/************************************************************************** * * Copyright 2008 VMware, Inc. * All Rights Reserved. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the * "Software"), to deal in the Software without restriction, including * without limitation the rights to use, copy, modify, merge, publish, * distribute, sub license, and/or sell copies of the Software, and to * permit persons to whom the Software is furnished to do so, subject to * the following conditions: * * The above copyright notice and this permission notice (including the * next paragraph) shall be included in all copies or substantial portions * of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. * IN NO EVENT SHALL VMWARE AND/OR ITS SUPPLIERS BE LIABLE FOR * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. * **************************************************************************/ /** * Math utilities and approximations for common math functions. * Reduced precision is usually acceptable in shaders... * * "fast" is used in the names of functions which are low-precision, * or at least lower-precision than the normal C lib functions. */ #ifndef U_MATH_H #define U_MATH_H #include "pipe/p_compiler.h" #ifdef __cplusplus extern "C" { #endif #include #include #include #ifdef PIPE_OS_UNIX #include /* for ffs */ #endif #ifndef M_SQRT2 #define M_SQRT2 1.41421356237309504880 #endif #if defined(_MSC_VER) #if _MSC_VER < 1400 && !defined(__cplusplus) static INLINE float cosf( float f ) { return (float) cos( (double) f ); } static INLINE float sinf( float f ) { return (float) sin( (double) f ); } static INLINE float ceilf( float f ) { return (float) ceil( (double) f ); } static INLINE float floorf( float f ) { return (float) floor( (double) f ); } static INLINE float powf( float f, float g ) { return (float) pow( (double) f, (double) g ); } static INLINE float sqrtf( float f ) { return (float) sqrt( (double) f ); } static INLINE float fabsf( float f ) { return (float) fabs( (double) f ); } static INLINE float logf( float f ) { return (float) log( (double) f ); } #else /* Work-around an extra semi-colon in VS 2005 logf definition */ #ifdef logf #undef logf #define logf(x) ((float)log((double)(x))) #endif /* logf */ #if _MSC_VER < 1800 #define isfinite(x) _finite((double)(x)) #define isnan(x) _isnan((double)(x)) #endif /* _MSC_VER < 1800 */ #endif /* _MSC_VER < 1400 && !defined(__cplusplus) */ #if _MSC_VER < 1800 static INLINE double log2( double x ) { const double invln2 = 1.442695041; return log( x ) * invln2; } static INLINE double round(double x) { return x >= 0.0 ? floor(x + 0.5) : ceil(x - 0.5); } static INLINE float roundf(float x) { return x >= 0.0f ? floorf(x + 0.5f) : ceilf(x - 0.5f); } #endif #ifndef INFINITY #define INFINITY (DBL_MAX + DBL_MAX) #endif #ifndef NAN #define NAN (INFINITY - INFINITY) #endif #endif /* _MSC_VER */ #if __STDC_VERSION__ < 199901L && (!defined(__cplusplus) || defined(_MSC_VER)) static INLINE long int lrint(double d) { long int rounded = (long int)(d + 0.5); if (d - floor(d) == 0.5) { if (rounded % 2 != 0) rounded += (d > 0) ? -1 : 1; } return rounded; } static INLINE long int lrintf(float f) { long int rounded = (long int)(f + 0.5f); if (f - floorf(f) == 0.5f) { if (rounded % 2 != 0) rounded += (f > 0) ? -1 : 1; } return rounded; } static INLINE long long int llrint(double d) { long long int rounded = (long long int)(d + 0.5); if (d - floor(d) == 0.5) { if (rounded % 2 != 0) rounded += (d > 0) ? -1 : 1; } return rounded; } static INLINE long long int llrintf(float f) { long long int rounded = (long long int)(f + 0.5f); if (f - floorf(f) == 0.5f) { if (rounded % 2 != 0) rounded += (f > 0) ? -1 : 1; } return rounded; } #endif /* C99 */ #define POW2_TABLE_SIZE_LOG2 9 #define POW2_TABLE_SIZE (1 << POW2_TABLE_SIZE_LOG2) #define POW2_TABLE_OFFSET (POW2_TABLE_SIZE/2) #define POW2_TABLE_SCALE ((float)(POW2_TABLE_SIZE/2)) extern float pow2_table[POW2_TABLE_SIZE]; /** * Initialize math module. This should be called before using any * other functions in this module. */ extern void util_init_math(void); union fi { float f; int32_t i; uint32_t ui; }; union di { double d; int64_t i; uint64_t ui; }; /** * Extract the IEEE float32 exponent. */ static INLINE signed util_get_float32_exponent(float x) { union fi f; f.f = x; return ((f.ui >> 23) & 0xff) - 127; } /** * Fast version of 2^x * Identity: exp2(a + b) = exp2(a) * exp2(b) * Let ipart = int(x) * Let fpart = x - ipart; * So, exp2(x) = exp2(ipart) * exp2(fpart) * Compute exp2(ipart) with i << ipart * Compute exp2(fpart) with lookup table. */ static INLINE float util_fast_exp2(float x) { int32_t ipart; float fpart, mpart; union fi epart; if(x > 129.00000f) return 3.402823466e+38f; if (x < -126.99999f) return 0.0f; ipart = (int32_t) x; fpart = x - (float) ipart; /* same as * epart.f = (float) (1 << ipart) * but faster and without integer overflow for ipart > 31 */ epart.i = (ipart + 127 ) << 23; mpart = pow2_table[POW2_TABLE_OFFSET + (int)(fpart * POW2_TABLE_SCALE)]; return epart.f * mpart; } /** * Fast approximation to exp(x). */ static INLINE float util_fast_exp(float x) { const float k = 1.44269f; /* = log2(e) */ return util_fast_exp2(k * x); } #define LOG2_TABLE_SIZE_LOG2 16 #define LOG2_TABLE_SCALE (1 << LOG2_TABLE_SIZE_LOG2) #define LOG2_TABLE_SIZE (LOG2_TABLE_SCALE + 1) extern float log2_table[LOG2_TABLE_SIZE]; /** * Fast approximation to log2(x). */ static INLINE float util_fast_log2(float x) { union fi num; float epart, mpart; num.f = x; epart = (float)(((num.i & 0x7f800000) >> 23) - 127); /* mpart = log2_table[mantissa*LOG2_TABLE_SCALE + 0.5] */ mpart = log2_table[((num.i & 0x007fffff) + (1 << (22 - LOG2_TABLE_SIZE_LOG2))) >> (23 - LOG2_TABLE_SIZE_LOG2)]; return epart + mpart; } /** * Fast approximation to x^y. */ static INLINE float util_fast_pow(float x, float y) { return util_fast_exp2(util_fast_log2(x) * y); } /* Note that this counts zero as a power of two. */ static INLINE boolean util_is_power_of_two( unsigned v ) { return (v & (v-1)) == 0; } /** * Floor(x), returned as int. */ static INLINE int util_ifloor(float f) { int ai, bi; double af, bf; union fi u; af = (3 << 22) + 0.5 + (double) f; bf = (3 << 22) + 0.5 - (double) f; u.f = (float) af; ai = u.i; u.f = (float) bf; bi = u.i; return (ai - bi) >> 1; } /** * Round float to nearest int. */ static INLINE int util_iround(float f) { #if defined(PIPE_CC_GCC) && defined(PIPE_ARCH_X86) int r; __asm__ ("fistpl %0" : "=m" (r) : "t" (f) : "st"); return r; #elif defined(PIPE_CC_MSVC) && defined(PIPE_ARCH_X86) int r; _asm { fld f fistp r } return r; #else if (f >= 0.0f) return (int) (f + 0.5f); else return (int) (f - 0.5f); #endif } /** * Approximate floating point comparison */ static INLINE boolean util_is_approx(float a, float b, float tol) { return fabs(b - a) <= tol; } /** * util_is_X_inf_or_nan = test if x is NaN or +/- Inf * util_is_X_nan = test if x is NaN * util_X_inf_sign = return +1 for +Inf, -1 for -Inf, or 0 for not Inf * * NaN can be checked with x != x, however this fails with the fast math flag **/ /** * Single-float */ static INLINE boolean util_is_inf_or_nan(float x) { union fi tmp; tmp.f = x; return (tmp.ui & 0x7f800000) == 0x7f800000; } static INLINE boolean util_is_nan(float x) { union fi tmp; tmp.f = x; return (tmp.ui & 0x7fffffff) > 0x7f800000; } static INLINE int util_inf_sign(float x) { union fi tmp; tmp.f = x; if ((tmp.ui & 0x7fffffff) != 0x7f800000) { return 0; } return (x < 0) ? -1 : 1; } /** * Double-float */ static INLINE boolean util_is_double_inf_or_nan(double x) { union di tmp; tmp.d = x; return (tmp.ui & 0x7ff0000000000000ULL) == 0x7ff0000000000000ULL; } static INLINE boolean util_is_double_nan(double x) { union di tmp; tmp.d = x; return (tmp.ui & 0x7fffffffffffffffULL) > 0x7ff0000000000000ULL; } static INLINE int util_double_inf_sign(double x) { union di tmp; tmp.d = x; if ((tmp.ui & 0x7fffffffffffffffULL) != 0x7ff0000000000000ULL) { return 0; } return (x < 0) ? -1 : 1; } /** * Half-float */ static INLINE boolean util_is_half_inf_or_nan(int16_t x) { return (x & 0x7c00) == 0x7c00; } static INLINE boolean util_is_half_nan(int16_t x) { return (x & 0x7fff) > 0x7c00; } static INLINE int util_half_inf_sign(int16_t x) { if ((x & 0x7fff) != 0x7c00) { return 0; } return (x < 0) ? -1 : 1; } /** * Find first bit set in word. Least significant bit is 1. * Return 0 if no bits set. */ #ifndef FFS_DEFINED #define FFS_DEFINED 1 #if defined(_MSC_VER) && _MSC_VER >= 1300 && (_M_IX86 || _M_AMD64 || _M_IA64) unsigned char _BitScanForward(unsigned long* Index, unsigned long Mask); #pragma intrinsic(_BitScanForward) static INLINE unsigned long ffs( unsigned long u ) { unsigned long i; if (_BitScanForward(&i, u)) return i + 1; else return 0; } #elif defined(PIPE_CC_MSVC) && defined(PIPE_ARCH_X86) static INLINE unsigned ffs( unsigned u ) { unsigned i; if (u == 0) { return 0; } __asm bsf eax, [u] __asm inc eax __asm mov [i], eax return i; } #elif defined(__MINGW32__) || defined(PIPE_OS_ANDROID) #define ffs __builtin_ffs #endif #endif /* FFS_DEFINED */ /** * Find last bit set in a word. The least significant bit is 1. * Return 0 if no bits are set. */ static INLINE unsigned util_last_bit(unsigned u) { #if defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 304) return u == 0 ? 0 : 32 - __builtin_clz(u); #else unsigned r = 0; while (u) { r++; u >>= 1; } return r; #endif } /** * Find last bit in a word that does not match the sign bit. The least * significant bit is 1. * Return 0 if no bits are set. */ static INLINE unsigned util_last_bit_signed(int i) { #if defined(__GNUC__) && ((__GNUC__ * 100 + __GNUC_MINOR__) >= 407) && !defined(__INTEL_COMPILER) return 31 - __builtin_clrsb(i); #else if (i >= 0) return util_last_bit(i); else return util_last_bit(~(unsigned)i); #endif } /* Destructively loop over all of the bits in a mask as in: * * while (mymask) { * int i = u_bit_scan(&mymask); * ... process element i * } * */ static INLINE int u_bit_scan(unsigned *mask) { int i = ffs(*mask) - 1; *mask &= ~(1 << i); return i; } /** * Return float bits. */ static INLINE unsigned fui( float f ) { union fi fi; fi.f = f; return fi.ui; } static INLINE float uif(uint32_t ui) { union fi fi; fi.ui = ui; return fi.f; } /** * Convert ubyte to float in [0, 1]. * XXX a 256-entry lookup table would be slightly faster. */ static INLINE float ubyte_to_float(ubyte ub) { return (float) ub * (1.0f / 255.0f); } /** * Convert float in [0,1] to ubyte in [0,255] with clamping. */ static INLINE ubyte float_to_ubyte(float f) { union fi tmp; tmp.f = f; if (tmp.i < 0) { return (ubyte) 0; } else if (tmp.i >= 0x3f800000 /* 1.0f */) { return (ubyte) 255; } else { tmp.f = tmp.f * (255.0f/256.0f) + 32768.0f; return (ubyte) tmp.i; } } static INLINE float byte_to_float_tex(int8_t b) { return (b == -128) ? -1.0F : b * 1.0F / 127.0F; } static INLINE int8_t float_to_byte_tex(float f) { return (int8_t) (127.0F * f); } /** * Calc log base 2 */ static INLINE unsigned util_logbase2(unsigned n) { #if defined(PIPE_CC_GCC) && (PIPE_CC_GCC_VERSION >= 304) return ((sizeof(unsigned) * 8 - 1) - __builtin_clz(n | 1)); #else unsigned pos = 0; if (n >= 1<<16) { n >>= 16; pos += 16; } if (n >= 