32-NEXT:    adcl $0, %ecx
; X32-NEXT:    calll _test_argReti64
; X32-NEXT:    addl $1, %eax
; X32-NEXT:    adcl $0, %ecx
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargReti64:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incq %rax
; WIN64-NEXT:    callq test_argReti64
; WIN64-NEXT:    incq %rax
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargReti64:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incq %rax
; LINUXOSX64-NEXT:    callq test_argReti64
; LINUXOSX64-NEXT:    incq %rax
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = add i64 %a, 1
  %c = call x86_regcallcc i64 @test_argReti64(i64 %b)
  %d = add i64 %c, 1
  ret i64 %d
}

; Test regcall when receiving/returning float
define dso_local x86_regcallcc float @test_argRetFloat(float %a)  {
; X32-LABEL: test_argRetFloat:
; X32:       # %bb.0:
; X32-NEXT:    vaddss __real@3f800000, %xmm0, %xmm0
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRetFloat:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vaddss __real@{{.*}}(%rip), %xmm0, %xmm0
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRetFloat:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vaddss {{.*}}(%rip), %xmm0, %xmm0
; LINUXOSX64-NEXT:    retq
  %add = fadd float 1.0, %a
  ret float %add
}

; Test regcall when passing/retrieving float
define dso_local x86_regcallcc float @test_CallargRetFloat(float %a)  {
; X32-LABEL: test_CallargRetFloat:
; X32:       # %bb.0:
; X32-NEXT:    subl $28, %esp
; X32-NEXT:    vmovups %xmm4, (%esp) # 16-byte Spill
; X32-NEXT:    vmovss {{.*#+}} xmm4 = mem[0],zero,zero,zero
; X32-NEXT:    vaddss %xmm4, %xmm0, %xmm0
; X32-NEXT:    calll _test_argRetFloat
; X32-NEXT:    vaddss %xmm4, %xmm0, %xmm0
; X32-NEXT:    vmovups (%esp), %xmm4 # 16-byte Reload
; X32-NEXT:    addl $28, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRetFloat:
; WIN64:       # %bb.0:
; WIN64-NEXT:    subq $24, %rsp
; WIN64-NEXT:    .seh_stackalloc 24
; WIN64-NEXT:    vmovaps %xmm8, (%rsp) # 16-byte Spill
; WIN64-NEXT:    .seh_savexmm %xmm8, 0
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    vmovss {{.*#+}} xmm8 = mem[0],zero,zero,zero
; WIN64-NEXT:    vaddss %xmm0, %xmm8, %xmm0
; WIN64-NEXT:    callq test_argRetFloat
; WIN64-NEXT:    vaddss %xmm0, %xmm8, %xmm0
; WIN64-NEXT:    vmovaps (%rsp), %xmm8 # 16-byte Reload
; WIN64-NEXT:    addq $24, %rsp
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRetFloat:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    subq $24, %rsp
; LINUXOSX64-NEXT:    vmovaps %xmm8, (%rsp) # 16-byte Spill
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 32
; LINUXOSX64-NEXT:    .cfi_offset %xmm8, -32
; LINUXOSX64-NEXT:    vmovss {{.*#+}} xmm8 = mem[0],zero,zero,zero
; LINUXOSX64-NEXT:    vaddss %xmm0, %xmm8, %xmm0
; LINUXOSX64-NEXT:    callq test_argRetFloat
; LINUXOSX64-NEXT:    vaddss %xmm0, %xmm8, %xmm0
; LINUXOSX64-NEXT:    vmovaps (%rsp), %xmm8 # 16-byte Reload
; LINUXOSX64-NEXT:    addq $24, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = fadd float 1.0, %a
  %c = call x86_regcallcc float @test_argRetFloat(float %b)
  %d = fadd float 1.0, %c
  ret float %d
}

; Test regcall when receiving/returning double
define dso_local x86_regcallcc double @test_argRetDouble(double %a)  {
; X32-LABEL: test_argRetDouble:
; X32:       # %bb.0:
; X32-NEXT:    vaddsd __real@3ff0000000000000, %xmm0, %xmm0
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRetDouble:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vaddsd __real@{{.*}}(%rip), %xmm0, %xmm0
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRetDouble:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vaddsd {{.*}}(%rip), %xmm0, %xmm0
; LINUXOSX64-NEXT:    retq
  %add = fadd double %a, 1.0
  ret double %add
}

; Test regcall when passing/retrieving double
define dso_local x86_regcallcc double @test_CallargRetDouble(double %a)  {
; X32-LABEL: test_CallargRetDouble:
; X32:       # %bb.0:
; X32-NEXT:    subl $28, %esp
; X32-NEXT:    vmovups %xmm4, (%esp) # 16-byte Spill
; X32-NEXT:    vmovsd {{.*#+}} xmm4 = mem[0],zero
; X32-NEXT:    vaddsd %xmm4, %xmm0, %xmm0
; X32-NEXT:    calll _test_argRetDouble
; X32-NEXT:    vaddsd %xmm4, %xmm0, %xmm0
; X32-NEXT:    vmovups (%esp), %xmm4 # 16-byte Reload
; X32-NEXT:    addl $28, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRetDouble:
; WIN64:       # %bb.0:
; WIN64-NEXT:    subq $24, %rsp
; WIN64-NEXT:    .seh_stackalloc 24
; WIN64-NEXT:    vmovaps %xmm8, (%rsp) # 16-byte Spill
; WIN64-NEXT:    .seh_savexmm %xmm8, 0
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    vmovsd {{.*#+}} xmm8 = mem[0],zero
; WIN64-NEXT:    vaddsd %xmm0, %xmm8, %xmm0
; WIN64-NEXT:    callq test_argRetDouble
; WIN64-NEXT:    vaddsd %xmm0, %xmm8, %xmm0
; WIN64-NEXT:    vmovaps (%rsp), %xmm8 # 16-byte Reload
; WIN64-NEXT:    addq $24, %rsp
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRetDouble:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    subq $24, %rsp
; LINUXOSX64-NEXT:    vmovaps %xmm8, (%rsp) # 16-byte Spill
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 32
; LINUXOSX64-NEXT:    .cfi_offset %xmm8, -32
; LINUXOSX64-NEXT:    vmovsd {{.*#+}} xmm8 = mem[0],zero
; LINUXOSX64-NEXT:    vaddsd %xmm0, %xmm8, %xmm0
; LINUXOSX64-NEXT:    callq test_argRetDouble
; LINUXOSX64-NEXT:    vaddsd %xmm0, %xmm8, %xmm0
; LINUXOSX64-NEXT:    vmovaps (%rsp), %xmm8 # 16-byte Reload
; LINUXOSX64-NEXT:    addq $24, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = fadd double 1.0, %a
  %c = call x86_regcallcc double @test_argRetDouble(double %b)
  %d = fadd double 1.0, %c
  ret double %d
}

