.code

asm01 PROC a:dword, b:dword, c:dword
; xmm0 <- a
; xmm1 <- b
; xmm2 <- c
local maxi:qword
   movss a, xmm0
   movss b, xmm1
   movss c, xmm2

   fld a            ;a
   fld b            ;b, a
   fld c            ;c, b, a
   fcomi st, st(1)  ;c, b , a  (c>b)
   ja CC
   fxch st(1)       ;b, c, a
CC:
   fcomi st, st(2)  ;max(b, c), c|b, a
   ja KONIEC
   fxch st(2)       ;a, max(b,c), b|c
KONIEC:
    fstp maxi       ; a|b|c, a|b|c 
    movsd xmm0, maxi
    fstp st         ;a|b|c
    fstp st         ;<empty>

    ret
asm01 ENDP

asm02 PROC a:qword, i:dword, b:qword, j:dword, c:qword, k:dword
; xmm0 <- a
;rdx <- i
;xmm2 <- b
;r9 <- j
;stack <- c
;stack <- k
    
    movsd a, xmm0
    mov i, edx
    movsd b, xmm2
    mov j, r9d

    fld a               ;a
    fld st              ;a, a
    fmul                ;a^2
    fld st              ;a^2, a^2
    fmul                ;a^4
    fild i              ;i, a^4
    mov i, 16
    fild i              ;16, i, a^4
    fmul                ;16i, a^4
    fmul                ;16ia^4

    fld b               ;b, 16ia^4
    fld st              ;b, b, 16ia^4
    fld st              ;b, b, b, 16ia^4
    fmul                ;bb, b, 16ia^4
    fmul                ;b^3, 16ia^4
    fild j              ;j, b^3, 16ia^4
    mov i, 32           
    fild i              ;32, j, b^3, 16ia^4
    fmul                ;32j, b^3, 16ia^4
    fmul                ;32jb^3, 16ia^4
    fadd                ;32jb^3 + 16ia^4

    fld c               ;c, 32jb^3 + 16ia^4
    fld st              ;c, c, 32jb^3 + 16ia^4
    fmul                ;c^2, 32jb^3 + 16ia^4
    fild k              ;k, c^2, 32jb^3 + 16ia^4
    mov i, 64
    fild i              ;64, k, c^2, 32jb^3 + 16ia^4
    fmul                ;64k, c^2, 32jb^3 + 16ia^4
    fmul                ;64kc^2, 32jb^3 + 16ia^4
    
    mov i, 128
    fild i              ;128, 64kc^2, 32jb^3 + 16ia^4
    fldl2t              ;log2(10), 128, 64kc^2, 32jb^3 + 16ia^4
    fmul                ;log2(10)*128, 64kc^2, 32jb^3 + 16ia^4
    fsub                ;64kc^2 -log2(10)*128, 32jb^3 + 16ia^4
    fdiv                ;32jb^3 + 16ia^4/64kc^2 -log2(10)*128 

    fstp a
    movsd xmm0, a

    ret
asm02 ENDP

asm03 PROC a:dword, b:dword, x:dword
;xmm0 <- a
;xmm1 <- b
;xmm2 <- x
local sto:qword

    movss a, xmm0
    movss b, xmm1
    movss x, xmm2
    
    mov sto, 180

    fild sto            ;180
    fldpi               ;pi, 180
    fdivr               ;pi/180
    fld st              ;pi/180, pi/180

    fld x               ;x, pi/180, pi/180
    fld b               ;b, x, pi/180, pi/180
    fmul                ;bx, pi/180, pi/180
    fmul                ;bx*pi/180, pi/180
    fcos                ;cos(bx*pi/180), pi/180
    fld st              ;cos(bx*pi/180), cos(bx*pi/180), pi/180
    fmul                ;cos^2(bx*pi/180), pi/180
    fxch st(1)          ;pi/180, cos^2(bx*pi/180)

    fld x               ;x, pi/180, cos^2(bx*pi/180)
    fld a               ;a, x, pi/180, cos^2(bx*pi/180)
    fmul                ;ax, pi/180, cos^2(bx*pi/180)
    fmul                ;axpi/180, cos^2(bx*pi/180)
    fld st              ;axpi/180, axpi/180, cos^2(bx*pi/180)
    fmul                ;(axpi/180)^2, cos^2(bx*pi/180)
    fsin                ;sin(axpi/180)^2, cos^2(bx*pi/180)
    fsubr               ;sin(axpi/180)^2 - cos^2(bx*pi/180)

    fstp sto

    movsd xmm0, sto

    ret
asm03 ENDP

asm04 PROC x:qword
;xmm0 <- x
    
    movsd x, xmm0

    fld x               ;x
    fldz                ;0, x
    fcomip st, st(1)    ;x
    jbe WIEKSZY

    fldln2              ;ln2, x
    fmul                ;x*ln2
    jmp KONIEC
WIEKSZY:
    fld1                ;1, x
    fxch st(1)          ;x, 1
    fyl2xp1             ;1*log(x+1)

