Sweden-Number/dlls/wined3d/baseshader.c

700 lines
23 KiB
C

/*
* shaders implementation
*
* Copyright 2002-2003 Jason Edmeades
* Copyright 2002-2003 Raphael Junqueira
* Copyright 2005 Oliver Stieber
* Copyright 2006 Ivan Gyurdiev
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with this library; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*/
#include "config.h"
#include <string.h>
#include <stdio.h>
#include "wined3d_private.h"
WINE_DEFAULT_DEBUG_CHANNEL(d3d_shader);
#define GLNAME_REQUIRE_GLSL ((const char *)1)
inline static BOOL shader_is_version_token(DWORD token) {
return shader_is_pshader_version(token) ||
shader_is_vshader_version(token);
}
int shader_addline(
SHADER_BUFFER* buffer,
const char *format, ...) {
char* base = buffer->buffer + buffer->bsize;
int rc;
va_list args;
va_start(args, format);
rc = vsnprintf(base, SHADER_PGMSIZE - 1 - buffer->bsize, format, args);
va_end(args);
if (rc < 0 || /* C89 */
rc > SHADER_PGMSIZE - 1 - buffer->bsize) { /* C99 */
ERR("The buffer allocated for the shader program string "
"is too small at %d bytes.\n", SHADER_PGMSIZE);
buffer->bsize = SHADER_PGMSIZE - 1;
return -1;
}
buffer->bsize += rc;
buffer->lineNo++;
TRACE("GL HW (%u, %u) : %s", buffer->lineNo, buffer->bsize, base);
return 0;
}
const SHADER_OPCODE* shader_get_opcode(
IWineD3DBaseShader *iface, const DWORD code) {
IWineD3DBaseShaderImpl *This = (IWineD3DBaseShaderImpl*) iface;
DWORD i = 0;
DWORD version = This->baseShader.version;
DWORD hex_version = This->baseShader.hex_version;
const SHADER_OPCODE *shader_ins = This->baseShader.shader_ins;
/** TODO: use dichotomic search */
while (NULL != shader_ins[i].name) {
if (((code & D3DSI_OPCODE_MASK) == shader_ins[i].opcode) &&
(((hex_version >= shader_ins[i].min_version) && (hex_version <= shader_ins[i].max_version)) ||
((shader_ins[i].min_version == 0) && (shader_ins[i].max_version == 0)))) {
return &shader_ins[i];
}
++i;
}
FIXME("Unsupported opcode %lx(%ld) masked %lx version %ld\n",
code, code, code & D3DSI_OPCODE_MASK, version);
return NULL;
}
/* Read a parameter opcode from the input stream,
* and possibly a relative addressing token.
* Return the number of tokens read */
int shader_get_param(
IWineD3DBaseShader* iface,
const DWORD* pToken,
DWORD* param,
DWORD* addr_token) {
/* PS >= 3.0 have relative addressing (with token)
* VS >= 2.0 have relative addressing (with token)
* VS >= 1.0 < 2.0 have relative addressing (without token)
* The version check below should work in general */
IWineD3DBaseShaderImpl* This = (IWineD3DBaseShaderImpl*) iface;
char rel_token = D3DSHADER_VERSION_MAJOR(This->baseShader.hex_version) >= 2 &&
((*pToken & D3DSHADER_ADDRESSMODE_MASK) == D3DSHADER_ADDRMODE_RELATIVE);
*param = *pToken;
*addr_token = rel_token? *(pToken + 1): 0;
return rel_token? 2:1;
}
/* Return the number of parameters to skip for an opcode */
static inline int shader_skip_opcode(
IWineD3DBaseShaderImpl* This,
const SHADER_OPCODE* curOpcode,
DWORD opcode_token) {
/* Shaders >= 2.0 may contain address tokens, but fortunately they
* have a useful legnth mask - use it here. Shaders 1.0 contain no such tokens */
return (D3DSHADER_VERSION_MAJOR(This->baseShader.hex_version) >= 2)?
((opcode_token & D3DSI_INSTLENGTH_MASK) >> D3DSI_INSTLENGTH_SHIFT):
curOpcode->num_params;
}
/* Read the parameters of an unrecognized opcode from the input stream
* Return the number of tokens read.
*
* Note: This function assumes source or destination token format.
* It will not work with specially-formatted tokens like DEF or DCL,
* but hopefully those would be recognized */
int shader_skip_unrecognized(
IWineD3DBaseShader* iface,
const DWORD* pToken) {
int tokens_read = 0;
int i = 0;
/* TODO: Think of a good name for 0x80000000 and replace it with a constant */
while (*pToken & 0x80000000) {
DWORD param, addr_token;
tokens_read += shader_get_param(iface, pToken, &param, &addr_token);
pToken += tokens_read;
FIXME("Unrecognized opcode param: token=%08lX "
"addr_token=%08lX name=", param, addr_token);
shader_dump_param(iface, param, addr_token, i);
FIXME("\n");
++i;
}
return tokens_read;
}
/* Note: For vertex shaders,
* texUsed = addrUsed, and
* D3DSPR_TEXTURE = D3DSPR_ADDR.
*
* Also note that this does not count the loop register
* as an address register. */
void shader_get_registers_used(
IWineD3DBaseShader *iface,
CONST DWORD* pToken) {
IWineD3DBaseShaderImpl* This = (IWineD3DBaseShaderImpl*) iface;
DWORD* tempsUsed = &This->baseShader.temps_used;
DWORD* texUsed = &This->baseShader.textures_used;
if (pToken == NULL)
return;
*tempsUsed = 0;
*texUsed = 0;
while (D3DVS_END() != *pToken) {
CONST SHADER_OPCODE* curOpcode;
DWORD opcode_token;
/* Skip version */
if (shader_is_version_token(*pToken)) {
++pToken;
continue;
/* Skip comments */
} else if (shader_is_comment(*pToken)) {
DWORD comment_len = (*pToken & D3DSI_COMMENTSIZE_MASK) >> D3DSI_COMMENTSIZE_SHIFT;
++pToken;
pToken += comment_len;
continue;
}
/* Fetch opcode */
opcode_token = *pToken++;
curOpcode = shader_get_opcode(iface, opcode_token);
/* Unhandled opcode, and its parameters */
if (NULL == curOpcode) {
while (*pToken & 0x80000000)
++pToken;
continue;
/* Skip declarations (for now) */
} else if (D3DSIO_DCL == curOpcode->opcode) {
pToken += curOpcode->num_params;
continue;
/* Skip definitions (for now) */
} else if (D3DSIO_DEF == curOpcode->opcode) {
pToken += curOpcode->num_params;
continue;
/* Set texture registers, and temporary registers */
} else {
int i, limit;
/* This will loop over all the registers and try to
* make a bitmask of the ones we're interested in.
*
* Relative addressing tokens are ignored, but that's
* okay, since we'll catch any address registers when
* they are initialized (required by spec) */
limit = (opcode_token & D3DSHADER_INSTRUCTION_PREDICATED)?
curOpcode->num_params + 1: curOpcode->num_params;
for (i = 0; i < limit; ++i) {
DWORD param, addr_token, reg, regtype;
pToken += shader_get_param(iface, pToken, &param, &addr_token);
regtype = (param & D3DSP_REGTYPE_MASK) >> D3DSP_REGTYPE_SHIFT;
reg = param & D3DSP_REGNUM_MASK;
if (D3DSPR_TEXTURE == regtype)
*texUsed |= (1 << reg);
if (D3DSPR_TEMP == regtype)
*tempsUsed |= (1 << reg);
}
}
}
}
void shader_program_dump_decl_usage(
DWORD decl,
DWORD param) {
DWORD regtype = shader_get_regtype(param);
TRACE("dcl_");
if (regtype == D3DSPR_SAMPLER) {
DWORD ttype = decl & D3DSP_TEXTURETYPE_MASK;
switch (ttype) {
case D3DSTT_2D: TRACE("2d"); break;
case D3DSTT_CUBE: TRACE("cube"); break;
case D3DSTT_VOLUME: TRACE("volume"); break;
default: TRACE("unknown_ttype(%08lx)", ttype);
}
} else {
DWORD usage = decl & D3DSP_DCL_USAGE_MASK;
DWORD idx = (decl & D3DSP_DCL_USAGEINDEX_MASK) >> D3DSP_DCL_USAGEINDEX_SHIFT;
switch(usage) {
case D3DDECLUSAGE_POSITION:
TRACE("%s%ld", "position", idx);
break;
case D3DDECLUSAGE_BLENDINDICES:
TRACE("%s", "blend");
break;
case D3DDECLUSAGE_BLENDWEIGHT:
TRACE("%s", "weight");
break;
case D3DDECLUSAGE_NORMAL:
TRACE("%s%ld", "normal", idx);
break;
case D3DDECLUSAGE_PSIZE:
TRACE("%s", "psize");
break;
case D3DDECLUSAGE_COLOR:
if(idx == 0) {
TRACE("%s", "color");
} else {
TRACE("%s%ld", "specular", (idx - 1));
}
break;
case D3DDECLUSAGE_TEXCOORD:
TRACE("%s%ld", "texture", idx);
break;
case D3DDECLUSAGE_TANGENT:
TRACE("%s", "tangent");
break;
case D3DDECLUSAGE_BINORMAL:
TRACE("%s", "binormal");
break;
case D3DDECLUSAGE_TESSFACTOR:
TRACE("%s", "tessfactor");
break;
case D3DDECLUSAGE_POSITIONT:
TRACE("%s%ld", "positionT", idx);
break;
case D3DDECLUSAGE_FOG:
TRACE("%s", "fog");
break;
case D3DDECLUSAGE_DEPTH:
TRACE("%s", "depth");
break;
case D3DDECLUSAGE_SAMPLE:
TRACE("%s", "sample");
break;
default:
FIXME("unknown_semantics(%08lx)", usage);
}
}
}
static void shader_dump_arr_entry(
IWineD3DBaseShader *iface,
const DWORD param,
const DWORD addr_token,
int input) {
DWORD reg = param & D3DSP_REGNUM_MASK;
char relative =
((param & D3DSHADER_ADDRESSMODE_MASK) == D3DSHADER_ADDRMODE_RELATIVE);
TRACE("[");
if (relative) {
if (addr_token)
shader_dump_param(iface, addr_token, 0, input);
else
TRACE("a0.x");
TRACE(" + ");
}
TRACE("%lu]", reg);
}
void shader_dump_param(
IWineD3DBaseShader *iface,
const DWORD param,
const DWORD addr_token,
int input) {
IWineD3DBaseShaderImpl* This = (IWineD3DBaseShaderImpl*) iface;
static const char* rastout_reg_names[] = { "oPos", "oFog", "oPts" };
char swizzle_reg_chars[4];
DWORD reg = param & D3DSP_REGNUM_MASK;
DWORD regtype = shader_get_regtype(param);
/* There are some minor differences between pixel and vertex shaders */
BOOL pshader = shader_is_pshader_version(This->baseShader.hex_version);
/* For one, we'd prefer color components to be shown for pshaders.