; Test regcall when receiving/returning long double
define dso_local x86_regcallcc x86_fp80 @test_argRetf80(x86_fp80 %a0) nounwind {
; X32-LABEL: test_argRetf80:
; X32:       # %bb.0:
; X32-NEXT:    fadd %st, %st(0)
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRetf80:
; WIN64:       # %bb.0:
; WIN64-NEXT:    fadd %st, %st(0)
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRetf80:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    fadd %st, %st(0)
; LINUXOSX64-NEXT:    retq
  %r0 = fadd x86_fp80 %a0, %a0
  ret x86_fp80 %r0
}

; Test regcall when receiving/returning long double
define dso_local x86_regcallcc double @test_argParamf80(x86_fp80 %a0) nounwind {
; X32-LABEL: test_argParamf80:
; X32:       # %bb.0:
; X32-NEXT:    pushl %ebp
; X32-NEXT:    movl %esp, %ebp
; X32-NEXT:    andl $-8, %esp
; X32-NEXT:    subl $8, %esp
; X32-NEXT:    fstpl (%esp)
; X32-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
; X32-NEXT:    movl %ebp, %esp
; X32-NEXT:    popl %ebp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argParamf80:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    fstpl (%rsp)
; WIN64-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
; WIN64-NEXT:    popq %rax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argParamf80:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    fstpl -{{[0-9]+}}(%rsp)
; LINUXOSX64-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
; LINUXOSX64-NEXT:    retq
  %r0 = fptrunc x86_fp80 %a0 to double
  ret double %r0
}

; Test regcall when passing/retrieving long double
define x86_regcallcc x86_fp80 @test_CallargRetf80(x86_fp80 %a)  {
; X32-LABEL: test_CallargRetf80:
; X32:       # %bb.0:
; X32-NEXT:    fadd %st, %st(0)
; X32-NEXT:    calll _test_argRetf80
; X32-NEXT:    fadd %st, %st(0)
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRetf80:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    fadd %st, %st(0)
; WIN64-NEXT:    callq test_argRetf80
; WIN64-NEXT:    fadd %st, %st(0)
; WIN64-NEXT:    popq %rax
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRetf80:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    fadd %st, %st(0)
; LINUXOSX64-NEXT:    callq test_argRetf80
; LINUXOSX64-NEXT:    fadd %st, %st(0)
; LINUXOSX64-NEXT:    popq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = fadd x86_fp80 %a, %a
  %c = call x86_regcallcc x86_fp80 @test_argRetf80(x86_fp80 %b)
  %d = fadd x86_fp80 %c, %c
  ret x86_fp80 %d
}

define dso_local x86_regcallcc double @test_CallargParamf80(x86_fp80 %a)  {
; X32-LABEL: test_CallargParamf80:
; X32:       # %bb.0:
; X32-NEXT:    fadd %st, %st(0)
; X32-NEXT:    calll _test_argParamf80
; X32-NEXT:    vaddsd %xmm0, %xmm0, %xmm0
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargParamf80:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    fadd %st, %st(0)
; WIN64-NEXT:    callq test_argParamf80
; WIN64-NEXT:    vaddsd %xmm0, %xmm0, %xmm0
; WIN64-NEXT:    popq %rax
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargParamf80:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    fadd %st, %st(0)
; LINUXOSX64-NEXT:    callq test_argParamf80
; LINUXOSX64-NEXT:    vaddsd %xmm0, %xmm0, %xmm0
; LINUXOSX64-NEXT:    popq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = fadd x86_fp80 %a, %a
  %c = call x86_regcallcc double @test_argParamf80(x86_fp80 %b)
  %d = fadd double %c, %c
  ret double %d
}

; Test regcall when receiving/returning pointer
define dso_local x86_regcallcc [4 x i32]* @test_argRetPointer([4 x i32]* %a)  {
; X32-LABEL: test_argRetPointer:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRetPointer:
; WIN64:       # %bb.0:
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRetPointer:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    retq
  %b = ptrtoint [4 x i32]* %a to i32
  %c = add i32 %b, 1
  %d = inttoptr i32 %c to [4 x i32]*
  ret [4 x i32]* %d
}

; Test regcall when passing/retrieving pointer
define dso_local x86_regcallcc [4 x i32]* @test_CallargRetPointer([4 x i32]* %a)  {
; X32-LABEL: test_CallargRetPointer:
; X32:       # %bb.0:
; X32-NEXT:    incl %eax
; X32-NEXT:    calll _test_argRetPointer
; X32-NEXT:    incl %eax
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRetPointer:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rax
; WIN64-NEXT:    .seh_stackalloc 8
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    callq test_argRetPointer
; WIN64-NEXT:    incl %eax
; WIN64-NEXT:    popq %rcx
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRetPointer:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rax
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 16
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    callq test_argRetPointer
; LINUXOSX64-NEXT:    incl %eax
; LINUXOSX64-NEXT:    popq %rcx
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = ptrtoint [4 x i32]* %a to i32
  %c = add i32 %b, 1
  %d = inttoptr i32 %c to [4 x i32]*
  %e = call x86_regcallcc [4 x i32]* @test_argRetPointer([4 x i32]* %d)
  %f = ptrtoint [4 x i32]* %e to i32
  %g = add i32 %f, 1
  %h = inttoptr i32 %g to [4 x i32]*
  ret [4 x i32]* %h
}

; Test regcall when receiving/returning 128 bit vector
define dso_local x86_regcallcc <4 x i32> @test_argRet128Vector(<4 x i1> %x, <4 x i32> %a, <4 x i32> %b)  {
; X32-LABEL: test_argRet128Vector:
; X32:       # %bb.0:
; X32-NEXT:    vpslld $31, %xmm0, %xmm0
; X32-NEXT:    vpmovd2m %xmm0, %k1
; X32-NEXT:    vpblendmd %xmm1, %xmm2, %xmm0 {%k1}
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRet128Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vpslld $31, %xmm0, %xmm0
; WIN64-NEXT:    vpmovd2m %xmm0, %k1
; WIN64-NEXT:    vpblendmd %xmm1, %xmm2, %xmm0 {%k1}
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRet128Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vpslld $31, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vpmovd2m %xmm0, %k1
; LINUXOSX64-NEXT:    vpblendmd %xmm1, %xmm2, %xmm0 {%k1}
; LINUXOSX64-NEXT:    retq
  %d = select <4 x i1> %x, <4 x i32> %a, <4 x i32> %b
  ret <4 x i32> %d
}

; Test regcall when passing/retrieving 128 bit vector
define dso_local x86_regcallcc <4 x i32> @test_CallargRet128Vector(<4 x i1> %x, <4 x i32> %a)  {
; X32-LABEL: test_CallargRet128Vector:
; X32:       # %bb.0:
; X32-NEXT:    subl $44, %esp
; X32-NEXT:    vmovups %xmm4, {{[-0-9]+}}(%e{{[sb]}}p) # 16-byte Spill
; X32-NEXT:    vmovdqa %xmm1, %xmm4
; X32-NEXT:    vpslld $31, %xmm0, %xmm1
; X32-NEXT:    vpmovd2m %xmm1, %k1
; X32-NEXT:    kmovw %k1, {{[-0-9]+}}(%e{{[sb]}}p) # 2-byte Spill
; X32-NEXT:    vmovdqa %xmm4, %xmm1
; X32-NEXT:    vmovdqa %xmm4, %xmm2
; X32-NEXT:    calll _test_argRet128Vector
; X32-NEXT:    kmovw {{[-0-9]+}}(%e{{[sb]}}p), %k1 # 2-byte Reload
; X32-NEXT:    vmovdqa32 %xmm4, %xmm0 {%k1}
; X32-NEXT:    vmovups {{[-0-9]+}}(%e{{[sb]}}p), %xmm4 # 16-byte Reload
; X32-NEXT:    addl $44, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRet128Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    subq $40, %rsp
; WIN64-NEXT:    .seh_stackalloc 40
; WIN64-NEXT:    vmovaps %xmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; WIN64-NEXT:    .seh_savexmm %xmm8, 16
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    vmovdqa %xmm1, %xmm8
; WIN64-NEXT:    vpslld $31, %xmm0, %xmm1
; WIN64-NEXT:    vpmovd2m %xmm1, %k1
; WIN64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; WIN64-NEXT:    vmovdqa %xmm8, %xmm1
; WIN64-NEXT:    vmovdqa %xmm8, %xmm2
; WIN64-NEXT:    callq test_argRet128Vector
; WIN64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; WIN64-NEXT:    vmovdqa32 %xmm8, %xmm0 {%k1}
; WIN64-NEXT:    vmovaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
; WIN64-NEXT:    addq $40, %rsp
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRet128Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    subq $40, %rsp
; LINUXOSX64-NEXT:    vmovaps %xmm8, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 48
; LINUXOSX64-NEXT:    .cfi_offset %xmm8, -32
; LINUXOSX64-NEXT:    vmovdqa %xmm1, %xmm8
; LINUXOSX64-NEXT:    vpslld $31, %xmm0, %xmm1
; LINUXOSX64-NEXT:    vpmovd2m %xmm1, %k1
; LINUXOSX64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; LINUXOSX64-NEXT:    vmovdqa %xmm8, %xmm1
; LINUXOSX64-NEXT:    vmovdqa %xmm8, %xmm2
; LINUXOSX64-NEXT:    callq test_argRet128Vector
; LINUXOSX64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; LINUXOSX64-NEXT:    vmovdqa32 %xmm8, %xmm0 {%k1}
; LINUXOSX64-NEXT:    vmovaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm8 # 16-byte Reload
; LINUXOSX64-NEXT:    addq $40, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = call x86_regcallcc <4 x i32> @test_argRet128Vector(<4 x i1> %x, <4 x i32> %a, <4 x i32> %a)
  %c = select <4 x i1> %x, <4 x i32> %a, <4 x i32> %b
  ret <4 x i32> %c
}

; Test regcall when receiving/returning 256 bit vector
define dso_local x86_regcallcc <8 x i32> @test_argRet256Vector(<8 x i1> %x, <8 x i32> %a, <8 x i32> %b)  {
; X32-LABEL: test_argRet256Vector:
; X32:       # %bb.0:
; X32-NEXT:    kmovd %eax, %k1
; X32-NEXT:    vpblendmd %ymm0, %ymm1, %ymm0 {%k1}
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRet256Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    kmovd %eax, %k1
; WIN64-NEXT:    vpblendmd %ymm0, %ymm1, %ymm0 {%k1}
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRet256Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    kmovd %eax, %k1
; LINUXOSX64-NEXT:    vpblendmd %ymm0, %ymm1, %ymm0 {%k1}
; LINUXOSX64-NEXT:    retq
  %d = select <8 x i1> %x, <8 x i32> %a, <8 x i32> %b
  ret <8 x i32> %d
}

