+++ /dev/null
-/*
- * Mesa 3-D graphics library
- *
- * Copyright (C) 1999-2008 Brian Paul 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, 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
- * THE AUTHORS OR COPYRIGHT HOLDERS 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 prog_execute.c
- * Software interpreter for vertex/fragment programs.
- * \author Brian Paul
- */
-
-/*
- * NOTE: we do everything in single-precision floating point; we don't
- * currently observe the single/half/fixed-precision qualifiers.
- *
- */
-
-
-#include "c99_math.h"
-#include "main/errors.h"
-#include "main/glheader.h"
-#include "main/macros.h"
-#include "main/mtypes.h"
-#include "prog_execute.h"
-#include "prog_instruction.h"
-#include "prog_parameter.h"
-#include "prog_print.h"
-#include "prog_noise.h"
-
-
-/* debug predicate */
-#define DEBUG_PROG 0
-
-
-/**
- * Set x to positive or negative infinity.
- */
-#define SET_POS_INFINITY(x) \
- do { \
- fi_type fi; \
- fi.i = 0x7F800000; \
- x = fi.f; \
- } while (0)
-#define SET_NEG_INFINITY(x) \
- do { \
- fi_type fi; \
- fi.i = 0xFF800000; \
- x = fi.f; \
- } while (0)
-
-#define SET_FLOAT_BITS(x, bits) ((fi_type *) (void *) &(x))->i = bits
-
-
-static const GLfloat ZeroVec[4] = { 0.0F, 0.0F, 0.0F, 0.0F };
-
-
-/**
- * Return a pointer to the 4-element float vector specified by the given
- * source register.
- */
-static inline const GLfloat *
-get_src_register_pointer(const struct prog_src_register *source,
- const struct gl_program_machine *machine)
-{
- const struct gl_program *prog = machine->CurProgram;
- GLint reg = source->Index;
-
- if (source->RelAddr) {
- /* add address register value to src index/offset */
- reg += machine->AddressReg[0][0];
- if (reg < 0) {
- return ZeroVec;
- }
- }
-
- switch (source->File) {
- case PROGRAM_TEMPORARY:
- if (reg >= MAX_PROGRAM_TEMPS)
- return ZeroVec;
- return machine->Temporaries[reg];
-
- case PROGRAM_INPUT:
- if (prog->Target == GL_VERTEX_PROGRAM_ARB) {
- if (reg >= VERT_ATTRIB_MAX)
- return ZeroVec;
- return machine->VertAttribs[reg];
- }
- else {
- if (reg >= VARYING_SLOT_MAX)
- return ZeroVec;
- return machine->Attribs[reg][machine->CurElement];
- }
-
- case PROGRAM_OUTPUT:
- if (reg >= MAX_PROGRAM_OUTPUTS)
- return ZeroVec;
- return machine->Outputs[reg];
-
- case PROGRAM_STATE_VAR:
- FALLTHROUGH;
- case PROGRAM_CONSTANT:
- FALLTHROUGH;
- case PROGRAM_UNIFORM: {
- if (reg >= (GLint) prog->Parameters->NumParameters)
- return ZeroVec;
-
- unsigned pvo = prog->Parameters->Parameters[reg].ValueOffset;
- return (GLfloat *) prog->Parameters->ParameterValues + pvo;
- }
- case PROGRAM_SYSTEM_VALUE:
- assert(reg < (GLint) ARRAY_SIZE(machine->SystemValues));
- return machine->SystemValues[reg];
-
- default:
- _mesa_problem(NULL,
- "Invalid src register file %d in get_src_register_pointer()",
- source->File);
- return ZeroVec;
- }
-}
-
-
-/**
- * Return a pointer to the 4-element float vector specified by the given
- * destination register.
- */
-static inline GLfloat *
-get_dst_register_pointer(const struct prog_dst_register *dest,
- struct gl_program_machine *machine)
-{
- static GLfloat dummyReg[4];
- GLint reg = dest->Index;
-
- if (dest->RelAddr) {
- /* add address register value to src index/offset */
- reg += machine->AddressReg[0][0];
- if (reg < 0) {
- return dummyReg;
- }
- }
-
- switch (dest->File) {
- case PROGRAM_TEMPORARY:
- if (reg >= MAX_PROGRAM_TEMPS)
- return dummyReg;
- return machine->Temporaries[reg];
-
- case PROGRAM_OUTPUT:
- if (reg >= MAX_PROGRAM_OUTPUTS)
- return dummyReg;
- return machine->Outputs[reg];
-
- default:
- _mesa_problem(NULL,
- "Invalid dest register file %d in get_dst_register_pointer()",
- dest->File);
- return dummyReg;
- }
-}
-
-
-
-/**
- * Fetch a 4-element float vector from the given source register.
- * Apply swizzling and negating as needed.
- */
-static void
-fetch_vector4(const struct prog_src_register *source,
- const struct gl_program_machine *machine, GLfloat result[4])
-{
- const GLfloat *src = get_src_register_pointer(source, machine);
-
- if (source->Swizzle == SWIZZLE_NOOP) {
- /* no swizzling */
- COPY_4V(result, src);
- }
- else {
- assert(GET_SWZ(source->Swizzle, 0) <= 3);
- assert(GET_SWZ(source->Swizzle, 1) <= 3);
- assert(GET_SWZ(source->Swizzle, 2) <= 3);
- assert(GET_SWZ(source->Swizzle, 3) <= 3);
- result[0] = src[GET_SWZ(source->Swizzle, 0)];
- result[1] = src[GET_SWZ(source->Swizzle, 1)];
- result[2] = src[GET_SWZ(source->Swizzle, 2)];
- result[3] = src[GET_SWZ(source->Swizzle, 3)];
- }
-
- if (source->Negate) {
- assert(source->Negate == NEGATE_XYZW);
- result[0] = -result[0];
- result[1] = -result[1];
- result[2] = -result[2];
- result[3] = -result[3];
- }
-
-#ifdef NAN_CHECK
- assert(!util_is_inf_or_nan(result[0]));
- assert(!util_is_inf_or_nan(result[0]));
- assert(!util_is_inf_or_nan(result[0]));
- assert(!util_is_inf_or_nan(result[0]));
-#endif
-}
-
-
-/**
- * Fetch the derivative with respect to X or Y for the given register.
- * XXX this currently only works for fragment program input attribs.
- */
-static void
-fetch_vector4_deriv(const struct prog_src_register *source,
- const struct gl_program_machine *machine,
- char xOrY, GLfloat result[4])
-{
- if (source->File == PROGRAM_INPUT &&
- source->Index < (GLint) machine->NumDeriv) {
- const GLint col = machine->CurElement;
- const GLfloat w = machine->Attribs[VARYING_SLOT_POS][col][3];
- const GLfloat invQ = 1.0f / w;
- GLfloat deriv[4];
-
- if (xOrY == 'X') {
- deriv[0] = machine->DerivX[source->Index][0] * invQ;
- deriv[1] = machine->DerivX[source->Index][1] * invQ;
- deriv[2] = machine->DerivX[source->Index][2] * invQ;
- deriv[3] = machine->DerivX[source->Index][3] * invQ;
- }
- else {
- deriv[0] = machine->DerivY[source->Index][0] * invQ;
- deriv[1] = machine->DerivY[source->Index][1] * invQ;
- deriv[2] = machine->DerivY[source->Index][2] * invQ;
- deriv[3] = machine->DerivY[source->Index][3] * invQ;
- }
-
- result[0] = deriv[GET_SWZ(source->Swizzle, 0)];
- result[1] = deriv[GET_SWZ(source->Swizzle, 1)];
- result[2] = deriv[GET_SWZ(source->Swizzle, 2)];
- result[3] = deriv[GET_SWZ(source->Swizzle, 3)];
-
- if (source->Negate) {
- assert(source->Negate == NEGATE_XYZW);
- result[0] = -result[0];
- result[1] = -result[1];
- result[2] = -result[2];
- result[3] = -result[3];
- }
- }
- else {
- ASSIGN_4V(result, 0.0, 0.0, 0.0, 0.0);
- }
-}
-
-
-/**
- * As above, but only return result[0] element.
- */
-static void
-fetch_vector1(const struct prog_src_register *source,
- const struct gl_program_machine *machine, GLfloat result[4])
-{
- const GLfloat *src = get_src_register_pointer(source, machine);
-
- result[0] = src[GET_SWZ(source->Swizzle, 0)];
-
- if (source->Negate) {
- result[0] = -result[0];
- }
-}
-
-
-/**
- * Fetch texel from texture. Use partial derivatives when possible.
- */
-static inline void
-fetch_texel(struct gl_context *ctx,
- const struct gl_program_machine *machine,
- const struct prog_instruction *inst,
- const GLfloat texcoord[4], GLfloat lodBias,
- GLfloat color[4])
-{
- const GLuint unit = machine->Samplers[inst->TexSrcUnit];
-
- /* Note: we only have the right derivatives for fragment input attribs.
- */
- if (machine->NumDeriv > 0 &&
- inst->SrcReg[0].File == PROGRAM_INPUT &&
- inst->SrcReg[0].Index == VARYING_SLOT_TEX0 + inst->TexSrcUnit) {
- /* simple texture fetch for which we should have derivatives */
- GLuint attr = inst->SrcReg[0].Index;
- machine->FetchTexelDeriv(ctx, texcoord,
- machine->DerivX[attr],
- machine->DerivY[attr],
- lodBias, unit, color);
- }
- else {
- machine->FetchTexelLod(ctx, texcoord, lodBias, unit, color);
- }
-}
-
-
-/**
- * Store 4 floats into a register. Observe the instructions saturate and
- * set-condition-code flags.
- */
-static void
-store_vector4(const struct prog_instruction *inst,
- struct gl_program_machine *machine, const GLfloat value[4])
-{
- const struct prog_dst_register *dstReg = &(inst->DstReg);
- const GLboolean clamp = inst->Saturate;
- GLuint writeMask = dstReg->WriteMask;
- GLfloat clampedValue[4];
- GLfloat *dst = get_dst_register_pointer(dstReg, machine);
-
-#if 0
- if (value[0] > 1.0e10 ||
- util_is_inf_or_nan(value[0]) ||
- util_is_inf_or_nan(value[1]) ||
- util_is_inf_or_nan(value[2]) || util_is_inf_or_nan(value[3]))
- printf("store %g %g %g %g\n", value[0], value[1], value[2], value[3]);
-#endif
-
- if (clamp) {
- clampedValue[0] = CLAMP(value[0], 0.0F, 1.0F);
- clampedValue[1] = CLAMP(value[1], 0.0F, 1.0F);
- clampedValue[2] = CLAMP(value[2], 0.0F, 1.0F);
- clampedValue[3] = CLAMP(value[3], 0.0F, 1.0F);
- value = clampedValue;
- }
-
-#ifdef NAN_CHECK
- assert(!util_is_inf_or_nan(value[0]));
- assert(!util_is_inf_or_nan(value[0]));
- assert(!util_is_inf_or_nan(value[0]));
- assert(!util_is_inf_or_nan(value[0]));
-#endif
-
- if (writeMask & WRITEMASK_X)
- dst[0] = value[0];
- if (writeMask & WRITEMASK_Y)
- dst[1] = value[1];
- if (writeMask & WRITEMASK_Z)
- dst[2] = value[2];
- if (writeMask & WRITEMASK_W)
- dst[3] = value[3];
-}
-
-
-/**
- * Execute the given vertex/fragment program.
- *
- * \param ctx rendering context
- * \param program the program to execute
- * \param machine machine state (must be initialized)
- * \return GL_TRUE if program completed or GL_FALSE if program executed KIL.
- */
-GLboolean
-_mesa_execute_program(struct gl_context * ctx,
- const struct gl_program *program,
- struct gl_program_machine *machine)
-{
- const GLuint numInst = program->arb.NumInstructions;
- const GLuint maxExec = 65536;
- GLuint pc, numExec = 0;
-
- machine->CurProgram = program;
-
- if (DEBUG_PROG) {
- printf("execute program %u --------------------\n", program->Id);
- }
-
- if (program->Target == GL_VERTEX_PROGRAM_ARB) {
- machine->EnvParams = ctx->VertexProgram.Parameters;
- }
- else {
- machine->EnvParams = ctx->FragmentProgram.Parameters;
- }
-
- for (pc = 0; pc < numInst; pc++) {
- const struct prog_instruction *inst = program->arb.Instructions + pc;
-
- if (DEBUG_PROG) {
- _mesa_print_instruction(inst);
- }
-
- switch (inst->Opcode) {
- case OPCODE_ABS:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] = fabsf(a[0]);
- result[1] = fabsf(a[1]);
- result[2] = fabsf(a[2]);
- result[3] = fabsf(a[3]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_ADD:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = a[0] + b[0];
- result[1] = a[1] + b[1];
- result[2] = a[2] + b[2];
- result[3] = a[3] + b[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("ADD (%g %g %g %g) = (%g %g %g %g) + (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3], b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_ARL:
- {
- GLfloat t[4];
- fetch_vector4(&inst->SrcReg[0], machine, t);
- machine->AddressReg[0][0] = util_ifloor(t[0]);
- if (DEBUG_PROG) {
- printf("ARL %d\n", machine->AddressReg[0][0]);
- }
- }
- break;
- case OPCODE_BGNLOOP:
- /* no-op */
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ENDLOOP);
- break;
- case OPCODE_ENDLOOP:
- /* subtract 1 here since pc is incremented by for(pc) loop */
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_BGNLOOP);
- pc = inst->BranchTarget - 1; /* go to matching BNGLOOP */
- break;
- case OPCODE_BGNSUB: /* begin subroutine */
- break;
- case OPCODE_ENDSUB: /* end subroutine */
- break;
- case OPCODE_BRK: /* break out of loop (conditional) */
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ENDLOOP);
- /* break out of loop */
- /* pc++ at end of for-loop will put us after the ENDLOOP inst */
- pc = inst->BranchTarget;
- break;
- case OPCODE_CONT: /* continue loop (conditional) */
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ENDLOOP);
- /* continue at ENDLOOP */
- /* Subtract 1 here since we'll do pc++ at end of for-loop */
- pc = inst->BranchTarget - 1;
- break;
- case OPCODE_CAL: /* Call subroutine (conditional) */
- /* call the subroutine */
- if (machine->StackDepth >= MAX_PROGRAM_CALL_DEPTH) {
- return GL_TRUE; /* Per GL_NV_vertex_program2 spec */
- }
- machine->CallStack[machine->StackDepth++] = pc + 1; /* next inst */
- /* Subtract 1 here since we'll do pc++ at end of for-loop */
- pc = inst->BranchTarget - 1;
- break;
- case OPCODE_CMP:
- {
- GLfloat a[4], b[4], c[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- fetch_vector4(&inst->SrcReg[2], machine, c);
- result[0] = a[0] < 0.0F ? b[0] : c[0];
- result[1] = a[1] < 0.0F ? b[1] : c[1];
- result[2] = a[2] < 0.0F ? b[2] : c[2];
- result[3] = a[3] < 0.0F ? b[3] : c[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("CMP (%g %g %g %g) = (%g %g %g %g) < 0 ? (%g %g %g %g) : (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3],
- c[0], c[1], c[2], c[3]);
- }
- }
- break;
- case OPCODE_COS:
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- result[0] = result[1] = result[2] = result[3]
- = cosf(a[0]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_DDX: /* Partial derivative with respect to X */
- {
- GLfloat result[4];
- fetch_vector4_deriv(&inst->SrcReg[0], machine, 'X', result);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_DDY: /* Partial derivative with respect to Y */
- {
- GLfloat result[4];
- fetch_vector4_deriv(&inst->SrcReg[0], machine, 'Y', result);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_DP2:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = result[1] = result[2] = result[3] = DOT2(a, b);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("DP2 %g = (%g %g) . (%g %g)\n",
- result[0], a[0], a[1], b[0], b[1]);
- }
- }
- break;
- case OPCODE_DP3:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = result[1] = result[2] = result[3] = DOT3(a, b);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("DP3 %g = (%g %g %g) . (%g %g %g)\n",
- result[0], a[0], a[1], a[2], b[0], b[1], b[2]);
- }
- }
- break;
- case OPCODE_DP4:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = result[1] = result[2] = result[3] = DOT4(a, b);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("DP4 %g = (%g, %g %g %g) . (%g, %g %g %g)\n",
- result[0], a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_DPH:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = result[1] = result[2] = result[3] = DOT3(a, b) + b[3];
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_DST: /* Distance vector */
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = 1.0F;
- result[1] = a[1] * b[1];
- result[2] = a[2];
- result[3] = b[3];
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_EXP:
- {
- GLfloat t[4], q[4], floor_t0;
- fetch_vector1(&inst->SrcReg[0], machine, t);
- floor_t0 = floorf(t[0]);
- if (floor_t0 > FLT_MAX_EXP) {
- SET_POS_INFINITY(q[0]);
- SET_POS_INFINITY(q[2]);
- }
- else if (floor_t0 < FLT_MIN_EXP) {
- q[0] = 0.0F;
- q[2] = 0.0F;
- }
- else {
- q[0] = ldexpf(1.0, (int) floor_t0);
- /* Note: GL_NV_vertex_program expects
- * result.z = result.x * APPX(result.y)
- * We do what the ARB extension says.
- */
- q[2] = exp2f(t[0]);
- }
- q[1] = t[0] - floor_t0;
- q[3] = 1.0F;
- store_vector4( inst, machine, q );
- }
- break;
- case OPCODE_EX2: /* Exponential base 2 */
- {
- GLfloat a[4], result[4], val;
- fetch_vector1(&inst->SrcReg[0], machine, a);
- val = exp2f(a[0]);
- /*
- if (util_is_inf_or_nan(val))
- val = 1.0e10;
- */
- result[0] = result[1] = result[2] = result[3] = val;
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_FLR:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] = floorf(a[0]);
- result[1] = floorf(a[1]);
- result[2] = floorf(a[2]);
- result[3] = floorf(a[3]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_FRC:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] = a[0] - floorf(a[0]);
- result[1] = a[1] - floorf(a[1]);
- result[2] = a[2] - floorf(a[2]);
- result[3] = a[3] - floorf(a[3]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_IF:
- {
- GLboolean cond;
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ELSE ||
- program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ENDIF);
- /* eval condition */
- GLfloat a[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- cond = (a[0] != 0.0F);
- if (DEBUG_PROG) {
- printf("IF: %d\n", cond);
- }
- /* do if/else */
- if (cond) {
- /* do if-clause (just continue execution) */
- }
- else {
- /* go to the instruction after ELSE or ENDIF */
- assert(inst->BranchTarget >= 0);
- pc = inst->BranchTarget;
- }
- }
- break;
- case OPCODE_ELSE:
- /* goto ENDIF */
- assert(program->arb.Instructions[inst->BranchTarget].Opcode
- == OPCODE_ENDIF);
- assert(inst->BranchTarget >= 0);
- pc = inst->BranchTarget;
- break;
- case OPCODE_ENDIF:
- /* nothing */
- break;
- case OPCODE_KIL: /* ARB_f_p only */
- {
- GLfloat a[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- if (DEBUG_PROG) {
- printf("KIL if (%g %g %g %g) <= 0.0\n",
- a[0], a[1], a[2], a[3]);
- }
-
- if (a[0] < 0.0F || a[1] < 0.0F || a[2] < 0.0F || a[3] < 0.0F) {
- return GL_FALSE;
- }
- }
- break;
- case OPCODE_LG2: /* log base 2 */
- {
- GLfloat a[4], result[4], val;
- fetch_vector1(&inst->SrcReg[0], machine, a);
- /* The fast LOG2 macro doesn't meet the precision requirements.
- */
- if (a[0] == 0.0F) {
- val = -FLT_MAX;
- }
- else {
- val = logf(a[0]) * 1.442695F;
- }
- result[0] = result[1] = result[2] = result[3] = val;
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_LIT:
- {
- const GLfloat epsilon = 1.0F / 256.0F; /* from NV VP spec */
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- a[0] = MAX2(a[0], 0.0F);
- a[1] = MAX2(a[1], 0.0F);
- /* XXX ARB version clamps a[3], NV version doesn't */
- a[3] = CLAMP(a[3], -(128.0F - epsilon), (128.0F - epsilon));
- result[0] = 1.0F;
- result[1] = a[0];
- /* XXX we could probably just use pow() here */
- if (a[0] > 0.0F) {
- if (a[1] == 0.0F && a[3] == 0.0F)
- result[2] = 1.0F;
- else
- result[2] = powf(a[1], a[3]);
- }
- else {
- result[2] = 0.0F;
- }
- result[3] = 1.0F;
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("LIT (%g %g %g %g) : (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3]);
- }
- }
- break;
- case OPCODE_LOG:
- {
- GLfloat t[4], q[4], abs_t0;
- fetch_vector1(&inst->SrcReg[0], machine, t);
- abs_t0 = fabsf(t[0]);
- if (abs_t0 != 0.0F) {
- if (util_is_inf_or_nan(abs_t0))
- {
- SET_POS_INFINITY(q[0]);
- q[1] = 1.0F;
- SET_POS_INFINITY(q[2]);
- }
- else {
- int exponent;
- GLfloat mantissa = frexpf(t[0], &exponent);
- q[0] = (GLfloat) (exponent - 1);
- q[1] = 2.0F * mantissa; /* map [.5, 1) -> [1, 2) */
-
- /* The fast LOG2 macro doesn't meet the precision
- * requirements.
- */
- q[2] = logf(t[0]) * 1.442695F;
- }
- }
- else {
- SET_NEG_INFINITY(q[0]);
- q[1] = 1.0F;
- SET_NEG_INFINITY(q[2]);
- }
- q[3] = 1.0;
- store_vector4(inst, machine, q);
- }
- break;
- case OPCODE_LRP:
- {
- GLfloat a[4], b[4], c[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- fetch_vector4(&inst->SrcReg[2], machine, c);
- result[0] = a[0] * b[0] + (1.0F - a[0]) * c[0];
- result[1] = a[1] * b[1] + (1.0F - a[1]) * c[1];
- result[2] = a[2] * b[2] + (1.0F - a[2]) * c[2];
- result[3] = a[3] * b[3] + (1.0F - a[3]) * c[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("LRP (%g %g %g %g) = (%g %g %g %g), "
- "(%g %g %g %g), (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3], c[0], c[1], c[2], c[3]);
- }
- }
- break;
- case OPCODE_MAD:
- {
- GLfloat a[4], b[4], c[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- fetch_vector4(&inst->SrcReg[2], machine, c);
- result[0] = a[0] * b[0] + c[0];
- result[1] = a[1] * b[1] + c[1];
- result[2] = a[2] * b[2] + c[2];
- result[3] = a[3] * b[3] + c[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("MAD (%g %g %g %g) = (%g %g %g %g) * "
- "(%g %g %g %g) + (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3], c[0], c[1], c[2], c[3]);
- }
- }
- break;
- case OPCODE_MAX:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = MAX2(a[0], b[0]);
- result[1] = MAX2(a[1], b[1]);
- result[2] = MAX2(a[2], b[2]);
- result[3] = MAX2(a[3], b[3]);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("MAX (%g %g %g %g) = (%g %g %g %g), (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3], b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_MIN:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = MIN2(a[0], b[0]);
- result[1] = MIN2(a[1], b[1]);
- result[2] = MIN2(a[2], b[2]);
- result[3] = MIN2(a[3], b[3]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_MOV:
- {
- GLfloat result[4];
- fetch_vector4(&inst->SrcReg[0], machine, result);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("MOV (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3]);
- }
- }
- break;
- case OPCODE_MUL:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = a[0] * b[0];
- result[1] = a[1] * b[1];
- result[2] = a[2] * b[2];
- result[3] = a[3] * b[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("MUL (%g %g %g %g) = (%g %g %g %g) * (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3], b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_NOISE1:
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- result[0] =
- result[1] =
- result[2] =
- result[3] = _mesa_noise1(a[0]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_NOISE2:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] =
- result[1] =
- result[2] = result[3] = _mesa_noise2(a[0], a[1]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_NOISE3:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] =
- result[1] =
- result[2] =
- result[3] = _mesa_noise3(a[0], a[1], a[2]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_NOISE4:
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] =
- result[1] =
- result[2] =
- result[3] = _mesa_noise4(a[0], a[1], a[2], a[3]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_NOP:
- break;
- case OPCODE_POW:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- fetch_vector1(&inst->SrcReg[1], machine, b);
- result[0] = result[1] = result[2] = result[3]
- = powf(a[0], b[0]);
- store_vector4(inst, machine, result);
- }
- break;
-
- case OPCODE_RCP:
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- if (DEBUG_PROG) {
- if (a[0] == 0)
- printf("RCP(0)\n");
- else if (util_is_inf_or_nan(a[0]))
- printf("RCP(inf)\n");
- }
- result[0] = result[1] = result[2] = result[3] = 1.0F / a[0];
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_RET: /* return from subroutine (conditional) */
- if (machine->StackDepth == 0) {
- return GL_TRUE; /* Per GL_NV_vertex_program2 spec */
- }
- /* subtract one because of pc++ in the for loop */
- pc = machine->CallStack[--machine->StackDepth] - 1;
- break;
- case OPCODE_RSQ: /* 1 / sqrt() */
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- a[0] = fabsf(a[0]);
- result[0] = result[1] = result[2] = result[3] = 1.0f / sqrtf(a[0]);
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("RSQ %g = 1/sqrt(|%g|)\n", result[0], a[0]);
- }
- }
- break;
- case OPCODE_SCS: /* sine and cos */
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- result[0] = cosf(a[0]);
- result[1] = sinf(a[0]);
- result[2] = 0.0F; /* undefined! */
- result[3] = 0.0F; /* undefined! */
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_SGE: /* set on greater or equal */
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = (a[0] >= b[0]) ? 1.0F : 0.0F;
- result[1] = (a[1] >= b[1]) ? 1.0F : 0.0F;
- result[2] = (a[2] >= b[2]) ? 1.0F : 0.0F;
- result[3] = (a[3] >= b[3]) ? 1.0F : 0.0F;
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("SGE (%g %g %g %g) = (%g %g %g %g) >= (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_SIN:
- {
- GLfloat a[4], result[4];
- fetch_vector1(&inst->SrcReg[0], machine, a);
- result[0] = result[1] = result[2] = result[3]
- = sinf(a[0]);
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_SLT: /* set on less */
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = (a[0] < b[0]) ? 1.0F : 0.0F;
- result[1] = (a[1] < b[1]) ? 1.0F : 0.0F;
- result[2] = (a[2] < b[2]) ? 1.0F : 0.0F;
- result[3] = (a[3] < b[3]) ? 1.0F : 0.0F;
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("SLT (%g %g %g %g) = (%g %g %g %g) < (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3],
- b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_SSG: /* set sign (-1, 0 or +1) */
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] = (GLfloat) ((a[0] > 0.0F) - (a[0] < 0.0F));
- result[1] = (GLfloat) ((a[1] > 0.0F) - (a[1] < 0.0F));
- result[2] = (GLfloat) ((a[2] > 0.0F) - (a[2] < 0.0F));
- result[3] = (GLfloat) ((a[3] > 0.0F) - (a[3] < 0.0F));
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_SUB:
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = a[0] - b[0];
- result[1] = a[1] - b[1];
- result[2] = a[2] - b[2];
- result[3] = a[3] - b[3];
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("SUB (%g %g %g %g) = (%g %g %g %g) - (%g %g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], a[3], b[0], b[1], b[2], b[3]);
- }
- }
- break;
- case OPCODE_SWZ: /* extended swizzle */
- {
- const struct prog_src_register *source = &inst->SrcReg[0];
- const GLfloat *src = get_src_register_pointer(source, machine);
- GLfloat result[4];
- GLuint i;
- for (i = 0; i < 4; i++) {
- const GLuint swz = GET_SWZ(source->Swizzle, i);
- if (swz == SWIZZLE_ZERO)
- result[i] = 0.0;
- else if (swz == SWIZZLE_ONE)
- result[i] = 1.0;
- else {
- assert(swz <= 3);
- result[i] = src[swz];
- }
- if (source->Negate & (1 << i))
- result[i] = -result[i];
- }
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_TEX: /* Both ARB and NV frag prog */
- /* Simple texel lookup */
- {
- GLfloat texcoord[4], color[4];
- fetch_vector4(&inst->SrcReg[0], machine, texcoord);
-
- /* For TEX, texcoord.Q should not be used and its value should not
- * matter (at most, we pass coord.xyz to texture3D() in GLSL).
- * Set Q=1 so that FetchTexelDeriv() doesn't get a garbage value
- * which is effectively what happens when the texcoord swizzle
- * is .xyzz
- */
- texcoord[3] = 1.0f;
-
- fetch_texel(ctx, machine, inst, texcoord, 0.0, color);
-
- if (DEBUG_PROG) {
- printf("TEX (%g, %g, %g, %g) = texture[%d][%g, %g, %g, %g]\n",
- color[0], color[1], color[2], color[3],
- inst->TexSrcUnit,
- texcoord[0], texcoord[1], texcoord[2], texcoord[3]);
- }
- store_vector4(inst, machine, color);
- }
- break;
- case OPCODE_TXB: /* GL_ARB_fragment_program only */
- /* Texel lookup with LOD bias */
- {
- GLfloat texcoord[4], color[4], lodBias;
-
- fetch_vector4(&inst->SrcReg[0], machine, texcoord);
-
- /* texcoord[3] is the bias to add to lambda */
- lodBias = texcoord[3];
-
- fetch_texel(ctx, machine, inst, texcoord, lodBias, color);
-
- if (DEBUG_PROG) {
- printf("TXB (%g, %g, %g, %g) = texture[%d][%g %g %g %g]"
- " bias %g\n",
- color[0], color[1], color[2], color[3],
- inst->TexSrcUnit,
- texcoord[0],
- texcoord[1],
- texcoord[2],
- texcoord[3],
- lodBias);
- }
-
- store_vector4(inst, machine, color);
- }
- break;
- case OPCODE_TXD:
- /* Texture lookup w/ partial derivatives for LOD */
- {
- GLfloat texcoord[4], dtdx[4], dtdy[4], color[4];
- fetch_vector4(&inst->SrcReg[0], machine, texcoord);
- fetch_vector4(&inst->SrcReg[1], machine, dtdx);
- fetch_vector4(&inst->SrcReg[2], machine, dtdy);
- machine->FetchTexelDeriv(ctx, texcoord, dtdx, dtdy,
- 0.0, /* lodBias */
- inst->TexSrcUnit, color);
- store_vector4(inst, machine, color);
- }
- break;
- case OPCODE_TXL:
- /* Texel lookup with explicit LOD */
- {
- GLfloat texcoord[4], color[4], lod;
-
- fetch_vector4(&inst->SrcReg[0], machine, texcoord);
-
- /* texcoord[3] is the LOD */
- lod = texcoord[3];
-
- machine->FetchTexelLod(ctx, texcoord, lod,
- machine->Samplers[inst->TexSrcUnit], color);
-
- store_vector4(inst, machine, color);
- }
- break;
- case OPCODE_TXP: /* GL_ARB_fragment_program only */
- /* Texture lookup w/ projective divide */
- {
- GLfloat texcoord[4], color[4];
-
- fetch_vector4(&inst->SrcReg[0], machine, texcoord);
- /* Not so sure about this test - if texcoord[3] is
- * zero, we'd probably be fine except for an assert in
- * IROUND_POS() which gets triggered by the inf values created.
- */
- if (texcoord[3] != 0.0F) {
- texcoord[0] /= texcoord[3];
- texcoord[1] /= texcoord[3];
- texcoord[2] /= texcoord[3];
- }
-
- fetch_texel(ctx, machine, inst, texcoord, 0.0, color);
-
- store_vector4(inst, machine, color);
- }
- break;
- case OPCODE_TRUNC: /* truncate toward zero */
- {
- GLfloat a[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- result[0] = (GLfloat) (GLint) a[0];
- result[1] = (GLfloat) (GLint) a[1];
- result[2] = (GLfloat) (GLint) a[2];
- result[3] = (GLfloat) (GLint) a[3];
- store_vector4(inst, machine, result);
- }
- break;
- case OPCODE_XPD: /* cross product */
- {
- GLfloat a[4], b[4], result[4];
- fetch_vector4(&inst->SrcReg[0], machine, a);
- fetch_vector4(&inst->SrcReg[1], machine, b);
- result[0] = a[1] * b[2] - a[2] * b[1];
- result[1] = a[2] * b[0] - a[0] * b[2];
- result[2] = a[0] * b[1] - a[1] * b[0];
- result[3] = 1.0;
- store_vector4(inst, machine, result);
- if (DEBUG_PROG) {
- printf("XPD (%g %g %g %g) = (%g %g %g) X (%g %g %g)\n",
- result[0], result[1], result[2], result[3],
- a[0], a[1], a[2], b[0], b[1], b[2]);
- }
- }
- break;
- case OPCODE_END:
- return GL_TRUE;
- default:
- _mesa_problem(ctx, "Bad opcode %d in _mesa_execute_program",
- inst->Opcode);
- return GL_TRUE; /* return value doesn't matter */
- }
-
- numExec++;
- if (numExec > maxExec) {
- static GLboolean reported = GL_FALSE;
- if (!reported) {
- _mesa_problem(ctx, "Infinite loop detected in fragment program");
- reported = GL_TRUE;
- }
- return GL_TRUE;
- }
-
- } /* for pc */
-
- return GL_TRUE;
-}
+++ /dev/null
-/*
- * Mesa 3-D graphics library
- *
- * Copyright (C) 2006 Brian Paul 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, 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
- * THE AUTHORS OR COPYRIGHT HOLDERS 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.
- */
-
-/*
- * SimplexNoise1234
- * Copyright (c) 2003-2005, Stefan Gustavson
- *
- * Contact: stegu@itn.liu.se
- */
-
-/**
- * \file
- * \brief C implementation of Perlin Simplex Noise over 1, 2, 3 and 4 dims.
- * \author Stefan Gustavson (stegu@itn.liu.se)
- *
- *
- * This implementation is "Simplex Noise" as presented by
- * Ken Perlin at a relatively obscure and not often cited course
- * session "Real-Time Shading" at Siggraph 2001 (before real
- * time shading actually took on), under the title "hardware noise".
- * The 3D function is numerically equivalent to his Java reference
- * code available in the PDF course notes, although I re-implemented
- * it from scratch to get more readable code. The 1D, 2D and 4D cases
- * were implemented from scratch by me from Ken Perlin's text.
- *
- * This file has no dependencies on any other file, not even its own
- * header file. The header file is made for use by external code only.
- */
-
-
-
-#include "prog_noise.h"
-
-#define FASTFLOOR(x) ( ((x)>0) ? ((int)x) : (((int)x)-1) )
-
-/*
- * ---------------------------------------------------------------------
- * Static data
- */
-
-/**
- * Permutation table. This is just a random jumble of all numbers 0-255,
- * repeated twice to avoid wrapping the index at 255 for each lookup.
- * This needs to be exactly the same for all instances on all platforms,
- * so it's easiest to just keep it as static explicit data.
- * This also removes the need for any initialisation of this class.
- *
- * Note that making this an int[] instead of a char[] might make the
- * code run faster on platforms with a high penalty for unaligned single
- * byte addressing. Intel x86 is generally single-byte-friendly, but
- * some other CPUs are faster with 4-aligned reads.
- * However, a char[] is smaller, which avoids cache trashing, and that
- * is probably the most important aspect on most architectures.
- * This array is accessed a *lot* by the noise functions.
- * A vector-valued noise over 3D accesses it 96 times, and a
- * float-valued 4D noise 64 times. We want this to fit in the cache!
- */
-static const unsigned char perm[512] = { 151, 160, 137, 91, 90, 15,
- 131, 13, 201, 95, 96, 53, 194, 233, 7, 225, 140, 36, 103, 30, 69, 142, 8,
- 99, 37, 240, 21, 10, 23,
- 190, 6, 148, 247, 120, 234, 75, 0, 26, 197, 62, 94, 252, 219, 203, 117, 35,
- 11, 32, 57, 177, 33,
- 88, 237, 149, 56, 87, 174, 20, 125, 136, 171, 168, 68, 175, 74, 165, 71,
- 134, 139, 48, 27, 166,
- 77, 146, 158, 231, 83, 111, 229, 122, 60, 211, 133, 230, 220, 105, 92, 41,
- 55, 46, 245, 40, 244,
- 102, 143, 54, 65, 25, 63, 161, 1, 216, 80, 73, 209, 76, 132, 187, 208, 89,
- 18, 169, 200, 196,
- 135, 130, 116, 188, 159, 86, 164, 100, 109, 198, 173, 186, 3, 64, 52, 217,
- 226, 250, 124, 123,
- 5, 202, 38, 147, 118, 126, 255, 82, 85, 212, 207, 206, 59, 227, 47, 16, 58,
- 17, 182, 189, 28, 42,
- 223, 183, 170, 213, 119, 248, 152, 2, 44, 154, 163, 70, 221, 153, 101, 155,
- 167, 43, 172, 9,
- 129, 22, 39, 253, 19, 98, 108, 110, 79, 113, 224, 232, 178, 185, 112, 104,
- 218, 246, 97, 228,
- 251, 34, 242, 193, 238, 210, 144, 12, 191, 179, 162, 241, 81, 51, 145, 235,
- 249, 14, 239, 107,
- 49, 192, 214, 31, 181, 199, 106, 157, 184, 84, 204, 176, 115, 121, 50, 45,
- 127, 4, 150, 254,
- 138, 236, 205, 93, 222, 114, 67, 29, 24, 72, 243, 141, 128, 195, 78, 66,
- 215, 61, 156, 180,
- 151, 160, 137, 91, 90, 15,
- 131, 13, 201, 95, 96, 53, 194, 233, 7, 225, 140, 36, 103, 30, 69, 142, 8,
- 99, 37, 240, 21, 10, 23,
- 190, 6, 148, 247, 120, 234, 75, 0, 26, 197, 62, 94, 252, 219, 203, 117, 35,
- 11, 32, 57, 177, 33,
- 88, 237, 149, 56, 87, 174, 20, 125, 136, 171, 168, 68, 175, 74, 165, 71,
- 134, 139, 48, 27, 166,
- 77, 146, 158, 231, 83, 111, 229, 122, 60, 211, 133, 230, 220, 105, 92, 41,
- 55, 46, 245, 40, 244,
- 102, 143, 54, 65, 25, 63, 161, 1, 216, 80, 73, 209, 76, 132, 187, 208, 89,
- 18, 169, 200, 196,
- 135, 130, 116, 188, 159, 86, 164, 100, 109, 198, 173, 186, 3, 64, 52, 217,
- 226, 250, 124, 123,
- 5, 202, 38, 147, 118, 126, 255, 82, 85, 212, 207, 206, 59, 227, 47, 16, 58,
- 17, 182, 189, 28, 42,
- 223, 183, 170, 213, 119, 248, 152, 2, 44, 154, 163, 70, 221, 153, 101, 155,
- 167, 43, 172, 9,
- 129, 22, 39, 253, 19, 98, 108, 110, 79, 113, 224, 232, 178, 185, 112, 104,
- 218, 246, 97, 228,
- 251, 34, 242, 193, 238, 210, 144, 12, 191, 179, 162, 241, 81, 51, 145, 235,
- 249, 14, 239, 107,
- 49, 192, 214, 31, 181, 199, 106, 157, 184, 84, 204, 176, 115, 121, 50, 45,
- 127, 4, 150, 254,
- 138, 236, 205, 93, 222, 114, 67, 29, 24, 72, 243, 141, 128, 195, 78, 66,
- 215, 61, 156, 180
-};
-
-/*
- * ---------------------------------------------------------------------
- */
-
-/*
- * Helper functions to compute gradients-dot-residualvectors (1D to 4D)
- * Note that these generate gradients of more than unit length. To make
- * a close match with the value range of classic Perlin noise, the final
- * noise values need to be rescaled to fit nicely within [-1,1].
- * (The simplex noise functions as such also have different scaling.)
- * Note also that these noise functions are the most practical and useful
- * signed version of Perlin noise. To return values according to the
- * RenderMan specification from the SL noise() and pnoise() functions,
- * the noise values need to be scaled and offset to [0,1], like this:
- * float SLnoise = (SimplexNoise1234::noise(x,y,z) + 1.0) * 0.5;
- */
-
-static float
-grad1(int hash, float x)
-{
- int h = hash & 15;
- float grad = 1.0f + (h & 7); /* Gradient value 1.0, 2.0, ..., 8.0 */
- if (h & 8)
- grad = -grad; /* Set a random sign for the gradient */
- return (grad * x); /* Multiply the gradient with the distance */
-}
-
-static float
-grad2(int hash, float x, float y)
-{
- int h = hash & 7; /* Convert low 3 bits of hash code */
- float u = h < 4 ? x : y; /* into 8 simple gradient directions, */
- float v = h < 4 ? y : x; /* and compute the dot product with (x,y). */
- return ((h & 1) ? -u : u) + ((h & 2) ? -2.0f * v : 2.0f * v);
-}
-
-static float
-grad3(int hash, float x, float y, float z)
-{
- int h = hash & 15; /* Convert low 4 bits of hash code into 12 simple */
- float u = h < 8 ? x : y; /* gradient directions, and compute dot product. */
- float v = h < 4 ? y : h == 12 || h == 14 ? x : z; /* Fix repeats at h = 12 to 15 */
- return ((h & 1) ? -u : u) + ((h & 2) ? -v : v);
-}
-
-static float
-grad4(int hash, float x, float y, float z, float t)
-{
- int h = hash & 31; /* Convert low 5 bits of hash code into 32 simple */
- float u = h < 24 ? x : y; /* gradient directions, and compute dot product. */
- float v = h < 16 ? y : z;
- float w = h < 8 ? z : t;
- return ((h & 1) ? -u : u) + ((h & 2) ? -v : v) + ((h & 4) ? -w : w);
-}
-
-/**
- * A lookup table to traverse the simplex around a given point in 4D.
- * Details can be found where this table is used, in the 4D noise method.
- * TODO: This should not be required, backport it from Bill's GLSL code!
- */
-static const unsigned char simplex[64][4] = {
- {0, 1, 2, 3}, {0, 1, 3, 2}, {0, 0, 0, 0}, {0, 2, 3, 1},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {1, 2, 3, 0},
- {0, 2, 1, 3}, {0, 0, 0, 0}, {0, 3, 1, 2}, {0, 3, 2, 1},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {1, 3, 2, 0},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {1, 2, 0, 3}, {0, 0, 0, 0}, {1, 3, 0, 2}, {0, 0, 0, 0},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {2, 3, 0, 1}, {2, 3, 1, 0},
- {1, 0, 2, 3}, {1, 0, 3, 2}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {0, 0, 0, 0}, {2, 0, 3, 1}, {0, 0, 0, 0}, {2, 1, 3, 0},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {2, 0, 1, 3}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {3, 0, 1, 2}, {3, 0, 2, 1}, {0, 0, 0, 0}, {3, 1, 2, 0},
- {2, 1, 0, 3}, {0, 0, 0, 0}, {0, 0, 0, 0}, {0, 0, 0, 0},
- {3, 1, 0, 2}, {0, 0, 0, 0}, {3, 2, 0, 1}, {3, 2, 1, 0}
-};
-
-
-/** 1D simplex noise */
-GLfloat
-_mesa_noise1(GLfloat x)
-{
- int i0 = FASTFLOOR(x);
- int i1 = i0 + 1;
- float x0 = x - i0;
- float x1 = x0 - 1.0f;
- float t1 = 1.0f - x1 * x1;
- float n0, n1;
-
- float t0 = 1.0f - x0 * x0;
-/* if(t0 < 0.0f) t0 = 0.0f; // this never happens for the 1D case */
- t0 *= t0;
- n0 = t0 * t0 * grad1(perm[i0 & 0xff], x0);
-
-/* if(t1 < 0.0f) t1 = 0.0f; // this never happens for the 1D case */
- t1 *= t1;
- n1 = t1 * t1 * grad1(perm[i1 & 0xff], x1);
- /* The maximum value of this noise is 8*(3/4)^4 = 2.53125 */
- /* A factor of 0.395 would scale to fit exactly within [-1,1], but */
- /* we want to match PRMan's 1D noise, so we scale it down some more. */
- return 0.25f * (n0 + n1);
-}
-
-
-/** 2D simplex noise */
-GLfloat
-_mesa_noise2(GLfloat x, GLfloat y)
-{
-#define F2 0.366025403f /* F2 = 0.5*(sqrt(3.0)-1.0) */
-#define G2 0.211324865f /* G2 = (3.0-Math.sqrt(3.0))/6.0 */
-
- float n0, n1, n2; /* Noise contributions from the three corners */
-
- /* Skew the input space to determine which simplex cell we're in */
- float s = (x + y) * F2; /* Hairy factor for 2D */
- float xs = x + s;
- float ys = y + s;
- int i = FASTFLOOR(xs);
- int j = FASTFLOOR(ys);
-
- float t = (float) (i + j) * G2;
- float X0 = i - t; /* Unskew the cell origin back to (x,y) space */
- float Y0 = j - t;
- float x0 = x - X0; /* The x,y distances from the cell origin */
- float y0 = y - Y0;
-
- float x1, y1, x2, y2;
- unsigned int ii, jj;
- float t0, t1, t2;
-
- /* For the 2D case, the simplex shape is an equilateral triangle. */
- /* Determine which simplex we are in. */
- unsigned int i1, j1; /* Offsets for second (middle) corner of simplex in (i,j) coords */
- if (x0 > y0) {
- i1 = 1;
- j1 = 0;
- } /* lower triangle, XY order: (0,0)->(1,0)->(1,1) */
- else {
- i1 = 0;
- j1 = 1;
- } /* upper triangle, YX order: (0,0)->(0,1)->(1,1) */
-
- /* A step of (1,0) in (i,j) means a step of (1-c,-c) in (x,y), and */
- /* a step of (0,1) in (i,j) means a step of (-c,1-c) in (x,y), where */
- /* c = (3-sqrt(3))/6 */
-
- x1 = x0 - i1 + G2; /* Offsets for middle corner in (x,y) unskewed coords */
- y1 = y0 - j1 + G2;
- x2 = x0 - 1.0f + 2.0f * G2; /* Offsets for last corner in (x,y) unskewed coords */
- y2 = y0 - 1.0f + 2.0f * G2;
-
- /* Wrap the integer indices at 256, to avoid indexing perm[] out of bounds */
- ii = i & 0xff;
- jj = j & 0xff;
-
- /* Calculate the contribution from the three corners */
- t0 = 0.5f - x0 * x0 - y0 * y0;
- if (t0 < 0.0f)
- n0 = 0.0f;
- else {
- t0 *= t0;
- n0 = t0 * t0 * grad2(perm[ii + perm[jj]], x0, y0);
- }
-
- t1 = 0.5f - x1 * x1 - y1 * y1;
- if (t1 < 0.0f)
- n1 = 0.0f;
- else {
- t1 *= t1;
- n1 = t1 * t1 * grad2(perm[ii + i1 + perm[jj + j1]], x1, y1);
- }
-
- t2 = 0.5f - x2 * x2 - y2 * y2;
- if (t2 < 0.0f)
- n2 = 0.0f;
- else {
- t2 *= t2;
- n2 = t2 * t2 * grad2(perm[ii + 1 + perm[jj + 1]], x2, y2);
- }
-
- /* Add contributions from each corner to get the final noise value. */
- /* The result is scaled to return values in the interval [-1,1]. */
- return 40.0f * (n0 + n1 + n2); /* TODO: The scale factor is preliminary! */
-}
-
-
-/** 3D simplex noise */
-GLfloat
-_mesa_noise3(GLfloat x, GLfloat y, GLfloat z)
-{
-/* Simple skewing factors for the 3D case */
-#define F3 0.333333333f
-#define G3 0.166666667f
-
- float n0, n1, n2, n3; /* Noise contributions from the four corners */
-
- /* Skew the input space to determine which simplex cell we're in */
- float s = (x + y + z) * F3; /* Very nice and simple skew factor for 3D */
- float xs = x + s;
- float ys = y + s;
- float zs = z + s;
- int i = FASTFLOOR(xs);
- int j = FASTFLOOR(ys);
- int k = FASTFLOOR(zs);
-
- float t = (float) (i + j + k) * G3;
- float X0 = i - t; /* Unskew the cell origin back to (x,y,z) space */
- float Y0 = j - t;
- float Z0 = k - t;
- float x0 = x - X0; /* The x,y,z distances from the cell origin */
- float y0 = y - Y0;
- float z0 = z - Z0;
-
- float x1, y1, z1, x2, y2, z2, x3, y3, z3;
- unsigned int ii, jj, kk;
- float t0, t1, t2, t3;
-
- /* For the 3D case, the simplex shape is a slightly irregular tetrahedron. */
- /* Determine which simplex we are in. */
- unsigned int i1, j1, k1; /* Offsets for second corner of simplex in (i,j,k) coords */
- unsigned int i2, j2, k2; /* Offsets for third corner of simplex in (i,j,k) coords */
-
-/* This code would benefit from a backport from the GLSL version! */
- if (x0 >= y0) {
- if (y0 >= z0) {
- i1 = 1;
- j1 = 0;
- k1 = 0;
- i2 = 1;
- j2 = 1;
- k2 = 0;
- } /* X Y Z order */
- else if (x0 >= z0) {
- i1 = 1;
- j1 = 0;
- k1 = 0;
- i2 = 1;
- j2 = 0;
- k2 = 1;
- } /* X Z Y order */
- else {
- i1 = 0;
- j1 = 0;
- k1 = 1;
- i2 = 1;
- j2 = 0;
- k2 = 1;
- } /* Z X Y order */
- }
- else { /* x0<y0 */
- if (y0 < z0) {
- i1 = 0;
- j1 = 0;
- k1 = 1;
- i2 = 0;
- j2 = 1;
- k2 = 1;
- } /* Z Y X order */
- else if (x0 < z0) {
- i1 = 0;
- j1 = 1;
- k1 = 0;
- i2 = 0;
- j2 = 1;
- k2 = 1;
- } /* Y Z X order */
- else {
- i1 = 0;
- j1 = 1;
- k1 = 0;
- i2 = 1;
- j2 = 1;
- k2 = 0;
- } /* Y X Z order */
- }
-
- /* A step of (1,0,0) in (i,j,k) means a step of (1-c,-c,-c) in
- * (x,y,z), a step of (0,1,0) in (i,j,k) means a step of
- * (-c,1-c,-c) in (x,y,z), and a step of (0,0,1) in (i,j,k) means a
- * step of (-c,-c,1-c) in (x,y,z), where c = 1/6.
- */
-
- x1 = x0 - i1 + G3; /* Offsets for second corner in (x,y,z) coords */
- y1 = y0 - j1 + G3;
- z1 = z0 - k1 + G3;
- x2 = x0 - i2 + 2.0f * G3; /* Offsets for third corner in (x,y,z) coords */
- y2 = y0 - j2 + 2.0f * G3;
- z2 = z0 - k2 + 2.0f * G3;
- x3 = x0 - 1.0f + 3.0f * G3;/* Offsets for last corner in (x,y,z) coords */
- y3 = y0 - 1.0f + 3.0f * G3;
- z3 = z0 - 1.0f + 3.0f * G3;
-
- /* Wrap the integer indices at 256 to avoid indexing perm[] out of bounds */
- ii = i & 0xff;
- jj = j & 0xff;
- kk = k & 0xff;
-
- /* Calculate the contribution from the four corners */
- t0 = 0.6f - x0 * x0 - y0 * y0 - z0 * z0;
- if (t0 < 0.0f)
- n0 = 0.0f;
- else {
- t0 *= t0;
- n0 = t0 * t0 * grad3(perm[ii + perm[jj + perm[kk]]], x0, y0, z0);
- }
-
- t1 = 0.6f - x1 * x1 - y1 * y1 - z1 * z1;
- if (t1 < 0.0f)
- n1 = 0.0f;
- else {
- t1 *= t1;
- n1 =
- t1 * t1 * grad3(perm[ii + i1 + perm[jj + j1 + perm[kk + k1]]], x1,
- y1, z1);
- }
-
- t2 = 0.6f - x2 * x2 - y2 * y2 - z2 * z2;
- if (t2 < 0.0f)
- n2 = 0.0f;
- else {
- t2 *= t2;
- n2 =
- t2 * t2 * grad3(perm[ii + i2 + perm[jj + j2 + perm[kk + k2]]], x2,
- y2, z2);
- }
-
- t3 = 0.6f - x3 * x3 - y3 * y3 - z3 * z3;
- if (t3 < 0.0f)
- n3 = 0.0f;
- else {
- t3 *= t3;
- n3 =
- t3 * t3 * grad3(perm[ii + 1 + perm[jj + 1 + perm[kk + 1]]], x3, y3,
- z3);
- }
-
- /* Add contributions from each corner to get the final noise value.
- * The result is scaled to stay just inside [-1,1]
- */
- return 32.0f * (n0 + n1 + n2 + n3); /* TODO: The scale factor is preliminary! */
-}
-
-
-/** 4D simplex noise */
-GLfloat
-_mesa_noise4(GLfloat x, GLfloat y, GLfloat z, GLfloat w)
-{
- /* The skewing and unskewing factors are hairy again for the 4D case */
-#define F4 0.309016994f /* F4 = (Math.sqrt(5.0)-1.0)/4.0 */
-#define G4 0.138196601f /* G4 = (5.0-Math.sqrt(5.0))/20.0 */
-
- float n0, n1, n2, n3, n4; /* Noise contributions from the five corners */
-
- /* Skew the (x,y,z,w) space to determine which cell of 24 simplices we're in */
- float s = (x + y + z + w) * F4; /* Factor for 4D skewing */
- float xs = x + s;
- float ys = y + s;
- float zs = z + s;
- float ws = w + s;
- int i = FASTFLOOR(xs);
- int j = FASTFLOOR(ys);
- int k = FASTFLOOR(zs);
- int l = FASTFLOOR(ws);
-
- float t = (i + j + k + l) * G4; /* Factor for 4D unskewing */
- float X0 = i - t; /* Unskew the cell origin back to (x,y,z,w) space */
- float Y0 = j - t;
- float Z0 = k - t;
- float W0 = l - t;
-
- float x0 = x - X0; /* The x,y,z,w distances from the cell origin */
- float y0 = y - Y0;
- float z0 = z - Z0;
- float w0 = w - W0;
-
- /* For the 4D case, the simplex is a 4D shape I won't even try to describe.
- * To find out which of the 24 possible simplices we're in, we need to
- * determine the magnitude ordering of x0, y0, z0 and w0.
- * The method below is a good way of finding the ordering of x,y,z,w and
- * then find the correct traversal order for the simplex we're in.
- * First, six pair-wise comparisons are performed between each possible pair
- * of the four coordinates, and the results are used to add up binary bits
- * for an integer index.
- */
- int c1 = (x0 > y0) ? 32 : 0;
- int c2 = (x0 > z0) ? 16 : 0;
- int c3 = (y0 > z0) ? 8 : 0;
- int c4 = (x0 > w0) ? 4 : 0;
- int c5 = (y0 > w0) ? 2 : 0;
- int c6 = (z0 > w0) ? 1 : 0;
- int c = c1 + c2 + c3 + c4 + c5 + c6;
-
- unsigned int i1, j1, k1, l1; /* The integer offsets for the second simplex corner */
- unsigned int i2, j2, k2, l2; /* The integer offsets for the third simplex corner */
- unsigned int i3, j3, k3, l3; /* The integer offsets for the fourth simplex corner */
-
- float x1, y1, z1, w1, x2, y2, z2, w2, x3, y3, z3, w3, x4, y4, z4, w4;
- unsigned int ii, jj, kk, ll;
- float t0, t1, t2, t3, t4;
-
- /*
- * simplex[c] is a 4-vector with the numbers 0, 1, 2 and 3 in some
- * order. Many values of c will never occur, since e.g. x>y>z>w
- * makes x<z, y<w and x<w impossible. Only the 24 indices which
- * have non-zero entries make any sense. We use a thresholding to
- * set the coordinates in turn from the largest magnitude. The
- * number 3 in the "simplex" array is at the position of the
- * largest coordinate.
- */
- i1 = simplex[c][0] >= 3 ? 1 : 0;
- j1 = simplex[c][1] >= 3 ? 1 : 0;
- k1 = simplex[c][2] >= 3 ? 1 : 0;
- l1 = simplex[c][3] >= 3 ? 1 : 0;
- /* The number 2 in the "simplex" array is at the second largest coordinate. */
- i2 = simplex[c][0] >= 2 ? 1 : 0;
- j2 = simplex[c][1] >= 2 ? 1 : 0;
- k2 = simplex[c][2] >= 2 ? 1 : 0;
- l2 = simplex[c][3] >= 2 ? 1 : 0;
- /* The number 1 in the "simplex" array is at the second smallest coordinate. */
- i3 = simplex[c][0] >= 1 ? 1 : 0;
- j3 = simplex[c][1] >= 1 ? 1 : 0;
- k3 = simplex[c][2] >= 1 ? 1 : 0;
- l3 = simplex[c][3] >= 1 ? 1 : 0;
- /* The fifth corner has all coordinate offsets = 1, so no need to look that up. */
-
- x1 = x0 - i1 + G4; /* Offsets for second corner in (x,y,z,w) coords */
- y1 = y0 - j1 + G4;
- z1 = z0 - k1 + G4;
- w1 = w0 - l1 + G4;
- x2 = x0 - i2 + 2.0f * G4; /* Offsets for third corner in (x,y,z,w) coords */
- y2 = y0 - j2 + 2.0f * G4;
- z2 = z0 - k2 + 2.0f * G4;
- w2 = w0 - l2 + 2.0f * G4;
- x3 = x0 - i3 + 3.0f * G4; /* Offsets for fourth corner in (x,y,z,w) coords */
- y3 = y0 - j3 + 3.0f * G4;
- z3 = z0 - k3 + 3.0f * G4;
- w3 = w0 - l3 + 3.0f * G4;
- x4 = x0 - 1.0f + 4.0f * G4; /* Offsets for last corner in (x,y,z,w) coords */
- y4 = y0 - 1.0f + 4.0f * G4;
- z4 = z0 - 1.0f + 4.0f * G4;
- w4 = w0 - 1.0f + 4.0f * G4;
-
- /* Wrap the integer indices at 256, to avoid indexing perm[] out of bounds */
- ii = i & 0xff;
- jj = j & 0xff;
- kk = k & 0xff;
- ll = l & 0xff;
-
- /* Calculate the contribution from the five corners */
- t0 = 0.6f - x0 * x0 - y0 * y0 - z0 * z0 - w0 * w0;
- if (t0 < 0.0f)
- n0 = 0.0f;
- else {
- t0 *= t0;
- n0 =
- t0 * t0 * grad4(perm[ii + perm[jj + perm[kk + perm[ll]]]], x0, y0,
- z0, w0);
- }
-
- t1 = 0.6f - x1 * x1 - y1 * y1 - z1 * z1 - w1 * w1;
- if (t1 < 0.0f)
- n1 = 0.0f;
- else {
- t1 *= t1;
- n1 =
- t1 * t1 *
- grad4(perm[ii + i1 + perm[jj + j1 + perm[kk + k1 + perm[ll + l1]]]],
- x1, y1, z1, w1);
- }
-
- t2 = 0.6f - x2 * x2 - y2 * y2 - z2 * z2 - w2 * w2;
- if (t2 < 0.0f)
- n2 = 0.0f;
- else {
- t2 *= t2;
- n2 =
- t2 * t2 *
- grad4(perm[ii + i2 + perm[jj + j2 + perm[kk + k2 + perm[ll + l2]]]],
- x2, y2, z2, w2);
- }
-
- t3 = 0.6f - x3 * x3 - y3 * y3 - z3 * z3 - w3 * w3;
- if (t3 < 0.0f)
- n3 = 0.0f;
- else {
- t3 *= t3;
- n3 =
- t3 * t3 *
- grad4(perm[ii + i3 + perm[jj + j3 + perm[kk + k3 + perm[ll + l3]]]],
- x3, y3, z3, w3);
- }
-
- t4 = 0.6f - x4 * x4 - y4 * y4 - z4 * z4 - w4 * w4;
- if (t4 < 0.0f)
- n4 = 0.0f;
- else {
- t4 *= t4;
- n4 =
- t4 * t4 *
- grad4(perm[ii + 1 + perm[jj + 1 + perm[kk + 1 + perm[ll + 1]]]], x4,
- y4, z4, w4);
- }
-
- /* Sum up and scale the result to cover the range [-1,1] */
- return 27.0f * (n0 + n1 + n2 + n3 + n4); /* TODO: The scale factor is preliminary! */
-}