# stack32 out_stack

# int32 out

# stack32 u_stack

# int32 u

# int32 eax

# int32 ebx

# int32 esi

# int32 edi

# int32 ebp

# float80 out0

# float80 out1

# float80 out2

# float80 out3

# float80 out4

# float80 carry0

# float80 carry1

# float80 carry2

# float80 carry3

# float80 carry4

# float80 u22

# float80 u32

# float80 u42

# stack64 u22_stack

# stack64 u32_stack

# float80 u0u0

# float80 u0u1

# float80 u0u2

# float80 u0u3

# float80 u1u1

# float80 u1u2

# float80 u1u3

# float80 u2u2

# float80 u2u3

# float80 u3u3

# enter gaudry_pm_square
.text
.p2align 5
.globl _gaudry_pm_square
.globl gaudry_pm_square
_gaudry_pm_square:
gaudry_pm_square:
mov %esp,%eax
and $31,%eax
add $32,%eax
sub %eax,%esp

# input out_stack

# input u_stack

# caller eax

# caller ebx

# caller esi

# caller edi

# caller ebp

#   u = u_stack
# movl <u_stack=stack32#-2,>u=int32#3
# movl <u_stack=8(%esp,%eax),>u=%edx
movl 8(%esp,%eax),%edx

#   out = out_stack
# movl <out_stack=stack32#-1,>out=int32#2
# movl <out_stack=4(%esp,%eax),>out=%ecx
movl 4(%esp,%eax),%ecx

#   out3 = *(float64 *) (u + 32)
# fldl 32(<u=int32#3)
# fldl 32(<u=%edx)
fldl 32(%edx)

#   out3 *= out3
# fmul <out3=float80#1,<out3=float80#1
# fmul <out3=%st(0),<out3=%st(0)
fmul %st(0),%st(0)

# u22 = *(float64 *) (u + 16)
# fldl 16(<u=int32#3)
# fldl 16(<u=%edx)
fldl 16(%edx)

# u22 += u22
# fadd <u22=float80#1,<u22=float80#1
# fadd <u22=%st(0),<u22=%st(0)
fadd %st(0),%st(0)

# u32 = *(float64 *) (u + 24)
# fldl 24(<u=int32#3)
# fldl 24(<u=%edx)
fldl 24(%edx)

# u32 += u32
# fadd <u32=float80#1,<u32=float80#1
# fadd <u32=%st(0),<u32=%st(0)
fadd %st(0),%st(0)

# u42 = *(float64 *) (u + 32)
# fldl 32(<u=int32#3)
# fldl 32(<u=%edx)
fldl 32(%edx)

# u42 += u42
# fadd <u42=float80#1,<u42=float80#1
# fadd <u42=%st(0),<u42=%st(0)
fadd %st(0),%st(0)

# stacktop out3
# fxch <out3=float80#4
# fxch <out3=%st(3)
fxch %st(3)

#   out3 *= *(float64 *) &gaudry_pm_scale
fmull gaudry_pm_scale

#   u1u2 = *(float64 *) (u + 8)
# fldl 8(<u=int32#3)
# fldl 8(<u=%edx)
fldl 8(%edx)

#   u1u2 *= u22
# fmul <u22=float80#4,<u1u2=float80#1
# fmul <u22=%st(3),<u1u2=%st(0)
fmul %st(3),%st(0)

#   out3 += u1u2
# faddp <u1u2=float80#1,<out3=float80#2
# faddp <u1u2=%st(0),<out3=%st(1)
faddp %st(0),%st(1)

#   u0u3 = *(float64 *) (u + 0)
# fldl 0(<u=int32#3)
# fldl 0(<u=%edx)
fldl 0(%edx)

#   u0u3 *= u32
# fmul <u32=float80#3,<u0u3=float80#1
# fmul <u32=%st(2),<u0u3=%st(0)
fmul %st(2),%st(0)

#   out3 += u0u3
# faddp <u0u3=float80#1,<out3=float80#2
# faddp <u0u3=%st(0),<out3=%st(1)
faddp %st(0),%st(1)

#             carry3 = *(float64 *) &gaudry_pm_alpha102
fldl gaudry_pm_alpha102

#             carry3 += out3
# fadd <out3=float80#2,<carry3=float80#1
# fadd <out3=%st(1),<carry3=%st(0)
fadd %st(1),%st(0)

