| 1 | /*************************************************************************** |
| 2 | * PCSX-Revolution - PlayStation Emulator for Nintendo Wii * |
| 3 | * Copyright (C) 2009-2010 PCSX-Revolution Dev Team * |
| 4 | * <http://code.google.com/p/pcsx-revolution/> * |
| 5 | * * |
| 6 | * This program is free software; you can redistribute it and/or modify * |
| 7 | * it under the terms of the GNU General Public License as published by * |
| 8 | * the Free Software Foundation; either version 2 of the License, or * |
| 9 | * (at your option) any later version. * |
| 10 | * * |
| 11 | * This program is distributed in the hope that it will be useful, * |
| 12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of * |
| 13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * |
| 14 | * GNU General Public License for more details. * |
| 15 | * * |
| 16 | * You should have received a copy of the GNU General Public License * |
| 17 | * along with this program; if not, write to the * |
| 18 | * Free Software Foundation, Inc., * |
| 19 | * 51 Franklin Street, Fifth Floor, Boston, MA 02111-1307 USA. * |
| 20 | ***************************************************************************/ |
| 21 | |
| 22 | /* |
| 23 | * GTE functions. |
| 24 | */ |
| 25 | |
| 26 | #include "gte.h" |
| 27 | #include "psxmem.h" |
| 28 | |
| 29 | #define VX(n) (n < 3 ? regs->CP2D.p[n << 1].sw.l : regs->CP2D.p[9].sw.l) |
| 30 | #define VY(n) (n < 3 ? regs->CP2D.p[n << 1].sw.h : regs->CP2D.p[10].sw.l) |
| 31 | #define VZ(n) (n < 3 ? regs->CP2D.p[(n << 1) + 1].sw.l : regs->CP2D.p[11].sw.l) |
| 32 | #define MX11(n) (n < 3 ? regs->CP2C.p[(n << 3)].sw.l : 0) |
| 33 | #define MX12(n) (n < 3 ? regs->CP2C.p[(n << 3)].sw.h : 0) |
| 34 | #define MX13(n) (n < 3 ? regs->CP2C.p[(n << 3) + 1].sw.l : 0) |
| 35 | #define MX21(n) (n < 3 ? regs->CP2C.p[(n << 3) + 1].sw.h : 0) |
| 36 | #define MX22(n) (n < 3 ? regs->CP2C.p[(n << 3) + 2].sw.l : 0) |
| 37 | #define MX23(n) (n < 3 ? regs->CP2C.p[(n << 3) + 2].sw.h : 0) |
| 38 | #define MX31(n) (n < 3 ? regs->CP2C.p[(n << 3) + 3].sw.l : 0) |
| 39 | #define MX32(n) (n < 3 ? regs->CP2C.p[(n << 3) + 3].sw.h : 0) |
| 40 | #define MX33(n) (n < 3 ? regs->CP2C.p[(n << 3) + 4].sw.l : 0) |
| 41 | #define CV1(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 5] : 0) |
| 42 | #define CV2(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 6] : 0) |
| 43 | #define CV3(n) (n < 3 ? (s32)regs->CP2C.r[(n << 3) + 7] : 0) |
| 44 | |
| 45 | #define fSX(n) ((regs->CP2D.p)[((n) + 12)].sw.l) |
| 46 | #define fSY(n) ((regs->CP2D.p)[((n) + 12)].sw.h) |
| 47 | #define fSZ(n) ((regs->CP2D.p)[((n) + 17)].w.l) /* (n == 0) => SZ1; */ |
| 48 | |
| 49 | #define gteVXY0 (regs->CP2D.r[0]) |
| 50 | #define gteVX0 (regs->CP2D.p[0].sw.l) |
| 51 | #define gteVY0 (regs->CP2D.p[0].sw.h) |
| 52 | #define gteVZ0 (regs->CP2D.p[1].sw.l) |
| 53 | #define gteVXY1 (regs->CP2D.r[2]) |
| 54 | #define gteVX1 (regs->CP2D.p[2].sw.l) |
| 55 | #define gteVY1 (regs->CP2D.p[2].sw.h) |
| 56 | #define gteVZ1 (regs->CP2D.p[3].sw.l) |
| 57 | #define gteVXY2 (regs->CP2D.r[4]) |
| 58 | #define gteVX2 (regs->CP2D.p[4].sw.l) |
| 59 | #define gteVY2 (regs->CP2D.p[4].sw.h) |
| 60 | #define gteVZ2 (regs->CP2D.p[5].sw.l) |
| 61 | #define gteRGB (regs->CP2D.r[6]) |
| 62 | #define gteR (regs->CP2D.p[6].b.l) |
| 63 | #define gteG (regs->CP2D.p[6].b.h) |
| 64 | #define gteB (regs->CP2D.p[6].b.h2) |
| 65 | #define gteCODE (regs->CP2D.p[6].b.h3) |
| 66 | #define gteOTZ (regs->CP2D.p[7].w.l) |
| 67 | #define gteIR0 (regs->CP2D.p[8].sw.l) |
| 68 | #define gteIR1 (regs->CP2D.p[9].sw.l) |
| 69 | #define gteIR2 (regs->CP2D.p[10].sw.l) |
| 70 | #define gteIR3 (regs->CP2D.p[11].sw.l) |
| 71 | #define gteSXY0 (regs->CP2D.r[12]) |
| 72 | #define gteSX0 (regs->CP2D.p[12].sw.l) |
| 73 | #define gteSY0 (regs->CP2D.p[12].sw.h) |
| 74 | #define gteSXY1 (regs->CP2D.r[13]) |
| 75 | #define gteSX1 (regs->CP2D.p[13].sw.l) |
| 76 | #define gteSY1 (regs->CP2D.p[13].sw.h) |
| 77 | #define gteSXY2 (regs->CP2D.r[14]) |
| 78 | #define gteSX2 (regs->CP2D.p[14].sw.l) |
| 79 | #define gteSY2 (regs->CP2D.p[14].sw.h) |
| 80 | #define gteSXYP (regs->CP2D.r[15]) |
| 81 | #define gteSXP (regs->CP2D.p[15].sw.l) |
| 82 | #define gteSYP (regs->CP2D.p[15].sw.h) |
| 83 | #define gteSZ0 (regs->CP2D.p[16].w.l) |
| 84 | #define gteSZ1 (regs->CP2D.p[17].w.l) |
| 85 | #define gteSZ2 (regs->CP2D.p[18].w.l) |
| 86 | #define gteSZ3 (regs->CP2D.p[19].w.l) |
| 87 | #define gteRGB0 (regs->CP2D.r[20]) |
| 88 | #define gteR0 (regs->CP2D.p[20].b.l) |
| 89 | #define gteG0 (regs->CP2D.p[20].b.h) |
| 90 | #define gteB0 (regs->CP2D.p[20].b.h2) |
| 91 | #define gteCODE0 (regs->CP2D.p[20].b.h3) |
| 92 | #define gteRGB1 (regs->CP2D.r[21]) |
| 93 | #define gteR1 (regs->CP2D.p[21].b.l) |
| 94 | #define gteG1 (regs->CP2D.p[21].b.h) |
| 95 | #define gteB1 (regs->CP2D.p[21].b.h2) |
| 96 | #define gteCODE1 (regs->CP2D.p[21].b.h3) |
| 97 | #define gteRGB2 (regs->CP2D.r[22]) |
| 98 | #define gteR2 (regs->CP2D.p[22].b.l) |
| 99 | #define gteG2 (regs->CP2D.p[22].b.h) |
| 100 | #define gteB2 (regs->CP2D.p[22].b.h2) |
| 101 | #define gteCODE2 (regs->CP2D.p[22].b.h3) |
| 102 | #define gteRES1 (regs->CP2D.r[23]) |
| 103 | #define gteMAC0 (((s32 *)regs->CP2D.r)[24]) |
| 104 | #define gteMAC1 (((s32 *)regs->CP2D.r)[25]) |
| 105 | #define gteMAC2 (((s32 *)regs->CP2D.r)[26]) |
| 106 | #define gteMAC3 (((s32 *)regs->CP2D.r)[27]) |
| 107 | #define gteIRGB (regs->CP2D.r[28]) |
| 108 | #define gteORGB (regs->CP2D.r[29]) |
| 109 | #define gteLZCS (regs->CP2D.r[30]) |
| 110 | #define gteLZCR (regs->CP2D.r[31]) |
| 111 | |
| 112 | #define gteR11R12 (((s32 *)regs->CP2C.r)[0]) |
| 113 | #define gteR22R23 (((s32 *)regs->CP2C.r)[2]) |
| 114 | #define gteR11 (regs->CP2C.p[0].sw.l) |
| 115 | #define gteR12 (regs->CP2C.p[0].sw.h) |
| 116 | #define gteR13 (regs->CP2C.p[1].sw.l) |
| 117 | #define gteR21 (regs->CP2C.p[1].sw.h) |
| 118 | #define gteR22 (regs->CP2C.p[2].sw.l) |
| 119 | #define gteR23 (regs->CP2C.p[2].sw.h) |
| 120 | #define gteR31 (regs->CP2C.p[3].sw.l) |
| 121 | #define gteR32 (regs->CP2C.p[3].sw.h) |
| 122 | #define gteR33 (regs->CP2C.p[4].sw.l) |
| 123 | #define gteTRX (((s32 *)regs->CP2C.r)[5]) |
| 124 | #define gteTRY (((s32 *)regs->CP2C.r)[6]) |
| 125 | #define gteTRZ (((s32 *)regs->CP2C.r)[7]) |
| 126 | #define gteL11 (regs->CP2C.p[8].sw.l) |
| 127 | #define gteL12 (regs->CP2C.p[8].sw.h) |
| 128 | #define gteL13 (regs->CP2C.p[9].sw.l) |
| 129 | #define gteL21 (regs->CP2C.p[9].sw.h) |
| 130 | #define gteL22 (regs->CP2C.p[10].sw.l) |
| 131 | #define gteL23 (regs->CP2C.p[10].sw.h) |
| 132 | #define gteL31 (regs->CP2C.p[11].sw.l) |
| 133 | #define gteL32 (regs->CP2C.p[11].sw.h) |
| 134 | #define gteL33 (regs->CP2C.p[12].sw.l) |
| 135 | #define gteRBK (((s32 *)regs->CP2C.r)[13]) |
| 136 | #define gteGBK (((s32 *)regs->CP2C.r)[14]) |
| 137 | #define gteBBK (((s32 *)regs->CP2C.r)[15]) |
| 138 | #define gteLR1 (regs->CP2C.p[16].sw.l) |
| 139 | #define gteLR2 (regs->CP2C.p[16].sw.h) |
| 140 | #define gteLR3 (regs->CP2C.p[17].sw.l) |
| 141 | #define gteLG1 (regs->CP2C.p[17].sw.h) |
| 142 | #define gteLG2 (regs->CP2C.p[18].sw.l) |
| 143 | #define gteLG3 (regs->CP2C.p[18].sw.h) |
| 144 | #define gteLB1 (regs->CP2C.p[19].sw.l) |
| 145 | #define gteLB2 (regs->CP2C.p[19].sw.h) |
| 146 | #define gteLB3 (regs->CP2C.p[20].sw.l) |
| 147 | #define gteRFC (((s32 *)regs->CP2C.r)[21]) |
| 148 | #define gteGFC (((s32 *)regs->CP2C.r)[22]) |
| 149 | #define gteBFC (((s32 *)regs->CP2C.r)[23]) |
| 150 | #define gteOFX (((s32 *)regs->CP2C.r)[24]) |
| 151 | #define gteOFY (((s32 *)regs->CP2C.r)[25]) |
| 152 | // senquack - gteH register is u16, not s16, and used in GTE that way. |
| 153 | // HOWEVER when read back by CPU using CFC2, it will be incorrectly |
| 154 | // sign-extended by bug in original hardware, according to Nocash docs |
| 155 | // GTE section 'Screen Offset and Distance'. The emulator does this |
| 156 | // sign extension when it is loaded to GTE by CTC2. |
| 157 | //#define gteH (regs->CP2C.p[26].sw.l) |
| 158 | #define gteH (regs->CP2C.p[26].w.l) |
| 159 | #define gteDQA (regs->CP2C.p[27].sw.l) |
| 160 | #define gteDQB (((s32 *)regs->CP2C.r)[28]) |
| 161 | #define gteZSF3 (regs->CP2C.p[29].sw.l) |
| 162 | #define gteZSF4 (regs->CP2C.p[30].sw.l) |
| 163 | #define gteFLAG (regs->CP2C.r[31]) |
| 164 | |
| 165 | #define GTE_OP(op) ((op >> 20) & 31) |
| 166 | #define GTE_SF(op) ((op >> 19) & 1) |
| 167 | #define GTE_MX(op) ((op >> 17) & 3) |
| 168 | #define GTE_V(op) ((op >> 15) & 3) |
| 169 | #define GTE_CV(op) ((op >> 13) & 3) |
| 170 | #define GTE_CD(op) ((op >> 11) & 3) /* not used */ |
| 171 | #define GTE_LM(op) ((op >> 10) & 1) |
| 172 | #define GTE_CT(op) ((op >> 6) & 15) /* not used */ |
| 173 | #define GTE_FUNCT(op) (op & 63) |
| 174 | |
| 175 | #define gteop (psxRegs.code & 0x1ffffff) |
| 176 | |
| 177 | #ifndef FLAGLESS |
| 178 | |
| 179 | static inline s64 BOUNDS_(psxCP2Regs *regs, s64 n_value, s64 n_max, int n_maxflag, s64 n_min, int n_minflag) { |
| 180 | if (n_value > n_max) { |
| 181 | gteFLAG |= n_maxflag; |
| 182 | } else if (n_value < n_min) { |
| 183 | gteFLAG |= n_minflag; |
| 184 | } |
| 185 | return n_value; |
| 186 | } |
| 187 | |
| 188 | static inline s32 LIM_(psxCP2Regs *regs, s32 value, s32 max, s32 min, u32 flag) { |
| 189 | s32 ret = value; |
| 190 | if (value > max) { |
| 191 | gteFLAG |= flag; |
| 192 | ret = max; |
| 193 | } else if (value < min) { |
| 194 | gteFLAG |= flag; |
| 195 | ret = min; |
| 196 | } |
| 197 | return ret; |
| 198 | } |
| 199 | |
| 200 | static inline u32 limE_(psxCP2Regs *regs, u32 result) { |
| 201 | if (result > 0x1ffff) { |
| 202 | gteFLAG |= (1 << 31) | (1 << 17); |
| 203 | return 0x1ffff; |
| 204 | } |
| 205 | return result; |
| 206 | } |
| 207 | |
| 208 | #else |
| 209 | |
| 210 | #define BOUNDS_(regs, a, ...) (a) |
| 211 | |
| 212 | static inline s32 LIM_(psxCP2Regs *regs, s32 value, s32 max, s32 min, u32 flag_unused) { |
| 213 | s32 ret = value; |
| 214 | if (value > max) |
| 215 | ret = max; |
| 216 | else if (value < min) |
| 217 | ret = min; |
| 218 | return ret; |
| 219 | } |
| 220 | |
| 221 | static inline u32 limE_(psxCP2Regs *regs, u32 result) { |
| 222 | if (result > 0x1ffff) |
| 223 | return 0x1ffff; |
| 224 | return result; |
| 225 | } |
| 226 | |
| 227 | #endif |
| 228 | |
| 229 | #define BOUNDS(n_value,n_max,n_maxflag,n_min,n_minflag) \ |
| 230 | BOUNDS_(regs,n_value,n_max,n_maxflag,n_min,n_minflag) |
| 231 | #define LIM(value,max,min,flag) \ |
| 232 | LIM_(regs,value,max,min,flag) |
| 233 | #define limE(result) \ |
| 234 | limE_(regs,result) |
| 235 | |
| 236 | #define A1(a) BOUNDS((a), 0x7fffffff, (1 << 30), -(s64)0x80000000, (1 << 31) | (1 << 27)) |
| 237 | #define A2(a) BOUNDS((a), 0x7fffffff, (1 << 29), -(s64)0x80000000, (1 << 31) | (1 << 26)) |
| 238 | #define A3(a) BOUNDS((a), 0x7fffffff, (1 << 28), -(s64)0x80000000, (1 << 31) | (1 << 25)) |
| 239 | #define limB1(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 31) | (1 << 24)) |
| 240 | #define limB2(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 31) | (1 << 23)) |
| 241 | #define limB3(a, l) LIM((a), 0x7fff, -0x8000 * !l, (1 << 22)) |
| 242 | #define limC1(a) LIM((a), 0x00ff, 0x0000, (1 << 21)) |
| 243 | #define limC2(a) LIM((a), 0x00ff, 0x0000, (1 << 20)) |
| 244 | #define limC3(a) LIM((a), 0x00ff, 0x0000, (1 << 19)) |
| 245 | #define limD(a) LIM((a), 0xffff, 0x0000, (1 << 31) | (1 << 18)) |
| 246 | |
| 247 | #define F(a) BOUNDS((a), 0x7fffffff, (1 << 31) | (1 << 16), -(s64)0x80000000, (1 << 31) | (1 << 15)) |
| 248 | #define limG1(a) LIM((a), 0x3ff, -0x400, (1 << 31) | (1 << 14)) |
| 249 | #define limG2(a) LIM((a), 0x3ff, -0x400, (1 << 31) | (1 << 13)) |
| 250 | #define limH(a) LIM((a), 0x1000, 0x0000, (1 << 12)) |
| 251 | |
| 252 | #ifndef __arm__ |
| 253 | #define A1U A1 |
| 254 | #define A2U A2 |
| 255 | #define A3U A3 |
| 256 | #else |
| 257 | /* these are unlikely to be hit and usually waste cycles, don't want them on ARM */ |
| 258 | #define A1U(x) (x) |
| 259 | #define A2U(x) (x) |
| 260 | #define A3U(x) (x) |
| 261 | #endif |
| 262 | |
| 263 | //senquack - n param should be unsigned (will be 'gteH' reg which is u16) |
| 264 | #ifdef GTE_USE_NATIVE_DIVIDE |
| 265 | INLINE u32 DIVIDE(u16 n, u16 d) { |
| 266 | if (n < d * 2) { |
| 267 | return ((u32)n << 16) / d; |
| 268 | } |
| 269 | return 0xffffffff; |
| 270 | } |
| 271 | #else |
| 272 | #include "gte_divider.h" |
| 273 | #endif // GTE_USE_NATIVE_DIVIDE |
| 274 | |
| 275 | #ifndef FLAGLESS |
| 276 | |
| 277 | u32 MFC2(int reg) { |
| 278 | psxCP2Regs *regs = &psxRegs.CP2; |
| 279 | switch (reg) { |
| 280 | case 1: |
| 281 | case 3: |
| 282 | case 5: |
| 283 | case 8: |
| 284 | case 9: |
| 285 | case 10: |
| 286 | case 11: |
| 287 | psxRegs.CP2D.r[reg] = (s32)psxRegs.CP2D.p[reg].sw.l; |
| 288 | break; |
| 289 | |
| 290 | case 7: |
| 291 | case 16: |
| 292 | case 17: |
| 293 | case 18: |
| 294 | case 19: |
| 295 | psxRegs.CP2D.r[reg] = (u32)psxRegs.CP2D.p[reg].w.l; |
| 296 | break; |
| 297 | |
| 298 | case 15: |
| 299 | psxRegs.CP2D.r[reg] = gteSXY2; |
| 300 | break; |
| 301 | |
| 302 | case 28: |
| 303 | case 29: |
| 304 | psxRegs.CP2D.r[reg] = LIM(gteIR1 >> 7, 0x1f, 0, 0) | |
| 305 | (LIM(gteIR2 >> 7, 0x1f, 0, 0) << 5) | |
| 306 | (LIM(gteIR3 >> 7, 0x1f, 0, 0) << 10); |
| 307 | break; |
| 308 | } |
| 309 | return psxRegs.CP2D.r[reg]; |
| 310 | } |
| 311 | |
| 312 | void MTC2(u32 value, int reg) { |
| 313 | psxCP2Regs *regs = &psxRegs.CP2; |
| 314 | switch (reg) { |
| 315 | case 15: |
| 316 | gteSXY0 = gteSXY1; |
| 317 | gteSXY1 = gteSXY2; |
| 318 | gteSXY2 = value; |
| 319 | gteSXYP = value; |
| 320 | break; |
| 321 | |
| 322 | case 28: |
| 323 | gteIRGB = value; |
| 324 | gteIR1 = (value & 0x1f) << 7; |
| 325 | gteIR2 = (value & 0x3e0) << 2; |
| 326 | gteIR3 = (value & 0x7c00) >> 3; |
| 327 | break; |
| 328 | |
| 329 | case 30: |
| 330 | { |
| 331 | int a; |
| 332 | gteLZCS = value; |
| 333 | |
| 334 | a = gteLZCS; |
| 335 | if (a > 0) { |
| 336 | int i; |
| 337 | for (i = 31; (a & (1 << i)) == 0 && i >= 0; i--); |
| 338 | gteLZCR = 31 - i; |
| 339 | } else if (a < 0) { |
| 340 | int i; |
| 341 | a ^= 0xffffffff; |
| 342 | for (i = 31; (a & (1 << i)) == 0 && i >= 0; i--); |
| 343 | gteLZCR = 31 - i; |
| 344 | } else { |
| 345 | gteLZCR = 32; |
| 346 | } |
| 347 | } |
| 348 | break; |
| 349 | |
| 350 | case 31: |
| 351 | return; |
| 352 | |
| 353 | default: |
| 354 | psxRegs.CP2D.r[reg] = value; |
| 355 | } |
| 356 | } |
| 357 | |
| 358 | void CTC2(u32 value, int reg) { |
| 359 | switch (reg) { |
| 360 | case 4: |
| 361 | case 12: |
| 362 | case 20: |
| 363 | case 26: |
| 364 | case 27: |
| 365 | case 29: |
| 366 | case 30: |
| 367 | value = (s32)(s16)value; |
| 368 | break; |
| 369 | |
| 370 | case 31: |
| 371 | value = value & 0x7ffff000; |
| 372 | if (value & 0x7f87e000) value |= 0x80000000; |
| 373 | break; |
| 374 | } |
| 375 | |
| 376 | psxRegs.CP2C.r[reg] = value; |
| 377 | } |
| 378 | |
| 379 | void gteMFC2() { |
| 380 | psxRegs.cycle += 1; |
| 381 | if (!_Rt_) return; |
| 382 | psxRegs.GPR.r[_Rt_] = MFC2(_Rd_); |
| 383 | } |
| 384 | |
| 385 | void gteCFC2() { |
| 386 | psxRegs.cycle += 1; |
| 387 | if (!_Rt_) return; |
| 388 | psxRegs.GPR.r[_Rt_] = psxRegs.CP2C.r[_Rd_]; |
| 389 | } |
| 390 | |
| 391 | void gteMTC2() { |
| 392 | MTC2(psxRegs.GPR.r[_Rt_], _Rd_); |
| 393 | } |
| 394 | |
| 395 | void gteCTC2() { |
| 396 | CTC2(psxRegs.GPR.r[_Rt_], _Rd_); |
| 397 | } |
| 398 | |
| 399 | #define _oB_ (psxRegs.GPR.r[_Rs_] + _Imm_) |
| 400 | |
| 401 | void gteLWC2() { |
| 402 | MTC2(psxMemRead32(_oB_), _Rt_); |
| 403 | } |
| 404 | |
| 405 | void gteSWC2() { |
| 406 | //psxRegs.cycle += 1; |
