ReC98/th04/math/vector2_near.asm

public VECTOR2_NEAR
vector2_near	proc near

@@r	= word ptr  4
@@angle	= byte ptr  6
@@_ret	= word ptr  8

	push	bp
	mov	bp, sp
	movsx	edx, [bp+@@r]
	mov	bl, [bp+@@angle]
	mov	bh, 0
	add	bx, bx
	movsx	eax, _CosTable8[bx]
	movsx	ecx, _SinTable8[bx]
	imul	eax, edx
	sar	eax, 8
	imul	edx, ecx
	sar	edx, 8
	mov	bx, [bp+@@_ret]
	mov	[bx], ax
	mov	[bx+2],	dx
	pop	bp
	retn	6
vector2_near	endp
[Position independence] False positives in item functions Welcome to this new [brand] of reverse-engineering, where we only fix things that might be mistaken for addresses, without looking deeper into what the actual functions do… unless that directly leaps into the eye. Part of P0057, funded by [Anonymous] and -Tom-. 2019-11-09 19:09:33 +00:00			`public VECTOR2_NEAR`
[Reduction] Separate functions for 1D and 2D vector construction Look at that TH05 vector2_at_opt function. What the hell, the caller is supposed to set up the stack frame for the function? How do you even get a compiler to do this (and no, I haven't found a compiler switch)? No way around writing a separate "optimizer" as part of the compilation pipeline, it seems. 2017-01-08 20:01:13 +00:00			`vector2_near proc near`

			`@@r = word ptr 4`
			`@@angle = byte ptr 6`
			`@@_ret = word ptr 8`

			`push bp`
			`mov bp, sp`
			`movsx edx, [bp+@@r]`
			`mov bl, [bp+@@angle]`
			`mov bh, 0`
			`add bx, bx`
			`movsx eax, _CosTable8[bx]`
			`movsx ecx, _SinTable8[bx]`
			`imul eax, edx`
			`sar eax, 8`
			`imul edx, ecx`
			`sar edx, 8`
			`mov bx, [bp+@@_ret]`
			`mov [bx], ax`
			`mov [bx+2], dx`
			`pop bp`
			`retn 6`
			`vector2_near endp`