Newer
Older
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
if (*ptr >= sm1->end)
return;
remaining = sm1->end - *ptr;
token = **ptr;
while ((token & VKD3D_SM1_OPCODE_MASK) == VKD3D_SM1_OP_COMMENT)
{
size = (token & VKD3D_SM1_COMMENT_SIZE_MASK) >> VKD3D_SM1_COMMENT_SIZE_SHIFT;
if (size > --remaining)
{
vkd3d_shader_parser_error(&sm1->p, VKD3D_SHADER_ERROR_D3DBC_UNEXPECTED_EOF,
"Encountered a %u token comment, but only %zu token(s) is/are remaining.",
size, remaining);
return;
}
comment = (const char *)++(*ptr);
remaining -= size;
*ptr += size;
if (size > 1 && *(const uint32_t *)comment == TAG_TEXT)
{
const char *end = comment + size * sizeof(token);
const char *p = comment + sizeof(token);
const char *line;
TRACE("// TEXT\n");
for (line = p; line < end; line = p)
{
if (!(p = memchr(line, '\n', end - line)))
p = end;
else
++p;
TRACE("// %s\n", debugstr_an(line, p - line));
}
}
else if (size)
{
TRACE("// %s\n", debugstr_an(comment, size * sizeof(token)));
}
else
break;
if (!remaining)
break;
token = **ptr;
}
}
static void shader_sm1_validate_instruction(struct vkd3d_shader_sm1_parser *sm1, struct vkd3d_shader_instruction *ins)
{
if ((ins->handler_idx == VKD3DSIH_BREAKP || ins->handler_idx == VKD3DSIH_IF) && ins->flags)
{
vkd3d_shader_parser_warning(&sm1->p, VKD3D_SHADER_WARNING_D3DBC_IGNORED_INSTRUCTION_FLAGS,
"Ignoring unexpected instruction flags %#x.", ins->flags);
ins->flags = 0;
}
}
static unsigned int mask_from_swizzle(unsigned int swizzle)
{
return (1u << vkd3d_swizzle_get_component(swizzle, 0))
| (1u << vkd3d_swizzle_get_component(swizzle, 1))
| (1u << vkd3d_swizzle_get_component(swizzle, 2))
| (1u << vkd3d_swizzle_get_component(swizzle, 3));
}
Henri Verbeet
committed
static void shader_sm1_read_instruction(struct vkd3d_shader_sm1_parser *sm1, struct vkd3d_shader_instruction *ins)
{
struct vkd3d_shader_src_param *src_params, *predicate;
const struct vkd3d_sm1_opcode_info *opcode_info;
struct vkd3d_shader_dst_param *dst_param;
const uint32_t **ptr = &sm1->ptr;
uint32_t opcode_token;
const uint32_t *p;
bool predicated;
unsigned int i;
shader_sm1_read_comment(sm1);
if (*ptr >= sm1->end)
{
WARN("End of byte-code, failed to read opcode.\n");
goto fail;
}
Henri Verbeet
committed
++sm1->p.location.line;
opcode_token = read_u32(ptr);
if (!(opcode_info = shader_sm1_get_opcode_info(sm1, opcode_token & VKD3D_SM1_OPCODE_MASK)))
{
Henri Verbeet
committed
vkd3d_shader_parser_error(&sm1->p, VKD3D_SHADER_ERROR_D3DBC_INVALID_OPCODE,
"Invalid opcode %#x (token 0x%08x, shader version %u.%u).",
opcode_token & VKD3D_SM1_OPCODE_MASK, opcode_token,
sm1->p.shader_version.major, sm1->p.shader_version.minor);
goto fail;
}
vsir_instruction_init(ins, &sm1->p.location, opcode_info->vkd3d_opcode);
ins->flags = (opcode_token & VKD3D_SM1_INSTRUCTION_FLAGS_MASK) >> VKD3D_SM1_INSTRUCTION_FLAGS_SHIFT;
ins->coissue = opcode_token & VKD3D_SM1_COISSUE;
ins->raw = false;
ins->structured = false;
predicated = !!(opcode_token & VKD3D_SM1_INSTRUCTION_PREDICATED);
Henri Verbeet
committed
ins->predicate = predicate = predicated ? shader_parser_get_src_params(&sm1->p, 1) : NULL;
ins->dst_count = opcode_info->dst_count;
Henri Verbeet
committed
ins->dst = dst_param = shader_parser_get_dst_params(&sm1->p, ins->dst_count);
ins->src_count = opcode_info->src_count;
Henri Verbeet
committed
ins->src = src_params = shader_parser_get_src_params(&sm1->p, ins->src_count);
if ((!predicate && predicated) || (!src_params && ins->src_count) || (!dst_param && ins->dst_count))
{
Henri Verbeet
committed
vkd3d_shader_parser_error(&sm1->p, VKD3D_SHADER_ERROR_D3DBC_OUT_OF_MEMORY, "Out of memory.");
