[PATCH v2 0/3] MR382: vkd3d-shader/tpf: Replace sm4 register structs with vkd3d-shader register structs, part 4.
-- v2: vkd3d-shader/tpf: Replace sm4_src_register with vkd3d_shader_src_param. vkd3d-shader/tpf: Remove sm4_src_register.swizzle_type. vkd3d-shader/tpf: Make sm4_src_register.mod a vkd3d_shader_src_modifier enum.
From: Francisco Casas fcasas@codeweavers.com
--- libs/vkd3d-shader/tpf.c | 57 +++++++++++++++++++++++++++++++---------- 1 file changed, 44 insertions(+), 13 deletions(-)
diff --git a/libs/vkd3d-shader/tpf.c b/libs/vkd3d-shader/tpf.c index 63771736..0b68f0f1 100644 --- a/libs/vkd3d-shader/tpf.c +++ b/libs/vkd3d-shader/tpf.c @@ -3628,7 +3628,7 @@ struct sm4_instruction struct vkd3d_shader_register reg; enum vkd3d_sm4_swizzle_type swizzle_type; DWORD swizzle; - unsigned int mod; + enum vkd3d_shader_src_modifier modifiers; } srcs[5]; unsigned int src_count;
@@ -3897,7 +3897,7 @@ static void sm4_write_src_register(const struct tpf_writer *tpf, const struct sm const struct vkd3d_sm4_register_type_info *register_type_info; struct vkd3d_bytecode_buffer *buffer = tpf->buffer; uint32_t sm4_reg_type, reg_dim; - uint32_t token = 0; + uint32_t token = 0, mod_token = 0; unsigned int j;
register_type_info = get_info_from_vkd3d_register_type(&tpf->lookup, src->reg.type); @@ -3921,13 +3921,44 @@ static void sm4_write_src_register(const struct tpf_writer *tpf, const struct sm token |= (uint32_t)src->swizzle_type << VKD3D_SM4_SWIZZLE_TYPE_SHIFT; token |= swizzle_to_sm4(src->swizzle) << VKD3D_SM4_SWIZZLE_SHIFT; } - if (src->mod) - token |= VKD3D_SM4_EXTENDED_OPERAND; - put_u32(buffer, token);
- if (src->mod) - put_u32(buffer, (src->mod << VKD3D_SM4_REGISTER_MODIFIER_SHIFT) - | VKD3D_SM4_EXTENDED_OPERAND_MODIFIER); + switch (src->modifiers) + { + case VKD3DSPSM_NONE: + mod_token = VKD3D_SM4_REGISTER_MODIFIER_NONE; + break; + + case VKD3DSPSM_ABS: + mod_token = (VKD3D_SM4_REGISTER_MODIFIER_ABS << VKD3D_SM4_REGISTER_MODIFIER_SHIFT) + | VKD3D_SM4_EXTENDED_OPERAND_MODIFIER; + break; + + case VKD3DSPSM_NEG: + mod_token = (VKD3D_SM4_REGISTER_MODIFIER_NEGATE << VKD3D_SM4_REGISTER_MODIFIER_SHIFT) + | VKD3D_SM4_EXTENDED_OPERAND_MODIFIER; + break; + + case VKD3DSPSM_ABSNEG: + mod_token = (VKD3D_SM4_REGISTER_MODIFIER_ABS_NEGATE << VKD3D_SM4_REGISTER_MODIFIER_SHIFT) + | VKD3D_SM4_EXTENDED_OPERAND_MODIFIER; + break; + + default: + ERR("Unhandled register modifier %#x.\n", src->modifiers); + vkd3d_unreachable(); + break; + } + + if (src->modifiers) + { + token |= VKD3D_SM4_EXTENDED_OPERAND; + put_u32(buffer, token); + put_u32(buffer, mod_token); + } + else + { + put_u32(buffer, token); + }
for (j = 0; j < src->reg.idx_count; ++j) { @@ -3962,7 +3993,7 @@ static uint32_t sm4_src_register_order(const struct sm4_src_register *src) if (src->reg.type == VKD3DSPR_IMMCONST) order += src->reg.dimension == VSIR_DIMENSION_VEC4 ? 4 : 1; order += src->reg.idx_count; - if (src->mod) + if (src->modifiers) ++order; return order; } @@ -4280,7 +4311,7 @@ static void write_sm4_ret(const struct tpf_writer *tpf) }
static void write_sm4_unary_op(const struct tpf_writer *tpf, enum vkd3d_sm4_opcode opcode, - const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src, unsigned int src_mod) + const struct hlsl_ir_node *dst, const struct hlsl_ir_node *src, enum vkd3d_shader_src_modifier src_mod) { struct sm4_instruction instr;
@@ -4291,7 +4322,7 @@ static void write_sm4_unary_op(const struct tpf_writer *tpf, enum vkd3d_sm4_opco instr.dst_count = 1;
sm4_src_from_node(&instr.srcs[0], src, instr.dsts[0].write_mask); - instr.srcs[0].mod = src_mod; + instr.srcs[0].modifiers = src_mod; instr.src_count = 1;
write_sm4_instruction(tpf, &instr); @@ -4776,7 +4807,7 @@ static void write_sm4_expr(const struct tpf_writer *tpf, const struct hlsl_ir_ex switch (dst_type->base_type) { case HLSL_TYPE_FLOAT: - write_sm4_unary_op(tpf, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3D_SM4_REGISTER_MODIFIER_ABS); + write_sm4_unary_op(tpf, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3DSPSM_ABS); break;
default: @@ -4857,7 +4888,7 @@ static void write_sm4_expr(const struct tpf_writer *tpf, const struct hlsl_ir_ex switch (dst_type->base_type) { case HLSL_TYPE_FLOAT: - write_sm4_unary_op(tpf, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3D_SM4_REGISTER_MODIFIER_NEGATE); + write_sm4_unary_op(tpf, VKD3D_SM4_OP_MOV, &expr->node, arg1, VKD3DSPSM_NEG); break;
case HLSL_TYPE_INT:
From: Francisco Casas fcasas@codeweavers.com
--- libs/vkd3d-shader/hlsl.h | 2 +- libs/vkd3d-shader/hlsl_codegen.c | 2 +- libs/vkd3d-shader/tpf.c | 235 +++++++++++++++---------------- 3 files changed, 115 insertions(+), 124 deletions(-)
