incdir include: incdir lvo: incdir powerasm:source/powerpc_lib/warprace/ include powerpc/powerpc.i include warprace.i include exec/memory.i include graphics/gfx.i include exec_lib.i include graphics_lib.i xref _LinkerDB xref C2P_PPC xdef bitmap xdef ChunkyBuffer VERSION = 1 REVISION = 0 REVISIONCOUNT = 1 merge macro move.l \2,\4 move.l #\5,\3 and.l \3,\2 and.l \1,\3 eor.l \3,\1 eor.l \2,\4 ifeq \6-1 add.l \3,\3 else lsl.l #\6,\3 endc lsr.l #\6,\4 or.l \3,\2 or.l \4,\1 endm section "",code mc68020 start movem.l d1-a6,-(sp) lea _LinkerDB,a4 cmp.l #WR_ID,d0 beq.b .go movem.l (sp)+,d1-a6 moveq #0,d0 rts .go lea ErrParams,a3 move.l a0,a2 move.l $4.w,_SysBase lea graf_name,a1 move.l a1,(a3) moveq #0,d0 move.l d0,4(a3) lea LibErr,a5 CALLEXEC OpenLibrary move.l d0,_GfxBase beq.b .error move.l #320*256,d0 move.l d0,(a3) move.l #MEMF_PUBLIC,d1 lea MemErr,a5 CALLEXEC AllocVec move.l d0,ChunkyBuffer beq.b .error lea ChipErr,a5 bsr SetupBitmaps tst.l d0 beq.b .error move.l WRI_POWERPCBASE(a2),_PowerPCBase tst WRI_68K(a2) bne .notPPC move.l a2,a0 RUNPOWERPC C2P_PPC bra.w .done .notPPC move.l WRI_STARTTIMER_68K(a2),a1 jsr (a1) moveq #50-1,d7 .loop move.l ChunkyBuffer,a0 lea bitmap,a1 bsr ChunkyToPlanar dbra d7,.loop move.l WRI_STOPTIMER_68K(a2),a1 jsr (a1) divu.l #50,d0 .done move.l d0,Result divul.l #1000,d1:d0 move.l d0,ResultParams move.l d1,ResultParams+4 move.l #STATUS_SUCCESS,Status bra.b .end .error move.l #STATUS_ERROR,Status move.l a5,ErrStr .end move.l ChunkyBuffer,d0 beq.b .nofree move.l d0,a1 CALLEXEC FreeVec .nofree bsr FreeBitmaps move.l _GfxBase,d0 beq.b .nograf move.l d0,a1 CALLEXEC CloseLibrary .nograf move.l #WRO,d0 movem.l (sp)+,d1-a6 rts ****************************************************************************** * * d0 = SetupBitmaps * * prepares one bitmap * * Out: * d0 = error code * * error codes: -1 = success * 0 = not enough memory ****************************************************************************** SetupBitmaps movem.l d1/d5-a2,-(sp) lea bitmap,a0 move.l a0,ActualBitmap .loop move.l a0,a2 lea bm_Planes(a2),a2 ;a2 -> planeptrs moveq #8,d0 move.l #320,d1 move.l #256,d2 CALLGRAF InitBitMap ;init bitmap structures move.l #320,d0 move.l #256*8,d1 CALLGRAF AllocRaster ;allocate bitmap memory tst.l d0 beq.b .error move.l d0,a0 move.l #(256*8*320/8/4),d5 .clear clr.l (a0)+ ;clear one plane subq.l #1,d5 bne.b .clear moveq #8-1,d6 .loop2 move.l d0,(a2)+ ;and insert into bitmap struct add.l #320/8*256,d0 dbra d6,.loop2 moveq #-1,d0 bra.b .end .error moveq #0,d0 .end movem.l (sp)+,d1/d5-a2 rts ****************************************************************************** * * FreeBitmaps * * frees all the memory allocated by 'AllocRaster' * ****************************************************************************** FreeBitmaps