; Test regcall when passing/retrieving 256 bit vector
define dso_local x86_regcallcc <8 x i32> @test_CallargRet256Vector(<8 x i1> %x, <8 x i32> %a)  {
; X32-LABEL: test_CallargRet256Vector:
; X32:       # %bb.0:
; X32-NEXT:    subl $92, %esp
; X32-NEXT:    vmovdqu %ymm0, {{[-0-9]+}}(%e{{[sb]}}p) # 32-byte Spill
; X32-NEXT:    kmovd %eax, %k1
; X32-NEXT:    kmovw %k1, {{[-0-9]+}}(%e{{[sb]}}p) # 2-byte Spill
; X32-NEXT:    vmovdqa %ymm0, %ymm1
; X32-NEXT:    calll _test_argRet256Vector
; X32-NEXT:    kmovw {{[-0-9]+}}(%e{{[sb]}}p), %k1 # 2-byte Reload
; X32-NEXT:    vmovdqu {{[-0-9]+}}(%e{{[sb]}}p), %ymm1 # 32-byte Reload
; X32-NEXT:    vmovdqa32 %ymm1, %ymm0 {%k1}
; X32-NEXT:    addl $92, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRet256Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    subq $88, %rsp
; WIN64-NEXT:    .seh_stackalloc 88
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; WIN64-NEXT:    kmovd %eax, %k1
; WIN64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; WIN64-NEXT:    vmovdqa %ymm0, %ymm1
; WIN64-NEXT:    callq test_argRet256Vector
; WIN64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; WIN64-NEXT:    vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; WIN64-NEXT:    vmovdqa32 %ymm1, %ymm0 {%k1}
; WIN64-NEXT:    addq $88, %rsp
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRet256Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    subq $88, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 96
; LINUXOSX64-NEXT:    vmovdqu %ymm0, {{[-0-9]+}}(%r{{[sb]}}p) # 32-byte Spill
; LINUXOSX64-NEXT:    kmovd %eax, %k1
; LINUXOSX64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; LINUXOSX64-NEXT:    vmovdqa %ymm0, %ymm1
; LINUXOSX64-NEXT:    callq test_argRet256Vector
; LINUXOSX64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; LINUXOSX64-NEXT:    vmovdqu {{[-0-9]+}}(%r{{[sb]}}p), %ymm1 # 32-byte Reload
; LINUXOSX64-NEXT:    vmovdqa32 %ymm1, %ymm0 {%k1}
; LINUXOSX64-NEXT:    addq $88, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = call x86_regcallcc <8 x i32> @test_argRet256Vector(<8 x i1> %x, <8 x i32> %a, <8 x i32> %a)
  %c = select <8 x i1> %x, <8 x i32> %a, <8 x i32> %b
  ret <8 x i32> %c
}

; Test regcall when receiving/returning 512 bit vector
define dso_local x86_regcallcc <16 x i32> @test_argRet512Vector(<16 x i1> %x, <16 x i32> %a, <16 x i32> %b)  {
; X32-LABEL: test_argRet512Vector:
; X32:       # %bb.0:
; X32-NEXT:    kmovd %eax, %k1
; X32-NEXT:    vpblendmd %zmm0, %zmm1, %zmm0 {%k1}
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRet512Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    kmovd %eax, %k1
; WIN64-NEXT:    vpblendmd %zmm0, %zmm1, %zmm0 {%k1}
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRet512Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    kmovd %eax, %k1
; LINUXOSX64-NEXT:    vpblendmd %zmm0, %zmm1, %zmm0 {%k1}
; LINUXOSX64-NEXT:    retq
  %d = select <16 x i1> %x, <16 x i32> %a, <16 x i32> %b
  ret <16 x i32> %d
}

; Test regcall when passing/retrieving 512 bit vector
define dso_local x86_regcallcc <16 x i32> @test_CallargRet512Vector(<16 x i1> %x, <16 x i32> %a)  {
; X32-LABEL: test_CallargRet512Vector:
; X32:       # %bb.0:
; X32-NEXT:    subl $188, %esp
; X32-NEXT:    vmovdqu64 %zmm0, {{[-0-9]+}}(%e{{[sb]}}p) # 64-byte Spill
; X32-NEXT:    kmovd %eax, %k1
; X32-NEXT:    kmovw %k1, {{[-0-9]+}}(%e{{[sb]}}p) # 2-byte Spill
; X32-NEXT:    vmovdqa64 %zmm0, %zmm1
; X32-NEXT:    calll _test_argRet512Vector
; X32-NEXT:    kmovw {{[-0-9]+}}(%e{{[sb]}}p), %k1 # 2-byte Reload
; X32-NEXT:    vmovdqu64 {{[-0-9]+}}(%e{{[sb]}}p), %zmm1 # 64-byte Reload
; X32-NEXT:    vmovdqa32 %zmm1, %zmm0 {%k1}
; X32-NEXT:    addl $188, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: test_CallargRet512Vector:
; WIN64:       # %bb.0:
; WIN64-NEXT:    subq $184, %rsp
; WIN64-NEXT:    .seh_stackalloc 184
; WIN64-NEXT:    .seh_endprologue
; WIN64-NEXT:    vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; WIN64-NEXT:    kmovd %eax, %k1
; WIN64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; WIN64-NEXT:    vmovdqa64 %zmm0, %zmm1
; WIN64-NEXT:    callq test_argRet512Vector
; WIN64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; WIN64-NEXT:    vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; WIN64-NEXT:    vmovdqa32 %zmm1, %zmm0 {%k1}
; WIN64-NEXT:    addq $184, %rsp
; WIN64-NEXT:    retq
; WIN64-NEXT:    .seh_endproc
;
; LINUXOSX64-LABEL: test_CallargRet512Vector:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    subq $184, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 192
; LINUXOSX64-NEXT:    vmovdqu64 %zmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 64-byte Spill
; LINUXOSX64-NEXT:    kmovd %eax, %k1
; LINUXOSX64-NEXT:    kmovw %k1, {{[-0-9]+}}(%r{{[sb]}}p) # 2-byte Spill
; LINUXOSX64-NEXT:    vmovdqa64 %zmm0, %zmm1
; LINUXOSX64-NEXT:    callq test_argRet512Vector
; LINUXOSX64-NEXT:    kmovw {{[-0-9]+}}(%r{{[sb]}}p), %k1 # 2-byte Reload
; LINUXOSX64-NEXT:    vmovdqu64 {{[-0-9]+}}(%r{{[sb]}}p), %zmm1 # 64-byte Reload
; LINUXOSX64-NEXT:    vmovdqa32 %zmm1, %zmm0 {%k1}
; LINUXOSX64-NEXT:    addq $184, %rsp
; LINUXOSX64-NEXT:    .cfi_def_cfa_offset 8
; LINUXOSX64-NEXT:    retq
  %b = call x86_regcallcc <16 x i32> @test_argRet512Vector(<16 x i1> %x, <16 x i32> %a, <16 x i32> %a)
  %c = select <16 x i1> %x, <16 x i32> %a, <16 x i32> %b
  ret <16 x i32> %c
}