KONIEC:
    
    fstp x
    movsd xmm0, x
    
    ret
asm04 ENDP

asm05 PROC x:qword, y:qword, sizee:dword
; rcx <- address of y vector
; rdx <- address of x vector
; r8d <- size
local help: qword
    
    movsxd r8, r8d
    mov help, 16

PETLA:
    fld dword ptr[rdx + 4* r8 - 4]      ;x[i]
    fld st                              ;x[i], x[i]
    fld st                              ;x[i], x[i], x[i]
    fmul                                ;x[i]^2, x[i]
    fsubr                               ;x[i]^2 - x[i]
    fld1                                ;1, x[i]^2 - x[i]
    fadd                                ;x[i]^2 - x[i] + 1 
    fsqrt                               ;sqrt(x[i]^2 - x[i] + 1 )
    fild help                           ;16, sqrt(x[i]^2 - x[i] + 1 )
    fmul                                ;16*sqrt(x[i]^2 - x[i] + 1 )
    fstp dword ptr[rcx + 4*r8 - 4]      ;<empty>
    
    dec r8
    jnz PETLA

    ret
asm05 ENDP

asm06 PROC a: qword, x:qword
;xmm0 <- a
;xmm1 <- x

    movsd a, xmm0
    movsd x, xmm1

    fld1                ;1
    fld a               ;a, 1
    fyl2x               ;1*log2(a)
    fld1                ;1, log2(a)
    fld x               ;x, 1, log2(a)
    fyl2x               ;1*log2(x), log2(a)
    fdiv                ;log2(a)/ log2(x)

    fstp a
    movsd xmm0, a

    ret
asm06 ENDP

asm07 PROC
    ret
asm07 ENDP

asm08 PROC x:qword, y:qword, a:dword, sizee:dword
;rcx <- float* x
;rdx <- float* y
;xmm2 <- float a
;r9d <- size

    movsxd r9, r9d
    shl r9, 2
    vbroadcastss ymm1, xmm2
PETLA:
    sub r9, 32

    vmovups ymm0, ymmword ptr[rcx + r9]
    vmulps ymm0, ymm0, ymm0
    vaddps ymm0, ymm0, ymm1
    vmovups ymmword ptr[rdx + r9], ymm0
    jnz PETLA
    ;   vmulps ymm0, ymm0, ymm0 

    ret
asm08 ENDP

asm09 PROC a:qword, b:qword, c:qword, d:qword, x:qword
; xmm0 <- a
; xmm1 <- b
; xmm2 <- c
; xmm3 <- d
; stack <- x
; y = 15ax^3 + 10bx^2 + 5cx + d

    mov rax, 5
    vcvtsi2sd xmm6, xmm6, rax       ;xmm6 = 5
    vmovsd xmm5, x                  ;xmm5 = x
    
    vmulsd xmm2, xmm2, xmm5         ;cx
    vmulsd xmm2, xmm2, xmm6         ;5cx
    vaddsd xmm2, xmm2, xmm3         ;5cx+d

    mov rax, 10
    vcvtsi2sd xmm6, xmm6, rax       ;xmm6 = 10
    vmulsd xmm1, xmm1, xmm5         ;bx
    vmulsd xmm1, xmm1, xmm5         ;bx^2
    vmulsd xmm1, xmm1, xmm6         ;10*bx^2
    vaddsd xmm1, xmm1, xmm2         ;10*bx^2 + 5cx+d

    mov rax, 15
    vcvtsi2sd xmm6, xmm6, rax       ;xmm6 = 15
    vmulsd xmm0, xmm0, xmm5         ;ax
    vmulsd xmm0, xmm0, xmm5         ;ax^2
    vmulsd xmm0, xmm0, xmm5         ;ax^3
    vmulsd xmm0, xmm0, xmm6         ;15ax^3
    vaddsd xmm0, xmm0, xmm1         ;15ax^3, 10*bx^2 + 5cx+d


    ret
asm09 ENDP

END