goto fail;
}
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
ins->resource_type = VKD3D_SHADER_RESOURCE_NONE;
ins->resource_stride = 0;
ins->resource_data_type[0] = VKD3D_DATA_FLOAT;
ins->resource_data_type[1] = VKD3D_DATA_FLOAT;
ins->resource_data_type[2] = VKD3D_DATA_FLOAT;
ins->resource_data_type[3] = VKD3D_DATA_FLOAT;
memset(&ins->texel_offset, 0, sizeof(ins->texel_offset));
p = *ptr;
shader_sm1_skip_opcode(sm1, ptr, opcode_info, opcode_token);
if (*ptr > sm1->end)
{
vkd3d_shader_parser_error(&sm1->p, VKD3D_SHADER_ERROR_D3DBC_UNEXPECTED_EOF,
"The current instruction ends %zu token(s) past the end of the shader.",
*ptr - sm1->end);
goto fail;
}
if (ins->handler_idx == VKD3DSIH_DCL)
{
shader_sm1_read_semantic(sm1, &p, &ins->declaration.semantic);
}
else if (ins->handler_idx == VKD3DSIH_DEF)
{
shader_sm1_read_dst_param(sm1, &p, dst_param);
shader_sm1_read_immconst(sm1, &p, &src_params[0], VKD3D_IMMCONST_VEC4, VKD3D_DATA_FLOAT);
shader_sm1_scan_register(sm1, &dst_param->reg, dst_param->write_mask, true);
}
else if (ins->handler_idx == VKD3DSIH_DEFB)
{
shader_sm1_read_dst_param(sm1, &p, dst_param);
shader_sm1_read_immconst(sm1, &p, &src_params[0], VKD3D_IMMCONST_SCALAR, VKD3D_DATA_UINT);
shader_sm1_scan_register(sm1, &dst_param->reg, dst_param->write_mask, true);
}
else if (ins->handler_idx == VKD3DSIH_DEFI)
{
shader_sm1_read_dst_param(sm1, &p, dst_param);
shader_sm1_read_immconst(sm1, &p, &src_params[0], VKD3D_IMMCONST_VEC4, VKD3D_DATA_INT);
shader_sm1_scan_register(sm1, &dst_param->reg, dst_param->write_mask, true);
}
else
{
/* Destination token */
if (ins->dst_count)
shader_sm1_read_dst_param(sm1, &p, dst_param);
shader_sm1_scan_register(sm1, &dst_param->reg, dst_param->write_mask, false);
/* Predication token */
if (ins->predicate)
shader_sm1_read_src_param(sm1, &p, predicate);
/* Other source tokens */
for (i = 0; i < ins->src_count; ++i)
shader_sm1_read_src_param(sm1, &p, &src_params[i]);
shader_sm1_scan_register(sm1, &src_params[i].reg, mask_from_swizzle(src_params[i].swizzle), false);
}
if (sm1->abort)
{
sm1->abort = false;
goto fail;
}
shader_sm1_validate_instruction(sm1, ins);
return;
fail:
ins->handler_idx = VKD3DSIH_INVALID;
*ptr = sm1->end;
}
Henri Verbeet
committed
static bool shader_sm1_is_end(struct vkd3d_shader_sm1_parser *sm1)
{
const uint32_t **ptr = &sm1->ptr;
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
shader_sm1_read_comment(sm1);
if (*ptr >= sm1->end)
return true;
if (**ptr == VKD3D_SM1_END)
{
++(*ptr);
return true;
}
return false;
}
const struct vkd3d_shader_parser_ops shader_sm1_parser_ops =
{
.parser_destroy = shader_sm1_destroy,
};
static enum vkd3d_result shader_sm1_init(struct vkd3d_shader_sm1_parser *sm1,
const struct vkd3d_shader_compile_info *compile_info, struct vkd3d_shader_message_context *message_context)
{
const struct vkd3d_shader_location location = {.source_name = compile_info->source_name};
const uint32_t *code = compile_info->source.code;
size_t code_size = compile_info->source.size;
struct vkd3d_shader_desc *shader_desc;
struct vkd3d_shader_version version;
uint16_t shader_type;
size_t token_count;
token_count = code_size / sizeof(*sm1->start);
if (token_count < 2)
{
vkd3d_shader_error(message_context, &location, VKD3D_SHADER_ERROR_D3DBC_UNEXPECTED_EOF,
"Invalid shader size %zu (token count %zu). At least 2 tokens are required.",
code_size, token_count);
return VKD3D_ERROR_INVALID_SHADER;
}
TRACE("Version: 0x%08x.\n", code[0]);
shader_type = code[0] >> 16;
version.major = VKD3D_SM1_VERSION_MAJOR(code[0]);
version.minor = VKD3D_SM1_VERSION_MINOR(code[0]);