; Test regcall when running multiple input parameters - callee saved xmms
define dso_local x86_regcallcc <32 x float> @testf32_inp(<32 x float> %a, <32 x float> %b, <32 x float> %c) nounwind {
; X32-LABEL: testf32_inp:
; X32:       # %bb.0:
; X32-NEXT:    subl $44, %esp
; X32-NEXT:    vmovups %xmm7, {{[-0-9]+}}(%e{{[sb]}}p) # 16-byte Spill
; X32-NEXT:    vmovups %xmm6, (%esp) # 16-byte Spill
; X32-NEXT:    vaddps %zmm2, %zmm0, %zmm6
; X32-NEXT:    vaddps %zmm3, %zmm1, %zmm7
; X32-NEXT:    vmulps %zmm2, %zmm0, %zmm0
; X32-NEXT:    vsubps %zmm0, %zmm6, %zmm0
; X32-NEXT:    vmulps %zmm3, %zmm1, %zmm1
; X32-NEXT:    vsubps %zmm1, %zmm7, %zmm1
; X32-NEXT:    vaddps %zmm4, %zmm0, %zmm0
; X32-NEXT:    vaddps %zmm5, %zmm1, %zmm1
; X32-NEXT:    vmovups (%esp), %xmm6 # 16-byte Reload
; X32-NEXT:    vmovups {{[-0-9]+}}(%e{{[sb]}}p), %xmm7 # 16-byte Reload
; X32-NEXT:    addl $44, %esp
; X32-NEXT:    retl
;
; WIN64-LABEL: testf32_inp:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vaddps %zmm2, %zmm0, %zmm6
; WIN64-NEXT:    vaddps %zmm3, %zmm1, %zmm7
; WIN64-NEXT:    vmulps %zmm2, %zmm0, %zmm0
; WIN64-NEXT:    vsubps %zmm0, %zmm6, %zmm0
; WIN64-NEXT:    vmulps %zmm3, %zmm1, %zmm1
; WIN64-NEXT:    vsubps %zmm1, %zmm7, %zmm1
; WIN64-NEXT:    vaddps %zmm4, %zmm0, %zmm0
; WIN64-NEXT:    vaddps %zmm5, %zmm1, %zmm1
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: testf32_inp:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vaddps %zmm2, %zmm0, %zmm6
; LINUXOSX64-NEXT:    vaddps %zmm3, %zmm1, %zmm7
; LINUXOSX64-NEXT:    vmulps %zmm2, %zmm0, %zmm0
; LINUXOSX64-NEXT:    vsubps %zmm0, %zmm6, %zmm0
; LINUXOSX64-NEXT:    vmulps %zmm3, %zmm1, %zmm1
; LINUXOSX64-NEXT:    vsubps %zmm1, %zmm7, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm4, %zmm0, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm5, %zmm1, %zmm1
; LINUXOSX64-NEXT:    retq
  %x1 = fadd <32 x float> %a, %b
  %x2 = fmul <32 x float> %a, %b
  %x3 = fsub <32 x float> %x1, %x2
  %x4 = fadd <32 x float> %x3, %c
  ret <32 x float> %x4
}