| 407 | psxMemWrite32(_oB_, MFC2(_Rt_)); |
| 408 | } |
| 409 | |
| 410 | #endif // FLAGLESS |
| 411 | |
| 412 | #if 0 |
| 413 | #define DIVIDE DIVIDE_ |
| 414 | static u32 DIVIDE_(s16 n, u16 d) { |
| 415 | if (n >= 0 && n < d * 2) { |
| 416 | s32 n_ = n; |
| 417 | return ((n_ << 16) + d / 2) / d; |
| 418 | //return (u32)((float)(n_ << 16) / (float)d + (float)0.5); |
| 419 | } |
| 420 | return 0xffffffff; |
| 421 | } |
| 422 | #endif |
| 423 | |
| 424 | void gteRTPS(psxCP2Regs *regs) { |
| 425 | int quotient; |
| 426 | s64 tmp; |
| 427 | |
| 428 | #ifdef GTE_LOG |
| 429 | GTE_LOG("GTE RTPS\n"); |
| 430 | #endif |
| 431 | psxRegs.cycle += 15; |
| 432 | gteFLAG = 0; |
| 433 | |
| 434 | gteMAC1 = A1((((s64)gteTRX << 12) + (gteR11 * gteVX0) + (gteR12 * gteVY0) + (gteR13 * gteVZ0)) >> 12); |
| 435 | gteMAC2 = A2((((s64)gteTRY << 12) + (gteR21 * gteVX0) + (gteR22 * gteVY0) + (gteR23 * gteVZ0)) >> 12); |
| 436 | gteMAC3 = A3((((s64)gteTRZ << 12) + (gteR31 * gteVX0) + (gteR32 * gteVY0) + (gteR33 * gteVZ0)) >> 12); |
| 437 | gteIR1 = limB1(gteMAC1, 0); |
| 438 | gteIR2 = limB2(gteMAC2, 0); |
| 439 | gteIR3 = limB3(gteMAC3, 0); |
| 440 | gteSZ0 = gteSZ1; |
| 441 | gteSZ1 = gteSZ2; |
| 442 | gteSZ2 = gteSZ3; |
| 443 | gteSZ3 = limD(gteMAC3); |
| 444 | quotient = limE(DIVIDE(gteH, gteSZ3)); |
| 445 | gteSXY0 = gteSXY1; |
| 446 | gteSXY1 = gteSXY2; |
| 447 | gteSX2 = limG1(F((s64)gteOFX + ((s64)gteIR1 * quotient)) >> 16); |
| 448 | gteSY2 = limG2(F((s64)gteOFY + ((s64)gteIR2 * quotient)) >> 16); |
| 449 | |
| 450 | tmp = (s64)gteDQB + ((s64)gteDQA * quotient); |
| 451 | gteMAC0 = F(tmp); |
| 452 | gteIR0 = limH(tmp >> 12); |
| 453 | } |
| 454 | |
| 455 | void gteRTPT(psxCP2Regs *regs) { |
| 456 | int quotient; |
| 457 | int v; |
| 458 | s32 vx, vy, vz; |
| 459 | s64 tmp; |
| 460 | |
| 461 | #ifdef GTE_LOG |
| 462 | GTE_LOG("GTE RTPT\n"); |
| 463 | #endif |
| 464 | psxRegs.cycle += 23; |
| 465 | gteFLAG = 0; |
| 466 | |
| 467 | gteSZ0 = gteSZ3; |
| 468 | for (v = 0; v < 3; v++) { |
| 469 | vx = VX(v); |
| 470 | vy = VY(v); |
| 471 | vz = VZ(v); |
| 472 | gteMAC1 = A1((((s64)gteTRX << 12) + (gteR11 * vx) + (gteR12 * vy) + (gteR13 * vz)) >> 12); |
| 473 | gteMAC2 = A2((((s64)gteTRY << 12) + (gteR21 * vx) + (gteR22 * vy) + (gteR23 * vz)) >> 12); |
| 474 | gteMAC3 = A3((((s64)gteTRZ << 12) + (gteR31 * vx) + (gteR32 * vy) + (gteR33 * vz)) >> 12); |
| 475 | gteIR1 = limB1(gteMAC1, 0); |
| 476 | gteIR2 = limB2(gteMAC2, 0); |
| 477 | gteIR3 = limB3(gteMAC3, 0); |
| 478 | fSZ(v) = limD(gteMAC3); |
| 479 | quotient = limE(DIVIDE(gteH, fSZ(v))); |
| 480 | fSX(v) = limG1(F((s64)gteOFX + ((s64)gteIR1 * quotient)) >> 16); |
| 481 | fSY(v) = limG2(F((s64)gteOFY + ((s64)gteIR2 * quotient)) >> 16); |
| 482 | } |
| 483 | |
| 484 | tmp = (s64)gteDQB + ((s64)gteDQA * quotient); |
| 485 | gteMAC0 = F(tmp); |
| 486 | gteIR0 = limH(tmp >> 12); |
| 487 | } |
| 488 | |
| 489 | void gteMVMVA(psxCP2Regs *regs) { |
| 490 | int shift = 12 * GTE_SF(gteop); |
| 491 | int mx = GTE_MX(gteop); |
| 492 | int v = GTE_V(gteop); |
| 493 | int cv = GTE_CV(gteop); |
| 494 | int lm = GTE_LM(gteop); |
| 495 | s32 vx = VX(v); |
| 496 | s32 vy = VY(v); |
| 497 | s32 vz = VZ(v); |
| 498 | |
| 499 | #ifdef GTE_LOG |
| 500 | GTE_LOG("GTE MVMVA\n"); |
| 501 | #endif |
| 502 | gteFLAG = 0; |
| 503 | psxRegs.cycle += 8; |
| 504 | |
| 505 | gteMAC1 = A1((((s64)CV1(cv) << 12) + (MX11(mx) * vx) + (MX12(mx) * vy) + (MX13(mx) * vz)) >> shift); |
| 506 | gteMAC2 = A2((((s64)CV2(cv) << 12) + (MX21(mx) * vx) + (MX22(mx) * vy) + (MX23(mx) * vz)) >> shift); |
| 507 | gteMAC3 = A3((((s64)CV3(cv) << 12) + (MX31(mx) * vx) + (MX32(mx) * vy) + (MX33(mx) * vz)) >> shift); |
| 508 | |
| 509 | gteIR1 = limB1(gteMAC1, lm); |
| 510 | gteIR2 = limB2(gteMAC2, lm); |
| 511 | gteIR3 = limB3(gteMAC3, lm); |
| 512 | } |
| 513 | |
| 514 | void gteNCLIP(psxCP2Regs *regs) { |
| 515 | #ifdef GTE_LOG |
| 516 | GTE_LOG("GTE NCLIP\n"); |
| 517 | #endif |
| 518 | gteFLAG = 0; |
| 519 | psxRegs.cycle += 8; |
| 520 | |
| 521 | gteMAC0 = F((s64)gteSX0 * (gteSY1 - gteSY2) + |
| 522 | gteSX1 * (gteSY2 - gteSY0) + |
| 523 | gteSX2 * (gteSY0 - gteSY1)); |
| 524 | } |
| 525 | |
| 526 | void gteAVSZ3(psxCP2Regs *regs) { |
| 527 | #ifdef GTE_LOG |
| 528 | GTE_LOG("GTE AVSZ3\n"); |
| 529 | #endif |
| 530 | gteFLAG = 0; |
| 531 | psxRegs.cycle += 5; |
| 532 | |
| 533 | gteMAC0 = F((s64)gteZSF3 * (gteSZ1 + gteSZ2 + gteSZ3)); |
| 534 | gteOTZ = limD(gteMAC0 >> 12); |
| 535 | } |
| 536 | |
| 537 | void gteAVSZ4(psxCP2Regs *regs) { |
| 538 | #ifdef GTE_LOG |
| 539 | GTE_LOG("GTE AVSZ4\n"); |
| 540 | #endif |
| 541 | gteFLAG = 0; |
| 542 | psxRegs.cycle += 6; |
| 543 | |
| 544 | gteMAC0 = F((s64)gteZSF4 * (gteSZ0 + gteSZ1 + gteSZ2 + gteSZ3)); |
| 545 | gteOTZ = limD(gteMAC0 >> 12); |
| 546 | } |
| 547 | |
| 548 | void gteSQR(psxCP2Regs *regs) { |
| 549 | int shift = 12 * GTE_SF(gteop); |
| 550 | int lm = GTE_LM(gteop); |
| 551 | |
| 552 | #ifdef GTE_LOG |
| 553 | GTE_LOG("GTE SQR\n"); |
| 554 | #endif |
| 555 | gteFLAG = 0; |
| 556 | psxRegs.cycle += 5; |
| 557 | |
| 558 | gteMAC1 = (gteIR1 * gteIR1) >> shift; |
| 559 | gteMAC2 = (gteIR2 * gteIR2) >> shift; |
| 560 | gteMAC3 = (gteIR3 * gteIR3) >> shift; |
| 561 | gteIR1 = limB1(gteMAC1, lm); |
