what are the possible values/expressions there? and how do you call that? displacement?

lea eax,[xxx+xxx*x]
mov [xxx+xxx*x], 0

i can just at max plus and mult one time? also, it doesnt let me set > *8, neither *3 or *5?

i saw intel manual and i cant find the XXX+XXX*X formula, just XXX+XXX or XXX*X. explain me!!

You mean the X's ?

I think maybe he wants to learn the LEA multiplier tricks?

[reg + reg * (1,2,4,8) + const]

I think.

no no!! i just wanna know why i cannot use others patterns than XXX+YYY*Z. And why i cant find that in the intel manuals


hmmm it seems to accept that const too!

so its XXX+YYY*Z+C !!

The instruction encoding has limited variability - this is outlined in the manual. Three bits for X and Y, two bits for Z, and an extension dword for C.

The "LEA-trick" mentioned means you can use LEA to multiply with small constants easily
where REG1=REG2=REG3 and C=0 in LEA REG1,[REG2*{0|1}+REG3*{1|2|4|8}+C]
For example:
lea eax,[eax*{0|1}+eax*{1|2|4|8}]
combining you can replace MUL (which is sometimes more expensive) with LEA in a few cases:
mul eax,{0,1,2,3,4,5,8,9}

You can see that it is impossible to encode 6 or 7 with this trick, but there are two ways to encode 1 and 2.
It is not very efficient to use LEA to multiply by 0,1,2,4 or 8 (ADD and SHL are better) so you are left with 3,5 and 9.

FASM allows this simple markup:

^^Good for obfuscation