1<< 8) { n >>= 8; pos += 8; } if (n >= 1<< 4) { n >>= 4; pos += 4; } if (n >= 1<< 2) { n >>= 2; pos += 2; } if (n >= 1<< 1) { pos += 1; } return pos; #endif } /** * Returns the smallest power of two >= x */ static INLINE unsigned util_next_power_of_two(unsigned x) { #if defined(PIPE_CC_GCC) && (PIPE_CC_GCC_VERSION >= 304) if (x <= 1) return 1; return (1 << ((sizeof(unsigned) * 8) - __builtin_clz(x - 1))); #else unsigned val = x; if (x <= 1) return 1; if (util_is_power_of_two(x)) return x; val--; val = (val >> 1) | val; val = (val >> 2) | val; val = (val >> 4) | val; val = (val >> 8) | val; val = (val >> 16) | val; val++; return val; #endif } /** * Return number of bits set in n. */ static INLINE unsigned util_bitcount(unsigned n) { #if defined(PIPE_CC_GCC) && (PIPE_CC_GCC_VERSION >= 304) return __builtin_popcount(n); #else /* K&R classic bitcount. * * For each iteration, clear the LSB from the bitfield. * Requires only one iteration per set bit, instead of * one iteration per bit less than highest set bit. */ unsigned bits = 0; for (bits; n; bits++) { n &= n - 1; } return bits; #endif } /** * Reverse bits in n * Algorithm taken from: * http://stackoverflow.com/questions/9144800/c-reverse-bits-in-unsigned-integer */ static INLINE unsigned util_bitreverse(unsigned n) { n = ((n >> 1) & 0x55555555u) | ((n & 0x55555555u) << 1); n = ((n >> 2) & 0x33333333u) | ((n & 0x33333333u) << 2); n = ((n >> 4) & 0x0f0f0f0fu) | ((n & 0x0f0f0f0fu) << 4); n = ((n >> 8) & 0x00ff00ffu) | ((n & 0x00ff00ffu) << 8); n = ((n >> 16) & 0xffffu) | ((n & 0xffffu) << 16); return n; } /** * Convert from little endian to CPU byte order. */ #ifdef PIPE_ARCH_BIG_ENDIAN #define util_le64_to_cpu(x) util_bswap64(x) #define util_le32_to_cpu(x) util_bswap32(x) #define util_le16_to_cpu(x) util_bswap16(x) #else #define util_le64_to_cpu(x) (x) #define util_le32_to_cpu(x) (x) #define util_le16_to_cpu(x) (x) #endif #define util_cpu_to_le64(x) util_le64_to_cpu(x) #define util_cpu_to_le32(x) util_le32_to_cpu(x) #define util_cpu_to_le16(x) util_le16_to_cpu(x) /** * Reverse byte order of a 32 bit word. */ static INLINE uint32_t util_bswap32(uint32_t n) { /* We need the gcc version checks for non-autoconf build system */ #if defined(HAVE___BUILTIN_BSWAP32) || (defined(PIPE_CC_GCC) && (PIPE_CC_GCC_VERSION >= 403)) return __builtin_bswap32(n); #else return (n >> 24) | ((n >> 8) & 0x0000ff00) | ((n << 8) & 0x00ff0000) | (n << 24); #endif } /** * Reverse byte order of a 64bit word. */ static INLINE uint64_t util_bswap64(uint64_t n) { #if defined(HAVE___BUILTIN_BSWAP64) return __builtin_bswap64(n); #else return ((uint64_t)util_bswap32(n) << 32) | util_bswap32((n >> 32)); #endif } /** * Reverse byte order of a 16 bit word. */ static INLINE uint16_t util_bswap16(uint16_t n) { return (n >> 8) | (n << 8); } static INLINE void* util_memcpy_cpu_to_le32(void * restrict dest, const void * restrict src, size_t n) { #ifdef PIPE_ARCH_BIG_ENDIAN size_t i, e; assert(n % 4 == 0); for (i = 0, e = n / 4; i < e; i++) { uint32_t * restrict d = (uint32_t* restrict)dest; const uint32_t * restrict s = (const uint32_t* restrict)src; d[i] = util_bswap32(s[i]); } return dest; #else return memcpy(dest, src, n); #endif } /** * Clamp X to [MIN, MAX]. * This is a macro to allow float, int, uint, etc. types. */ #define CLAMP( X, MIN, MAX ) ( (X)<(MIN) ? (MIN) : ((X)>(MAX) ? (MAX) : (X)) ) #define MIN2( A, B ) ( (A)<(B) ? (A) : (B) ) #define MAX2( A, B ) ( (A)>(B) ? (A) : (B) ) #define MIN3( A, B, C ) ((A) < (B) ? MIN2(A, C) : MIN2(B, C)) #define MAX3( A, B, C ) ((A) > (B) ? MAX2(A, C) : MAX2(B, C)) #define MIN4( A, B, C, D ) ((A) < (B) ? MIN3(A, C, D) : MIN3(B, C, D)) #define MAX4( A, B, C, D ) ((A) > (B) ? MAX3(A, C, D) : MAX3(B, C, D)) /** * Align a value, only works pot alignemnts. */ static INLINE int align(int value, int alignment) { return (value + alignment - 1) & ~(alignment - 1); } /** * Works like align but on npot alignments. */ static INLINE size_t util_align_npot(size_t value, size_t alignment) { if (value % alignment) return value + (alignment - (value % alignment)); return value; } static INLINE unsigned u_minify(unsigned value, unsigned levels) { return MAX2(1, value >> levels); } #ifndef COPY_4V #define COPY_4V( DST, SRC ) \ do { \ (DST)[0] = (SRC)[0]; \ (DST)[1] = (SRC)[1]; \ (DST)[2] = (SRC)[2]; \ (DST)[3] = (SRC)[3]; \ } while (0) #endif #ifndef COPY_4FV #define COPY_4FV( DST, SRC ) COPY_4V(DST, SRC) #endif #ifndef ASSIGN_4V #define ASSIGN_4V( DST, V0, V1, V2, V3 ) \ do { \ (DST)[0] = (V0); \ (DST)[1] = (V1); \ (DST)[2] = (V2); \ (DST)[3] = (V3); \ } while (0) #endif static INLINE uint32_t util_unsigned_fixed(float value, unsigned frac_bits) { return value < 0 ? 0 : (uint32_t)(value * (1< 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 
/*
 * Mesa 3-D graphics library
 * Version:  7.7
 *
 * Copyright (C) 1999-2008  Brian Paul   All Rights Reserved.
 * Copyright (c) 2008-2009  VMware, Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a
 * copy of this software and associated documentation files (the "Software"),
 * to deal in the Software without restriction, including without limitation
 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
 * and/or sell copies of the Software, and to permit persons to whom the
 * Software is furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included
 * in all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
 * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
 * BRIAN PAUL BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN
 * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 */


/**
 * \file texfetch_tmp.h
 * Texel fetch functions template.
 * 
 * This template file is used by texfetch.c to generate texel fetch functions
 * for 1-D, 2-D and 3-D texture images. 
 *
 * It should be expanded by defining \p DIM as the number texture dimensions
 * (1, 2 or 3).  According to the value of \p DIM a series of macros is defined
 * for the texel lookup in the gl_texture_image::Data.
 * 
 * \author Gareth Hughes
 * \author Brian Paul
 */


#if DIM == 1

#define TEXEL_ADDR( type, image, i, j, k, size ) \
	((void) (j), (void) (k), ((type *)(image)->Map + (i) * (size)))

#define FETCH(x) fetch_texel_1d_##x

#elif DIM == 2

#define TEXEL_ADDR( type, image, i, j, k, size )			\
	((void) (k),							\
	 ((type *)(image)->Map + ((image)->RowStride * (j) + (i)) * (size)))

#define FETCH(x) fetch_texel_2d_##x

#elif DIM == 3

#define TEXEL_ADDR( type, image, i, j, k, size )			\
	((type *)(image)->Map + ((image)->ImageOffsets[k]		\
             + (image)->RowStride * (j) + (i)) * (size))

#define FETCH(x) fetch_texel_3d_##x

#else
#error	illegal number of texture dimensions
#endif


/* MESA_FORMAT_Z32 ***********************************************************/

/* Fetch depth texel from 1D, 2D or 3D 32-bit depth texture,
 * returning 1 GLfloat.
 * Note: no GLchan version of this function.
 */
static void FETCH(f_z32)( const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint *src = TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[0] = src[0] * (1.0F / 0xffffffff);
}


/* MESA_FORMAT_Z16 ***********************************************************/

/* Fetch depth texel from 1D, 2D or 3D 16-bit depth texture,
 * returning 1 GLfloat.
 * Note: no GLchan version of this function.
 */
static void FETCH(f_z16)(const struct swrast_texture_image *texImage,
                         GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[0] = src[0] * (1.0F / 65535.0F);
}



/* MESA_FORMAT_RGBA_F32 ******************************************************/

/* Fetch texel from 1D, 2D or 3D RGBA_FLOAT32 texture, returning 4 GLfloats.
 */
static void FETCH(f_rgba_f32)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 4);
   texel[RCOMP] = src[0];
   texel[GCOMP] = src[1];
   texel[BCOMP] = src[2];
   texel[ACOMP] = src[3];
}




/* MESA_FORMAT_RGBA_F16 ******************************************************/

/* Fetch texel from 1D, 2D or 3D RGBA_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_rgba_f16)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 4);
   texel[RCOMP] = _mesa_half_to_float(src[0]);
   texel[GCOMP] = _mesa_half_to_float(src[1]);
   texel[BCOMP] = _mesa_half_to_float(src[2]);
   texel[ACOMP] = _mesa_half_to_float(src[3]);
}



/* MESA_FORMAT_RGB_F32 *******************************************************/

/* Fetch texel from 1D, 2D or 3D RGB_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_rgb_f32)( const struct swrast_texture_image *texImage,
                              GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 3);
   texel[RCOMP] = src[0];
   texel[GCOMP] = src[1];
   texel[BCOMP] = src[2];
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_RGB_F16 *******************************************************/

/* Fetch texel from 1D, 2D or 3D RGB_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_rgb_f16)( const struct swrast_texture_image *texImage,
                              GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 3);
   texel[RCOMP] = _mesa_half_to_float(src[0]);
   texel[GCOMP] = _mesa_half_to_float(src[1]);
   texel[BCOMP] = _mesa_half_to_float(src[2]);
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_ALPHA_F32 *****************************************************/

/* Fetch texel from 1D, 2D or 3D ALPHA_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_alpha_f32)( const struct swrast_texture_image *texImage,
                              GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = src[0];
}




/* MESA_FORMAT_ALPHA_F32 *****************************************************/

/* Fetch texel from 1D, 2D or 3D ALPHA_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_alpha_f16)( const struct swrast_texture_image *texImage,
                              GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = _mesa_half_to_float(src[0]);
}




/* MESA_FORMAT_LUMINANCE_F32 *************************************************/

/* Fetch texel from 1D, 2D or 3D LUMINANCE_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_luminance_f32)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = src[0];
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_LUMINANCE_F16 *************************************************/

/* Fetch texel from 1D, 2D or 3D LUMINANCE_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_luminance_f16)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = _mesa_half_to_float(src[0]);
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_LUMINANCE_ALPHA_F32 *******************************************/

/* Fetch texel from 1D, 2D or 3D LUMINANCE_ALPHA_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_luminance_alpha_f32)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 2);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = src[0];
   texel[ACOMP] = src[1];
}




/* MESA_FORMAT_LUMINANCE_ALPHA_F16 *******************************************/

/* Fetch texel from 1D, 2D or 3D LUMINANCE_ALPHA_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_luminance_alpha_f16)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 2);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = _mesa_half_to_float(src[0]);
   texel[ACOMP] = _mesa_half_to_float(src[1]);
}




/* MESA_FORMAT_INTENSITY_F32 *************************************************/

/* Fetch texel from 1D, 2D or 3D INTENSITY_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_intensity_f32)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] =
   texel[ACOMP] = src[0];
}




/* MESA_FORMAT_INTENSITY_F16 *************************************************/

/* Fetch texel from 1D, 2D or 3D INTENSITY_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_intensity_f16)( const struct swrast_texture_image *texImage,
                                    GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] =
   texel[ACOMP] = _mesa_half_to_float(src[0]);
}




/* MESA_FORMAT_R_FLOAT32 *****************************************************/

/* Fetch texel from 1D, 2D or 3D R_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_r_f32)( const struct swrast_texture_image *texImage,
                            GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 1);
   texel[RCOMP] = src[0];
   texel[GCOMP] = 0.0F;
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_R_FLOAT16 *****************************************************/

/* Fetch texel from 1D, 2D or 3D R_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_r_f16)( const struct swrast_texture_image *texImage,
                            GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 1);
   texel[RCOMP] = _mesa_half_to_float(src[0]);
   texel[GCOMP] = 0.0F;
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_RG_FLOAT32 ****************************************************/

/* Fetch texel from 1D, 2D or 3D RG_FLOAT32 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_rg_f32)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLfloat *src = TEXEL_ADDR(GLfloat, texImage, i, j, k, 2);
   texel[RCOMP] = src[0];
   texel[GCOMP] = src[1];
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_RG_FLOAT16 ****************************************************/

/* Fetch texel from 1D, 2D or 3D RG_FLOAT16 texture,
 * returning 4 GLfloats.
 */
static void FETCH(f_rg_f16)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLhalfARB *src = TEXEL_ADDR(GLhalfARB, texImage, i, j, k, 2);
   texel[RCOMP] = _mesa_half_to_float(src[0]);
   texel[GCOMP] = _mesa_half_to_float(src[1]);
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = 1.0F;
}




/*
 * Begin Hardware formats
 */

/* MESA_FORMAT_RGBA8888 ******************************************************/

/* Fetch texel from 1D, 2D or 3D rgba8888 texture, return 4 GLfloats */
static void FETCH(f_rgba8888)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
}






/* MESA_FORMAT_RGBA888_REV ***************************************************/

/* Fetch texel from 1D, 2D or 3D abgr8888 texture, return 4 GLchans */
static void FETCH(f_rgba8888_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
}




/* MESA_FORMAT_ARGB8888 ******************************************************/

/* Fetch texel from 1D, 2D or 3D argb8888 texture, return 4 GLchans */
static void FETCH(f_argb8888)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
}




/* MESA_FORMAT_ARGB8888_REV **************************************************/

/* Fetch texel from 1D, 2D or 3D argb8888_rev texture, return 4 GLfloats */
static void FETCH(f_argb8888_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
}




/* MESA_FORMAT_RGBX8888 ******************************************************/

/* Fetch texel from 1D, 2D or 3D rgbx8888 texture, return 4 GLfloats */
static void FETCH(f_rgbx8888)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[ACOMP] = 1.0f;
}




/* MESA_FORMAT_RGBX888_REV ***************************************************/

/* Fetch texel from 1D, 2D or 3D rgbx8888_rev texture, return 4 GLchans */
static void FETCH(f_rgbx8888_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[ACOMP] = 1.0f;
}




/* MESA_FORMAT_XRGB8888 ******************************************************/

/* Fetch texel from 1D, 2D or 3D xrgb8888 texture, return 4 GLchans */
static void FETCH(f_xrgb8888)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff );
   texel[ACOMP] = 1.0f;
}