| 562 | gteIR2 = limB2(gteMAC2, lm); |
| 563 | gteIR3 = limB3(gteMAC3, lm); |
| 564 | } |
| 565 | |
| 566 | void gteNCCS(psxCP2Regs *regs) { |
| 567 | #ifdef GTE_LOG |
| 568 | GTE_LOG("GTE NCCS\n"); |
| 569 | #endif |
| 570 | gteFLAG = 0; |
| 571 | psxRegs.cycle += 17; |
| 572 | |
| 573 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
| 574 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; |
| 575 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; |
| 576 | gteIR1 = limB1(gteMAC1, 1); |
| 577 | gteIR2 = limB2(gteMAC2, 1); |
| 578 | gteIR3 = limB3(gteMAC3, 1); |
| 579 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 580 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 581 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 582 | gteIR1 = limB1(gteMAC1, 1); |
| 583 | gteIR2 = limB2(gteMAC2, 1); |
| 584 | gteIR3 = limB3(gteMAC3, 1); |
| 585 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
| 586 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; |
| 587 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; |
| 588 | gteIR1 = gteMAC1; |
| 589 | gteIR2 = gteMAC2; |
| 590 | gteIR3 = gteMAC3; |
| 591 | |
| 592 | gteRGB0 = gteRGB1; |
| 593 | gteRGB1 = gteRGB2; |
| 594 | gteCODE2 = gteCODE; |
| 595 | gteR2 = limC1(gteMAC1 >> 4); |
| 596 | gteG2 = limC2(gteMAC2 >> 4); |
| 597 | gteB2 = limC3(gteMAC3 >> 4); |
| 598 | } |
| 599 | |
| 600 | void gteNCCT(psxCP2Regs *regs) { |
| 601 | int v; |
| 602 | s32 vx, vy, vz; |
| 603 | |
| 604 | #ifdef GTE_LOG |
| 605 | GTE_LOG("GTE NCCT\n"); |
| 606 | #endif |
| 607 | gteFLAG = 0; |
| 608 | psxRegs.cycle += 39; |
| 609 | |
| 610 | for (v = 0; v < 3; v++) { |
| 611 | vx = VX(v); |
| 612 | vy = VY(v); |
| 613 | vz = VZ(v); |
| 614 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
| 615 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; |
| 616 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; |
| 617 | gteIR1 = limB1(gteMAC1, 1); |
| 618 | gteIR2 = limB2(gteMAC2, 1); |
| 619 | gteIR3 = limB3(gteMAC3, 1); |
| 620 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 621 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 622 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 623 | gteIR1 = limB1(gteMAC1, 1); |
| 624 | gteIR2 = limB2(gteMAC2, 1); |
| 625 | gteIR3 = limB3(gteMAC3, 1); |
| 626 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
| 627 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; |
| 628 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; |
| 629 | |
| 630 | gteRGB0 = gteRGB1; |
| 631 | gteRGB1 = gteRGB2; |
| 632 | gteCODE2 = gteCODE; |
| 633 | gteR2 = limC1(gteMAC1 >> 4); |
| 634 | gteG2 = limC2(gteMAC2 >> 4); |
| 635 | gteB2 = limC3(gteMAC3 >> 4); |
| 636 | } |
| 637 | gteIR1 = gteMAC1; |
| 638 | gteIR2 = gteMAC2; |
| 639 | gteIR3 = gteMAC3; |
| 640 | } |
| 641 | |
| 642 | void gteNCDS(psxCP2Regs *regs) { |
| 643 | #ifdef GTE_LOG |
| 644 | GTE_LOG("GTE NCDS\n"); |
| 645 | #endif |
| 646 | gteFLAG = 0; |
| 647 | psxRegs.cycle += 19; |
| 648 | |
| 649 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
| 650 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; |
| 651 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; |
| 652 | gteIR1 = limB1(gteMAC1, 1); |
| 653 | gteIR2 = limB2(gteMAC2, 1); |
| 654 | gteIR3 = limB3(gteMAC3, 1); |
| 655 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 656 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 657 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 658 | gteIR1 = limB1(gteMAC1, 1); |
| 659 | gteIR2 = limB2(gteMAC2, 1); |
| 660 | gteIR3 = limB3(gteMAC3, 1); |
| 661 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
| 662 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; |
| 663 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; |
| 664 | gteIR1 = limB1(gteMAC1, 1); |
| 665 | gteIR2 = limB2(gteMAC2, 1); |
| 666 | gteIR3 = limB3(gteMAC3, 1); |
| 667 | |
| 668 | gteRGB0 = gteRGB1; |
| 669 | gteRGB1 = gteRGB2; |
| 670 | gteCODE2 = gteCODE; |
| 671 | gteR2 = limC1(gteMAC1 >> 4); |
| 672 | gteG2 = limC2(gteMAC2 >> 4); |
| 673 | gteB2 = limC3(gteMAC3 >> 4); |
| 674 | } |
| 675 | |
| 676 | void gteNCDT(psxCP2Regs *regs) { |
| 677 | int v; |
| 678 | s32 vx, vy, vz; |
| 679 | |
| 680 | #ifdef GTE_LOG |
| 681 | GTE_LOG("GTE NCDT\n"); |
| 682 | #endif |
| 683 | gteFLAG = 0; |
| 684 | psxRegs.cycle += 44; |
| 685 | |
| 686 | for (v = 0; v < 3; v++) { |
| 687 | vx = VX(v); |
| 688 | vy = VY(v); |
| 689 | vz = VZ(v); |
| 690 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
| 691 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; |
| 692 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; |
| 693 | gteIR1 = limB1(gteMAC1, 1); |
| 694 | gteIR2 = limB2(gteMAC2, 1); |
| 695 | gteIR3 = limB3(gteMAC3, 1); |
| 696 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 697 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 698 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 699 | gteIR1 = limB1(gteMAC1, 1); |
| 700 | gteIR2 = limB2(gteMAC2, 1); |
| 701 | gteIR3 = limB3(gteMAC3, 1); |
| 702 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
| 703 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; |
| 704 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; |
