unsigned divide, round up?

flat assembler
Message board for the users of flat assembler.

Index > Main > unsigned divide, round up?

Author

Thread

bitshifter

Joined: 04 Dec 2007
Posts: 796
Location: Massachusetts, USA

bitshifter 08 Feb 2011, 13:11

Hi, i need to divide 2 16bit numbers and round up to the next integer.

For example:
16 / 16 = 1
17 / 16 = 2

Now i explain the purpose:
I am writing a real-mode paragraph allocator.
The client code may ask for byte-size memory block.
The minimum allocation block is one paragraph.
My allocator should return a segment ptr to paragraph (offset = 0)
(Kinda like a _seg* in C language)

So it needs to be 8086 instructions.
I would like not to use DIV if possible (for speed reasons)
But if it turns out to be a better way then so be it...
I thought of using SHR/ADC but SHR only set CF with last bit shifted out.
Maybe you have a simple solution?

Thanks for your time.

_________________
Coding a 3D game engine with fasm is like trying to eat an elephant,
you just have to keep focused and take it one 'byte' at a time.

08 Feb 2011, 13:11

edfed

Joined: 20 Feb 2006
Posts: 4336
Location: Now

edfed 08 Feb 2011, 13:21

div 16
cmp dx,0
je @f
inc ax
@@:

mov dx,ax
and dx,0fh
je @f
add ax,16
@@:
shr ax,4

08 Feb 2011, 13:21

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20343
Location: In your JS exploiting you and your system

revolution 08 Feb 2011, 13:46

Add 15 to your numerator.

Code:

mov eax,[myvalue]
add eax,15
shr eax,4
mov [myresult],eax

08 Feb 2011, 13:46

edfed

Joined: 20 Feb 2006
Posts: 4336
Location: Now

edfed 08 Feb 2011, 14:12

no jcc, cool!!!

08 Feb 2011, 14:12

b1528932

Joined: 21 May 2010
Posts: 287

b1528932 08 Feb 2011, 20:47

So it needs to be 8086 code, so no shr using imm8 there. and no eAX

considering that you divide only by 16:

;ax = input
test ax, 0x000F
mov cl, 4
ror ax, cl
adc ax, 0
and ax, 0x0FFF
;ax = output

08 Feb 2011, 20:47

Hrstka

Joined: 05 May 2008
Posts: 57
Location: Czech republic

Hrstka 14 Feb 2011, 20:12

My favourite way for rounding up:

Code:

neg ax
and ax, -16    ; round up to next paragraph
neg ax

I'm not sure if the neg instruction exists on 8086, but neg ax can be replaced with

Code:

xor ax, -1    ; = not ax
inc ax

EDIT: I overlooked the div part, in this case use revolution's way. If shr ax, imm8 is not allowed,

Code:

mov cl, 4
shr ax, cl

should work.

14 Feb 2011, 20:12

< Last Thread | Next Thread >

Forum Rules:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum