path: root/src/mesa/drivers/dri/r300/compiler/r3xx_vertprog.c

/*
 * Copyright 2009 Nicolai Hähnle <nhaehnle@gmail.com>
 *
 * 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
 * on 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
 * THE AUTHOR(S) AND/OR THEIR 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. */

#include "radeon_compiler.h"

#include "../r300_reg.h"

#include "radeon_nqssadce.h"
#include "radeon_program.h"
#include "radeon_program_alu.h"

#include "shader/prog_print.h"


/*
 * Take an already-setup and valid source then swizzle it appropriately to
 * obtain a constant ZERO or ONE source.
 */
#define __CONST(x, y)	\
	(PVS_SRC_OPERAND(t_src_index(vp, &vpi->SrcReg[x]),	\
			   t_swizzle(y),	\
			   t_swizzle(y),	\
			   t_swizzle(y),	\
			   t_swizzle(y),	\
			   t_src_class(vpi->SrcReg[x].File), \
			   NEGATE_NONE) | (vpi->SrcReg[x].RelAddr << 4))


static unsigned long t_dst_mask(GLuint mask)
{
	/* WRITEMASK_* is equivalent to VSF_FLAG_* */
	return mask & WRITEMASK_XYZW;
}

static unsigned long t_dst_class(gl_register_file file)
{

	switch (file) {
	case PROGRAM_TEMPORARY:
		return PVS_DST_REG_TEMPORARY;
	case PROGRAM_OUTPUT:
		return PVS_DST_REG_OUT;
	case PROGRAM_ADDRESS:
		return PVS_DST_REG_A0;
		/*
		   case PROGRAM_INPUT:
		   case PROGRAM_LOCAL_PARAM:
		   case PROGRAM_ENV_PARAM:
		   case PROGRAM_NAMED_PARAM:
		   case PROGRAM_STATE_VAR:
		   case PROGRAM_WRITE_ONLY:
		   case PROGRAM_ADDRESS:
		 */
	default:
		fprintf(stderr, "problem in %s", __FUNCTION__);
		_mesa_exit(-1);
		return -1;
	}
}

static unsigned long t_dst_index(struct r300_vertex_program_code *vp,
				 struct prog_dst_register *dst)
{
	if (dst->File == PROGRAM_OUTPUT)
		return vp->outputs[dst->Index];

	return dst->Index;
}

static unsigned long t_src_class(gl_register_file file)
{
	switch (file) {
	case PROGRAM_BUILTIN:
	case PROGRAM_TEMPORARY:
		return PVS_SRC_REG_TEMPORARY;
	case PROGRAM_INPUT:
		return PVS_SRC_REG_INPUT;
	case PROGRAM_LOCAL_PARAM:
	case PROGRAM_ENV_PARAM:
	case PROGRAM_NAMED_PARAM:
	case PROGRAM_CONSTANT:
	case PROGRAM_STATE_VAR:
		return PVS_SRC_REG_CONSTANT;
		/*
		   case PROGRAM_OUTPUT:
		   case PROGRAM_WRITE_ONLY:
		   case PROGRAM_ADDRESS:
		 */
	default:
		fprintf(stderr, "problem in %s", __FUNCTION__);
		_mesa_exit(-1);
		return -1;
	}
}

static GLboolean t_src_conflict(struct prog_src_register a, struct prog_src_register b)
{
	unsigned long aclass = t_src_class(a.File);
	unsigned long bclass = t_src_class(b.File);

	if (aclass != bclass)
		return GL_FALSE;
	if (aclass == PVS_SRC_REG_TEMPORARY)
		return GL_FALSE;

	if (a.RelAddr || b.RelAddr)
		return GL_TRUE;
	if (a.Index != b.Index)
		return GL_TRUE;

	return GL_FALSE;
}

static INLINE unsigned long t_swizzle(GLubyte swizzle)
{
	/* this is in fact a NOP as the Mesa SWIZZLE_* are all identical to VSF_IN_COMPONENT_* */
	return swizzle;
}

static unsigned long t_src_index(struct r300_vertex_program_code *vp,
				 struct prog_src_register *src)
{
	if (src->File == PROGRAM_INPUT) {
		assert(vp->inputs[src->Index] != -1);
		return vp->inputs[src->Index];
	} else {
		if (src->Index < 0) {
			fprintf(stderr,
				"negative offsets for indirect addressing do not work.\n");
			return 0;
		}
		return src->Index;
	}
}