; Test regcall when running multiple input parameters - callee saved GPRs
define dso_local x86_regcallcc i32 @testi32_inp(i32 %a1, i32 %a2, i32 %a3, i32 %a4, i32 %a5, i32 %a6, i32 %b1, i32 %b2, i32 %b3, i32 %b4, i32 %b5, i32 %b6) nounwind {
; X32-LABEL: testi32_inp:
; X32:       # %bb.0:
; X32-NEXT:    pushl %ebp
; X32-NEXT:    pushl %ebx
; X32-NEXT:    subl $20, %esp
; X32-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; X32-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; X32-NEXT:    movl %edx, (%esp) # 4-byte Spill
; X32-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; X32-NEXT:    movl %eax, %ebx
; X32-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
; X32-NEXT:    subl %ecx, %ebx
; X32-NEXT:    movl {{[0-9]+}}(%esp), %esi
; X32-NEXT:    movl %esi, %ebp
; X32-NEXT:    subl {{[0-9]+}}(%esp), %ebp
; X32-NEXT:    imull %ebp, %ebx
; X32-NEXT:    movl %edx, %ebp
; X32-NEXT:    subl %edi, %ebp
; X32-NEXT:    movl {{[0-9]+}}(%esp), %edx
; X32-NEXT:    movl %edx, %ecx
; X32-NEXT:    subl {{[0-9]+}}(%esp), %ecx
; X32-NEXT:    imull %ebp, %ecx
; X32-NEXT:    addl %ecx, %ebx
; X32-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
; X32-NEXT:    movl %edi, %ebp
; X32-NEXT:    subl {{[0-9]+}}(%esp), %ebp
; X32-NEXT:    movl {{[0-9]+}}(%esp), %ecx
; X32-NEXT:    movl %ecx, %eax
; X32-NEXT:    subl {{[0-9]+}}(%esp), %eax
; X32-NEXT:    imull %ebp, %eax
; X32-NEXT:    addl %eax, %ebx
; X32-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
; X32-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
; X32-NEXT:    movl (%esp), %ebp # 4-byte Reload
; X32-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %ebp # 4-byte Folded Reload
; X32-NEXT:    addl {{[0-9]+}}(%esp), %edi
; X32-NEXT:    addl {{[0-9]+}}(%esp), %esi
; X32-NEXT:    imull %eax, %esi
; X32-NEXT:    addl {{[0-9]+}}(%esp), %edx
; X32-NEXT:    imull %ebp, %edx
; X32-NEXT:    addl %esi, %edx
; X32-NEXT:    addl {{[0-9]+}}(%esp), %ecx
; X32-NEXT:    imull %edi, %ecx
; X32-NEXT:    addl %edx, %ecx
; X32-NEXT:    addl %ecx, %ebx
; X32-NEXT:    movl %ebx, %eax
; X32-NEXT:    addl $20, %esp
; X32-NEXT:    popl %ebx
; X32-NEXT:    popl %ebp
; X32-NEXT:    retl
;
; WIN64-LABEL: testi32_inp:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %r13
; WIN64-NEXT:    pushq %rbp
; WIN64-NEXT:    pushq %rbx
; WIN64-NEXT:    movl %eax, %r13d
; WIN64-NEXT:    subl %ecx, %eax
; WIN64-NEXT:    movl %edx, %ebp
; WIN64-NEXT:    subl %edi, %ebp
; WIN64-NEXT:    movl %r9d, %ebx
; WIN64-NEXT:    subl %r10d, %ebx
; WIN64-NEXT:    imull %ebx, %eax
; WIN64-NEXT:    movl %r11d, %ebx
; WIN64-NEXT:    subl %r12d, %ebx
; WIN64-NEXT:    imull %ebp, %ebx
; WIN64-NEXT:    movl %esi, %ebp
; WIN64-NEXT:    subl %r8d, %ebp
; WIN64-NEXT:    addl %ebx, %eax
; WIN64-NEXT:    movl %r14d, %ebx
; WIN64-NEXT:    subl %r15d, %ebx
; WIN64-NEXT:    imull %ebp, %ebx
; WIN64-NEXT:    addl %ebx, %eax
; WIN64-NEXT:    addl %ecx, %r13d
; WIN64-NEXT:    addl %edi, %edx
; WIN64-NEXT:    addl %r8d, %esi
; WIN64-NEXT:    addl %r10d, %r9d
; WIN64-NEXT:    imull %r13d, %r9d
; WIN64-NEXT:    addl %r12d, %r11d
; WIN64-NEXT:    imull %edx, %r11d
; WIN64-NEXT:    addl %r9d, %r11d
; WIN64-NEXT:    addl %r15d, %r14d
; WIN64-NEXT:    imull %esi, %r14d
; WIN64-NEXT:    addl %r11d, %r14d
; WIN64-NEXT:    addl %r14d, %eax
; WIN64-NEXT:    popq %rbx
; WIN64-NEXT:    popq %rbp
; WIN64-NEXT:    popq %r13
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: testi32_inp:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rbp
; LINUXOSX64-NEXT:    pushq %rbx
; LINUXOSX64-NEXT:    movl %eax, %r10d
; LINUXOSX64-NEXT:    movl {{[0-9]+}}(%rsp), %r11d
; LINUXOSX64-NEXT:    subl %ecx, %eax
; LINUXOSX64-NEXT:    movl %edx, %ebx
; LINUXOSX64-NEXT:    subl %edi, %ebx
; LINUXOSX64-NEXT:    movl %r9d, %ebp
; LINUXOSX64-NEXT:    subl %r12d, %ebp
; LINUXOSX64-NEXT:    imull %ebp, %eax
; LINUXOSX64-NEXT:    movl %r13d, %ebp
; LINUXOSX64-NEXT:    subl %r14d, %ebp
; LINUXOSX64-NEXT:    imull %ebx, %ebp
; LINUXOSX64-NEXT:    movl %esi, %ebx
; LINUXOSX64-NEXT:    subl %r8d, %ebx
; LINUXOSX64-NEXT:    addl %ebp, %eax
; LINUXOSX64-NEXT:    movl %r15d, %ebp
; LINUXOSX64-NEXT:    subl %r11d, %ebp
; LINUXOSX64-NEXT:    imull %ebx, %ebp
; LINUXOSX64-NEXT:    addl %ebp, %eax
; LINUXOSX64-NEXT:    addl %ecx, %r10d
; LINUXOSX64-NEXT:    addl %edi, %edx
; LINUXOSX64-NEXT:    addl %r8d, %esi
; LINUXOSX64-NEXT:    addl %r12d, %r9d
; LINUXOSX64-NEXT:    imull %r10d, %r9d
; LINUXOSX64-NEXT:    addl %r14d, %r13d
; LINUXOSX64-NEXT:    imull %edx, %r13d
; LINUXOSX64-NEXT:    addl %r9d, %r13d
; LINUXOSX64-NEXT:    addl %r11d, %r15d
; LINUXOSX64-NEXT:    imull %esi, %r15d
; LINUXOSX64-NEXT:    addl %r13d, %r15d
; LINUXOSX64-NEXT:    addl %r15d, %eax
; LINUXOSX64-NEXT:    popq %rbx
; LINUXOSX64-NEXT:    popq %rbp
; LINUXOSX64-NEXT:    retq
  %x1 = sub i32 %a1, %a2
  %x2 = sub i32 %a3, %a4
  %x3 = sub i32 %a5, %a6
  %y1 = sub i32 %b1, %b2
  %y2 = sub i32 %b3, %b4
  %y3 = sub i32 %b5, %b6
  %v1 = add i32 %a1, %a2
  %v2 = add i32 %a3, %a4
  %v3 = add i32 %a5, %a6
  %w1 = add i32 %b1, %b2
  %w2 = add i32 %b3, %b4
  %w3 = add i32 %b5, %b6
  %s1 = mul i32 %x1, %y1
  %s2 = mul i32 %x2, %y2
  %s3 = mul i32 %x3, %y3
  %t1 = mul i32 %v1, %w1
  %t2 = mul i32 %v2, %w2
  %t3 = mul i32 %v3, %w3
  %m1 = add i32 %s1, %s2
  %m2 = add i32 %m1, %s3
  %n1 = add i32 %t1, %t2
  %n2 = add i32 %n1, %t3
  %r1 = add i32 %m2, %n2
  ret i32 %r1
}