switch (shader_type)
{
case VKD3D_SM1_VS:
version.type = VKD3D_SHADER_TYPE_VERTEX;
sm1->opcode_table = vs_opcode_table;
break;
case VKD3D_SM1_PS:
version.type = VKD3D_SHADER_TYPE_PIXEL;
sm1->opcode_table = ps_opcode_table;
break;
default:
vkd3d_shader_error(message_context, &location, VKD3D_SHADER_ERROR_D3DBC_INVALID_VERSION_TOKEN,
"Invalid shader type %#x (token 0x%08x).", shader_type, code[0]);
return VKD3D_ERROR_INVALID_SHADER;
}
if (!shader_ver_le(&version, 3, 0))
{
vkd3d_shader_error(message_context, &location, VKD3D_SHADER_ERROR_D3DBC_INVALID_VERSION_TOKEN,
"Invalid shader version %u.%u (token 0x%08x).", version.major, version.minor, code[0]);
return VKD3D_ERROR_INVALID_SHADER;
}
sm1->start = &code[1];
sm1->end = &code[token_count];
/* Estimate instruction count to avoid reallocation in most shaders. */
Conor McCarthy
committed
if (!vkd3d_shader_parser_init(&sm1->p, message_context, compile_info->source_name, &version, &shader_sm1_parser_ops,
code_size != ~(size_t)0 ? token_count / 4u + 4 : 16))
return VKD3D_ERROR_OUT_OF_MEMORY;
shader_desc = &sm1->p.shader_desc;
shader_desc->byte_code = code;
shader_desc->byte_code_size = code_size;
sm1->ptr = sm1->start;
return VKD3D_OK;
}
static uint32_t get_external_constant_count(struct vkd3d_shader_sm1_parser *sm1,
enum vkd3d_shader_d3dbc_constant_register set)
{
unsigned int j;
/* Find the highest constant index which is not written by a DEF
* instruction. We can't (easily) use an FFZ function for this since it
* needs to be limited by the highest used register index. */
for (j = sm1->p.shader_desc.flat_constant_count[set].used; j > 0; --j)
{
if (!bitmap_is_set(sm1->constant_def_mask[set], j - 1))
return j;
}
return 0;
}
int vkd3d_shader_sm1_parser_create(const struct vkd3d_shader_compile_info *compile_info,
struct vkd3d_shader_message_context *message_context, struct vkd3d_shader_parser **parser)
{
struct vkd3d_shader_instruction_array *instructions;
struct vkd3d_shader_instruction *ins;
struct vkd3d_shader_sm1_parser *sm1;
unsigned int i;
int ret;
if (!(sm1 = vkd3d_calloc(1, sizeof(*sm1))))
{
ERR("Failed to allocate parser.\n");
return VKD3D_ERROR_OUT_OF_MEMORY;
}
if ((ret = shader_sm1_init(sm1, compile_info, message_context)) < 0)
{
WARN("Failed to initialise shader parser, ret %d.\n", ret);
vkd3d_free(sm1);
return ret;
}
instructions = &sm1->p.instructions;
Henri Verbeet
committed
while (!shader_sm1_is_end(sm1))
{
if (!shader_instruction_array_reserve(instructions, instructions->count + 1))
{
ERR("Failed to allocate instructions.\n");
vkd3d_shader_parser_error(&sm1->p, VKD3D_SHADER_ERROR_D3DBC_OUT_OF_MEMORY, "Out of memory.");
shader_sm1_destroy(&sm1->p);
return VKD3D_ERROR_OUT_OF_MEMORY;
}
ins = &instructions->elements[instructions->count];
Henri Verbeet
committed
shader_sm1_read_instruction(sm1, ins);
if (ins->handler_idx == VKD3DSIH_INVALID)
{
WARN("Encountered unrecognized or invalid instruction.\n");
shader_sm1_destroy(&sm1->p);
return VKD3D_ERROR_INVALID_SHADER;
}
++instructions->count;
}
for (i = 0; i < ARRAY_SIZE(sm1->p.shader_desc.flat_constant_count); ++i)
sm1->p.shader_desc.flat_constant_count[i].external = get_external_constant_count(sm1, i);
Giovanni Mascellani
committed
if (!sm1->p.failed)
vsir_validate(&sm1->p);
if (sm1->p.failed)
{
WARN("Failed to parse shader.\n");
shader_sm1_destroy(&sm1->p);
return VKD3D_ERROR_INVALID_SHADER;
}
*parser = &sm1->p;
return VKD3D_OK;
}
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
bool hlsl_sm1_register_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_semantic *semantic,
bool output, D3DSHADER_PARAM_REGISTER_TYPE *type, unsigned int *reg)
