--- /dev/null
+/*
+* Glide64 - Glide video plugin for Nintendo 64 emulators.
+* Copyright (c) 2002 Dave2001
+* Copyright (c) 2003-2009 Sergey 'Gonetz' Lipski
+*
+* This program is free software; you can redistribute it and/or modify
+* it under the terms of the GNU General Public License as published by
+* the Free Software Foundation; either version 2 of the License, or
+* any later version.
+*
+* This program is distributed in the hope that it will be useful,
+* but WITHOUT ANY WARRANTY; without even the implied warranty of
+* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+* GNU General Public License for more details.
+*
+* You should have received a copy of the GNU General Public License
+* along with this program; if not, write to the Free Software
+* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+*/
+
+//****************************************************************
+//
+// Glide64 - Glide Plugin for Nintendo 64 emulators
+// Project started on December 29th, 2001
+//
+// Authors:
+// Dave2001, original author, founded the project in 2001, left it in 2002
+// Gugaman, joined the project in 2002, left it in 2002
+// Sergey 'Gonetz' Lipski, joined the project in 2002, main author since fall of 2002
+// Hiroshi 'KoolSmoky' Morii, joined the project in 2007
+//
+//****************************************************************
+//
+// To modify Glide64:
+// * Write your name and (optional)email, commented by your work, so I know who did it, and so that you can find which parts you modified when it comes time to send it to me.
+// * Do NOT send me the whole project or file that you modified. Take out your modified code sections, and tell me where to put them. If people sent the whole thing, I would have many different versions, but no idea how to combine them all.
+//
+//****************************************************************
+//
+// Software rendering into N64 depth buffer
+// Idea and N64 depth value format by Orkin
+// Polygon rasterization algorithm is taken from FATMAP2 engine by Mats Byggmastar, mri@penti.sit.fi
+//
+// Created by Gonetz, Dec 2004
+//
+//****************************************************************
+
+#include "Gfx_1.3.h"
+#include "rdp.h"
+#include "DepthBufferRender.h"
+
+wxUint16 * zLUT = 0;
+
+void ZLUT_init()
+{
+ if (zLUT)
+ return;
+ zLUT = new wxUint16[0x40000];
+ for(int i=0; i<0x40000; i++)
+ {
+ wxUint32 exponent = 0;
+ wxUint32 testbit = 1 << 17;
+ while((i & testbit) && (exponent < 7))
+ {
+ exponent++;
+ testbit = 1 << (17 - exponent);
+ }
+
+ wxUint32 mantissa = (i >> (6 - (6 < exponent ? 6 : exponent))) & 0x7ff;
+ zLUT[i] = (wxUint16)(((exponent << 11) | mantissa) << 2);
+ }
+}
+
+void ZLUT_release()
+{
+ delete[] zLUT;
+ zLUT = 0;
+}
+
+static vertexi * max_vtx; // Max y vertex (ending vertex)
+static vertexi * start_vtx, * end_vtx; // First and last vertex in array
+static vertexi * right_vtx, * left_vtx; // Current right and left vertex
+
+#ifdef HAVE_GLES
+// using float should be faster than int in fact, especialy if using NEON
+static float right_height, left_height;
+static float right_x, right_dxdy, left_x, left_dxdy;
+static float left_z, left_dzdy;
+
+__inline float iceil(int x)
+{
+ return cellf((float)x*(1.0f/65536.0f));
+}
+__inline float i2float(int x)
+{
+ return (float)x*(1.0f/65536.0f);