; Test that parameters, overflowing register capacity, are passed through the stack
define dso_local x86_regcallcc <32 x float> @testf32_stack(<32 x float> %a0, <32 x float> %b0, <32 x float> %c0, <32 x float> %a1, <32 x float> %b1, <32 x float> %c1, <32 x float> %a2, <32 x float> %b2, <32 x float> %c2) nounwind {
; X32-LABEL: testf32_stack:
; X32:       # %bb.0:
; X32-NEXT:    pushl %ebp
; X32-NEXT:    movl %esp, %ebp
; X32-NEXT:    andl $-64, %esp
; X32-NEXT:    subl $64, %esp
; X32-NEXT:    vaddps %zmm3, %zmm1, %zmm1
; X32-NEXT:    vaddps %zmm2, %zmm0, %zmm0
; X32-NEXT:    vaddps %zmm0, %zmm4, %zmm0
; X32-NEXT:    vaddps %zmm1, %zmm5, %zmm1
; X32-NEXT:    vaddps %zmm1, %zmm7, %zmm1
; X32-NEXT:    vaddps %zmm0, %zmm6, %zmm0
; X32-NEXT:    vaddps 8(%ebp), %zmm0, %zmm0
; X32-NEXT:    vaddps 72(%ebp), %zmm1, %zmm1
; X32-NEXT:    vaddps 200(%ebp), %zmm1, %zmm1
; X32-NEXT:    vaddps 136(%ebp), %zmm0, %zmm0
; X32-NEXT:    vaddps 264(%ebp), %zmm0, %zmm0
; X32-NEXT:    vaddps 328(%ebp), %zmm1, %zmm1
; X32-NEXT:    vaddps 456(%ebp), %zmm1, %zmm1
; X32-NEXT:    vaddps 392(%ebp), %zmm0, %zmm0
; X32-NEXT:    vaddps 520(%ebp), %zmm0, %zmm0
; X32-NEXT:    vaddps 584(%ebp), %zmm1, %zmm1
; X32-NEXT:    movl %ebp, %esp
; X32-NEXT:    popl %ebp
; X32-NEXT:    retl
;
; WIN64-LABEL: testf32_stack:
; WIN64:       # %bb.0:
; WIN64-NEXT:    pushq %rbp
; WIN64-NEXT:    subq $48, %rsp
; WIN64-NEXT:    leaq {{[0-9]+}}(%rsp), %rbp
; WIN64-NEXT:    andq $-64, %rsp
; WIN64-NEXT:    vaddps %zmm3, %zmm1, %zmm1
; WIN64-NEXT:    vaddps %zmm2, %zmm0, %zmm0
; WIN64-NEXT:    vaddps %zmm0, %zmm4, %zmm0
; WIN64-NEXT:    vaddps %zmm1, %zmm5, %zmm1
; WIN64-NEXT:    vaddps %zmm1, %zmm7, %zmm1
; WIN64-NEXT:    vaddps %zmm0, %zmm6, %zmm0
; WIN64-NEXT:    vaddps %zmm0, %zmm8, %zmm0
; WIN64-NEXT:    vaddps %zmm1, %zmm9, %zmm1
; WIN64-NEXT:    vaddps %zmm1, %zmm11, %zmm1
; WIN64-NEXT:    vaddps %zmm0, %zmm10, %zmm0
; WIN64-NEXT:    vaddps %zmm0, %zmm12, %zmm0
; WIN64-NEXT:    vaddps %zmm1, %zmm13, %zmm1
; WIN64-NEXT:    vaddps %zmm1, %zmm15, %zmm1
; WIN64-NEXT:    vaddps %zmm0, %zmm14, %zmm0
; WIN64-NEXT:    vaddps 16(%rbp), %zmm0, %zmm0
; WIN64-NEXT:    vaddps 80(%rbp), %zmm1, %zmm1
; WIN64-NEXT:    movq %rbp, %rsp
; WIN64-NEXT:    popq %rbp
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: testf32_stack:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    pushq %rbp
; LINUXOSX64-NEXT:    movq %rsp, %rbp
; LINUXOSX64-NEXT:    andq $-64, %rsp
; LINUXOSX64-NEXT:    subq $64, %rsp
; LINUXOSX64-NEXT:    vaddps %zmm3, %zmm1, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm2, %zmm0, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm4, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm5, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm7, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm6, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm8, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm9, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm11, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm10, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm12, %zmm0
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm13, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm1, %zmm15, %zmm1
; LINUXOSX64-NEXT:    vaddps %zmm0, %zmm14, %zmm0
; LINUXOSX64-NEXT:    vaddps 16(%rbp), %zmm0, %zmm0
; LINUXOSX64-NEXT:    vaddps 80(%rbp), %zmm1, %zmm1
; LINUXOSX64-NEXT:    movq %rbp, %rsp
; LINUXOSX64-NEXT:    popq %rbp
; LINUXOSX64-NEXT:    retq
  %x1 = fadd <32 x float> %a0, %b0
  %x2 = fadd <32 x float> %c0, %x1
  %x3 = fadd <32 x float> %a1, %x2
  %x4 = fadd <32 x float> %b1, %x3
  %x5 = fadd <32 x float> %c1, %x4
  %x6 = fadd <32 x float> %a2, %x5
  %x7 = fadd <32 x float> %b2, %x6
  %x8 = fadd <32 x float> %c2, %x7
  ret <32 x float> %x8
}