{
unsigned int i;
static const struct
{
const char *semantic;
bool output;
enum vkd3d_shader_type shader_type;
unsigned int major_version;
D3DSHADER_PARAM_REGISTER_TYPE type;
DWORD offset;
}
register_table[] =
{
{"color", false, VKD3D_SHADER_TYPE_PIXEL, 1, D3DSPR_INPUT},
{"texcoord", false, VKD3D_SHADER_TYPE_PIXEL, 1, D3DSPR_TEXTURE},
{"color", true, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_COLOROUT},
{"depth", true, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_DEPTHOUT},
{"sv_depth", true, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_DEPTHOUT},
{"sv_target", true, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_COLOROUT},
{"color", false, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_INPUT},
{"texcoord", false, VKD3D_SHADER_TYPE_PIXEL, 2, D3DSPR_TEXTURE},
{"color", true, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_COLOROUT},
{"depth", true, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_DEPTHOUT},
{"sv_depth", true, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_DEPTHOUT},
{"sv_target", true, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_COLOROUT},
{"sv_position", false, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_MISCTYPE, D3DSMO_POSITION},
{"vface", false, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_MISCTYPE, D3DSMO_FACE},
{"vpos", false, VKD3D_SHADER_TYPE_PIXEL, 3, D3DSPR_MISCTYPE, D3DSMO_POSITION},
{"color", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_ATTROUT},
{"fog", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_RASTOUT, D3DSRO_FOG},
{"position", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_RASTOUT, D3DSRO_POSITION},
{"psize", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_RASTOUT, D3DSRO_POINT_SIZE},
{"sv_position", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_RASTOUT, D3DSRO_POSITION},
{"texcoord", true, VKD3D_SHADER_TYPE_VERTEX, 1, D3DSPR_TEXCRDOUT},
{"color", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_ATTROUT},
{"fog", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_RASTOUT, D3DSRO_FOG},
{"position", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_RASTOUT, D3DSRO_POSITION},
{"psize", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_RASTOUT, D3DSRO_POINT_SIZE},
{"sv_position", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_RASTOUT, D3DSRO_POSITION},
{"texcoord", true, VKD3D_SHADER_TYPE_VERTEX, 2, D3DSPR_TEXCRDOUT},
};
for (i = 0; i < ARRAY_SIZE(register_table); ++i)
{
if (!ascii_strcasecmp(semantic->name, register_table[i].semantic)
&& output == register_table[i].output
&& ctx->profile->type == register_table[i].shader_type
&& ctx->profile->major_version == register_table[i].major_version)
{
*type = register_table[i].type;
if (register_table[i].type == D3DSPR_MISCTYPE || register_table[i].type == D3DSPR_RASTOUT)
*reg = register_table[i].offset;
else
*reg = semantic->index;
return true;
}
}
return false;
}
bool hlsl_sm1_usage_from_semantic(const struct hlsl_semantic *semantic, D3DDECLUSAGE *usage, uint32_t *usage_idx)
{
static const struct
{
const char *name;
D3DDECLUSAGE usage;
}
semantics[] =
{
{"binormal", D3DDECLUSAGE_BINORMAL},
{"blendindices", D3DDECLUSAGE_BLENDINDICES},
{"blendweight", D3DDECLUSAGE_BLENDWEIGHT},
{"color", D3DDECLUSAGE_COLOR},
{"depth", D3DDECLUSAGE_DEPTH},
{"fog", D3DDECLUSAGE_FOG},
{"normal", D3DDECLUSAGE_NORMAL},
{"position", D3DDECLUSAGE_POSITION},
{"positiont", D3DDECLUSAGE_POSITIONT},
{"psize", D3DDECLUSAGE_PSIZE},
{"sample", D3DDECLUSAGE_SAMPLE},
{"sv_depth", D3DDECLUSAGE_DEPTH},
{"sv_position", D3DDECLUSAGE_POSITION},
{"sv_target", D3DDECLUSAGE_COLOR},
{"tangent", D3DDECLUSAGE_TANGENT},
{"tessfactor", D3DDECLUSAGE_TESSFACTOR},
{"texcoord", D3DDECLUSAGE_TEXCOORD},
};
unsigned int i;
for (i = 0; i < ARRAY_SIZE(semantics); ++i)
{
if (!ascii_strcasecmp(semantic->name, semantics[i].name))
{
*usage = semantics[i].usage;
*usage_idx = semantic->index;
return true;
}
}
return false;
}
static uint32_t sm1_version(enum vkd3d_shader_type type, unsigned int major, unsigned int minor)
{
if (type == VKD3D_SHADER_TYPE_VERTEX)
return D3DVS_VERSION(major, minor);
else
return D3DPS_VERSION(major, minor);
}
static D3DXPARAMETER_CLASS sm1_class(const struct hlsl_type *type)
{
switch (type->class)
{
case HLSL_CLASS_ARRAY:
return sm1_class(type->e.array.type);
case HLSL_CLASS_MATRIX:
assert(type->modifiers & HLSL_MODIFIERS_MAJORITY_MASK);
if (type->modifiers & HLSL_MODIFIER_COLUMN_MAJOR)
return D3DXPC_MATRIX_COLUMNS;
else
return D3DXPC_MATRIX_ROWS;
case HLSL_CLASS_OBJECT:
return D3DXPC_OBJECT;
case HLSL_CLASS_SCALAR:
return D3DXPC_SCALAR;
case HLSL_CLASS_STRUCT:
return D3DXPC_STRUCT;
case HLSL_CLASS_VECTOR:
return D3DXPC_VECTOR;
default:
ERR("Invalid class %#x.\n", type->class);
vkd3d_unreachable();
}
}
static D3DXPARAMETER_TYPE sm1_base_type(const struct hlsl_type *type)
{
switch (type->base_type)
{
case HLSL_TYPE_BOOL:
return D3DXPT_BOOL;
case HLSL_TYPE_FLOAT:
case HLSL_TYPE_HALF:
return D3DXPT_FLOAT;
case HLSL_TYPE_INT:
case HLSL_TYPE_UINT:
return D3DXPT_INT;
case HLSL_TYPE_PIXELSHADER:
return D3DXPT_PIXELSHADER;
case HLSL_TYPE_SAMPLER:
switch (type->sampler_dim)
{
case HLSL_SAMPLER_DIM_1D:
return D3DXPT_SAMPLER1D;
case HLSL_SAMPLER_DIM_2D:
return D3DXPT_SAMPLER2D;
case HLSL_SAMPLER_DIM_3D:
return D3DXPT_SAMPLER3D;
case HLSL_SAMPLER_DIM_CUBE:
return D3DXPT_SAMPLERCUBE;
case HLSL_SAMPLER_DIM_GENERIC:
return D3DXPT_SAMPLER;
default:
ERR("Invalid dimension %#x.\n", type->sampler_dim);
vkd3d_unreachable();
}
break;
case HLSL_TYPE_STRING:
return D3DXPT_STRING;
case HLSL_TYPE_TEXTURE:
switch (type->sampler_dim)
{
case HLSL_SAMPLER_DIM_1D:
return D3DXPT_TEXTURE1D;
case HLSL_SAMPLER_DIM_2D:
return D3DXPT_TEXTURE2D;
case HLSL_SAMPLER_DIM_3D:
return D3DXPT_TEXTURE3D;
case HLSL_SAMPLER_DIM_CUBE:
return D3DXPT_TEXTURECUBE;
case HLSL_SAMPLER_DIM_GENERIC:
return D3DXPT_TEXTURE;
default:
ERR("Invalid dimension %#x.\n", type->sampler_dim);
vkd3d_unreachable();
}
break;
case HLSL_TYPE_VERTEXSHADER:
return D3DXPT_VERTEXSHADER;
case HLSL_TYPE_VOID:
return D3DXPT_VOID;
default:
vkd3d_unreachable();
}
}
static void write_sm1_type(struct vkd3d_bytecode_buffer *buffer, struct hlsl_type *type, unsigned int ctab_start)
{
const struct hlsl_type *array_type = hlsl_get_multiarray_element_type(type);
unsigned int array_size = hlsl_get_multiarray_size(type);
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
unsigned int field_count = 0;
size_t fields_offset = 0;
size_t i;
if (type->bytecode_offset)
return;
if (array_type->class == HLSL_CLASS_STRUCT)
{
field_count = array_type->e.record.field_count;
for (i = 0; i < field_count; ++i)
{
struct hlsl_struct_field *field = &array_type->e.record.fields[i];
field->name_bytecode_offset = put_string(buffer, field->name);
write_sm1_type(buffer, field->type, ctab_start);
}
fields_offset = bytecode_align(buffer) - ctab_start;
for (i = 0; i < field_count; ++i)
{
struct hlsl_struct_field *field = &array_type->e.record.fields[i];
put_u32(buffer, field->name_bytecode_offset - ctab_start);
put_u32(buffer, field->type->bytecode_offset - ctab_start);
}
}
type->bytecode_offset = put_u32(buffer, vkd3d_make_u32(sm1_class(type), sm1_base_type(array_type)));
put_u32(buffer, vkd3d_make_u32(type->dimy, type->dimx));
put_u32(buffer, vkd3d_make_u32(array_size, field_count));
put_u32(buffer, fields_offset);
}
static void sm1_sort_extern(struct list *sorted, struct hlsl_ir_var *to_sort)
{
struct hlsl_ir_var *var;
list_remove(&to_sort->extern_entry);
LIST_FOR_EACH_ENTRY(var, sorted, struct hlsl_ir_var, extern_entry)
{
if (strcmp(to_sort->name, var->name) < 0)
{
list_add_before(&var->extern_entry, &to_sort->extern_entry);
return;
}
}
list_add_tail(sorted, &to_sort->extern_entry);
}
static void sm1_sort_externs(struct hlsl_ctx *ctx)
{
struct list sorted = LIST_INIT(sorted);
struct hlsl_ir_var *var, *next;
LIST_FOR_EACH_ENTRY_SAFE(var, next, &ctx->extern_vars, struct hlsl_ir_var, extern_entry)
{
if (var->is_uniform)
sm1_sort_extern(&sorted, var);
}
list_move_tail(&ctx->extern_vars, &sorted);
}
static void write_sm1_uniforms(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
struct hlsl_ir_function_decl *entry_func)
{
size_t ctab_offset, ctab_start, ctab_end, vars_start, size_offset, creator_offset, offset;
unsigned int uniform_count = 0;
struct hlsl_ir_var *var;
LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry)
{
unsigned int r;
for (r = 0; r <= HLSL_REGSET_LAST; ++r)
if (var->semantic.name || !var->regs[r].allocated)
continue;
++uniform_count;
if (var->is_param && var->is_uniform)
{
char *new_name;
if (!(new_name = hlsl_sprintf_alloc(ctx, "$%s", var->name)))
return;
vkd3d_free((char *)var->name);
var->name = new_name;
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
}
}
}
sm1_sort_externs(ctx);
size_offset = put_u32(buffer, 0);
ctab_offset = put_u32(buffer, VKD3D_MAKE_TAG('C','T','A','B'));
ctab_start = put_u32(buffer, sizeof(D3DXSHADER_CONSTANTTABLE));
creator_offset = put_u32(buffer, 0);
put_u32(buffer, sm1_version(ctx->profile->type, ctx->profile->major_version, ctx->profile->minor_version));
put_u32(buffer, uniform_count);
put_u32(buffer, sizeof(D3DXSHADER_CONSTANTTABLE)); /* offset of constants */
put_u32(buffer, 0); /* FIXME: flags */
put_u32(buffer, 0); /* FIXME: target string */
vars_start = bytecode_align(buffer);
LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry)
{
unsigned int r;
for (r = 0; r <= HLSL_REGSET_LAST; ++r)
if (var->semantic.name || !var->regs[r].allocated)
continue;
put_u32(buffer, 0); /* name */
if (r == HLSL_REGSET_NUMERIC)
put_u32(buffer, vkd3d_make_u32(D3DXRS_FLOAT4, var->regs[r].id));
put_u32(buffer, var->data_type->reg_size[r] / 4);
put_u32(buffer, vkd3d_make_u32(D3DXRS_SAMPLER, var->regs[r].id));

Francisco Casas
committed
put_u32(buffer, var->bind_count[r]);
}
put_u32(buffer, 0); /* type */
put_u32(buffer, 0); /* FIXME: default value */
}
}
uniform_count = 0;
LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry)
{
unsigned int r;
for (r = 0; r <= HLSL_REGSET_LAST; ++r)
size_t var_offset, name_offset;
if (var->semantic.name || !var->regs[r].allocated)
continue;
var_offset = vars_start + (uniform_count * 5 * sizeof(uint32_t));
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
name_offset = put_string(buffer, var->name);
set_u32(buffer, var_offset, name_offset - ctab_start);
write_sm1_type(buffer, var->data_type, ctab_start);
set_u32(buffer, var_offset + 3 * sizeof(uint32_t), var->data_type->bytecode_offset - ctab_start);
++uniform_count;
}
}
offset = put_string(buffer, vkd3d_shader_get_version(NULL, NULL));
set_u32(buffer, creator_offset, offset - ctab_start);
ctab_end = bytecode_align(buffer);
set_u32(buffer, size_offset, vkd3d_make_u32(D3DSIO_COMMENT, (ctab_end - ctab_offset) / sizeof(uint32_t)));
}
static uint32_t sm1_encode_register_type(D3DSHADER_PARAM_REGISTER_TYPE type)
{
return ((type << D3DSP_REGTYPE_SHIFT) & D3DSP_REGTYPE_MASK)
| ((type << D3DSP_REGTYPE_SHIFT2) & D3DSP_REGTYPE_MASK2);
}
struct sm1_instruction
{
D3DSHADER_INSTRUCTION_OPCODE_TYPE opcode;
struct sm1_dst_register
{
D3DSHADER_PARAM_REGISTER_TYPE type;
D3DSHADER_PARAM_DSTMOD_TYPE mod;
unsigned int writemask;
uint32_t reg;
} dst;
struct sm1_src_register
{
D3DSHADER_PARAM_REGISTER_TYPE type;
D3DSHADER_PARAM_SRCMOD_TYPE mod;
unsigned int swizzle;
uint32_t reg;
} srcs[3];
unsigned int src_count;
unsigned int has_dst;
};
static void write_sm1_dst_register(struct vkd3d_bytecode_buffer *buffer, const struct sm1_dst_register *reg)
{
assert(reg->writemask);
put_u32(buffer, (1u << 31) | sm1_encode_register_type(reg->type) | reg->mod | (reg->writemask << 16) | reg->reg);
}
static void write_sm1_src_register(struct vkd3d_bytecode_buffer *buffer,
const struct sm1_src_register *reg)
{
put_u32(buffer, (1u << 31) | sm1_encode_register_type(reg->type) | reg->mod | (reg->swizzle << 16) | reg->reg);
}
static void write_sm1_instruction(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
const struct sm1_instruction *instr)
{
uint32_t token = instr->opcode;
unsigned int i;
if (ctx->profile->major_version > 1)
token |= (instr->has_dst + instr->src_count) << D3DSI_INSTLENGTH_SHIFT;
put_u32(buffer, token);
if (instr->has_dst)
write_sm1_dst_register(buffer, &instr->dst);
for (i = 0; i < instr->src_count; ++i)
write_sm1_src_register(buffer, &instr->srcs[i]);
};
static void sm1_map_src_swizzle(struct sm1_src_register *src, unsigned int map_writemask)
{
src->swizzle = hlsl_map_swizzle(src->swizzle, map_writemask);
}
static void write_sm1_dp2add(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
const struct hlsl_reg *dst, const struct hlsl_reg *src1, const struct hlsl_reg *src2,
const struct hlsl_reg *src3)
{
struct sm1_instruction instr =
{
.opcode = D3DSIO_DP2ADD,
.dst.type = D3DSPR_TEMP,
.dst.writemask = dst->writemask,
.dst.reg = dst->id,
.has_dst = 1,
.srcs[0].type = D3DSPR_TEMP,
.srcs[0].swizzle = hlsl_swizzle_from_writemask(src1->writemask),
.srcs[0].reg = src1->id,
.srcs[1].type = D3DSPR_TEMP,
.srcs[1].swizzle = hlsl_swizzle_from_writemask(src2->writemask),
.srcs[1].reg = src2->id,
.srcs[2].type = D3DSPR_TEMP,
.srcs[2].swizzle = hlsl_swizzle_from_writemask(src3->writemask),
.srcs[2].reg = src3->id,
.src_count = 3,
};
write_sm1_instruction(ctx, buffer, &instr);
}
static void write_sm1_binary_op(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
D3DSHADER_INSTRUCTION_OPCODE_TYPE opcode, const struct hlsl_reg *dst,
const struct hlsl_reg *src1, const struct hlsl_reg *src2)
{
struct sm1_instruction instr =
{
.opcode = opcode,
.dst.type = D3DSPR_TEMP,
.dst.writemask = dst->writemask,
.dst.reg = dst->id,
.has_dst = 1,
.srcs[0].type = D3DSPR_TEMP,
.srcs[0].swizzle = hlsl_swizzle_from_writemask(src1->writemask),
.srcs[0].reg = src1->id,
.srcs[1].type = D3DSPR_TEMP,
.srcs[1].swizzle = hlsl_swizzle_from_writemask(src2->writemask),
.srcs[1].reg = src2->id,
.src_count = 2,
};
sm1_map_src_swizzle(&instr.srcs[0], instr.dst.writemask);
sm1_map_src_swizzle(&instr.srcs[1], instr.dst.writemask);
write_sm1_instruction(ctx, buffer, &instr);
}
static void write_sm1_binary_op_dot(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
D3DSHADER_INSTRUCTION_OPCODE_TYPE opcode, const struct hlsl_reg *dst,
const struct hlsl_reg *src1, const struct hlsl_reg *src2)
{
struct sm1_instruction instr =
{
.opcode = opcode,
.dst.type = D3DSPR_TEMP,
.dst.writemask = dst->writemask,
.dst.reg = dst->id,
.has_dst = 1,
.srcs[0].type = D3DSPR_TEMP,
.srcs[0].swizzle = hlsl_swizzle_from_writemask(src1->writemask),
.srcs[0].reg = src1->id,
.srcs[1].type = D3DSPR_TEMP,
.srcs[1].swizzle = hlsl_swizzle_from_writemask(src2->writemask),
.srcs[1].reg = src2->id,
.src_count = 2,
};
write_sm1_instruction(ctx, buffer, &instr);
}
static void write_sm1_unary_op(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
D3DSHADER_INSTRUCTION_OPCODE_TYPE opcode, const struct hlsl_reg *dst,
const struct hlsl_reg *src, D3DSHADER_PARAM_SRCMOD_TYPE src_mod, D3DSHADER_PARAM_DSTMOD_TYPE dst_mod)
{
struct sm1_instruction instr =
{
.opcode = opcode,
.dst.type = D3DSPR_TEMP,
.dst.mod = dst_mod,
.dst.writemask = dst->writemask,
.dst.reg = dst->id,
.has_dst = 1,
.srcs[0].type = D3DSPR_TEMP,
.srcs[0].swizzle = hlsl_swizzle_from_writemask(src->writemask),
.srcs[0].reg = src->id,
.srcs[0].mod = src_mod,
.src_count = 1,
};
sm1_map_src_swizzle(&instr.srcs[0], instr.dst.writemask);
write_sm1_instruction(ctx, buffer, &instr);
}
static void write_sm1_constant_defs(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer)
{
unsigned int i, x;
for (i = 0; i < ctx->constant_defs.count; ++i)
{
const struct hlsl_constant_register *constant_reg = &ctx->constant_defs.regs[i];
uint32_t token = D3DSIO_DEF;
const struct sm1_dst_register reg =
{
.type = D3DSPR_CONST,
.writemask = VKD3DSP_WRITEMASK_ALL,
.reg = constant_reg->index,
};
if (ctx->profile->major_version > 1)
token |= 5 << D3DSI_INSTLENGTH_SHIFT;
put_u32(buffer, token);
write_sm1_dst_register(buffer, ®);
for (x = 0; x < 4; ++x)
put_f32(buffer, constant_reg->value.f[x]);
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
}
}
static void write_sm1_semantic_dcl(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
const struct hlsl_ir_var *var, bool output)
{
struct sm1_dst_register reg = {0};
uint32_t token, usage_idx;
D3DDECLUSAGE usage;
bool ret;
if (hlsl_sm1_register_from_semantic(ctx, &var->semantic, output, ®.type, ®.reg))
{
usage = 0;
usage_idx = 0;
}
else
{
ret = hlsl_sm1_usage_from_semantic(&var->semantic, &usage, &usage_idx);
assert(ret);
reg.type = output ? D3DSPR_OUTPUT : D3DSPR_INPUT;
reg.reg = var->regs[HLSL_REGSET_NUMERIC].id;
}
token = D3DSIO_DCL;
if (ctx->profile->major_version > 1)
token |= 2 << D3DSI_INSTLENGTH_SHIFT;
put_u32(buffer, token);
token = (1u << 31);
token |= usage << D3DSP_DCL_USAGE_SHIFT;
token |= usage_idx << D3DSP_DCL_USAGEINDEX_SHIFT;
put_u32(buffer, token);
reg.writemask = (1 << var->data_type->dimx) - 1;
write_sm1_dst_register(buffer, ®);
}
static void write_sm1_semantic_dcls(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer)
{
bool write_in = false, write_out = false;
struct hlsl_ir_var *var;
if (ctx->profile->type == VKD3D_SHADER_TYPE_PIXEL && ctx->profile->major_version >= 2)
write_in = true;
else if (ctx->profile->type == VKD3D_SHADER_TYPE_VERTEX && ctx->profile->major_version == 3)
write_in = write_out = true;
else if (ctx->profile->type == VKD3D_SHADER_TYPE_VERTEX && ctx->profile->major_version < 3)
write_in = true;
LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry)
{
if (write_in && var->is_input_semantic)
write_sm1_semantic_dcl(ctx, buffer, var, false);
if (write_out && var->is_output_semantic)
write_sm1_semantic_dcl(ctx, buffer, var, true);
}
}
static void write_sm1_sampler_dcl(struct hlsl_ctx *ctx, struct vkd3d_bytecode_buffer *buffer,
unsigned int reg_id, enum hlsl_sampler_dim sampler_dim)
{
struct sm1_dst_register reg = {0};
uint32_t token, res_type = 0;
token = D3DSIO_DCL;
if (ctx->profile->major_version > 1)
token |= 2 << D3DSI_INSTLENGTH_SHIFT;
put_u32(buffer, token);