movem.l d0/d1/d6-a2,-(sp) lea bitmap,a0 .loop move.l a0,a2 lea bm_Planes(a2),a2 moveq #8-1,d6 .loop2 move.l (a2)+,d0 ;read planeptr beq.b .next move.l d0,a0 move.l #320,d0 move.l #256,d1 CALLGRAF FreeRaster ;free the bitmap memory .next dbra d6,.loop2 movem.l (sp)+,d0/d1/d6-a2 rts ****************************************************************************** * * ChunkyToPlanar (a0,a1) * * converts chunky data into bitplane data (using Bitmap Structure) * works ONLY with a depth of 8 and resolution 320*256. * the 8 bitplanes MUST have been allocated in ONE piece!! * * original algorithm by James McCoull (I found the source somewhere * on the AMINET CD's) * * In: a0 -> Chunkybuffer * a1 -> Bitmap structure * ****************************************************************************** ChunkyToPlanar movem.l d0-a6,-(sp) sub.w #40,sp ; space for temporary variables move.l #256*40,d2 move.l bm_Planes(a1),a1 ; a0 = chunky buffer ; a1 = output area ; d2 = plsiz ext.l d2 movea.l d2,a3 ; a3 = plsiz move.l a0,a4 add.l #320*256,a4 ; a4 -> end of chunky data first_case move.l (0,a0),d1 move.l (4,a0),d3 move.l (8,a0),d0 move.l (12,a0),d2 move.l (2,a0),d4 move.l (10,a0),d5 move.l (6,a0),d6 move.l (14,a0),d7 move.w (16,a0),d1 move.w (24,a0),d0 move.w (20,a0),d3 move.w (28,a0),d2 move.w (18,a0),d4 move.w (26,a0),d5 move.w (22,a0),d6 move.w (30,a0),d7 adda.w #32,a0 move.l d6,a5 move.l d7,a6 merge d1,d0,d6,d7,$00ff00ff,8 merge d3,d2,d6,d7,$00ff00ff,8 merge d1,d3,d6,d7,$0f0f0f0f,4 merge d0,d2,d6,d7,$0f0f0f0f,4 exg d1,a5 exg d0,a6 merge d4,d5,d6,d7,$00ff00ff,8 merge d1,d0,d6,d7,$00ff00ff,8 merge d4,d1,d6,d7,$0f0f0f0f,4 merge d5,d0,d6,d7,$0f0f0f0f,4 merge d3,d1,d6,d7,$33333333,2 merge d2,d0,d6,d7,$33333333,2 merge d3,d2,d6,d7,$55555555,1 merge d1,d0,d6,d7,$55555555,1 move.l d0,(0*4,sp) ;plane0 move.l d1,(1*4,sp) ;plane1 move.l d2,(2*4,sp) ;plane2 move.l d3,(3*4,sp) ;plane3 move.l a5,d3 move.l a6,d2 merge d3,d4,d6,d7,$33333333,2 merge d2,d5,d6,d7,$33333333,2 merge d3,d2,d6,d7,$55555555,1 merge d4,d5,d6,d7,$55555555,1 move.l d5,(4*4,sp) ;plane4 move.l d4,(5*4,sp) ;plane5 move.l d2,(6*4,sp) ;plane6 move.l d3,(7*4,sp) ;plane7 move.l a1,(32,sp) ; save output address addq.l #4,a1 ; skip 32 pixels on output cmpa.l a0,a4 beq.w final_case main_case move.l a1,(36,sp) ; save current output address move.l (32,sp),a1 ; a1 = previous output address move.l (0,a0),d1 move.l (4,a0),d3 move.l (8,a0),d0 move.l (12,a0),d2 move.l (2,a0),d4 move.l (10,a0),d5 move.l (6,a0),d6 move.l (14,a0),d7 move.w (16,a0),d1 move.w (24,a0),d0 move.w (20,a0),d3 move.w (28,a0),d2 move.w (18,a0),d4 move.w (26,a0),d5 move.w (22,a0),d6 move.w (30,a0),d7 adda.w #32,a0 move.l d6,a5 move.l d7,a6 move.l (0*4,sp),(a1) ;plane0 adda.l a3,a1 ;a1+=plsiz merge d1,d0,d6,d7,$00ff00ff,8 merge d3,d2,d6,d7,$00ff00ff,8 move.l (1*4,sp),(a1) ;plane1 adda.l a3,a1 ;a1+=plsiz merge d1,d3,d6,d7,$0f0f0f0f,4 merge d0,d2,d6,d7,$0f0f0f0f,4 exg d1,a5 exg d0,a6 move.l (2*4,sp),(a1) ;plane2 adda.l a3,a1 ;a1+=plsiz merge d4,d5,d6,d7,$00ff00ff,8 merge d1,d0,d6,d7,$00ff00ff,8 move.l (3*4,sp),(a1) ;plane3 adda.l a3,a1 ;a1+=plsiz merge d4,d1,d6,d7,$0f0f0f0f,4 merge d5,d0,d6,d7,$0f0f0f0f,4 move.l (4*4,sp),(a1) ;plane4 adda.l a3,a1 ;a1+=plsiz merge d3,d1,d6,d7,$33333333,2 merge d2,d0,d6,d7,$33333333,2 move.l (5*4,sp),(a1) ;plane5 adda.l a3,a1 ;a1+=plsiz merge d3,d2,d6,d7,$55555555,1 merge d1,d0,d6,d7,$55555555,1 move.l d0,(0*4,sp) ;plane0 (movem.l is slower!) move.l d1,(1*4,sp) ;plane1 move.l d2,(2*4,sp) ;plane2 move.l d3,(3*4,sp) ;plane3 move.l a5,d3 move.l a6,d2 move.l (6*4,sp),(a1) ;plane6 adda.l a3,a1 ;a1+=plsiz merge d3,d4,d6,d7,$33333333,2 merge d2,d5,d6,d7,$33333333,2 move.l (7*4,sp),(a1) ;plane7 adda.l a3,a1 ;a1+=plsiz merge d3,d2,d6,d7,$55555555,1 merge d4,d5,d6,d7,$55555555,1 move.l d5,(4*4,sp) ;plane4 move.l d4,(5*4,sp) ;plane5 move.l d2,(6*4,sp) ;plane6 move.l d3,(7*4,sp) ;plane7 movea.l (36,sp),a1 ; restore current output address move.l a1,(32,sp) ; save output address addq.l #4,a1 cmpa.l a0,a4 bne.w main_case final_case move.l (32,sp),a1 ; a1 = previous output address move.l (0*4,sp),(a1) ;plane0 adda.l a3,a1 ;a1+=plsiz move.l (1*4,sp),(a1) ;plane1 adda.l a3,a1 ;a1+=plsiz move.l (2*4,sp),(a1) ;plane2 adda.l a3,a1 ;a1+=plsiz move.l (3*4,sp),(a1) ;plane3 adda.l a3,a1 ;a1+=plsiz move.l (4*4,sp),(a1) ;plane4 adda.l a3,a1 ;a1+=plsiz move.l (5*4,sp),(a1) ;plane5 adda.l a3,a1 ;a1+=plsiz move.l (6*4,sp),(a1) ;plane6 adda.l a3,a1 ;a1+=plsiz move.l (7*4,sp),(a1) ;plane7 exit add.w #40,sp movem.l (sp)+,d0-a6 rts section "",data graf_name GRAFNAME cnop 0,4 WRO dc.l ModName dc.l Short dc.l Description dc.l Author dc.l VERSION dc.l REVISION dc.l REVISIONCOUNT dc.l 0 Result dc.l 0 dc.l RES_MICROSECS dc.l ResultString dc.l ResultParams Status dc.l 0 ErrStr dc.l 0 dc.l ErrParams dc.l 0 ModName dc.b "C2P",0 Short dc.b "ChunkyToPlanar converter (320*256*8)",0 Description dc.b "This program measures the time for a ChunkyToPlanar-\n" dc.b "conversion to complete. The dimensions are 320*256*8.",0 Author dc.b "Sam Jordan (original C2P algorithm by James Mc Coull)",0 ResultString dc.b "Elapsed time: %ld.%03ld ms",0 LibErr dc.b "Failed to open %s V%ld",0 MemErr dc.b "Failed to allocate %ld bytes or memory",0 ChipErr dc.b "Failed to allocate 320*256 bytes of CHIP memory",0 section "",bss _SysBase ds.l 1 _GfxBase ds.l 1 _PowerPCBase ds.l 1 ActualBitmap ds.l 1 ChunkyBuffer ds.l 1 ResultParams ds.l 2 ErrParams ds.l 2 bitmap ds.b bm_SIZEOF