Category : Files from Magazines
Archive   : JAN94.ZIP
Filename : PENT.ASC

_OPTIMIZING PENTIUM CODE_
by Mike Schmit

Example 1:

; with without branch prediction
; (cycles are for a case of non-space character)
loop1:
mov al, [si] ; 1 1
inc si ; 0 0 (0 due to pairing)
cmp al, ' ' ; 1 1
jne foo ; 0 3
call space ;
jmp bar ;
foo:
inc dx ; 1 1
bar:
dec cx ; 0 0
jnz loop1 ; 1 3
; total 4 10



Example 2:


(a)
first subsequent
mov ax, 1 1 1 (numbers are CPU cycle numbers)
inc bx 2 1
mov cx, 1 2 2
call xyz 3 2


(b)

read-after-write (do not pair)
mov al, 1
add bh, ah

mov ax, 1
add bx, ax


Example 3:

(a)

pop bx ; two POPs pair
pop ax ; but then RET
ret ; causes AGI delay

inc bx ; two INCs pair
inc ax
mov cx, [si] ; two MOVs pair
mov dx, [bx] ; AGI delay because BX changed in previous cycle


(b)

add [bx], 2
add [si], 2

write-after-write (do not pair)
mov eax, 1
add eax, ebx

mov ax, 1

mov ax, 2

write-after-read (pairs)
mov ax, bx
inc bx

read-after-read (pairs)
mov eax, ebx
add ecx, ebx


Example 4:


(a)

mov ax, [bx]
add ax, 2
mov [bx], ax
mov ax, [si]
add ax, 2
mov [si], ax


(b)

mov ax, [bx]
add ax, 2
mov [bx], ax
add [si], 2


(c)

mov ax, [bx]
add ax, 2
mov [bx], ax
mov cx, [si]
add cx, 2
mov [si], cx


(d)


mov ax, [bx]
mov cx, [si]
add ax, 2
add cx, 2
mov [bx], ax
mov [si], cx




Example 5:

loop1: lodsb
stosb
or al, al
jne loop1


Figures 2: ASCIIZ string copy with max string length


(a)

; 8088 286 386 486 Pent. w/pair bytes
loop1:
lodsb ; 16 5 5 5 2 2 1
stosb ; 15 3 4 5 3 3 1
or al, al ; 3 2 2 1 1 1 2
jne loop1 ; 16 8 8 3 1 1 2
-------------------------- ---
50 18 19 14 7 7 6


(b)
; 8088 286 386 486 Pent. w/pair bytes
loop2:
mov al, [si] ; 17 5 4 1 1 1 2
inc si ; 3 2 2 1 1 0 1
mov [di], al ; 18 4 2 1 1 1 2
inc di ; 3 2 2 1 1 0 1
cmp al, 0 ; 4 3 2 1 1 1 2
jne loop2 ; 16 9 8 3 1 0 2
-------------------------- ---
61 25 20 8 6 3 11



Figures 4: ASCIIZ string copy with max string length


(a)
; 8088 286 386 486 Pent. w/pair
loop3:
lodsb ; 16 5 5 5 2 2
stosb ; 15 3 4 5 3 3
or al, al ; 3 2 2 1 1 1
loopne loop3 ; 19 10 13 9 8 8
;-----------------------------
53 20 24 20 14 14

(b)
; 8088 286 386 486 Pent. w/pair
loop4:
mov al, [si] ; 17 12 4 1 1 1
inc si ; 3 2 2 1 1 0
mov [di], al ; 18 9 2 1 1 1
inc di ; 3 2 2 1 1 0
cmp al, 0 ; 4 3 2 1 1 1
je exit4 ; 4 3 3 1 1 0
dec cx ; 3 2 2 1 1 1
jnz loop4 ; 16 9 8 3 1 1
exit4:
;-----------------------------
68 42 25 10 8 5