+}
+__inline int float2i(float x)
+{
+ return x*65536.0f;
+}
+
+#else
+
+static int right_height, left_height;
+static int right_x, right_dxdy, left_x, left_dxdy;
+static int left_z, left_dzdy;
+
+__inline int imul16(int x, int y) // (x * y) >> 16
+{
+ return (((long long)x) * ((long long)y)) >> 16;
+}
+
+__inline int imul14(int x, int y) // (x * y) >> 14
+{
+ return (((long long)x) * ((long long)y)) >> 14;
+}
+__inline int idiv16(int x, int y) // (x << 16) / y
+{
+ //x = (((long long)x) << 16) / ((long long)y);
+ /*
+ eax = x;
+ ebx = y;
+ edx = x;
+ (x << 16) | ()
+ */
+#if !defined(__GNUC__) && !defined(NO_ASM)
+ __asm {
+ mov eax, x
+ mov ebx, y
+ mov edx,eax
+ sar edx,16
+ shl eax,16
+ idiv ebx
+ mov x, eax
+ }
+#elif !defined(NO_ASM)
+ int reminder;
+ asm ("idivl %[divisor]"
+ : "=a" (x), "=d" (reminder)
+ : [divisor] "g" (y), "d" (x >> 16), "a" (x << 16));
+#else
+ x = (((long long)x) << 16) / ((long long)y);
+#endif
+ return x;
+}
+
+__inline int iceil(int x)
+{
+ x += 0xffff;
+ return (x >> 16);
+}
+#endif
+
+static void RightSection(void)
+{
+ // Walk backwards trough the vertex array
+
+ vertexi * v2, * v1 = right_vtx;
+ if(right_vtx > start_vtx) v2 = right_vtx-1;
+ else v2 = end_vtx; // Wrap to end of array
+ right_vtx = v2;
+
+ // v1 = top vertex
+ // v2 = bottom vertex
+
+ // Calculate number of scanlines in this section
+
+ right_height = iceil(v2->y) - iceil(v1->y);
+ if(right_height <= 0) return;
+
+ // Guard against possible div overflows
+
+ #ifdef HAVE_GLES
+ right_dxdy = i2float(v2->x - v1->x)/i2float(v2->y - v1->y);
+ #else
+ if(right_height > 1) {
+ // OK, no worries, we have a section that is at least
+ // one pixel high. Calculate slope as usual.
+
+ int height = v2->y - v1->y;
+ right_dxdy = idiv16(v2->x - v1->x, height);
+ }
+ else {
+ // Height is less or equal to one pixel.
+ // Calculate slope = width * 1/height
+ // using 18:14 bit precision to avoid overflows.
+
+ int inv_height = (0x10000 << 14) / (v2->y - v1->y);
+ right_dxdy = imul14(v2->x - v1->x, inv_height);
+ }
+ #endif
+
+ // Prestep initial values
+
+ #ifdef HAVE_GLES
+ float prestep = iceil(v1->y) - i2float(v1->y);
+ right_x = i2float(v1->x) + prestep*right_dxdy;
+ #else
+ int prestep = (iceil(v1->y) << 16) - v1->y;
+ right_x = v1->x + imul16(prestep, right_dxdy);
+ #endif
+}
+
+static void LeftSection(void)
+{
+ // Walk forward trough the vertex array
+
+ vertexi * v2, * v1 = left_vtx;
+ if(left_vtx < end_vtx) v2 = left_vtx+1;
+ else v2 = start_vtx; // Wrap to start of array
+ left_vtx = v2;
+
+ // v1 = top vertex
+ // v2 = bottom vertex
+
+ // Calculate number of scanlines in this section
+
+ left_height = iceil(v2->y) - iceil(v1->y);
+ if(left_height <= 0) return;
+
+ // Guard against possible div overflows
+
+ #ifdef HAVE_GLES
+ float invheight = 1.0f/i2float(v2->y - v1->y)
+ left_dxdy = i2float(v2->x - v1->x)*invheight;
+ left_dzdy = i2float(v2->z - v1->z)*invheight;
+ #else
+ if(left_height > 1) {
+ // OK, no worries, we have a section that is at least
+ // one pixel high. Calculate slope as usual.
+
+ int height = v2->y - v1->y;
+ left_dxdy = idiv16(v2->x - v1->x, height);
+ left_dzdy = idiv16(v2->z - v1->z, height);
+ }
+ else {
+ // Height is less or equal to one pixel.
+ // Calculate slope = width * 1/height
+ // using 18:14 bit precision to avoid overflows.