#             carry3 -= *(float64 *) &gaudry_pm_alpha102
fsubl gaudry_pm_alpha102

# stacktop u22
# fxch <u22=float80#4
# fxch <u22=%st(3)
fxch %st(3)

# *(float64 *) &u22_stack = u22
# fstpl >u22_stack=stack64#1
# fstpl >u22_stack=0(%esp)
fstpl 0(%esp)

# stacktop u32
# fxch <u32=float80#2
# fxch <u32=%st(1)
fxch %st(1)

# *(float64 *) &u32_stack = u32
# fstl >u32_stack=stack64#2
# fstl >u32_stack=8(%esp)
fstl 8(%esp)

#     out4 = *(float64 *) (u + 0)
# fldl 0(<u=int32#3)
# fldl 0(<u=%edx)
fldl 0(%edx)

#     out4 *= u42
# fmul <u42=float80#5,<out4=float80#1
# fmul <u42=%st(4),<out4=%st(0)
fmul %st(4),%st(0)

#     u1u3 = *(float64 *) (u + 8)
# fldl 8(<u=int32#3)
# fldl 8(<u=%edx)
fldl 8(%edx)

#     u1u3 *= u32
# fmulp <u32=float80#1,<u1u3=float80#3
# fmulp <u32=%st(0),<u1u3=%st(2)
fmulp %st(0),%st(2)

#     out4 += u1u3
# faddp <u1u3=float80#1,<out4=float80#2
# faddp <u1u3=%st(0),<out4=%st(1)
faddp %st(0),%st(1)

#     u2u2 = *(float64 *) (u + 16)
# fldl 16(<u=int32#3)
# fldl 16(<u=%edx)
fldl 16(%edx)

#     u2u2 *= u2u2
# fmul <u2u2=float80#1,<u2u2=float80#1
# fmul <u2u2=%st(0),<u2u2=%st(0)
fmul %st(0),%st(0)

#     out4 += u2u2
# faddp <u2u2=float80#1,<out4=float80#2
# faddp <u2u2=%st(0),<out4=%st(1)
faddp %st(0),%st(1)

#             out4 += carry3
# fadd <carry3=float80#3,<out4=float80#1
# fadd <carry3=%st(2),<out4=%st(0)
fadd %st(2),%st(0)

# stacktop carry3
# fxch <carry3=float80#3
# fxch <carry3=%st(2)
fxch %st(2)

#             out3 -= carry3
# fsubrp <carry3=float80#1,<out3=float80#2
# fsubrp <carry3=%st(0),<out3=%st(1)
fsubrp %st(0),%st(1)

#       out0 = *(float64 *) (u + 8)
# fldl 8(<u=int32#3)
# fldl 8(<u=%edx)
fldl 8(%edx)

#       out0 *= u42
# fmul <u42=float80#4,<out0=float80#1
# fmul <u42=%st(3),<out0=%st(0)
fmul %st(3),%st(0)

#       u2u3 = *(float64 *) (u + 16)
# fldl 16(<u=int32#3)
# fldl 16(<u=%edx)
fldl 16(%edx)

#       u2u3 *= *(float64 *) &u32_stack
# fmull <u32_stack=stack64#2
# fmull <u32_stack=8(%esp)
fmull 8(%esp)

#       out0 += u2u3
# faddp <u2u3=float80#1,<out0=float80#2
# faddp <u2u3=%st(0),<out0=%st(1)
faddp %st(0),%st(1)

#               carry4 = *(float64 *) &gaudry_pm_alpha127
fldl gaudry_pm_alpha127

#               carry4 += out4
# fadd <out4=float80#4,<carry4=float80#1
# fadd <out4=%st(3),<carry4=%st(0)
fadd %st(3),%st(0)

#               carry4 -= *(float64 *) &gaudry_pm_alpha127
fsubl gaudry_pm_alpha127

#               out0 += carry4
# fadd <carry4=float80#1,<out0=float80#2
# fadd <carry4=%st(0),<out0=%st(1)
fadd %st(0),%st(1)

#               out4 -= carry4
# fsubrp <carry4=float80#1,<out4=float80#4
# fsubrp <carry4=%st(0),<out4=%st(3)
fsubrp %st(0),%st(3)