diff --git a/libs/vkd3d-shader/hlsl.h b/libs/vkd3d-shader/hlsl.h index 44cebaaf..f81083c0 100644 --- a/libs/vkd3d-shader/hlsl.h +++ b/libs/vkd3d-shader/hlsl.h @@ -1273,7 +1273,7 @@ int hlsl_sm1_write(struct hlsl_ctx *ctx, struct hlsl_ir_function_decl *entry_fun bool hlsl_sm4_usage_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_semantic *semantic, bool output, D3D_NAME *usage); bool hlsl_sm4_register_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_semantic *semantic, - bool output, enum vkd3d_shader_register_type *type, enum vkd3d_sm4_swizzle_type *swizzle_type, bool *has_idx); + bool output, enum vkd3d_shader_register_type *type, bool *has_idx); int hlsl_sm4_write(struct hlsl_ctx *ctx, struct hlsl_ir_function_decl *entry_func, struct vkd3d_shader_code *out);
struct hlsl_ir_function_decl *hlsl_compile_internal_function(struct hlsl_ctx *ctx, const char *name, const char *hlsl); diff --git a/libs/vkd3d-shader/hlsl_codegen.c b/libs/vkd3d-shader/hlsl_codegen.c index 76572cf9..b113696e 100644 --- a/libs/vkd3d-shader/hlsl_codegen.c +++ b/libs/vkd3d-shader/hlsl_codegen.c @@ -3703,7 +3703,7 @@ static void allocate_semantic_register(struct hlsl_ctx *ctx, struct hlsl_ir_var "Invalid semantic '%s'.", var->semantic.name); return; } - if ((builtin = hlsl_sm4_register_from_semantic(ctx, &var->semantic, output, &type, NULL, &has_idx))) + if ((builtin = hlsl_sm4_register_from_semantic(ctx, &var->semantic, output, &type, &has_idx))) reg = has_idx ? var->semantic.index : 0; }
diff --git a/libs/vkd3d-shader/tpf.c b/libs/vkd3d-shader/tpf.c index 0b68f0f1..7581e1c2 100644 --- a/libs/vkd3d-shader/tpf.c +++ b/libs/vkd3d-shader/tpf.c @@ -510,6 +510,8 @@ enum vkd3d_sm4_swizzle_type VKD3D_SM4_SWIZZLE_NONE = 0x0, /* swizzle bitfield contains a mask */ VKD3D_SM4_SWIZZLE_VEC4 = 0x1, VKD3D_SM4_SWIZZLE_SCALAR = 0x2, + + VKD3D_SM4_SWIZZLE_INVALID = ~0u, };
enum vkd3d_sm4_dimension @@ -1526,6 +1528,9 @@ struct vkd3d_sm4_register_type_info { enum vkd3d_sm4_register_type sm4_type; enum vkd3d_shader_register_type vkd3d_type; + + /* Swizzle type to be used for src registers when their dimension is VKD3D_SM4_DIMENSION_VEC4. */ + enum vkd3d_sm4_swizzle_type default_src_swizzle_type; };
static const enum vkd3d_shader_register_precision register_precision_table[] = @@ -1565,42 +1570,42 @@ static void init_sm4_lookup_tables(struct vkd3d_sm4_lookup_tables *lookup)
static const struct vkd3d_sm4_register_type_info register_type_table[] = { - {VKD3D_SM4_RT_TEMP, VKD3DSPR_TEMP}, - {VKD3D_SM4_RT_INPUT, VKD3DSPR_INPUT}, - {VKD3D_SM4_RT_OUTPUT, VKD3DSPR_OUTPUT}, - {VKD3D_SM4_RT_INDEXABLE_TEMP, VKD3DSPR_IDXTEMP}, - {VKD3D_SM4_RT_IMMCONST, VKD3DSPR_IMMCONST}, - {VKD3D_SM4_RT_IMMCONST64, VKD3DSPR_IMMCONST64}, - {VKD3D_SM4_RT_SAMPLER, VKD3DSPR_SAMPLER}, - {VKD3D_SM4_RT_RESOURCE, VKD3DSPR_RESOURCE}, - {VKD3D_SM4_RT_CONSTBUFFER, VKD3DSPR_CONSTBUFFER}, - {VKD3D_SM4_RT_IMMCONSTBUFFER, VKD3DSPR_IMMCONSTBUFFER}, - {VKD3D_SM4_RT_PRIMID, VKD3DSPR_PRIMID}, - {VKD3D_SM4_RT_DEPTHOUT, VKD3DSPR_DEPTHOUT}, - {VKD3D_SM4_RT_NULL, VKD3DSPR_NULL}, - {VKD3D_SM4_RT_RASTERIZER, VKD3DSPR_RASTERIZER}, - {VKD3D_SM4_RT_OMASK, VKD3DSPR_SAMPLEMASK}, - {VKD3D_SM5_RT_STREAM, VKD3DSPR_STREAM}, - {VKD3D_SM5_RT_FUNCTION_BODY, VKD3DSPR_FUNCTIONBODY}, - {VKD3D_SM5_RT_FUNCTION_POINTER, VKD3DSPR_FUNCTIONPOINTER}, - {VKD3D_SM5_RT_OUTPUT_CONTROL_POINT_ID, VKD3DSPR_OUTPOINTID}, - {VKD3D_SM5_RT_FORK_INSTANCE_ID, VKD3DSPR_FORKINSTID}, - {VKD3D_SM5_RT_JOIN_INSTANCE_ID, VKD3DSPR_JOININSTID}, - {VKD3D_SM5_RT_INPUT_CONTROL_POINT, VKD3DSPR_INCONTROLPOINT}, - {VKD3D_SM5_RT_OUTPUT_CONTROL_POINT, VKD3DSPR_OUTCONTROLPOINT}, - {VKD3D_SM5_RT_PATCH_CONSTANT_DATA, VKD3DSPR_PATCHCONST}, - {VKD3D_SM5_RT_DOMAIN_LOCATION, VKD3DSPR_TESSCOORD}, - {VKD3D_SM5_RT_UAV, VKD3DSPR_UAV}, - {VKD3D_SM5_RT_SHARED_MEMORY, VKD3DSPR_GROUPSHAREDMEM}, - {VKD3D_SM5_RT_THREAD_ID, VKD3DSPR_THREADID}, - {VKD3D_SM5_RT_THREAD_GROUP_ID, VKD3DSPR_THREADGROUPID}, - {VKD3D_SM5_RT_LOCAL_THREAD_ID, VKD3DSPR_LOCALTHREADID}, - {VKD3D_SM5_RT_COVERAGE, VKD3DSPR_COVERAGE}, - {VKD3D_SM5_RT_LOCAL_THREAD_INDEX, VKD3DSPR_LOCALTHREADINDEX}, - {VKD3D_SM5_RT_GS_INSTANCE_ID, VKD3DSPR_GSINSTID}, - {VKD3D_SM5_RT_DEPTHOUT_GREATER_EQUAL, VKD3DSPR_DEPTHOUTGE}, - {VKD3D_SM5_RT_DEPTHOUT_LESS_EQUAL, VKD3DSPR_DEPTHOUTLE}, - {VKD3D_SM5_RT_OUTPUT_STENCIL_REF, VKD3DSPR_OUTSTENCILREF}, + {VKD3D_SM4_RT_TEMP, VKD3DSPR_TEMP, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_INPUT, VKD3DSPR_INPUT, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_OUTPUT, VKD3DSPR_OUTPUT, VKD3D_SM4_SWIZZLE_INVALID}, + {VKD3D_SM4_RT_INDEXABLE_TEMP, VKD3DSPR_IDXTEMP, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_IMMCONST, VKD3DSPR_IMMCONST, VKD3D_SM4_SWIZZLE_NONE}, + {VKD3D_SM4_RT_IMMCONST64, VKD3DSPR_IMMCONST64, VKD3D_SM4_SWIZZLE_NONE}, + {VKD3D_SM4_RT_SAMPLER, VKD3DSPR_SAMPLER, VKD3D_SM4_SWIZZLE_SCALAR}, + {VKD3D_SM4_RT_RESOURCE, VKD3DSPR_RESOURCE, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_CONSTBUFFER, VKD3DSPR_CONSTBUFFER, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_IMMCONSTBUFFER, VKD3DSPR_IMMCONSTBUFFER, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_PRIMID, VKD3DSPR_PRIMID, VKD3D_SM4_SWIZZLE_NONE}, + {VKD3D_SM4_RT_DEPTHOUT, VKD3DSPR_DEPTHOUT, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_NULL, VKD3DSPR_NULL, VKD3D_SM4_SWIZZLE_INVALID}, + {VKD3D_SM4_RT_RASTERIZER, VKD3DSPR_RASTERIZER, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM4_RT_OMASK, VKD3DSPR_SAMPLEMASK, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_STREAM, VKD3DSPR_STREAM, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_FUNCTION_BODY, VKD3DSPR_FUNCTIONBODY, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_FUNCTION_POINTER, VKD3DSPR_FUNCTIONPOINTER, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_OUTPUT_CONTROL_POINT_ID, VKD3DSPR_OUTPOINTID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_FORK_INSTANCE_ID, VKD3DSPR_FORKINSTID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_JOIN_INSTANCE_ID, VKD3DSPR_JOININSTID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_INPUT_CONTROL_POINT, VKD3DSPR_INCONTROLPOINT, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_OUTPUT_CONTROL_POINT, VKD3DSPR_OUTCONTROLPOINT, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_PATCH_CONSTANT_DATA, VKD3DSPR_PATCHCONST, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_DOMAIN_LOCATION, VKD3DSPR_TESSCOORD, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_UAV, VKD3DSPR_UAV, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_SHARED_MEMORY, VKD3DSPR_GROUPSHAREDMEM, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_THREAD_ID, VKD3DSPR_THREADID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_THREAD_GROUP_ID, VKD3DSPR_THREADGROUPID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_LOCAL_THREAD_ID, VKD3DSPR_LOCALTHREADID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_COVERAGE, VKD3DSPR_COVERAGE, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_LOCAL_THREAD_INDEX, VKD3DSPR_LOCALTHREADINDEX,VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_GS_INSTANCE_ID, VKD3DSPR_GSINSTID, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_DEPTHOUT_GREATER_EQUAL, VKD3DSPR_DEPTHOUTGE, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_DEPTHOUT_LESS_EQUAL, VKD3DSPR_DEPTHOUTLE, VKD3D_SM4_SWIZZLE_VEC4}, + {VKD3D_SM5_RT_OUTPUT_STENCIL_REF, VKD3DSPR_OUTSTENCILREF, VKD3D_SM4_SWIZZLE_VEC4}, };
memset(lookup, 0, sizeof(*lookup)); @@ -1636,6 +1641,16 @@ static const struct vkd3d_sm4_register_type_info *get_info_from_vkd3d_register_t return lookup->register_type_info_from_vkd3d[vkd3d_type]; }
+static enum vkd3d_sm4_swizzle_type vkd3d_sm4_get_default_swizzle_type( + const struct vkd3d_sm4_lookup_tables *lookup, enum vkd3d_shader_register_type vkd3d_type) +{ + const struct vkd3d_sm4_register_type_info *register_type_info = + get_info_from_vkd3d_register_type(lookup, vkd3d_type); + + assert(register_type_info); + return register_type_info->default_src_swizzle_type; +} + static void map_register(const struct vkd3d_shader_sm4_parser *sm4, struct vkd3d_shader_register *reg) { switch (sm4->p.shader_version.type) @@ -2700,7 +2715,7 @@ static bool type_is_integer(const struct hlsl_type *type) }
bool hlsl_sm4_register_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_semantic *semantic, - bool output, enum vkd3d_shader_register_type *type, enum vkd3d_sm4_swizzle_type *swizzle_type, bool *has_idx) + bool output, enum vkd3d_shader_register_type *type, bool *has_idx) { unsigned int i;
@@ -2709,25 +2724,24 @@ bool hlsl_sm4_register_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_sem const char *semantic; bool output; enum vkd3d_shader_type shader_type; - enum vkd3d_sm4_swizzle_type swizzle_type; enum vkd3d_shader_register_type type; bool has_idx; } register_table[] = { - {"sv_dispatchthreadid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_THREADID, false}, - {"sv_groupid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_THREADGROUPID, false}, - {"sv_groupthreadid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_LOCALTHREADID, false}, + {"sv_dispatchthreadid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3DSPR_THREADID, false}, + {"sv_groupid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3DSPR_THREADGROUPID, false}, + {"sv_groupthreadid", false, VKD3D_SHADER_TYPE_COMPUTE, VKD3DSPR_LOCALTHREADID, false},
- {"sv_primitiveid", false, VKD3D_SHADER_TYPE_GEOMETRY, VKD3D_SM4_SWIZZLE_NONE, VKD3DSPR_PRIMID, false}, + {"sv_primitiveid", false, VKD3D_SHADER_TYPE_GEOMETRY, VKD3DSPR_PRIMID, false},
/* Put sv_target in this table, instead of letting it fall through to * default varying allocation, so that the register index matches the * usage index. */ - {"color", true, VKD3D_SHADER_TYPE_PIXEL, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_OUTPUT, true}, - {"depth", true, VKD3D_SHADER_TYPE_PIXEL, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_DEPTHOUT, false}, - {"sv_depth", true, VKD3D_SHADER_TYPE_PIXEL, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_DEPTHOUT, false}, - {"sv_target", true, VKD3D_SHADER_TYPE_PIXEL, VKD3D_SM4_SWIZZLE_VEC4, VKD3DSPR_OUTPUT, true}, + {"color", true, VKD3D_SHADER_TYPE_PIXEL, VKD3DSPR_OUTPUT, true}, + {"depth", true, VKD3D_SHADER_TYPE_PIXEL, VKD3DSPR_DEPTHOUT, false}, + {"sv_depth", true, VKD3D_SHADER_TYPE_PIXEL, VKD3DSPR_DEPTHOUT, false}, + {"sv_target", true, VKD3D_SHADER_TYPE_PIXEL, VKD3DSPR_OUTPUT, true}, };
for (i = 0; i < ARRAY_SIZE(register_table); ++i) @@ -2738,8 +2752,6 @@ bool hlsl_sm4_register_from_semantic(struct hlsl_ctx *ctx, const struct hlsl_sem { if (type) *type = register_table[i].type; - if (swizzle_type) - *swizzle_type = register_table[i].swizzle_type; *has_idx = register_table[i].has_idx; return true; } @@ -2851,7 +2863,7 @@ static void write_sm4_signature(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc, continue; usage_idx = var->semantic.index;
- if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, output, NULL, NULL, &has_idx)) + if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, output, NULL, &has_idx)) { reg_idx = has_idx ? var->semantic.index : ~0u; } @@ -3639,8 +3651,7 @@ struct sm4_instruction };
static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_register *reg, - unsigned int *writemask, enum vkd3d_sm4_swizzle_type *swizzle_type, - const struct hlsl_deref *deref) + unsigned int *writemask, const struct hlsl_deref *deref) { const struct hlsl_type *data_type = hlsl_deref_get_type(ctx, deref); const struct hlsl_ir_var *var = deref->var; @@ -3653,8 +3664,6 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re { reg->type = VKD3DSPR_RESOURCE; reg->dimension = VSIR_DIMENSION_VEC4; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = var->regs[HLSL_REGSET_TEXTURES].id; reg->idx[0].offset += hlsl_offset_from_deref_safe(ctx, deref); assert(regset == HLSL_REGSET_TEXTURES); @@ -3665,8 +3674,6 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re { reg->type = VKD3DSPR_UAV; reg->dimension = VSIR_DIMENSION_VEC4; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = var->regs[HLSL_REGSET_UAVS].id; reg->idx[0].offset += hlsl_offset_from_deref_safe(ctx, deref); assert(regset == HLSL_REGSET_UAVS); @@ -3677,8 +3684,6 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re { reg->type = VKD3DSPR_SAMPLER; reg->dimension = VSIR_DIMENSION_NONE; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_NONE; reg->idx[0].offset = var->regs[HLSL_REGSET_SAMPLERS].id; reg->idx[0].offset += hlsl_offset_from_deref_safe(ctx, deref); assert(regset == HLSL_REGSET_SAMPLERS); @@ -3692,8 +3697,6 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re assert(data_type->class <= HLSL_CLASS_VECTOR); reg->type = VKD3DSPR_CONSTBUFFER; reg->dimension = VSIR_DIMENSION_VEC4; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = var->buffer->reg.id; reg->idx[1].offset = offset / 4; reg->idx_count = 2; @@ -3704,7 +3707,7 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re { bool has_idx;
- if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, false, ®->type, swizzle_type, &has_idx)) + if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, false, ®->type, &has_idx)) { unsigned int offset = hlsl_offset_from_deref_safe(ctx, deref);
@@ -3724,8 +3727,6 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re assert(hlsl_reg.allocated); reg->type = VKD3DSPR_INPUT; reg->dimension = VSIR_DIMENSION_VEC4; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = hlsl_reg.id; reg->idx_count = 1; *writemask = hlsl_reg.writemask; @@ -3735,7 +3736,7 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re { bool has_idx;
- if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, true, ®->type, swizzle_type, &has_idx)) + if (hlsl_sm4_register_from_semantic(ctx, &var->semantic, true, ®->type, &has_idx)) { unsigned int offset = hlsl_offset_from_deref_safe(ctx, deref);
@@ -3770,21 +3771,19 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re assert(hlsl_reg.allocated); reg->type = VKD3DSPR_TEMP; reg->dimension = VSIR_DIMENSION_VEC4; - if (swizzle_type) - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = hlsl_reg.id; reg->idx_count = 1; *writemask = hlsl_reg.writemask; } }
-static void sm4_src_from_deref(struct hlsl_ctx *ctx, struct sm4_src_register *src, +static void sm4_src_from_deref(const struct tpf_writer *tpf, struct sm4_src_register *src, const struct hlsl_deref *deref, unsigned int map_writemask) { unsigned int writemask, hlsl_swizzle;
- sm4_register_from_deref(ctx, &src->reg, &writemask, &src->swizzle_type, deref); - if (src->swizzle_type == VKD3D_SM4_SWIZZLE_VEC4) + sm4_register_from_deref(tpf->ctx, &src->reg, &writemask, deref); + if (vkd3d_sm4_get_default_swizzle_type(&tpf->lookup, src->reg.type) == VKD3D_SM4_SWIZZLE_VEC4) { hlsl_swizzle = hlsl_map_swizzle(hlsl_swizzle_from_writemask(writemask), map_writemask); src->swizzle = swizzle_from_sm4(hlsl_swizzle); @@ -3792,12 +3791,11 @@ static void sm4_src_from_deref(struct hlsl_ctx *ctx, struct sm4_src_register *sr }
static void sm4_register_from_node(struct vkd3d_shader_register *reg, unsigned int *writemask, - enum vkd3d_sm4_swizzle_type *swizzle_type, const struct hlsl_ir_node *instr) + const struct hlsl_ir_node *instr) { assert(instr->reg.allocated); reg->type = VKD3DSPR_TEMP; reg->dimension = VSIR_DIMENSION_VEC4; - *swizzle_type = VKD3D_SM4_SWIZZLE_VEC4; reg->idx[0].offset = instr->reg.id; reg->idx_count = 1; *writemask = instr->reg.writemask; @@ -3805,16 +3803,13 @@ static void sm4_register_from_node(struct vkd3d_shader_register *reg, unsigned i
static void sm4_dst_from_node(struct vkd3d_shader_dst_param *dst, const struct hlsl_ir_node *instr) { - unsigned int swizzle_type; - - sm4_register_from_node(&dst->reg, &dst->write_mask, &swizzle_type, instr); + sm4_register_from_node(&dst->reg, &dst->write_mask, instr); }
static void sm4_src_from_constant_value(struct sm4_src_register *src, const struct hlsl_constant_value *value, unsigned int width, unsigned int map_writemask) { src->swizzle = VKD3D_SHADER_NO_SWIZZLE; - src->swizzle_type = VKD3D_SM4_SWIZZLE_NONE; src->reg.type = VKD3DSPR_IMMCONST; if (width == 1) { @@ -3836,7 +3831,7 @@ static void sm4_src_from_constant_value(struct sm4_src_register *src, } }
-static void sm4_src_from_node(struct sm4_src_register *src, +static void sm4_src_from_node(const struct tpf_writer *tpf, struct sm4_src_register *src, const struct hlsl_ir_node *instr, unsigned int map_writemask) { unsigned int writemask, hlsl_swizzle; @@ -3849,8 +3844,8 @@ static void sm4_src_from_node(struct sm4_src_register *src, return; }
- sm4_register_from_node(&src->reg, &writemask, &src->swizzle_type, instr); - if (src->swizzle_type == VKD3D_SM4_SWIZZLE_VEC4) + sm4_register_from_node(&src->reg, &writemask, instr); + if (vkd3d_sm4_get_default_swizzle_type(&tpf->lookup, src->reg.type) == VKD3D_SM4_SWIZZLE_VEC4) { hlsl_swizzle = hlsl_map_swizzle(hlsl_swizzle_from_writemask(writemask), map_writemask); src->swizzle = swizzle_from_sm4(hlsl_swizzle); @@ -3918,7 +3913,7 @@ static void sm4_write_src_register(const struct tpf_writer *tpf, const struct sm token |= reg_dim << VKD3D_SM4_DIMENSION_SHIFT; if (reg_dim == VKD3D_SM4_DIMENSION_VEC4) { - token |= (uint32_t)src->swizzle_type << VKD3D_SM4_SWIZZLE_TYPE_SHIFT; + token |= (uint32_t)register_type_info->default_src_swizzle_type << VKD3D_SM4_SWIZZLE_TYPE_SHIFT; token |= swizzle_to_sm4(src->swizzle) << VKD3D_SM4_SWIZZLE_SHIFT; }
@@ -4078,7 +4073,6 @@ static void write_sm4_dcl_constant_buffer(const struct tpf_writer *tpf, const st .srcs[0].reg.idx[0].offset = cbuffer->reg.id, .srcs[0].reg.idx[1].offset = (cbuffer->used_size + 3) / 4, .srcs[0].reg.idx_count = 2, - .srcs[0].swizzle_type = VKD3D_SM4_SWIZZLE_VEC4, .srcs[0].swizzle = VKD3D_SHADER_NO_SWIZZLE, .src_count = 1, }; @@ -4185,7 +4179,7 @@ static void write_sm4_dcl_semantic(const struct tpf_writer *tpf, const struct hl .dst_count = 1, };
- if (hlsl_sm4_register_from_semantic(tpf->ctx, &var->semantic, output, &instr.dsts[0].reg.type, NULL, &has_idx)) + if (hlsl_sm4_register_from_semantic(tpf->ctx, &var->semantic, output, &instr.dsts[0].reg.type, &has_idx)) { if (has_idx) { @@ -4321,7 +4315,7 @@ static void write_sm4_unary_op(const struct tpf_writer *tpf, enum vkd3d_sm4_opco sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], src, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], src, instr.dsts[0].write_mask); instr.srcs[0].modifiers = src_mod; instr.src_count = 1;