/* MESA_FORMAT_XRGB8888_REV **************************************************/

/* Fetch texel from 1D, 2D or 3D xrgb8888_rev texture, return 4 GLfloats */
static void FETCH(f_xrgb8888_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( (s >>  8) & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( (s >> 16) & 0xff );
   texel[BCOMP] = UBYTE_TO_FLOAT( (s >> 24)        );
   texel[ACOMP] = 1.0f;
}




/* MESA_FORMAT_RGB888 ********************************************************/

/* Fetch texel from 1D, 2D or 3D rgb888 texture, return 4 GLchans */
static void FETCH(f_rgb888)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 3);
   texel[RCOMP] = UBYTE_TO_FLOAT( src[2] );
   texel[GCOMP] = UBYTE_TO_FLOAT( src[1] );
   texel[BCOMP] = UBYTE_TO_FLOAT( src[0] );
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_BGR888 ********************************************************/

/* Fetch texel from 1D, 2D or 3D bgr888 texture, return 4 GLchans */
static void FETCH(f_bgr888)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 3);
   texel[RCOMP] = UBYTE_TO_FLOAT( src[0] );
   texel[GCOMP] = UBYTE_TO_FLOAT( src[1] );
   texel[BCOMP] = UBYTE_TO_FLOAT( src[2] );
   texel[ACOMP] = 1.0F;
}




/* use color expansion like (g << 2) | (g >> 4) (does somewhat random rounding)
   instead of slow (g << 2) * 255 / 252 (always rounds down) */

/* MESA_FORMAT_RGB565 ********************************************************/

/* Fetch texel from 1D, 2D or 3D rgb565 texture, return 4 GLchans */
static void FETCH(f_rgb565)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = *src;
   texel[RCOMP] = ((s >> 11) & 0x1f) * (1.0F / 31.0F);
   texel[GCOMP] = ((s >> 5 ) & 0x3f) * (1.0F / 63.0F);
   texel[BCOMP] = ((s      ) & 0x1f) * (1.0F / 31.0F);
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_RGB565_REV ****************************************************/

/* Fetch texel from 1D, 2D or 3D rgb565_rev texture, return 4 GLchans */
static void FETCH(f_rgb565_rev)( const struct swrast_texture_image *texImage,
                                 GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = (*src >> 8) | (*src << 8); /* byte swap */
   texel[RCOMP] = UBYTE_TO_FLOAT( ((s >> 8) & 0xf8) | ((s >> 13) & 0x7) );
   texel[GCOMP] = UBYTE_TO_FLOAT( ((s >> 3) & 0xfc) | ((s >>  9) & 0x3) );
   texel[BCOMP] = UBYTE_TO_FLOAT( ((s << 3) & 0xf8) | ((s >>  2) & 0x7) );
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_ARGB4444 ******************************************************/

/* Fetch texel from 1D, 2D or 3D argb444 texture, return 4 GLchans */
static void FETCH(f_argb4444)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = *src;
   texel[RCOMP] = ((s >>  8) & 0xf) * (1.0F / 15.0F);
   texel[GCOMP] = ((s >>  4) & 0xf) * (1.0F / 15.0F);
   texel[BCOMP] = ((s      ) & 0xf) * (1.0F / 15.0F);
   texel[ACOMP] = ((s >> 12) & 0xf) * (1.0F / 15.0F);
}




/* MESA_FORMAT_ARGB4444_REV **************************************************/

/* Fetch texel from 1D, 2D or 3D argb4444_rev texture, return 4 GLchans */
static void FETCH(f_argb4444_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = ((s      ) & 0xf) * (1.0F / 15.0F);
   texel[GCOMP] = ((s >> 12) & 0xf) * (1.0F / 15.0F);
   texel[BCOMP] = ((s >>  8) & 0xf) * (1.0F / 15.0F);
   texel[ACOMP] = ((s >>  4) & 0xf) * (1.0F / 15.0F);
}



/* MESA_FORMAT_RGBA5551 ******************************************************/

/* Fetch texel from 1D, 2D or 3D argb1555 texture, return 4 GLchans */
static void FETCH(f_rgba5551)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = *src;
   texel[RCOMP] = ((s >> 11) & 0x1f) * (1.0F / 31.0F);
   texel[GCOMP] = ((s >>  6) & 0x1f) * (1.0F / 31.0F);
   texel[BCOMP] = ((s >>  1) & 0x1f) * (1.0F / 31.0F);
   texel[ACOMP] = ((s      ) & 0x01) * 1.0F;
}



/* MESA_FORMAT_ARGB1555 ******************************************************/

/* Fetch texel from 1D, 2D or 3D argb1555 texture, return 4 GLchans */
static void FETCH(f_argb1555)( const struct swrast_texture_image *texImage,
			     GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = *src;
   texel[RCOMP] = ((s >> 10) & 0x1f) * (1.0F / 31.0F);
   texel[GCOMP] = ((s >>  5) & 0x1f) * (1.0F / 31.0F);
   texel[BCOMP] = ((s >>  0) & 0x1f) * (1.0F / 31.0F);
   texel[ACOMP] = ((s >> 15) & 0x01) * 1.0F;
}




/* MESA_FORMAT_ARGB1555_REV **************************************************/

/* Fetch texel from 1D, 2D or 3D argb1555_rev texture, return 4 GLchans */
static void FETCH(f_argb1555_rev)( const struct swrast_texture_image *texImage,
                                   GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   const GLushort s = (*src << 8) | (*src >> 8); /* byteswap */
   texel[RCOMP] = UBYTE_TO_FLOAT( ((s >>  7) & 0xf8) | ((s >> 12) & 0x7) );
   texel[GCOMP] = UBYTE_TO_FLOAT( ((s >>  2) & 0xf8) | ((s >>  7) & 0x7) );
   texel[BCOMP] = UBYTE_TO_FLOAT( ((s <<  3) & 0xf8) | ((s >>  2) & 0x7) );
   texel[ACOMP] = UBYTE_TO_FLOAT( ((s >> 15) & 0x01) * 255 );
}