* FIXME: use the swizzle function for this */
swizzle_reg_chars[0] = pshader? 'r': 'x';
swizzle_reg_chars[1] = pshader? 'g': 'y';
swizzle_reg_chars[2] = pshader? 'b': 'z';
swizzle_reg_chars[3] = pshader? 'a': 'w';
if (input) {
if ( ((param & D3DSP_SRCMOD_MASK) == D3DSPSM_NEG) ||
((param & D3DSP_SRCMOD_MASK) == D3DSPSM_BIASNEG) ||
((param & D3DSP_SRCMOD_MASK) == D3DSPSM_SIGNNEG) ||
((param & D3DSP_SRCMOD_MASK) == D3DSPSM_X2NEG) )
TRACE("-");
else if ((param & D3DSP_SRCMOD_MASK) == D3DSPSM_COMP)
TRACE("1-");
else if ((param & D3DSP_SRCMOD_MASK) == D3DSPSM_NOT)
TRACE("!");
}
switch (regtype) {
case D3DSPR_TEMP:
TRACE("r%lu", reg);
break;
case D3DSPR_INPUT:
TRACE("v%lu", reg);
break;
case D3DSPR_CONST:
TRACE("c");
shader_dump_arr_entry(iface, param, addr_token, input);
break;
case D3DSPR_TEXTURE: /* vs: case D3DSPR_ADDR */
TRACE("%c%lu", (pshader? 't':'a'), reg);
break;
case D3DSPR_RASTOUT:
TRACE("%s", rastout_reg_names[reg]);
break;
case D3DSPR_COLOROUT:
TRACE("oC%lu", reg);
break;
case D3DSPR_DEPTHOUT:
TRACE("oDepth");
break;
case D3DSPR_ATTROUT:
TRACE("oD%lu", reg);
break;
case D3DSPR_TEXCRDOUT:
/* Vertex shaders >= 3.0 use general purpose output registers
* (D3DSPR_OUTPUT), which can include an address token */
if (D3DSHADER_VERSION_MAJOR(This->baseShader.hex_version) >= 3) {
TRACE("o");
shader_dump_arr_entry(iface, param, addr_token, input);
}
else
TRACE("oT%lu", reg);
break;
case D3DSPR_CONSTINT:
TRACE("i");
shader_dump_arr_entry(iface, param, addr_token, input);
break;
case D3DSPR_CONSTBOOL:
TRACE("b");
shader_dump_arr_entry(iface, param, addr_token, input);
break;
case D3DSPR_LABEL:
TRACE("l%lu", reg);
break;
case D3DSPR_LOOP:
TRACE("aL");
break;
case D3DSPR_SAMPLER:
TRACE("s%lu", reg);
break;
case D3DSPR_PREDICATE:
TRACE("p%lu", reg);
break;
default:
TRACE("unhandled_rtype(%#lx)", regtype);
break;
}
if (!input) {
/* operand output (for modifiers and shift, see dump_ins_modifiers) */
if ((param & D3DSP_WRITEMASK_ALL) != D3DSP_WRITEMASK_ALL) {
TRACE(".");
if (param & D3DSP_WRITEMASK_0) TRACE("%c", swizzle_reg_chars[0]);
if (param & D3DSP_WRITEMASK_1) TRACE("%c", swizzle_reg_chars[1]);
if (param & D3DSP_WRITEMASK_2) TRACE("%c", swizzle_reg_chars[2]);
if (param & D3DSP_WRITEMASK_3) TRACE("%c", swizzle_reg_chars[3]);
}
} else {
/** operand input */
DWORD swizzle = (param & D3DSP_SWIZZLE_MASK) >> D3DSP_SWIZZLE_SHIFT;
DWORD swizzle_r = swizzle & 0x03;
DWORD swizzle_g = (swizzle >> 2) & 0x03;
DWORD swizzle_b = (swizzle >> 4) & 0x03;
DWORD swizzle_a = (swizzle >> 6) & 0x03;
if (0 != (param & D3DSP_SRCMOD_MASK)) {
DWORD mask = param & D3DSP_SRCMOD_MASK;
switch (mask) {
case D3DSPSM_NONE: break;
case D3DSPSM_NEG: break;
case D3DSPSM_NOT: break;
case D3DSPSM_BIAS: TRACE("_bias"); break;
case D3DSPSM_BIASNEG: TRACE("_bias"); break;
case D3DSPSM_SIGN: TRACE("_bx2"); break;
case D3DSPSM_SIGNNEG: TRACE("_bx2"); break;
case D3DSPSM_COMP: break;
case D3DSPSM_X2: TRACE("_x2"); break;
case D3DSPSM_X2NEG: TRACE("_x2"); break;
case D3DSPSM_DZ: TRACE("_dz"); break;
case D3DSPSM_DW: TRACE("_dw"); break;
default:
TRACE("_unknown_modifier(%#lx)", mask >> D3DSP_SRCMOD_SHIFT);
}
}
/**
* swizzle bits fields:
* RRGGBBAA
*/
if ((D3DVS_NOSWIZZLE >> D3DVS_SWIZZLE_SHIFT) != swizzle) { /* ! D3DVS_NOSWIZZLE == 0xE4 << D3DVS_SWIZZLE_SHIFT */
if (swizzle_r == swizzle_g &&
swizzle_r == swizzle_b &&
swizzle_r == swizzle_a) {
TRACE(".%c", swizzle_reg_chars[swizzle_r]);
} else {
TRACE(".%c%c%c%c",
swizzle_reg_chars[swizzle_r],
swizzle_reg_chars[swizzle_g],
swizzle_reg_chars[swizzle_b],
swizzle_reg_chars[swizzle_a]);
}
}
}
}
/** Generate the variable & register declarations for the ARB_vertex_program
output target */
void generate_arb_declarations(IWineD3DBaseShader *iface, SHADER_BUFFER* buffer) {
IWineD3DBaseShaderImpl* This = (IWineD3DBaseShaderImpl*) iface;
DWORD i;
for(i = 0; i < This->baseShader.limits.temporary; i++) {
if (This->baseShader.temps_used & (1 << i))
shader_addline(buffer, "TEMP R%lu;\n", i);
}
for (i = 0; i < This->baseShader.limits.address; i++) {
if (This->baseShader.textures_used & (1 << i))
shader_addline(buffer, "ADDRESS A%ld;\n", i);
}
for(i = 0; i < This->baseShader.limits.texture; i++) {
if (This->baseShader.textures_used & (1 << i))
shader_addline(buffer,"TEMP T%lu;\n", i);
}
/* Texture coordinate registers must be pre-loaded */
for (i = 0; i < This->baseShader.limits.texture; i++) {
if (This->baseShader.textures_used & (1 << i))
shader_addline(buffer, "MOV T%lu, fragment.texcoord[%lu];\n", i, i);
}
}
/** Generate the variable & register declarations for the GLSL
output target */
void generate_glsl_declarations(IWineD3DBaseShader *iface, SHADER_BUFFER* buffer) {
FIXME("GLSL not fully implemented yet.\n");
}
/** Shared code in order to generate the bulk of the shader string.
Use the shader_header_fct & shader_footer_fct to add strings
that are specific to pixel or vertex functions
NOTE: A description of how to parse tokens can be found at:
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/graphics/hh/graphics/usermodedisplaydriver_shader_cc8e4e05-f5c3-4ec0-8853-8ce07c1551b2.xml.asp */
void generate_base_shader(
IWineD3DBaseShader *iface,
SHADER_BUFFER* buffer,
CONST DWORD* pFunction) {
IWineD3DBaseShaderImpl* This = (IWineD3DBaseShaderImpl*) iface;
const DWORD *pToken = pFunction;
const SHADER_OPCODE *curOpcode = NULL;
DWORD opcode_token;
DWORD i;
/* Initialize current parsing state */
This->baseShader.parse_state.current_row = 0;
/* First pass: figure out which temporary and texture registers are used */
shader_get_registers_used(iface, pToken);
TRACE("Texture/Address registers used: %#lx, Temp registers used %#lx\n",
This->baseShader.textures_used, This->baseShader.temps_used);
/* TODO: check register usage against GL/Directx limits, and fail if they're exceeded
nUseAddressRegister < = GL_MAX_PROGRAM_ADDRESS_REGISTERS_AR
nUseTempRegister <= GL_MAX_PROGRAM_LOCAL_PARAMETERS_ARB
*/
/* Pre-declare registers */
if (USING_GLSL)
generate_glsl_declarations(iface, buffer);
else
generate_arb_declarations(iface, buffer);
/* Second pass, process opcodes */
if (NULL != pToken) {
while (D3DPS_END() != *pToken) {
/* Skip version token */
if (shader_is_version_token(*pToken)) {
++pToken;
continue;
}
/* Skip comment tokens */
if (shader_is_comment(*pToken)) {
DWORD comment_len = (*pToken & D3DSI_COMMENTSIZE_MASK) >> D3DSI_COMMENTSIZE_SHIFT;
++pToken;
TRACE("#%s\n", (char*)pToken);
pToken += comment_len;
continue;
}
/* Read opcode */
opcode_token = *pToken++;
curOpcode = shader_get_opcode(iface, opcode_token);
/* Unknown opcode and its parameters */
if (NULL == curOpcode) {
FIXME("Unrecognized opcode: token=%08lX\n", opcode_token);
pToken += shader_skip_unrecognized(iface, pToken);
/* Using GLSL & no generator function exists */
} else if (USING_GLSL && curOpcode->hw_glsl_fct == NULL) {
FIXME("Token %s is not yet implemented with GLSL\n", curOpcode->name);
pToken += shader_skip_opcode(This, curOpcode, opcode_token);
/* Unhandled opcode in ARB */
} else if ( !USING_GLSL && GLNAME_REQUIRE_GLSL == curOpcode->glname) {
FIXME("Token %s requires greater functionality than "
"Vertex or Fragment_Program_ARB supports\n", curOpcode->name);
pToken += shader_skip_opcode(This, curOpcode, opcode_token);
/* If a generator function is set for current shader target, use it */
} else if ((!USING_GLSL && curOpcode->hw_fct != NULL) ||
(USING_GLSL && curOpcode->hw_glsl_fct != NULL)) {
SHADER_OPCODE_ARG hw_arg;
hw_arg.shader = iface;
hw_arg.opcode = curOpcode;
hw_arg.buffer = buffer;
if (curOpcode->num_params > 0) {
DWORD param, addr_token = 0;
/* DCL instruction has usage dst parameter, not register */
if (curOpcode->opcode == D3DSIO_DCL)
param = *pToken++;
else
pToken += shader_get_param(iface, pToken, &param, &addr_token);
hw_arg.dst = param;
hw_arg.dst_addr = addr_token;
if (opcode_token & D3DSHADER_INSTRUCTION_PREDICATED)
hw_arg.predicate = *pToken++;
for (i = 1; i < curOpcode->num_params; i++) {
/* DEF* instructions have constant src parameters, not registers */
if (curOpcode->opcode == D3DSIO_DEF ||
curOpcode->opcode == D3DSIO_DEFI ||
curOpcode->opcode == D3DSIO_DEFB) {
param = *pToken++;
} else
pToken += shader_get_param(iface, pToken, &param, &addr_token);
hw_arg.src[i-1] = param;
hw_arg.src_addr[i-1] = addr_token;
}
}
/* Call appropriate function for output target */
if (USING_GLSL)
curOpcode->hw_glsl_fct(&hw_arg);
else
curOpcode->hw_fct(&hw_arg);
} else {
/* Unless we encounter a no-op command, this opcode is unrecognized */
if (curOpcode->opcode != D3DSIO_NOP) {
FIXME("Can't handle opcode %s in hwShader\n", curOpcode->name);
pToken += shader_skip_opcode(This, curOpcode, opcode_token);
}
}
}
/* TODO: What about result.depth? */
}
}
void shader_dump_ins_modifiers(const DWORD output) {
DWORD shift = (output & D3DSP_DSTSHIFT_MASK) >> D3DSP_DSTSHIFT_SHIFT;
DWORD mmask = output & D3DSP_DSTMOD_MASK;
switch (shift) {
case 0: break;
case 13: TRACE("_d8"); break;
case 14: TRACE("_d4"); break;
case 15: TRACE("_d2"); break;
case 1: TRACE("_x2"); break;
case 2: TRACE("_x4"); break;
case 3: TRACE("_x8"); break;
default: TRACE("_unhandled_shift(%ld)", shift); break;
}
if (mmask & D3DSPDM_SATURATE) TRACE("_sat");
if (mmask & D3DSPDM_PARTIALPRECISION) TRACE("_pp");
if (mmask & D3DSPDM_MSAMPCENTROID) TRACE("_centroid");
mmask &= ~(D3DSPDM_SATURATE | D3DSPDM_PARTIALPRECISION | D3DSPDM_MSAMPCENTROID);
if (mmask)
FIXME("_unrecognized_modifier(%#lx)", mmask >> D3DSP_DSTMOD_SHIFT);
}
/* TODO: Move other shared code here */