@@ -4344,7 +4338,7 @@ static void write_sm4_unary_op_with_two_destinations(const struct tpf_writer *tp instr.dsts[1 - dst_idx].reg.idx_count = 0; instr.dst_count = 2;
- sm4_src_from_node(&instr.srcs[0], src, instr.dsts[dst_idx].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], src, instr.dsts[dst_idx].write_mask); instr.src_count = 1;
write_sm4_instruction(tpf, &instr); @@ -4361,8 +4355,8 @@ static void write_sm4_binary_op(const struct tpf_writer *tpf, enum vkd3d_sm4_opc sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], src1, instr.dsts[0].write_mask); - sm4_src_from_node(&instr.srcs[1], src2, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], src1, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[1], src2, instr.dsts[0].write_mask); instr.src_count = 2;
write_sm4_instruction(tpf, &instr); @@ -4380,8 +4374,8 @@ static void write_sm4_binary_op_dot(const struct tpf_writer *tpf, enum vkd3d_sm4 sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], src1, VKD3DSP_WRITEMASK_ALL); - sm4_src_from_node(&instr.srcs[1], src2, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[0], src1, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[1], src2, VKD3DSP_WRITEMASK_ALL); instr.src_count = 2;
write_sm4_instruction(tpf, &instr); @@ -4404,8 +4398,8 @@ static void write_sm4_binary_op_with_two_destinations(const struct tpf_writer *t instr.dsts[1 - dst_idx].reg.idx_count = 0; instr.dst_count = 2;
- sm4_src_from_node(&instr.srcs[0], src1, instr.dsts[dst_idx].write_mask); - sm4_src_from_node(&instr.srcs[1], src2, instr.dsts[dst_idx].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], src1, instr.dsts[dst_idx].write_mask); + sm4_src_from_node(tpf, &instr.srcs[1], src2, instr.dsts[dst_idx].write_mask); instr.src_count = 2;
write_sm4_instruction(tpf, &instr); @@ -4423,9 +4417,9 @@ static void write_sm4_ternary_op(const struct tpf_writer *tpf, enum vkd3d_sm4_op sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], src1, instr.dsts[0].write_mask); - sm4_src_from_node(&instr.srcs[1], src2, instr.dsts[0].write_mask); - sm4_src_from_node(&instr.srcs[2], src3, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], src1, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[1], src2, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[2], src3, instr.dsts[0].write_mask); instr.src_count = 3;
write_sm4_instruction(tpf, &instr); @@ -4474,9 +4468,9 @@ static void write_sm4_ld(const struct tpf_writer *tpf, const struct hlsl_ir_node coords_writemask = VKD3DSP_WRITEMASK_0 | VKD3DSP_WRITEMASK_1 | VKD3DSP_WRITEMASK_3; }
- sm4_src_from_node(&instr.srcs[0], coords, coords_writemask); + sm4_src_from_node(tpf, &instr.srcs[0], coords, coords_writemask);
- sm4_src_from_deref(tpf->ctx, &instr.srcs[1], resource, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[1], resource, instr.dsts[0].write_mask);
instr.src_count = 2;
@@ -4490,7 +4484,6 @@ static void write_sm4_ld(const struct tpf_writer *tpf, const struct hlsl_ir_node index = hlsl_ir_constant(sample_index);
memset(&instr.srcs[2], 0, sizeof(instr.srcs[2])); - instr.srcs[2].swizzle_type = VKD3D_SM4_SWIZZLE_NONE; reg->type = VKD3DSPR_IMMCONST; reg->dimension = VSIR_DIMENSION_SCALAR; reg->u.immconst_uint[0] = index->value.u[0].u; @@ -4501,7 +4494,7 @@ static void write_sm4_ld(const struct tpf_writer *tpf, const struct hlsl_ir_node } else { - sm4_src_from_node(&instr.srcs[2], sample_index, 0); + sm4_src_from_node(tpf, &instr.srcs[2], sample_index, 0); }
++instr.src_count; @@ -4563,27 +4556,27 @@ static void write_sm4_sample(const struct tpf_writer *tpf, const struct hlsl_ir_ sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], coords, VKD3DSP_WRITEMASK_ALL); - sm4_src_from_deref(tpf->ctx, &instr.srcs[1], resource, instr.dsts[0].write_mask); - sm4_src_from_deref(tpf->ctx, &instr.srcs[2], sampler, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[0], coords, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_deref(tpf, &instr.srcs[1], resource, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[2], sampler, VKD3DSP_WRITEMASK_ALL); instr.src_count = 3;