| 705 | |
| 706 | gteRGB0 = gteRGB1; |
| 707 | gteRGB1 = gteRGB2; |
| 708 | gteCODE2 = gteCODE; |
| 709 | gteR2 = limC1(gteMAC1 >> 4); |
| 710 | gteG2 = limC2(gteMAC2 >> 4); |
| 711 | gteB2 = limC3(gteMAC3 >> 4); |
| 712 | } |
| 713 | gteIR1 = limB1(gteMAC1, 1); |
| 714 | gteIR2 = limB2(gteMAC2, 1); |
| 715 | gteIR3 = limB3(gteMAC3, 1); |
| 716 | } |
| 717 | |
| 718 | void gteOP(psxCP2Regs *regs) { |
| 719 | int shift = 12 * GTE_SF(gteop); |
| 720 | int lm = GTE_LM(gteop); |
| 721 | |
| 722 | #ifdef GTE_LOG |
| 723 | GTE_LOG("GTE OP\n"); |
| 724 | #endif |
| 725 | gteFLAG = 0; |
| 726 | psxRegs.cycle += 6; |
| 727 | |
| 728 | gteMAC1 = ((gteR22 * gteIR3) - (gteR33 * gteIR2)) >> shift; |
| 729 | gteMAC2 = ((gteR33 * gteIR1) - (gteR11 * gteIR3)) >> shift; |
| 730 | gteMAC3 = ((gteR11 * gteIR2) - (gteR22 * gteIR1)) >> shift; |
| 731 | gteIR1 = limB1(gteMAC1, lm); |
| 732 | gteIR2 = limB2(gteMAC2, lm); |
| 733 | gteIR3 = limB3(gteMAC3, lm); |
| 734 | } |
| 735 | |
| 736 | void gteDCPL(psxCP2Regs *regs) { |
| 737 | int lm = GTE_LM(gteop); |
| 738 | |
| 739 | s32 RIR1 = ((s32)gteR * gteIR1) >> 8; |
| 740 | s32 GIR2 = ((s32)gteG * gteIR2) >> 8; |
| 741 | s32 BIR3 = ((s32)gteB * gteIR3) >> 8; |
| 742 | |
| 743 | #ifdef GTE_LOG |
| 744 | GTE_LOG("GTE DCPL\n"); |
| 745 | #endif |
| 746 | gteFLAG = 0; |
| 747 | psxRegs.cycle += 8; |
| 748 | |
| 749 | gteMAC1 = RIR1 + ((gteIR0 * limB1(A1U((s64)gteRFC - RIR1), 0)) >> 12); |
| 750 | gteMAC2 = GIR2 + ((gteIR0 * limB1(A2U((s64)gteGFC - GIR2), 0)) >> 12); |
| 751 | gteMAC3 = BIR3 + ((gteIR0 * limB1(A3U((s64)gteBFC - BIR3), 0)) >> 12); |
| 752 | |
| 753 | gteIR1 = limB1(gteMAC1, lm); |
| 754 | gteIR2 = limB2(gteMAC2, lm); |
| 755 | gteIR3 = limB3(gteMAC3, lm); |
| 756 | |
| 757 | gteRGB0 = gteRGB1; |
| 758 | gteRGB1 = gteRGB2; |
| 759 | gteCODE2 = gteCODE; |
| 760 | gteR2 = limC1(gteMAC1 >> 4); |
| 761 | gteG2 = limC2(gteMAC2 >> 4); |
| 762 | gteB2 = limC3(gteMAC3 >> 4); |
| 763 | } |
| 764 | |
| 765 | void gteGPF(psxCP2Regs *regs) { |
| 766 | int shift = 12 * GTE_SF(gteop); |
| 767 | |
| 768 | #ifdef GTE_LOG |
| 769 | GTE_LOG("GTE GPF\n"); |
| 770 | #endif |
| 771 | gteFLAG = 0; |
| 772 | psxRegs.cycle += 5; |
| 773 | |
| 774 | gteMAC1 = (gteIR0 * gteIR1) >> shift; |
| 775 | gteMAC2 = (gteIR0 * gteIR2) >> shift; |
| 776 | gteMAC3 = (gteIR0 * gteIR3) >> shift; |
| 777 | gteIR1 = limB1(gteMAC1, 0); |
| 778 | gteIR2 = limB2(gteMAC2, 0); |
| 779 | gteIR3 = limB3(gteMAC3, 0); |
| 780 | |
| 781 | gteRGB0 = gteRGB1; |
| 782 | gteRGB1 = gteRGB2; |
| 783 | gteCODE2 = gteCODE; |
| 784 | gteR2 = limC1(gteMAC1 >> 4); |
| 785 | gteG2 = limC2(gteMAC2 >> 4); |
| 786 | gteB2 = limC3(gteMAC3 >> 4); |
| 787 | } |
| 788 | |
| 789 | void gteGPL(psxCP2Regs *regs) { |
| 790 | int shift = 12 * GTE_SF(gteop); |
| 791 | |
| 792 | #ifdef GTE_LOG |
| 793 | GTE_LOG("GTE GPL\n"); |
| 794 | #endif |
| 795 | gteFLAG = 0; |
| 796 | psxRegs.cycle += 5; |
| 797 | |
| 798 | gteMAC1 = A1((((s64)gteMAC1 << shift) + (gteIR0 * gteIR1)) >> shift); |
| 799 | gteMAC2 = A2((((s64)gteMAC2 << shift) + (gteIR0 * gteIR2)) >> shift); |
| 800 | gteMAC3 = A3((((s64)gteMAC3 << shift) + (gteIR0 * gteIR3)) >> shift); |
| 801 | gteIR1 = limB1(gteMAC1, 0); |
| 802 | gteIR2 = limB2(gteMAC2, 0); |
| 803 | gteIR3 = limB3(gteMAC3, 0); |
| 804 | |
| 805 | gteRGB0 = gteRGB1; |
| 806 | gteRGB1 = gteRGB2; |
| 807 | gteCODE2 = gteCODE; |
| 808 | gteR2 = limC1(gteMAC1 >> 4); |
| 809 | gteG2 = limC2(gteMAC2 >> 4); |
| 810 | gteB2 = limC3(gteMAC3 >> 4); |
| 811 | } |
| 812 | |
| 813 | void gteDPCS(psxCP2Regs *regs) { |
| 814 | int shift = 12 * GTE_SF(gteop); |
| 815 | |
| 816 | #ifdef GTE_LOG |
| 817 | GTE_LOG("GTE DPCS\n"); |
| 818 | #endif |
| 819 | gteFLAG = 0; |
| 820 | psxRegs.cycle += 8; |
| 821 | |
| 822 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U(((s64)gteRFC - (gteR << 4)) << (12 - shift)), 0))) >> 12; |
| 823 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U(((s64)gteGFC - (gteG << 4)) << (12 - shift)), 0))) >> 12; |
| 824 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U(((s64)gteBFC - (gteB << 4)) << (12 - shift)), 0))) >> 12; |
| 825 | |
| 826 | gteIR1 = limB1(gteMAC1, 0); |
| 827 | gteIR2 = limB2(gteMAC2, 0); |
| 828 | gteIR3 = limB3(gteMAC3, 0); |
| 829 | gteRGB0 = gteRGB1; |
| 830 | gteRGB1 = gteRGB2; |
| 831 | gteCODE2 = gteCODE; |
| 832 | gteR2 = limC1(gteMAC1 >> 4); |
| 833 | gteG2 = limC2(gteMAC2 >> 4); |
| 834 | gteB2 = limC3(gteMAC3 >> 4); |
| 835 | } |
| 836 | |
| 837 | void gteDPCT(psxCP2Regs *regs) { |
| 838 | int v; |
| 839 | |
| 840 | #ifdef GTE_LOG |
| 841 | GTE_LOG("GTE DPCT\n"); |
| 842 | #endif |
| 843 | gteFLAG = 0; |
| 844 | psxRegs.cycle += 17; |
| 845 | |
| 846 | for (v = 0; v < 3; v++) { |
| 847 | gteMAC1 = ((gteR0 << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR0 << 4)), 0))) >> 12; |
| 848 | gteMAC2 = ((gteG0 << 16) + (gteIR0 * limB1(A2U((s64)gteGFC - (gteG0 << 4)), 0))) >> 12; |
| 849 | gteMAC3 = ((gteB0 << 16) + (gteIR0 * limB1(A3U((s64)gteBFC - (gteB0 << 4)), 0))) >> 12; |
| 850 | |
| 851 | gteRGB0 = gteRGB1; |
| 852 | gteRGB1 = gteRGB2; |
| 853 | gteCODE2 = gteCODE; |
| 854 | gteR2 = limC1(gteMAC1 >> 4); |
| 855 | gteG2 = limC2(gteMAC2 >> 4); |
| 856 | gteB2 = limC3(gteMAC3 >> 4); |