/* these two functions should probably be merged... */

static unsigned long t_src(struct r300_vertex_program_code *vp,
			   struct prog_src_register *src)
{
	/* src->Negate uses the NEGATE_ flags from program_instruction.h,
	 * which equal our VSF_FLAGS_ values, so it's safe to just pass it here.
	 */
	return PVS_SRC_OPERAND(t_src_index(vp, src),
			       t_swizzle(GET_SWZ(src->Swizzle, 0)),
			       t_swizzle(GET_SWZ(src->Swizzle, 1)),
			       t_swizzle(GET_SWZ(src->Swizzle, 2)),
			       t_swizzle(GET_SWZ(src->Swizzle, 3)),
			       t_src_class(src->File),
			       src->Negate) | (src->RelAddr << 4);
}

static unsigned long t_src_scalar(struct r300_vertex_program_code *vp,
				  struct prog_src_register *src)
{
	/* src->Negate uses the NEGATE_ flags from program_instruction.h,
	 * which equal our VSF_FLAGS_ values, so it's safe to just pass it here.
	 */
	return PVS_SRC_OPERAND(t_src_index(vp, src),
			       t_swizzle(GET_SWZ(src->Swizzle, 0)),
			       t_swizzle(GET_SWZ(src->Swizzle, 0)),
			       t_swizzle(GET_SWZ(src->Swizzle, 0)),
			       t_swizzle(GET_SWZ(src->Swizzle, 0)),
			       t_src_class(src->File),
			       src->Negate ? NEGATE_XYZW : NEGATE_NONE) |
	    (src->RelAddr << 4);
}

static GLboolean valid_dst(struct r300_vertex_program_code *vp,
			   struct prog_dst_register *dst)
{
	if (dst->File == PROGRAM_OUTPUT && vp->outputs[dst->Index] == -1) {
		return GL_FALSE;
	} else if (dst->File == PROGRAM_ADDRESS) {
		assert(dst->Index == 0);
	}

	return GL_TRUE;
}

static void ei_vector1(struct r300_vertex_program_code *vp,
				GLuint hw_opcode,
				struct prog_instruction *vpi,
				GLuint * inst)
{
	inst[0] = PVS_OP_DST_OPERAND(hw_opcode,
				     GL_FALSE,
				     GL_FALSE,
				     t_dst_index(vp, &vpi->DstReg),
				     t_dst_mask(vpi->DstReg.WriteMask),
				     t_dst_class(vpi->DstReg.File));
	inst[1] = t_src(vp, &vpi->SrcReg[0]);
	inst[2] = __CONST(0, SWIZZLE_ZERO);
	inst[3] = __CONST(0, SWIZZLE_ZERO);
}

static void ei_vector2(struct r300_vertex_program_code *vp,
				GLuint hw_opcode,
				struct prog_instruction *vpi,
				GLuint * inst)
{
	inst[0] = PVS_OP_DST_OPERAND(hw_opcode,
				     GL_FALSE,
				     GL_FALSE,
				     t_dst_index(vp, &vpi->DstReg),
				     t_dst_mask(vpi->DstReg.WriteMask),
				     t_dst_class(vpi->DstReg.File));
	inst[1] = t_src(vp, &vpi->SrcReg[0]);
	inst[2] = t_src(vp, &vpi->SrcReg[1]);
	inst[3] = __CONST(1, SWIZZLE_ZERO);
}

static void ei_math1(struct r300_vertex_program_code *vp,
				GLuint hw_opcode,
				struct prog_instruction *vpi,
				GLuint * inst)
{
	inst[0] = PVS_OP_DST_OPERAND(hw_opcode,
				     GL_TRUE,
				     GL_FALSE,
				     t_dst_index(vp, &vpi->DstReg),
				     t_dst_mask(vpi->DstReg.WriteMask),
				     t_dst_class(vpi->DstReg.File));
	inst[1] = t_src_scalar(vp, &vpi->SrcReg[0]);
	inst[2] = __CONST(0, SWIZZLE_ZERO);
	inst[3] = __CONST(0, SWIZZLE_ZERO);
}

static void ei_lit(struct r300_vertex_program_code *vp,
				      struct prog_instruction *vpi,
				      GLuint * inst)
{
	//LIT TMP 1.Y Z TMP 1{} {X W Z Y} TMP 1{} {Y W Z X} TMP 1{} {Y X Z W}

	inst[0] = PVS_OP_DST_OPERAND(ME_LIGHT_COEFF_DX,
				     GL_TRUE,
				     GL_FALSE,
				     t_dst_index(vp, &vpi->DstReg),
				     t_dst_mask(vpi->DstReg.WriteMask),
				     t_dst_class(vpi->DstReg.File));
	/* NOTE: Users swizzling might not work. */
	inst[1] = PVS_SRC_OPERAND(t_src_index(vp, &vpi->SrcReg[0]), t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 0)),	// X
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 3)),	// W
				  PVS_SRC_SELECT_FORCE_0,	// Z
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 1)),	// Y
				  t_src_class(vpi->SrcReg[0].File),
				  vpi->SrcReg[0].Negate ? NEGATE_XYZW : NEGATE_NONE) |
	    (vpi->SrcReg[0].RelAddr << 4);
	inst[2] = PVS_SRC_OPERAND(t_src_index(vp, &vpi->SrcReg[0]), t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 1)),	// Y
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 3)),	// W
				  PVS_SRC_SELECT_FORCE_0,	// Z
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 0)),	// X
				  t_src_class(vpi->SrcReg[0].File),
				  vpi->SrcReg[0].Negate ? NEGATE_XYZW : NEGATE_NONE) |
	    (vpi->SrcReg[0].RelAddr << 4);
	inst[3] = PVS_SRC_OPERAND(t_src_index(vp, &vpi->SrcReg[0]), t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 1)),	// Y
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 0)),	// X
				  PVS_SRC_SELECT_FORCE_0,	// Z
				  t_swizzle(GET_SWZ(vpi->SrcReg[0].Swizzle, 3)),	// W
				  t_src_class(vpi->SrcReg[0].File),
				  vpi->SrcReg[0].Negate ? NEGATE_XYZW : NEGATE_NONE) |
	    (vpi->SrcReg[0].RelAddr << 4);
}

static void ei_mad(struct r300_vertex_program_code *vp,
				      struct prog_instruction *vpi,
				      GLuint * inst)
{
	/* Remarks about hardware limitations of MAD
	 * (please preserve this comment, as this information is _NOT_
	 * in the documentation provided by AMD).
	 *
	 * As described in the documentation, MAD with three unique temporary
	 * source registers requires the use of the macro version.
	 *
	 * However (and this is not mentioned in the documentation), apparently
	 * the macro version is _NOT_ a full superset of the normal version.
	 * In particular, the macro version does not always work when relative
	 * addressing is used in the source operands.
	 *
	 * This limitation caused incorrect rendering in Sauerbraten's OpenGL
	 * assembly shader path when using medium quality animations
	 * (i.e. animations with matrix blending instead of quaternion blending).
	 *
	 * Unfortunately, I (nha) have been unable to extract a Piglit regression
	 * test for this issue - for some reason, it is possible to have vertex
	 * programs whose prefix is *exactly* the same as the prefix of the
	 * offending program in Sauerbraten up to the offending instruction
	 * without causing any trouble.
	 *
	 * Bottom line: Only use the macro version only when really necessary;
	 * according to AMD docs, this should improve performance by one clock
	 * as a nice side bonus.
	 */
	if (vpi->SrcReg[0].File == PROGRAM_TEMPORARY &&
	    vpi->SrcReg[1].File == PROGRAM_TEMPORARY &&
	    vpi->SrcReg[2].File == PROGRAM_TEMPORARY &&
	    vpi->SrcReg[0].Index != vpi->SrcReg[1].Index &&
	    vpi->SrcReg[0].Index != vpi->SrcReg[2].Index &&
	    vpi->SrcReg[1].Index != vpi->SrcReg[2].Index) {
		inst[0] = PVS_OP_DST_OPERAND(PVS_MACRO_OP_2CLK_MADD,
				GL_FALSE,
				GL_TRUE,
				t_dst_index(vp, &vpi->DstReg),
				t_dst_mask(vpi->DstReg.WriteMask),
				t_dst_class(vpi->DstReg.File));
	} else {
		inst[0] = PVS_OP_DST_OPERAND(VE_MULTIPLY_ADD,
				GL_FALSE,
				GL_FALSE,
				t_dst_index(vp, &vpi->DstReg),
				t_dst_mask(vpi->DstReg.WriteMask),
				t_dst_class(vpi->DstReg.File));
	}
	inst[1] = t_src(vp, &vpi->SrcReg[0]);
	inst[2] = t_src(vp, &vpi->SrcReg[1]);
	inst[3] = t_src(vp, &vpi->SrcReg[2]);
}

static void ei_pow(struct r300_vertex_program_code *vp,
				      struct prog_instruction *vpi,
				      GLuint * inst)
{
	inst[0] = PVS_OP_DST_OPERAND(ME_POWER_FUNC_FF,
				     GL_TRUE,
				     GL_FALSE,
				     t_dst_index(vp, &vpi->DstReg),
				     t_dst_mask(vpi->DstReg.WriteMask),
				     t_dst_class(vpi->DstReg.File));
	inst[1] = t_src_scalar(vp, &vpi->SrcReg[0]);
	inst[2] = __CONST(0, SWIZZLE_ZERO);
	inst[3] = t_src_scalar(vp, &vpi->SrcReg[1]);
}


static void translate_vertex_program(struct r300_vertex_program_compiler * compiler)
{
	struct rc_instruction *rci;

	compiler->code->pos_end = 0;	/* Not supported yet */
	compiler->code->length = 0;

	compiler->SetHwInputOutput(compiler);

	for(rci = compiler->Base.Program.Instructions.Next; rci != &compiler->Base.Program.Instructions; rci = rci->Next) {
		struct prog_instruction *vpi = &rci->I;
		GLuint *inst = compiler->code->body.d + compiler->code->length;

		/* Skip instructions writing to non-existing destination */
		if (!valid_dst(compiler->code, &vpi->DstReg))
			continue;

		if (compiler->code->length >= VSF_MAX_FRAGMENT_LENGTH) {
			rc_error(&compiler->Base, "Vertex program has too many instructions\n");
			return;
		}

		switch (vpi->Opcode) {
		case OPCODE_ADD: ei_vector2(compiler->code, VE_ADD, vpi, inst); break;
		case OPCODE_ARL: ei_vector1(compiler->code, VE_FLT2FIX_DX, vpi, inst); break;
		case OPCODE_DP4: ei_vector2(compiler->code, VE_DOT_PRODUCT, vpi, inst); break;
		case OPCODE_DST: ei_vector2(compiler->code, VE_DISTANCE_VECTOR, vpi, inst); break;
		case OPCODE_EX2: ei_math1(compiler->code, ME_EXP_BASE2_FULL_DX, vpi, inst); break;
		case OPCODE_EXP: ei_math1(compiler->code, ME_EXP_BASE2_DX, vpi, inst); break;
		case OPCODE_FRC: ei_vector1(compiler->code, VE_FRACTION, vpi, inst); break;
		case OPCODE_LG2: ei_math1(compiler->code, ME_LOG_BASE2_FULL_DX, vpi, inst); break;
		case OPCODE_LIT: ei_lit(compiler->code, vpi, inst); break;
		case OPCODE_LOG: ei_math1(compiler->code, ME_LOG_BASE2_DX, vpi, inst); break;
		case OPCODE_MAD: ei_mad(compiler->code, vpi, inst); break;
		case OPCODE_MAX: ei_vector2(compiler->code, VE_MAXIMUM, vpi, inst); break;
		case OPCODE_MIN: ei_vector2(compiler->code, VE_MINIMUM, vpi, inst); break;
		case OPCODE_MOV: ei_vector1(compiler->code, VE_ADD, vpi, inst); break;
		case OPCODE_MUL: ei_vector2(compiler->code, VE_MULTIPLY, vpi, inst); break;
		case OPCODE_POW: ei_pow(compiler->code, vpi, inst); break;
		case OPCODE_RCP: ei_math1(compiler->code, ME_RECIP_DX, vpi, inst); break;
		case OPCODE_RSQ: ei_math1(compiler->code, ME_RECIP_SQRT_DX, vpi, inst); break;
		case OPCODE_SGE: ei_vector2(compiler->code, VE_SET_GREATER_THAN_EQUAL, vpi, inst); break;
		case OPCODE_SLT: ei_vector2(compiler->code, VE_SET_LESS_THAN, vpi, inst); break;
		default:
			rc_error(&compiler->Base, "Unknown opcode %i\n", vpi->Opcode);
			return;
		}

		compiler->code->length += 4;

		if (compiler->Base.Error)
			return;
	}
}

struct temporary_allocation {
	GLuint Allocated:1;
	GLuint HwTemp:15;
	struct rc_instruction * LastRead;
};

static void allocate_temporary_registers(struct r300_vertex_program_compiler * compiler)
{
	struct rc_instruction *inst;
	GLuint num_orig_temps = 0;
	GLboolean hwtemps[VSF_MAX_FRAGMENT_TEMPS];
	struct temporary_allocation * ta;
	GLuint i, j;

	compiler->code->num_temporaries = 0;
	memset(hwtemps, 0, sizeof(hwtemps));

	/* Pass 1: Count original temporaries and allocate structures */
	for(inst = compiler->Base.Program.Instructions.Next; inst != &compiler->Base.Program.Instructions; inst = inst->Next) {
		GLuint numsrcs = _mesa_num_inst_src_regs(inst->I.Opcode);
		GLuint numdsts = _mesa_num_inst_dst_regs(inst->I.Opcode);

		for (i = 0; i < numsrcs; ++i) {
			if (inst->I.SrcReg[i].File == PROGRAM_TEMPORARY) {
				if (inst->I.SrcReg[i].Index >= num_orig_temps)
					num_orig_temps = inst->I.SrcReg[i].Index + 1;
			}
		}

		if (numdsts) {
			if (inst->I.DstReg.File == PROGRAM_TEMPORARY) {
				if (inst->I.DstReg.Index >= num_orig_temps)
					num_orig_temps = inst->I.DstReg.Index + 1;
			}
		}
	}

	ta = (struct temporary_allocation*)memory_pool_malloc(&compiler->Base.Pool,
			sizeof(struct temporary_allocation) * num_orig_temps);
	memset(ta, 0, sizeof(struct temporary_allocation) * num_orig_temps);

	/* Pass 2: Determine original temporary lifetimes */
	for(inst = compiler->Base.Program.Instructions.Next; inst != &compiler->Base.Program.Instructions; inst = inst->Next) {
		GLuint numsrcs = _mesa_num_inst_src_regs(inst->I.Opcode);

		for (i = 0; i < numsrcs; ++i) {
			if (inst->I.SrcReg[i].File == PROGRAM_TEMPORARY)
				ta[inst->I.SrcReg[i].Index].LastRead = inst;
		}
	}

	/* Pass 3: Register allocation */
	for(inst = compiler->Base.Program.Instructions.Next; inst != &compiler->Base.Program.Instructions; inst = inst->Next) {
		GLuint numsrcs = _mesa_num_inst_src_regs(inst->I.Opcode);
		GLuint numdsts = _mesa_num_inst_dst_regs(inst->I.Opcode);

		for (i = 0; i < numsrcs; ++i) {
			if (inst->I.SrcReg[i].File == PROGRAM_TEMPORARY) {
				GLuint orig = inst->I.SrcReg[i].Index;
				inst->I.SrcReg[i].Index = ta[orig].HwTemp;

				if (ta[orig].Allocated && inst == ta[orig].LastRead)
					hwtemps[ta[orig].HwTemp] = GL_FALSE;
			}
		}

		if (numdsts) {
			if (inst->I.DstReg.File == PROGRAM_TEMPORARY) {
				GLuint orig = inst->I.DstReg.Index;

				if (!ta[orig].Allocated) {
					for(j = 0; j < VSF_MAX_FRAGMENT_TEMPS; ++j) {
						if (!hwtemps[j])
							break;
					}
					if (j >= VSF_MAX_FRAGMENT_TEMPS) {
						fprintf(stderr, "Out of hw temporaries\n");
					} else {
						ta[orig].Allocated = GL_TRUE;
						ta[orig].HwTemp = j;
						hwtemps[j] = GL_TRUE;

						if (j >= compiler->code->num_temporaries)
							compiler->code->num_temporaries = j + 1;
					}
				}

				inst->I.DstReg.Index = ta[orig].HwTemp;
			}
		}
	}
}


/**
 * Vertex engine cannot read two inputs or two constants at the same time.
 * Introduce intermediate MOVs to temporary registers to account for this.
 */
static GLboolean transform_source_conflicts(
	struct radeon_compiler *c,
	struct rc_instruction* inst,
	void* unused)
{
	GLuint num_operands = _mesa_num_inst_src_regs(inst->I.Opcode);

	if (num_operands == 3) {
		if (t_src_conflict(inst->I.SrcReg[1], inst->I.SrcReg[2])
		    || t_src_conflict(inst->I.SrcReg[0], inst->I.SrcReg[2])) {
			int tmpreg = rc_find_free_temporary(c);
			struct rc_instruction * inst_mov = rc_insert_new_instruction(c, inst->Prev);
			inst_mov->I.Opcode = OPCODE_MOV;
			inst_mov->I.DstReg.File = PROGRAM_TEMPORARY;
			inst_mov->I.DstReg.Index = tmpreg;
			inst_mov->I.SrcReg[0] = inst->I.SrcReg[2];

			reset_srcreg(&inst->I.SrcReg[2]);
			inst->I.SrcReg[2].File = PROGRAM_TEMPORARY;
			inst->I.SrcReg[2].Index = tmpreg;
		}
	}

	if (num_operands >= 2) {
		if (t_src_conflict(inst->I.SrcReg[1], inst->I.SrcReg[0])) {
			int tmpreg = rc_find_free_temporary(c);
			struct rc_instruction * inst_mov = rc_insert_new_instruction(c, inst->Prev);
			inst_mov->I.Opcode = OPCODE_MOV;
			inst_mov->I.DstReg.File = PROGRAM_TEMPORARY;
			inst_mov->I.DstReg.Index = tmpreg;
			inst_mov->I.SrcReg[0] = inst->I.SrcReg[1];

			reset_srcreg(&inst->I.SrcReg[1]);
			inst->I.SrcReg[1].File = PROGRAM_TEMPORARY;
			inst->I.SrcReg[1].Index = tmpreg;
		}
	}

	return GL_TRUE;
}

static void addArtificialOutputs(struct r300_vertex_program_compiler * compiler)
{
	int i;

	for(i = 0; i < 32; ++i) {
		if ((compiler->RequiredOutputs & (1 << i)) &&
		    !(compiler->Base.Program.OutputsWritten & (1 << i))) {
			struct rc_instruction * inst = rc_insert_new_instruction(&compiler->Base, compiler->Base.Program.Instructions.Prev);
			inst->I.Opcode = OPCODE_MOV;

			inst->I.DstReg.File = PROGRAM_OUTPUT;
			inst->I.DstReg.Index = i;
			inst->I.DstReg.WriteMask = WRITEMASK_XYZW;

			inst->I.SrcReg[0].File = PROGRAM_CONSTANT;
			inst->I.SrcReg[0].Index = 0;
			inst->I.SrcReg[0].Swizzle = SWIZZLE_XYZW;

			compiler->Base.Program.OutputsWritten |= 1 << i;
		}
	}
}

static void nqssadceInit(struct nqssadce_state* s)
{
	struct r300_vertex_program_compiler * compiler = s->UserData;
	int i;

	for(i = 0; i < VERT_RESULT_MAX; ++i) {
		if (compiler->RequiredOutputs & (1 << i))
			s->Outputs[i].Sourced = WRITEMASK_XYZW;
	}
}

static GLboolean swizzleIsNative(GLuint opcode, struct prog_src_register reg)
{
	(void) opcode;
	(void) reg;

	return GL_TRUE;
}



void r3xx_compile_vertex_program(struct r300_vertex_program_compiler* compiler)
{
	addArtificialOutputs(compiler);

	{
		struct radeon_program_transformation transformations[] = {
			{ &r300_transform_vertex_alu, 0 },
		};
		radeonLocalTransform(&compiler->Base, 1, transformations);
	}

	if (compiler->Base.Debug) {
		fprintf(stderr, "Vertex program after native rewrite:\n");
		rc_print_program(&compiler->Base.Program);
		fflush(stderr);
	}

	{
		/* Note: This pass has to be done seperately from ALU rewrite,
		 * otherwise non-native ALU instructions with source conflits
		 * will not be treated properly.
		 */
		struct radeon_program_transformation transformations[] = {
			{ &transform_source_conflicts, 0 },
		};
		radeonLocalTransform(&compiler->Base, 1, transformations);
	}

	if (compiler->Base.Debug) {
		fprintf(stderr, "Vertex program after source conflict resolve:\n");
		rc_print_program(&compiler->Base.Program);
		fflush(stderr);
	}

	{
		struct radeon_nqssadce_descr nqssadce = {
			.Init = &nqssadceInit,
			.IsNativeSwizzle = &swizzleIsNative,
			.BuildSwizzle = NULL
		};
		radeonNqssaDce(&compiler->Base, &nqssadce, compiler);

		/* We need this step for reusing temporary registers */
		allocate_temporary_registers(compiler);

		if (compiler->Base.Debug) {
			fprintf(stderr, "Vertex program after NQSSADCE:\n");
			rc_print_program(&compiler->Base.Program);
			fflush(stderr);
		}
	}

	translate_vertex_program(compiler);

	rc_constants_copy(&compiler->code->constants, &compiler->Base.Program.Constants);

	compiler->code->InputsRead = compiler->Base.Program.InputsRead;
	compiler->code->OutputsWritten = compiler->Base.Program.OutputsWritten;

	if (compiler->Base.Debug) {
		fprintf(stderr, "Final vertex program code:\n");
		r300_vertex_program_dump(compiler->code);
	}
}