if (load->load_type == HLSL_RESOURCE_SAMPLE_LOD || load->load_type == HLSL_RESOURCE_SAMPLE_LOD_BIAS) { - sm4_src_from_node(&instr.srcs[3], load->lod.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[3], load->lod.node, VKD3DSP_WRITEMASK_ALL); ++instr.src_count; } else if (load->load_type == HLSL_RESOURCE_SAMPLE_GRAD) { - sm4_src_from_node(&instr.srcs[3], load->ddx.node, VKD3DSP_WRITEMASK_ALL); - sm4_src_from_node(&instr.srcs[4], load->ddy.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[3], load->ddx.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[4], load->ddy.node, VKD3DSP_WRITEMASK_ALL); instr.src_count += 2; } else if (load->load_type == HLSL_RESOURCE_SAMPLE_CMP || load->load_type == HLSL_RESOURCE_SAMPLE_CMP_LZ) { - sm4_src_from_node(&instr.srcs[3], load->cmp.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[3], load->cmp.node, VKD3DSP_WRITEMASK_ALL); ++instr.src_count; }
@@ -4606,7 +4599,7 @@ static void write_sm4_sampleinfo(const struct tpf_writer *tpf, const struct hlsl sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_deref(tpf->ctx, &instr.srcs[0], resource, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[0], resource, instr.dsts[0].write_mask); instr.src_count = 1;
write_sm4_instruction(tpf, &instr); @@ -4628,8 +4621,8 @@ static void write_sm4_resinfo(const struct tpf_writer *tpf, const struct hlsl_ir sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], load->lod.node, VKD3DSP_WRITEMASK_ALL); - sm4_src_from_deref(tpf->ctx, &instr.srcs[1], resource, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], load->lod.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_deref(tpf, &instr.srcs[1], resource, instr.dsts[0].write_mask); instr.src_count = 2;
write_sm4_instruction(tpf, &instr); @@ -4651,8 +4644,7 @@ static void write_sm4_cast_from_bool(const struct tpf_writer *tpf, const struct sm4_dst_from_node(&instr.dsts[0], &expr->node); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], arg, instr.dsts[0].write_mask); - instr.srcs[1].swizzle_type = VKD3D_SM4_SWIZZLE_NONE; + sm4_src_from_node(tpf, &instr.srcs[0], arg, instr.dsts[0].write_mask); instr.srcs[1].reg.type = VKD3DSPR_IMMCONST; instr.srcs[1].reg.dimension = VSIR_DIMENSION_SCALAR; instr.srcs[1].reg.u.immconst_uint[0] = mask; @@ -4778,11 +4770,11 @@ static void write_sm4_store_uav_typed(const struct tpf_writer *tpf, const struct memset(&instr, 0, sizeof(instr)); instr.opcode = VKD3D_SM5_OP_STORE_UAV_TYPED;
- sm4_register_from_deref(tpf->ctx, &instr.dsts[0].reg, &instr.dsts[0].write_mask, NULL, dst); + sm4_register_from_deref(tpf->ctx, &instr.dsts[0].reg, &instr.dsts[0].write_mask, dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], coords, VKD3DSP_WRITEMASK_ALL); - sm4_src_from_node(&instr.srcs[1], value, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[0], coords, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[1], value, VKD3DSP_WRITEMASK_ALL); instr.src_count = 2;
write_sm4_instruction(tpf, &instr); @@ -5239,7 +5231,7 @@ static void write_sm4_if(const struct tpf_writer *tpf, const struct hlsl_ir_if *
assert(iff->condition.node->data_type->dimx == 1);
- sm4_src_from_node(&instr.srcs[0], iff->condition.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[0], iff->condition.node, VKD3DSP_WRITEMASK_ALL); write_sm4_instruction(tpf, &instr);
write_sm4_block(tpf, &iff->then_block); @@ -5274,7 +5266,7 @@ static void write_sm4_jump(const struct tpf_writer *tpf, const struct hlsl_ir_ju
memset(&instr.srcs[0], 0, sizeof(*instr.srcs)); instr.src_count = 1; - sm4_src_from_node(&instr.srcs[0], jump->condition.node, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[0], jump->condition.node, VKD3DSP_WRITEMASK_ALL); break; }
@@ -5320,7 +5312,7 @@ static void write_sm4_load(const struct tpf_writer *tpf, const struct hlsl_ir_lo
instr.opcode = VKD3D_SM4_OP_MOVC;
- sm4_src_from_deref(tpf->ctx, &instr.srcs[0], &load->src, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[0], &load->src, instr.dsts[0].write_mask);
memset(&value, 0xff, sizeof(value)); sm4_src_from_constant_value(&instr.srcs[1], &value, type->dimx, instr.dsts[0].write_mask); @@ -5332,7 +5324,7 @@ static void write_sm4_load(const struct tpf_writer *tpf, const struct hlsl_ir_lo { instr.opcode = VKD3D_SM4_OP_MOV;
- sm4_src_from_deref(tpf->ctx, &instr.srcs[0], &load->src, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[0], &load->src, instr.dsts[0].write_mask); instr.src_count = 1; }
@@ -5368,7 +5360,7 @@ static void write_sm4_gather(const struct tpf_writer *tpf, const struct hlsl_ir_ sm4_dst_from_node(&instr.dsts[0], dst); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[instr.src_count++], coords, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[instr.src_count++], coords, VKD3DSP_WRITEMASK_ALL);
if (texel_offset) { @@ -5381,16 +5373,15 @@ static void write_sm4_gather(const struct tpf_writer *tpf, const struct hlsl_ir_ return; } instr.opcode = VKD3D_SM5_OP_GATHER4_PO; - sm4_src_from_node(&instr.srcs[instr.src_count++], texel_offset, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_node(tpf, &instr.srcs[instr.src_count++], texel_offset, VKD3DSP_WRITEMASK_ALL); } }