/* MESA_FORMAT_ARGB2101010 ***************************************************/

/* Fetch texel from 1D, 2D or 3D argb2101010 texture, return 4 GLchans */
static void FETCH(f_argb2101010)( const struct swrast_texture_image *texImage,
                                  GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint *src = TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   const GLuint s = *src;
   texel[RCOMP] = ((s >> 20) & 0x3ff) * (1.0F / 1023.0F);
   texel[GCOMP] = ((s >> 10) & 0x3ff) * (1.0F / 1023.0F);
   texel[BCOMP] = ((s >>  0) & 0x3ff) * (1.0F / 1023.0F);
   texel[ACOMP] = ((s >> 30) & 0x03) * (1.0F / 3.0F);
}




/* MESA_FORMAT_GR88 **********************************************************/

/* Fetch texel from 1D, 2D or 3D rg88 texture, return 4 GLchans */
static void FETCH(f_gr88)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( s & 0xff );
   texel[GCOMP] = UBYTE_TO_FLOAT( s >> 8 );
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_RG88 ******************************************************/

/* Fetch texel from 1D, 2D or 3D rg88_rev texture, return 4 GLchans */
static void FETCH(f_rg88)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT( s >> 8 );
   texel[GCOMP] = UBYTE_TO_FLOAT( s & 0xff );
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_AL44 **********************************************************/

/* Fetch texel from 1D, 2D or 3D al44 texture, return 4 GLchans */
static void FETCH(f_al44)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte s = *TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = (s & 0xf) * (1.0F / 15.0F);
   texel[ACOMP] = ((s >> 4) & 0xf) * (1.0F / 15.0F);
}




/* MESA_FORMAT_AL88 **********************************************************/

/* Fetch texel from 1D, 2D or 3D al88 texture, return 4 GLchans */
static void FETCH(f_al88)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = 
   texel[GCOMP] = 
   texel[BCOMP] = UBYTE_TO_FLOAT( s & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( s >> 8 );
}




/* MESA_FORMAT_R8 ************************************************************/

/* Fetch texel from 1D, 2D or 3D rg88 texture, return 4 GLchans */
static void FETCH(f_r8)(const struct swrast_texture_image *texImage,
			GLint i, GLint j, GLint k, GLfloat *texel)
{
   const GLubyte s = *TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] = UBYTE_TO_FLOAT(s);
   texel[GCOMP] = 0.0;
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_R16 ***********************************************************/

/* Fetch texel from 1D, 2D or 3D r16 texture, return 4 GLchans */
static void FETCH(f_r16)(const struct swrast_texture_image *texImage,
			GLint i, GLint j, GLint k, GLfloat *texel)
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = USHORT_TO_FLOAT(s);
   texel[GCOMP] = 0.0;
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_AL88_REV ******************************************************/

/* Fetch texel from 1D, 2D or 3D al88_rev texture, return 4 GLchans */
static void FETCH(f_al88_rev)( const struct swrast_texture_image *texImage,
                               GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort s = *TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] = 
   texel[GCOMP] = 
   texel[BCOMP] = UBYTE_TO_FLOAT( s >> 8 );
   texel[ACOMP] = UBYTE_TO_FLOAT( s & 0xff );
}




/* MESA_FORMAT_RG1616 ********************************************************/

/* Fetch texel from 1D, 2D or 3D rg1616 texture, return 4 GLchans */
static void FETCH(f_rg1616)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = USHORT_TO_FLOAT( s & 0xffff );
   texel[GCOMP] = USHORT_TO_FLOAT( s >> 16 );
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_RG1616_REV ****************************************************/

/* Fetch texel from 1D, 2D or 3D rg1616_rev texture, return 4 GLchans */
static void FETCH(f_rg1616_rev)( const struct swrast_texture_image *texImage,
                           GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = USHORT_TO_FLOAT( s >> 16 );
   texel[GCOMP] = USHORT_TO_FLOAT( s & 0xffff );
   texel[BCOMP] = 0.0;
   texel[ACOMP] = 1.0;
}




/* MESA_FORMAT_AL1616 ********************************************************/

/* Fetch texel from 1D, 2D or 3D al1616 texture, return 4 GLchans */
static void FETCH(f_al1616)( const struct swrast_texture_image *texImage,
			     GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = USHORT_TO_FLOAT( s & 0xffff );
   texel[ACOMP] = USHORT_TO_FLOAT( s >> 16 );
}




/* MESA_FORMAT_AL1616_REV ****************************************************/

/* Fetch texel from 1D, 2D or 3D al1616_rev texture, return 4 GLchans */
static void FETCH(f_al1616_rev)( const struct swrast_texture_image *texImage,
				 GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = USHORT_TO_FLOAT( s >> 16 );
   texel[ACOMP] = USHORT_TO_FLOAT( s & 0xffff );
}




/* MESA_FORMAT_RGB332 ********************************************************/

/* Fetch texel from 1D, 2D or 3D rgb332 texture, return 4 GLchans */
static void FETCH(f_rgb332)( const struct swrast_texture_image *texImage,
                             GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   const GLubyte s = *src;
   texel[RCOMP] = ((s >> 5) & 0x7) * (1.0F / 7.0F);
   texel[GCOMP] = ((s >> 2) & 0x7) * (1.0F / 7.0F);
   texel[BCOMP] = ((s     ) & 0x3) * (1.0F / 3.0F);
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_A8 ************************************************************/