#               out0 *= *(float64 *) &gaudry_pm_scale
fmull gaudry_pm_scale

#               u0u0 = *(float64 *) (u + 0)
# fldl 0(<u=int32#3)
# fldl 0(<u=%edx)
fldl 0(%edx)

#               u0u0 *= u0u0
# fmul <u0u0=float80#1,<u0u0=float80#1
# fmul <u0u0=%st(0),<u0u0=%st(0)
fmul %st(0),%st(0)

#               out0 += u0u0
# faddp <u0u0=float80#1,<out0=float80#2
# faddp <u0u0=%st(0),<out0=%st(1)
faddp %st(0),%st(1)

#         out1 = *(float64 *) (u + 16)
# fldl 16(<u=int32#3)
# fldl 16(<u=%edx)
fldl 16(%edx)

#         out1 *= u42
# fmul <u42=float80#5,<out1=float80#1
# fmul <u42=%st(4),<out1=%st(0)
fmul %st(4),%st(0)

#         u3u3 = *(float64 *) (u + 24)
# fldl 24(<u=int32#3)
# fldl 24(<u=%edx)
fldl 24(%edx)

#         u3u3 *= u3u3
# fmul <u3u3=float80#1,<u3u3=float80#1
# fmul <u3u3=%st(0),<u3u3=%st(0)
fmul %st(0),%st(0)

#         out1 += u3u3
# faddp <u3u3=float80#1,<out1=float80#2
# faddp <u3u3=%st(0),<out1=%st(1)
faddp %st(0),%st(1)

#         out1 *= *(float64 *) &gaudry_pm_scale
fmull gaudry_pm_scale

#         u0u1 = *(float64 *) (u + 0)
# fldl 0(<u=int32#3)
# fldl 0(<u=%edx)
fldl 0(%edx)

#         u0u1 += u0u1
# fadd <u0u1=float80#1,<u0u1=float80#1
# fadd <u0u1=%st(0),<u0u1=%st(0)
fadd %st(0),%st(0)

#         u0u1 *= *(float64 *) (u + 8)
# fmull 8(<u=int32#3)
# fmull 8(<u=%edx)
fmull 8(%edx)

#         out1 += u0u1
# faddp <u0u1=float80#1,<out1=float80#2
# faddp <u0u1=%st(0),<out1=%st(1)
faddp %st(0),%st(1)

#           out2 = *(float64 *) (u + 24)
# fldl 24(<u=int32#3)
# fldl 24(<u=%edx)
fldl 24(%edx)

#           out2 *= u42
# fmulp <u42=float80#1,<out2=float80#6
# fmulp <u42=%st(0),<out2=%st(5)
fmulp %st(0),%st(5)

# stacktop out2
# fxch <out2=float80#5
# fxch <out2=%st(4)
fxch %st(4)

#           out2 *= *(float64 *) &gaudry_pm_scale
fmull gaudry_pm_scale

#           u0u2 = *(float64 *) (u + 0)
# fldl 0(<u=int32#3)
# fldl 0(<u=%edx)
fldl 0(%edx)

#           u0u2 *= *(float64 *) &u22_stack
# fmull <u22_stack=stack64#1
# fmull <u22_stack=0(%esp)
fmull 0(%esp)

#           out2 += u0u2
# faddp <u0u2=float80#1,<out2=float80#2
# faddp <u0u2=%st(0),<out2=%st(1)
faddp %st(0),%st(1)

#           u1u1 = *(float64 *) (u + 8)
# fldl 8(<u=int32#3)
# fldl 8(<u=%edx)
fldl 8(%edx)

#           u1u1 *= u1u1
# fmul <u1u1=float80#1,<u1u1=float80#1
# fmul <u1u1=%st(0),<u1u1=%st(0)
fmul %st(0),%st(0)

#           out2 += u1u1
# faddp <u1u1=float80#1,<out2=float80#2
# faddp <u1u1=%st(0),<out2=%st(1)
faddp %st(0),%st(1)

#                 carry0 = *(float64 *) &gaudry_pm_alpha26
fldl gaudry_pm_alpha26

#                 carry0 += out0
# fadd <out0=float80#3,<carry0=float80#1
# fadd <out0=%st(2),<carry0=%st(0)
fadd %st(2),%st(0)