| 857 | } |
| 858 | gteIR1 = limB1(gteMAC1, 0); |
| 859 | gteIR2 = limB2(gteMAC2, 0); |
| 860 | gteIR3 = limB3(gteMAC3, 0); |
| 861 | } |
| 862 | |
| 863 | void gteNCS(psxCP2Regs *regs) { |
| 864 | #ifdef GTE_LOG |
| 865 | GTE_LOG("GTE NCS\n"); |
| 866 | #endif |
| 867 | gteFLAG = 0; |
| 868 | psxRegs.cycle += 14; |
| 869 | |
| 870 | gteMAC1 = ((s64)(gteL11 * gteVX0) + (gteL12 * gteVY0) + (gteL13 * gteVZ0)) >> 12; |
| 871 | gteMAC2 = ((s64)(gteL21 * gteVX0) + (gteL22 * gteVY0) + (gteL23 * gteVZ0)) >> 12; |
| 872 | gteMAC3 = ((s64)(gteL31 * gteVX0) + (gteL32 * gteVY0) + (gteL33 * gteVZ0)) >> 12; |
| 873 | gteIR1 = limB1(gteMAC1, 1); |
| 874 | gteIR2 = limB2(gteMAC2, 1); |
| 875 | gteIR3 = limB3(gteMAC3, 1); |
| 876 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 877 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 878 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 879 | gteIR1 = limB1(gteMAC1, 1); |
| 880 | gteIR2 = limB2(gteMAC2, 1); |
| 881 | gteIR3 = limB3(gteMAC3, 1); |
| 882 | |
| 883 | gteRGB0 = gteRGB1; |
| 884 | gteRGB1 = gteRGB2; |
| 885 | gteCODE2 = gteCODE; |
| 886 | gteR2 = limC1(gteMAC1 >> 4); |
| 887 | gteG2 = limC2(gteMAC2 >> 4); |
| 888 | gteB2 = limC3(gteMAC3 >> 4); |
| 889 | } |
| 890 | |
| 891 | void gteNCT(psxCP2Regs *regs) { |
| 892 | int v; |
| 893 | s32 vx, vy, vz; |
| 894 | |
| 895 | #ifdef GTE_LOG |
| 896 | GTE_LOG("GTE NCT\n"); |
| 897 | #endif |
| 898 | gteFLAG = 0; |
| 899 | psxRegs.cycle += 30; |
| 900 | |
| 901 | for (v = 0; v < 3; v++) { |
| 902 | vx = VX(v); |
| 903 | vy = VY(v); |
| 904 | vz = VZ(v); |
| 905 | gteMAC1 = ((s64)(gteL11 * vx) + (gteL12 * vy) + (gteL13 * vz)) >> 12; |
| 906 | gteMAC2 = ((s64)(gteL21 * vx) + (gteL22 * vy) + (gteL23 * vz)) >> 12; |
| 907 | gteMAC3 = ((s64)(gteL31 * vx) + (gteL32 * vy) + (gteL33 * vz)) >> 12; |
| 908 | gteIR1 = limB1(gteMAC1, 1); |
| 909 | gteIR2 = limB2(gteMAC2, 1); |
| 910 | gteIR3 = limB3(gteMAC3, 1); |
| 911 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 912 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 913 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 914 | gteRGB0 = gteRGB1; |
| 915 | gteRGB1 = gteRGB2; |
| 916 | gteCODE2 = gteCODE; |
| 917 | gteR2 = limC1(gteMAC1 >> 4); |
| 918 | gteG2 = limC2(gteMAC2 >> 4); |
| 919 | gteB2 = limC3(gteMAC3 >> 4); |
| 920 | } |
| 921 | gteIR1 = limB1(gteMAC1, 1); |
| 922 | gteIR2 = limB2(gteMAC2, 1); |
| 923 | gteIR3 = limB3(gteMAC3, 1); |
| 924 | } |
| 925 | |
| 926 | void gteCC(psxCP2Regs *regs) { |
| 927 | #ifdef GTE_LOG |
| 928 | GTE_LOG("GTE CC\n"); |
| 929 | #endif |
| 930 | gteFLAG = 0; |
| 931 | psxRegs.cycle += 11; |
| 932 | |
| 933 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 934 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 935 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 936 | gteIR1 = limB1(gteMAC1, 1); |
| 937 | gteIR2 = limB2(gteMAC2, 1); |
| 938 | gteIR3 = limB3(gteMAC3, 1); |
| 939 | gteMAC1 = ((s32)gteR * gteIR1) >> 8; |
| 940 | gteMAC2 = ((s32)gteG * gteIR2) >> 8; |
| 941 | gteMAC3 = ((s32)gteB * gteIR3) >> 8; |
| 942 | gteIR1 = limB1(gteMAC1, 1); |
| 943 | gteIR2 = limB2(gteMAC2, 1); |
| 944 | gteIR3 = limB3(gteMAC3, 1); |
| 945 | |
| 946 | gteRGB0 = gteRGB1; |
| 947 | gteRGB1 = gteRGB2; |
| 948 | gteCODE2 = gteCODE; |
| 949 | gteR2 = limC1(gteMAC1 >> 4); |
| 950 | gteG2 = limC2(gteMAC2 >> 4); |
| 951 | gteB2 = limC3(gteMAC3 >> 4); |
| 952 | } |
| 953 | |
| 954 | void gteINTPL(psxCP2Regs *regs) { |
| 955 | int shift = 12 * GTE_SF(gteop); |
| 956 | int lm = GTE_LM(gteop); |
| 957 | |
| 958 | #ifdef GTE_LOG |
| 959 | GTE_LOG("GTE INTPL\n"); |
| 960 | #endif |
| 961 | gteFLAG = 0; |
| 962 | psxRegs.cycle += 8; |
| 963 | |
| 964 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))) >> shift; |
| 965 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))) >> shift; |
| 966 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))) >> shift; |
| 967 | gteIR1 = limB1(gteMAC1, lm); |
| 968 | gteIR2 = limB2(gteMAC2, lm); |
| 969 | gteIR3 = limB3(gteMAC3, lm); |
| 970 | gteRGB0 = gteRGB1; |
| 971 | gteRGB1 = gteRGB2; |
| 972 | gteCODE2 = gteCODE; |
| 973 | gteR2 = limC1(gteMAC1 >> 4); |
| 974 | gteG2 = limC2(gteMAC2 >> 4); |
| 975 | gteB2 = limC3(gteMAC3 >> 4); |
| 976 | } |
| 977 | |
| 978 | void gteCDP(psxCP2Regs *regs) { |
| 979 | #ifdef GTE_LOG |
| 980 | GTE_LOG("GTE CDP\n"); |
| 981 | #endif |
| 982 | gteFLAG = 0; |
| 983 | psxRegs.cycle += 13; |
| 984 | |
| 985 | gteMAC1 = A1((((s64)gteRBK << 12) + (gteLR1 * gteIR1) + (gteLR2 * gteIR2) + (gteLR3 * gteIR3)) >> 12); |
| 986 | gteMAC2 = A2((((s64)gteGBK << 12) + (gteLG1 * gteIR1) + (gteLG2 * gteIR2) + (gteLG3 * gteIR3)) >> 12); |
| 987 | gteMAC3 = A3((((s64)gteBBK << 12) + (gteLB1 * gteIR1) + (gteLB2 * gteIR2) + (gteLB3 * gteIR3)) >> 12); |
| 988 | gteIR1 = limB1(gteMAC1, 1); |
| 989 | gteIR2 = limB2(gteMAC2, 1); |
| 990 | gteIR3 = limB3(gteMAC3, 1); |
| 991 | gteMAC1 = (((gteR << 4) * gteIR1) + (gteIR0 * limB1(A1U((s64)gteRFC - ((gteR * gteIR1) >> 8)), 0))) >> 12; |
| 992 | gteMAC2 = (((gteG << 4) * gteIR2) + (gteIR0 * limB2(A2U((s64)gteGFC - ((gteG * gteIR2) >> 8)), 0))) >> 12; |
| 993 | gteMAC3 = (((gteB << 4) * gteIR3) + (gteIR0 * limB3(A3U((s64)gteBFC - ((gteB * gteIR3) >> 8)), 0))) >> 12; |