/* Fetch texel from 1D, 2D or 3D a8 texture, return 4 GLchans */
static void FETCH(f_a8)( const struct swrast_texture_image *texImage,
                         GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = UBYTE_TO_FLOAT( src[0] );
}




/* MESA_FORMAT_A16 ************************************************************/

/* Fetch texel from 1D, 2D or 3D a8 texture, return 4 GLchans */
static void FETCH(f_a16)( const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = 0.0F;
   texel[ACOMP] = USHORT_TO_FLOAT( src[0] );
}




/* MESA_FORMAT_L8 ************************************************************/

/* Fetch texel from 1D, 2D or 3D l8 texture, return 4 GLchans */
static void FETCH(f_l8)( const struct swrast_texture_image *texImage,
                         GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = UBYTE_TO_FLOAT( src[0] );
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_L16 ***********************************************************/

/* Fetch texel from 1D, 2D or 3D l16 texture, return 4 GLchans */
static void FETCH(f_l16)( const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = USHORT_TO_FLOAT( src[0] );
   texel[ACOMP] = 1.0F;
}




/* MESA_FORMAT_I8 ************************************************************/

/* Fetch texel from 1D, 2D or 3D i8 texture, return 4 GLchans */
static void FETCH(f_i8)( const struct swrast_texture_image *texImage,
                         GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] =
   texel[ACOMP] = UBYTE_TO_FLOAT( src[0] );
}




/* MESA_FORMAT_I16 ***********************************************************/

/* Fetch texel from 1D, 2D or 3D i16 texture, return 4 GLchans */
static void FETCH(f_i16)( const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLushort *src = TEXEL_ADDR(GLushort, texImage, i, j, k, 1);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] =
   texel[ACOMP] = USHORT_TO_FLOAT( src[0] );
}




/* Fetch texel from 1D, 2D or 3D srgb8 texture, return 4 GLfloats */
/* Note: component order is same as for MESA_FORMAT_RGB888 */
static void FETCH(srgb8)(const struct swrast_texture_image *texImage,
                         GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 3);
   texel[RCOMP] = nonlinear_to_linear(src[2]);
   texel[GCOMP] = nonlinear_to_linear(src[1]);
   texel[BCOMP] = nonlinear_to_linear(src[0]);
   texel[ACOMP] = 1.0F;
}



/* Fetch texel from 1D, 2D or 3D srgba8 texture, return 4 GLfloats */
static void FETCH(srgba8)(const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = nonlinear_to_linear( (s >> 24) );
   texel[GCOMP] = nonlinear_to_linear( (s >> 16) & 0xff );
   texel[BCOMP] = nonlinear_to_linear( (s >>  8) & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s      ) & 0xff ); /* linear! */
}



/* Fetch texel from 1D, 2D or 3D sargb8 texture, return 4 GLfloats */
static void FETCH(sargb8)(const struct swrast_texture_image *texImage,
                          GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLuint s = *TEXEL_ADDR(GLuint, texImage, i, j, k, 1);
   texel[RCOMP] = nonlinear_to_linear( (s >> 16) & 0xff );
   texel[GCOMP] = nonlinear_to_linear( (s >>  8) & 0xff );
   texel[BCOMP] = nonlinear_to_linear( (s      ) & 0xff );
   texel[ACOMP] = UBYTE_TO_FLOAT( (s >> 24) ); /* linear! */
}



/* Fetch texel from 1D, 2D or 3D sl8 texture, return 4 GLfloats */
static void FETCH(sl8)(const struct swrast_texture_image *texImage,
                       GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 1);
   texel[RCOMP] = 
   texel[GCOMP] = 
   texel[BCOMP] = nonlinear_to_linear(src[0]);
   texel[ACOMP] = 1.0F;
}



/* Fetch texel from 1D, 2D or 3D sla8 texture, return 4 GLfloats */
static void FETCH(sla8)(const struct swrast_texture_image *texImage,
                       GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLubyte *src = TEXEL_ADDR(GLubyte, texImage, i, j, k, 2);
   texel[RCOMP] =
   texel[GCOMP] =
   texel[BCOMP] = nonlinear_to_linear(src[0]);
   texel[ACOMP] = UBYTE_TO_FLOAT(src[1]); /* linear */
}




/* MESA_FORMAT_RGBA_INT8 **************************************************/

static void
FETCH(rgba_int8)(const struct swrast_texture_image *texImage,
                 GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLbyte *src = TEXEL_ADDR(GLbyte, texImage, i, j, k, 4);
   texel[RCOMP] = (GLfloat) src[0];
   texel[GCOMP] = (GLfloat) src[1];
   texel[BCOMP] = (GLfloat) src[2];
   texel[ACOMP] = (GLfloat) src[3];
}




/* MESA_FORMAT_RGBA_INT16 **************************************************/

static void
FETCH(rgba_int16)(const struct swrast_texture_image *texImage,
                  GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLshort *src = TEXEL_ADDR(GLshort, texImage, i, j, k, 4);
   texel[RCOMP] = (GLfloat) src[0];
   texel[GCOMP] = (GLfloat) src[1];
   texel[BCOMP] = (GLfloat) src[2];
   texel[ACOMP] = (GLfloat) src[3];
}




/* MESA_FORMAT_RGBA_INT32 **************************************************/

static void
FETCH(rgba_int32)(const struct swrast_texture_image *texImage,
                  GLint i, GLint j, GLint k, GLfloat *texel )
{
   const GLint *src = TEXEL_ADDR(GLint, texImage, i, j, k, 4);
   texel[RCOMP] = (GLfloat) src[0];
   texel[GCOMP] = (GLfloat) src[1];
   texel[BCOMP] = (GLfloat) src[2];
   texel[ACOMP] = (GLfloat) src[3];