#                 carry0 -= *(float64 *) &gaudry_pm_alpha26
fsubl gaudry_pm_alpha26

#                 out1 += carry0
# fadd <carry0=float80#1,<out1=float80#6
# fadd <carry0=%st(0),<out1=%st(5)
fadd %st(0),%st(5)

#                 out0 -= carry0
# fsubrp <carry0=float80#1,<out0=float80#3
# fsubrp <carry0=%st(0),<out0=%st(2)
fsubrp %st(0),%st(2)

# stacktop out0
# fxch <out0=float80#2
# fxch <out0=%st(1)
fxch %st(1)

#                 *(float64 *) (out + 0) = out0
# fstpl 0(<out=int32#2)
# fstpl 0(<out=%ecx)
fstpl 0(%ecx)

#                 carry1 = *(float64 *) &gaudry_pm_alpha51
fldl gaudry_pm_alpha51

#                 carry1 += out1
# fadd <out1=float80#5,<carry1=float80#1
# fadd <out1=%st(4),<carry1=%st(0)
fadd %st(4),%st(0)

#                 carry1 -= *(float64 *) &gaudry_pm_alpha51
fsubl gaudry_pm_alpha51

#                   out2 += carry1
# fadd <carry1=float80#1,<out2=float80#2
# fadd <carry1=%st(0),<out2=%st(1)
fadd %st(0),%st(1)

#                   carry2 = *(float64 *) &gaudry_pm_alpha77
fldl gaudry_pm_alpha77

#                   carry2 += out2
# fadd <out2=float80#3,<carry2=float80#1
# fadd <out2=%st(2),<carry2=%st(0)
fadd %st(2),%st(0)

#                   carry2 -= *(float64 *) &gaudry_pm_alpha77
fsubl gaudry_pm_alpha77

#                     out3 += carry2
# fadd <carry2=float80#1,<out3=float80#4
# fadd <carry2=%st(0),<out3=%st(3)
fadd %st(0),%st(3)

#                     carry3 = *(float64 *) &gaudry_pm_alpha102
fldl gaudry_pm_alpha102

#                     carry3 += out3
# fadd <out3=float80#5,<carry3=float80#1
# fadd <out3=%st(4),<carry3=%st(0)
fadd %st(4),%st(0)

#                     carry3 -= *(float64 *) &gaudry_pm_alpha102
fsubl gaudry_pm_alpha102

# stacktop carry1
# fxch <carry1=float80#3
# fxch <carry1=%st(2)
fxch %st(2)

#                       out1 -= carry1
# fsubrp <carry1=float80#1,<out1=float80#7
# fsubrp <carry1=%st(0),<out1=%st(6)
fsubrp %st(0),%st(6)

# stacktop out1
# fxch <out1=float80#6
# fxch <out1=%st(5)
fxch %st(5)

#                       *(float64 *) (out + 8) = out1
# fstpl 8(<out=int32#2)
# fstpl 8(<out=%ecx)
fstpl 8(%ecx)

# stacktop carry2
# fxch <carry2=float80#5
# fxch <carry2=%st(4)
fxch %st(4)

#                       out2 -= carry2
# fsubrp <carry2=float80#1,<out2=float80#2
# fsubrp <carry2=%st(0),<out2=%st(1)
fsubrp %st(0),%st(1)

#                       *(float64 *) (out + 16) = out2
# fstpl 16(<out=int32#2)
# fstpl 16(<out=%ecx)
fstpl 16(%ecx)

#                       out3 -= carry3
# fsub <carry3=float80#3,<out3=float80#1
# fsub <carry3=%st(2),<out3=%st(0)
fsub %st(2),%st(0)

#                       *(float64 *) (out + 24) = out3
# fstpl 24(<out=int32#2)
# fstpl 24(<out=%ecx)
fstpl 24(%ecx)

#                       out4 += carry3
# faddp <carry3=float80#1,<out4=float80#2
# faddp <carry3=%st(0),<out4=%st(1)
faddp %st(0),%st(1)

#                       *(float64 *) (out + 32) = out4
# fstpl 32(<out=int32#2)
# fstpl 32(<out=%ecx)
fstpl 32(%ecx)

# leave
add %eax,%esp
ret