|Previous||Table of Contents||Next|
Listing 47.4 shows Mode X rectangle fill code. The plane is selected for each pixel in turn, with drawing cycling from plane 0 to plane 3, then wrapping back to plane 0. This is the sort of code that stems from a write-pixel line of thinking; it reflects not a whit of the unique perspective that Mode X demands, and although it looks reasonably efficient, it is in fact some of the slowest graphics code you will ever see. Ive provided Listing 47.4 partly for illustrative purposes, but mostly so well have a point of reference for the substantial speed-up thats possible with code thats designed from a Mode X perspective.
LISTING 47.4 L47-4.ASM
; Mode X (320x240, 256 colors) rectangle fill routine. Works on all ; VGAs. Uses slow approach that selects the plane explicitly for each ; pixel. Fills up to but not including the column at EndX and the row ; at EndY. No clipping is performed. ; C near-callable as: ; ; void FillRectangleX(int StartX, int StartY, int EndX, int EndY, ; unsigned int PageBase, int Color); SC_INDEX equ 03c4h ;Sequence Controller Index MAP_MASK equ 02h ;index in SC of Map Mask register SCREEN_SEG equ 0a000h ;segment of display memory in mode X SCREEN_WIDTH equ 80 ;width of screen in bytes from one scan line ; to the next parms struc dw 2 dup (?) ;pushed BP and return address StartX dw ? ;X coordinate of upper left corner of rect StartY dw ? ;Y coordinate of upper left corner of rect EndX dw ? ;X coordinate of lower right corner of rect ; (the row at EndX is not filled) EndY dw ? ;Y coordinate of lower right corner of rect ; (the column at EndY is not filled) PageBase dw ? ;base offset in display memory of page in ; which to fill rectangle Color dw ? ;color in which to draw pixel parms ends .model small .code public _FillRectangleX _FillRectangleX proc near push bp ;preserve callers stack frame mov bp,sp ;point to local stack frame push si ;preserve callers register variables push di mov ax,SCREEN_WIDTH mul [bp+StartY] ;offset in page of top rectangle scan line mov di,[bp+StartX] shr di,1 shr di,1 ;X/4 = offset of first rectangle pixel in scan ; line add di,ax ;offset of first rectangle pixel in page add di,[bp+PageBase] ;offset of first rectangle pixel in ; display memory mov ax,SCREEN_SEG mov es,ax ;point ES:DI to the first rectangle pixels ; address mov dx,SC_INDEX ;set the Sequence Controller Index to mov al,MAP_MASK ; point to the Map Mask register out dx,al inc dx ;point DX to the SC Data register mov cl,byte ptr [bp+StartX] and cl,011b ;CL = first rectangle pixels plane mov al,01h shl al,cl ;set only the bit for the pixels plane to 1 mov ah,byte ptr [bp+Color] ;color with which to fill mov bx,[bp+EndY] sub bx,[bp+StartY] ;BX = height of rectangle jle FillDone ;skip if 0 or negative height mov si,[bp+EndX] sub si,[bp+StartX] ;CX = width of rectangle jle FillDone ;skip if 0 or negative width FillRowsLoop: push ax ;remember the plane mask for the left edge push di ;remember the start offset of the scan line mov cx,si ;set count of pixels in this scan line FillScanLineLoop: out dx,al ;set the plane for this pixel mov es:[di],ah ;draw the pixel shl al,1 ;adjust the plane mask for the next pixels and al,01111b ; bit, modulo 4 jnz AddressSet ;advance address if we turned over from inc di ; plane 3 to plane 0 mov al,00001b ;set plane mask bit for plane 0 AddressSet: loop FillScanLineLoop pop di ;retrieve the start offset of the scan line add di,SCREEN_WIDTH ;point to the start of the next scan ; line of the rectangle pop ax ;retrieve the plane mask for the left edge dec bx ;count down scan lines jnz FillRowsLoop FillDone: pop di ;restore callers register variables pop si pop bp ;restore callers stack frame ret _FillRectangleX endp end
The two major weaknesses of Listing 47.4 both result from selecting the plane on a pixel by pixel basis. First, endless OUTs (which are particularly slow on 386s, 486s, and Pentiums, much slower than accesses to display memory) must be performed, and, second, REP STOS cant be used. Listing 47.5 overcomes both these problems by tailoring the fill technique to the organization of display memory. Each plane is filled in its entirety in one burst before the next plane is processed, so only five OUTs are required in all, and REP STOS can indeed be used; Ive used REP STOSB in Listings 47.5 and 47.6. REP STOSW could be used and would improve performance on most VGAs; however, REP STOSW requires extra overhead to set up, so it can be slower for small rectangles, especially on 8-bit VGAs. Note that doing an entire plane at a time can produce a fading-in effect for large images, because all columns for one plane are drawn before any columns for the next. If this is a problem, the four planes can be cycled through once for each scan line, rather than once for the entire rectangle.
Listing 47.5 is 2.5 times faster than Listing 47.4 at clearing the screen on a 20-MHz cached 386 with a Paradise VGA. Although Listing 47.5 is slightly slower than an equivalent mode 13H fill routine would be, its not grievously so.
|Previous||Table of Contents||Next|