+
+ int inv_height = (0x10000 << 14) / (v2->y - v1->y);
+ left_dxdy = imul14(v2->x - v1->x, inv_height);
+ left_dzdy = imul14(v2->z - v1->z, inv_height);
+ }
+ #endif
+
+ // Prestep initial values
+
+ #ifdef HAVE_GLES
+ float prestep = iceil(v1->y) - i2float(v1->y);
+ left_x = i2float(v1->x) + prestep*left_dxdy;
+ left_z = i2float(v1->z) + prestep*left_dzdy;
+ #else
+ int prestep = (iceil(v1->y) << 16) - v1->y;
+ left_x = v1->x + imul16(prestep, left_dxdy);
+ left_z = v1->z + imul16(prestep, left_dzdy);
+ #endif
+}
+
+
+void Rasterize(vertexi * vtx, int vertices, int dzdx)
+{
+ start_vtx = vtx; // First vertex in array
+
+ // Search trough the vtx array to find min y, max y
+ // and the location of these structures.
+
+ vertexi * min_vtx = vtx;
+ max_vtx = vtx;
+
+ int min_y = vtx->y;
+ int max_y = vtx->y;
+
+ vtx++;
+
+ for(int n=1; n<vertices; n++) {
+ if(vtx->y < min_y) {
+ min_y = vtx->y;
+ min_vtx = vtx;
+ }
+ else
+ if(vtx->y > max_y) {
+ max_y = vtx->y;
+ max_vtx = vtx;
+ }
+ vtx++;
+ }
+
+ // OK, now we know where in the array we should start and
+ // where to end while scanning the edges of the polygon
+
+ left_vtx = min_vtx; // Left side starting vertex
+ right_vtx = min_vtx; // Right side starting vertex
+ end_vtx = vtx-1; // Last vertex in array
+
+ // Search for the first usable right section
+
+ do {
+ if(right_vtx == max_vtx) return;
+ RightSection();
+ } while(right_height <= 0);
+
+ // Search for the first usable left section
+
+ do {
+ if(left_vtx == max_vtx) return;
+ LeftSection();
+ } while(left_height <= 0);
+
+ wxUint16 * destptr = (wxUint16*)(gfx.RDRAM+rdp.zimg);
+ int y1 = iceil(min_y);
+ if (y1 >= (int)rdp.scissor_o.lr_y) return;
+ int shift;
+
+ for(;;)
+ {
+ int x1 = iceil(left_x);
+ if (x1 < (int)rdp.scissor_o.ul_x)
+ x1 = rdp.scissor_o.ul_x;
+ int width = iceil(right_x) - x1;
+ if (x1+width >= (int)rdp.scissor_o.lr_x)
+ width = rdp.scissor_o.lr_x - x1 - 1;
+
+ if(width > 0 && y1 >= (int)rdp.scissor_o.ul_y) {
+
+
+ // Prestep initial z
+
+ #ifdef HAVE_GLES
+ float z = left_z + (x1 - left_x)*dzdx;
+ #else
+ int prestep = (x1 << 16) - left_x;
+ int z = left_z + imul16(prestep, dzdx);
+ #endif
+
+ shift = x1 + y1*rdp.zi_width;
+ //draw to depth buffer
+ int trueZ;
+ int idx;
+ wxUint16 encodedZ;
+ for (int x = 0; x < width; x++)
+ {
+ #ifdef HAVE_GLES
+ trueZ = (int)(z*8.0f);
+ #else
+ trueZ = z/8192;
+ #endif
+ if (trueZ < 0) trueZ = 0;
+ else if (trueZ > 0x3FFFF) trueZ = 0x3FFFF;
+ encodedZ = zLUT[trueZ];
+ idx = (shift+x)^1;
+ if(encodedZ < destptr[idx])
+ destptr[idx] = encodedZ;
+ z += dzdx;
+ }
+ }
+
+ //destptr += rdp.zi_width;
+ y1++;
+ if (y1 >= (int)rdp.scissor_o.lr_y) return;
+
+ // Scan the right side
+
+ if(--right_height <= 0) { // End of this section?
+ do {
+ if(right_vtx == max_vtx) return;
+ RightSection();
+ } while(right_height <= 0);
+ }
+ else
+ right_x += right_dxdy;
+
+ // Scan the left side
+
+ if(--left_height <= 0) { // End of this section?
+ do {
+ if(left_vtx == max_vtx) return;
+ LeftSection();
+ } while(left_height <= 0);
+ }
+ else {
+ left_x += left_dxdy;
+ left_z += left_dzdy;
+ }
+ }
+}