; Test regcall when passing/retrieving mixed types
define dso_local x86_regcallcc i32 @test_argRetMixTypes(double, float, i8 signext, i32, i64, i16 signext, i32*) #0 {
; X32-LABEL: test_argRetMixTypes:
; X32:       # %bb.0:
; X32-NEXT:    pushl %ebx
; X32-NEXT:    movl {{[0-9]+}}(%esp), %ebx
; X32-NEXT:    vcvtss2sd %xmm1, %xmm1, %xmm1
; X32-NEXT:    vaddsd %xmm0, %xmm1, %xmm0
; X32-NEXT:    vcvtsi2sd %eax, %xmm2, %xmm1
; X32-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; X32-NEXT:    vcvtsi2sd %ecx, %xmm2, %xmm1
; X32-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; X32-NEXT:    vmovd %edx, %xmm1
; X32-NEXT:    vpinsrd $1, %edi, %xmm1, %xmm1
; X32-NEXT:    vcvtqq2pd %ymm1, %ymm1
; X32-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; X32-NEXT:    vcvtsi2sd %esi, %xmm2, %xmm1
; X32-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; X32-NEXT:    vcvtsi2sdl (%ebx), %xmm2, %xmm1
; X32-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; X32-NEXT:    vcvttsd2si %xmm0, %eax
; X32-NEXT:    popl %ebx
; X32-NEXT:    vzeroupper
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argRetMixTypes:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vcvtss2sd %xmm1, %xmm1, %xmm1
; WIN64-NEXT:    vaddsd %xmm0, %xmm1, %xmm0
; WIN64-NEXT:    vcvtsi2sd %eax, %xmm2, %xmm1
; WIN64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; WIN64-NEXT:    vcvtsi2sd %ecx, %xmm2, %xmm1
; WIN64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; WIN64-NEXT:    vcvtsi2sd %rdx, %xmm2, %xmm1
; WIN64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; WIN64-NEXT:    vcvtsi2sd %edi, %xmm2, %xmm1
; WIN64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; WIN64-NEXT:    vcvtsi2sdl (%rsi), %xmm2, %xmm1
; WIN64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; WIN64-NEXT:    vcvttsd2si %xmm0, %eax
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argRetMixTypes:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vcvtss2sd %xmm1, %xmm1, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm0, %xmm1, %xmm0
; LINUXOSX64-NEXT:    vcvtsi2sd %eax, %xmm2, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vcvtsi2sd %ecx, %xmm2, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vcvtsi2sd %rdx, %xmm2, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vcvtsi2sd %edi, %xmm2, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vcvtsi2sdl (%rsi), %xmm2, %xmm1
; LINUXOSX64-NEXT:    vaddsd %xmm1, %xmm0, %xmm0
; LINUXOSX64-NEXT:    vcvttsd2si %xmm0, %eax
; LINUXOSX64-NEXT:    retq
  %8 = fpext float %1 to double
  %9 = fadd double %8, %0
  %10 = sitofp i8 %2 to double
  %11 = fadd double %9, %10
  %12 = sitofp i32 %3 to double
  %13 = fadd double %11, %12
  %14 = sitofp i64 %4 to double
  %15 = fadd double %13, %14
  %16 = sitofp i16 %5 to double
  %17 = fadd double %15, %16
  %18 = load i32, i32* %6, align 4
  %19 = sitofp i32 %18 to double
  %20 = fadd double %17, %19
  %21 = fptosi double %20 to i32
  ret i32 %21
}

%struct.complex = type { float, double, i32, i8, i64}

define x86_regcallcc %struct.complex @test_argMultiRet(float, double, i32, i8, i64) local_unnamed_addr #0 {
; X32-LABEL: test_argMultiRet:
; X32:       # %bb.0:
; X32-NEXT:    vaddsd __real@4014000000000000, %xmm1, %xmm1
; X32-NEXT:    movl $4, %eax
; X32-NEXT:    movb $7, %cl
; X32-NEXT:    movl $999, %edx # imm = 0x3E7
; X32-NEXT:    xorl %edi, %edi
; X32-NEXT:    retl
;
; WIN64-LABEL: test_argMultiRet:
; WIN64:       # %bb.0:
; WIN64-NEXT:    vaddsd __real@{{.*}}(%rip), %xmm1, %xmm1
; WIN64-NEXT:    movl $999, %edx # imm = 0x3E7
; WIN64-NEXT:    movl $4, %eax
; WIN64-NEXT:    movb $7, %cl
; WIN64-NEXT:    retq
;
; LINUXOSX64-LABEL: test_argMultiRet:
; LINUXOSX64:       # %bb.0:
; LINUXOSX64-NEXT:    vaddsd {{.*}}(%rip), %xmm1, %xmm1
; LINUXOSX64-NEXT:    movl $999, %edx # imm = 0x3E7
; LINUXOSX64-NEXT:    movl $4, %eax
; LINUXOSX64-NEXT:    movb $7, %cl
; LINUXOSX64-NEXT:    retq
  %6 = fadd double %1, 5.000000e+00
  %7 = insertvalue %struct.complex undef, float %0, 0
  %8 = insertvalue %struct.complex %7, double %6, 1
  %9 = insertvalue %struct.complex %8, i32 4, 2
  %10 = insertvalue %struct.complex %9, i8 7, 3
  %11 = insertvalue %struct.complex %10, i64 999, 4
  ret %struct.complex %11
}