| 994 | gteIR1 = limB1(gteMAC1, 1); |
| 995 | gteIR2 = limB2(gteMAC2, 1); |
| 996 | gteIR3 = limB3(gteMAC3, 1); |
| 997 | |
| 998 | gteRGB0 = gteRGB1; |
| 999 | gteRGB1 = gteRGB2; |
| 1000 | gteCODE2 = gteCODE; |
| 1001 | gteR2 = limC1(gteMAC1 >> 4); |
| 1002 | gteG2 = limC2(gteMAC2 >> 4); |
| 1003 | gteB2 = limC3(gteMAC3 >> 4); |
| 1004 | } |
| 1005 | |
| 1006 | /* decomposed/parametrized versions for the recompiler */ |
| 1007 | |
| 1008 | #ifndef FLAGLESS |
| 1009 | |
| 1010 | void gteSQR_part_noshift(psxCP2Regs *regs) { |
| 1011 | gteFLAG = 0; |
| 1012 | |
| 1013 | gteMAC1 = gteIR1 * gteIR1; |
| 1014 | gteMAC2 = gteIR2 * gteIR2; |
| 1015 | gteMAC3 = gteIR3 * gteIR3; |
| 1016 | } |
| 1017 | |
| 1018 | void gteSQR_part_shift(psxCP2Regs *regs) { |
| 1019 | gteFLAG = 0; |
| 1020 | |
| 1021 | gteMAC1 = (gteIR1 * gteIR1) >> 12; |
| 1022 | gteMAC2 = (gteIR2 * gteIR2) >> 12; |
| 1023 | gteMAC3 = (gteIR3 * gteIR3) >> 12; |
| 1024 | } |
| 1025 | |
| 1026 | void gteOP_part_noshift(psxCP2Regs *regs) { |
| 1027 | gteFLAG = 0; |
| 1028 | |
| 1029 | gteMAC1 = (gteR22 * gteIR3) - (gteR33 * gteIR2); |
| 1030 | gteMAC2 = (gteR33 * gteIR1) - (gteR11 * gteIR3); |
| 1031 | gteMAC3 = (gteR11 * gteIR2) - (gteR22 * gteIR1); |
| 1032 | } |
| 1033 | |
| 1034 | void gteOP_part_shift(psxCP2Regs *regs) { |
| 1035 | gteFLAG = 0; |
| 1036 | |
| 1037 | gteMAC1 = ((gteR22 * gteIR3) - (gteR33 * gteIR2)) >> 12; |
| 1038 | gteMAC2 = ((gteR33 * gteIR1) - (gteR11 * gteIR3)) >> 12; |
| 1039 | gteMAC3 = ((gteR11 * gteIR2) - (gteR22 * gteIR1)) >> 12; |
| 1040 | } |
| 1041 | |
| 1042 | void gteDCPL_part(psxCP2Regs *regs) { |
| 1043 | s32 RIR1 = ((s32)gteR * gteIR1) >> 8; |
| 1044 | s32 GIR2 = ((s32)gteG * gteIR2) >> 8; |
| 1045 | s32 BIR3 = ((s32)gteB * gteIR3) >> 8; |
| 1046 | |
| 1047 | gteFLAG = 0; |
| 1048 | |
| 1049 | gteMAC1 = RIR1 + ((gteIR0 * limB1(A1U((s64)gteRFC - RIR1), 0)) >> 12); |
| 1050 | gteMAC2 = GIR2 + ((gteIR0 * limB1(A2U((s64)gteGFC - GIR2), 0)) >> 12); |
| 1051 | gteMAC3 = BIR3 + ((gteIR0 * limB1(A3U((s64)gteBFC - BIR3), 0)) >> 12); |
| 1052 | } |
| 1053 | |
| 1054 | void gteGPF_part_noshift(psxCP2Regs *regs) { |
| 1055 | gteFLAG = 0; |
| 1056 | |
| 1057 | gteMAC1 = gteIR0 * gteIR1; |
| 1058 | gteMAC2 = gteIR0 * gteIR2; |
| 1059 | gteMAC3 = gteIR0 * gteIR3; |
| 1060 | } |
| 1061 | |
| 1062 | void gteGPF_part_shift(psxCP2Regs *regs) { |
| 1063 | gteFLAG = 0; |
| 1064 | |
| 1065 | gteMAC1 = (gteIR0 * gteIR1) >> 12; |
| 1066 | gteMAC2 = (gteIR0 * gteIR2) >> 12; |
| 1067 | gteMAC3 = (gteIR0 * gteIR3) >> 12; |
| 1068 | } |
| 1069 | |
| 1070 | #endif // !FLAGLESS |
| 1071 | |
| 1072 | void gteGPL_part_noshift(psxCP2Regs *regs) { |
| 1073 | gteFLAG = 0; |
| 1074 | |
| 1075 | gteMAC1 = A1((s64)gteMAC1 + (gteIR0 * gteIR1)); |
| 1076 | gteMAC2 = A2((s64)gteMAC2 + (gteIR0 * gteIR2)); |
| 1077 | gteMAC3 = A3((s64)gteMAC3 + (gteIR0 * gteIR3)); |
| 1078 | } |
| 1079 | |
| 1080 | void gteGPL_part_shift(psxCP2Regs *regs) { |
| 1081 | gteFLAG = 0; |
| 1082 | |
| 1083 | gteMAC1 = A1((s64)gteMAC1 + ((gteIR0 * gteIR1) >> 12)); |
| 1084 | gteMAC2 = A2((s64)gteMAC2 + ((gteIR0 * gteIR2) >> 12)); |
| 1085 | gteMAC3 = A3((s64)gteMAC3 + ((gteIR0 * gteIR3) >> 12)); |
| 1086 | } |
| 1087 | |
| 1088 | void gteDPCS_part_noshift(psxCP2Regs *regs) { |
| 1089 | int shift = 0; |
| 1090 | |
| 1091 | gteFLAG = 0; |
| 1092 | |
| 1093 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR << 4)) << (12 - shift), 0))) >> 12; |
| 1094 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U((s64)gteGFC - (gteG << 4)) << (12 - shift), 0))) >> 12; |
| 1095 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U((s64)gteBFC - (gteB << 4)) << (12 - shift), 0))) >> 12; |
| 1096 | } |
| 1097 | |
| 1098 | void gteDPCS_part_shift(psxCP2Regs *regs) { |
| 1099 | int shift = 12; |
| 1100 | |
| 1101 | gteFLAG = 0; |
| 1102 | |
| 1103 | gteMAC1 = ((gteR << 16) + (gteIR0 * limB1(A1U((s64)gteRFC - (gteR << 4)) << (12 - shift), 0))) >> 12; |
| 1104 | gteMAC2 = ((gteG << 16) + (gteIR0 * limB2(A2U((s64)gteGFC - (gteG << 4)) << (12 - shift), 0))) >> 12; |
| 1105 | gteMAC3 = ((gteB << 16) + (gteIR0 * limB3(A3U((s64)gteBFC - (gteB << 4)) << (12 - shift), 0))) >> 12; |
| 1106 | } |
| 1107 | |
| 1108 | void gteINTPL_part_noshift(psxCP2Regs *regs) { |
| 1109 | gteFLAG = 0; |
| 1110 | |
| 1111 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))); |
| 1112 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))); |
| 1113 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))); |
| 1114 | } |
| 1115 | |
| 1116 | void gteINTPL_part_shift(psxCP2Regs *regs) { |
| 1117 | gteFLAG = 0; |
| 1118 | |
| 1119 | gteMAC1 = ((gteIR1 << 12) + (gteIR0 * limB1(A1U((s64)gteRFC - gteIR1), 0))) >> 12; |
| 1120 | gteMAC2 = ((gteIR2 << 12) + (gteIR0 * limB2(A2U((s64)gteGFC - gteIR2), 0))) >> 12; |
| 1121 | gteMAC3 = ((gteIR3 << 12) + (gteIR0 * limB3(A3U((s64)gteBFC - gteIR3), 0))) >> 12; |
| 1122 | } |
| 1123 | |
| 1124 | void gteMACtoRGB(psxCP2Regs *regs) { |
| 1125 | gteRGB0 = gteRGB1; |
| 1126 | gteRGB1 = gteRGB2; |
| 1127 | gteCODE2 = gteCODE; |
| 1128 | gteR2 = limC1(gteMAC1 >> 4); |
| 1129 | gteG2 = limC2(gteMAC2 >> 4); |
| 1130 | gteB2 = limC3(gteMAC3 >> 4); |
| 1131 | } |
| 1132 | |