- sm4_src_from_deref(tpf->ctx, &instr.srcs[instr.src_count++], resource, instr.dsts[0].write_mask); + sm4_src_from_deref(tpf, &instr.srcs[instr.src_count++], resource, instr.dsts[0].write_mask);
src = &instr.srcs[instr.src_count++]; - sm4_src_from_deref(tpf->ctx, src, sampler, VKD3DSP_WRITEMASK_ALL); + sm4_src_from_deref(tpf, src, sampler, VKD3DSP_WRITEMASK_ALL); src->reg.dimension = VSIR_DIMENSION_VEC4; - src->swizzle_type = VKD3D_SM4_SWIZZLE_SCALAR; src->swizzle = swizzle;
write_sm4_instruction(tpf, &instr); @@ -5490,11 +5481,11 @@ static void write_sm4_store(const struct tpf_writer *tpf, const struct hlsl_ir_s memset(&instr, 0, sizeof(instr)); instr.opcode = VKD3D_SM4_OP_MOV;
- sm4_register_from_deref(tpf->ctx, &instr.dsts[0].reg, &writemask, NULL, &store->lhs); + sm4_register_from_deref(tpf->ctx, &instr.dsts[0].reg, &writemask, &store->lhs); instr.dsts[0].write_mask = hlsl_combine_writemasks(writemask, store->writemask); instr.dst_count = 1;
- sm4_src_from_node(&instr.srcs[0], rhs, instr.dsts[0].write_mask); + sm4_src_from_node(tpf, &instr.srcs[0], rhs, instr.dsts[0].write_mask); instr.src_count = 1;
write_sm4_instruction(tpf, &instr); @@ -5511,7 +5502,7 @@ static void write_sm4_swizzle(const struct tpf_writer *tpf, const struct hlsl_ir sm4_dst_from_node(&instr.dsts[0], &swizzle->node); instr.dst_count = 1;
- sm4_register_from_node(&instr.srcs[0].reg, &writemask, &instr.srcs[0].swizzle_type, swizzle->val.node); + sm4_register_from_node(&instr.srcs[0].reg, &writemask, swizzle->val.node); hlsl_swizzle = hlsl_map_swizzle(hlsl_combine_swizzles(hlsl_swizzle_from_writemask(writemask), swizzle->swizzle, swizzle->node.data_type->dimx), instr.dsts[0].write_mask); instr.srcs[0].swizzle = swizzle_from_sm4(hlsl_swizzle);
From: Francisco Casas fcasas@codeweavers.com
--- libs/vkd3d-shader/tpf.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-)
diff --git a/libs/vkd3d-shader/tpf.c b/libs/vkd3d-shader/tpf.c index 7581e1c2..d6902f8b 100644 --- a/libs/vkd3d-shader/tpf.c +++ b/libs/vkd3d-shader/tpf.c @@ -3635,13 +3635,7 @@ struct sm4_instruction struct vkd3d_shader_dst_param dsts[2]; unsigned int dst_count;
- struct sm4_src_register - { - struct vkd3d_shader_register reg; - enum vkd3d_sm4_swizzle_type swizzle_type; - DWORD swizzle; - enum vkd3d_shader_src_modifier modifiers; - } srcs[5]; + struct vkd3d_shader_src_param srcs[5]; unsigned int src_count;
unsigned int byte_stride; @@ -3777,7 +3771,7 @@ static void sm4_register_from_deref(struct hlsl_ctx *ctx, struct vkd3d_shader_re } }
-static void sm4_src_from_deref(const struct tpf_writer *tpf, struct sm4_src_register *src, +static void sm4_src_from_deref(const struct tpf_writer *tpf, struct vkd3d_shader_src_param *src, const struct hlsl_deref *deref, unsigned int map_writemask) { unsigned int writemask, hlsl_swizzle; @@ -3806,7 +3800,7 @@ static void sm4_dst_from_node(struct vkd3d_shader_dst_param *dst, const struct h sm4_register_from_node(&dst->reg, &dst->write_mask, instr); }
-static void sm4_src_from_constant_value(struct sm4_src_register *src, +static void sm4_src_from_constant_value(struct vkd3d_shader_src_param *src, const struct hlsl_constant_value *value, unsigned int width, unsigned int map_writemask) { src->swizzle = VKD3D_SHADER_NO_SWIZZLE; @@ -3831,7 +3825,7 @@ static void sm4_src_from_constant_value(struct sm4_src_register *src, } }
-static void sm4_src_from_node(const struct tpf_writer *tpf, struct sm4_src_register *src, +static void sm4_src_from_node(const struct tpf_writer *tpf, struct vkd3d_shader_src_param *src, const struct hlsl_ir_node *instr, unsigned int map_writemask) { unsigned int writemask, hlsl_swizzle; @@ -3887,7 +3881,7 @@ static void sm4_write_dst_register(const struct tpf_writer *tpf, const struct vk } }
-static void sm4_write_src_register(const struct tpf_writer *tpf, const struct sm4_src_register *src) +static void sm4_write_src_register(const struct tpf_writer *tpf, const struct vkd3d_shader_src_param *src) { const struct vkd3d_sm4_register_type_info *register_type_info; struct vkd3d_bytecode_buffer *buffer = tpf->buffer; @@ -3973,7 +3967,7 @@ static void sm4_write_src_register(const struct tpf_writer *tpf, const struct sm } }
-static uint32_t vkd3d_shader_dst_param_order(const struct vkd3d_shader_dst_param *dst) +static uint32_t sm4_token_count_from_dst_register(const struct vkd3d_shader_dst_param *dst) { uint32_t order = 1; if (dst->reg.type == VKD3DSPR_IMMCONST) @@ -3982,7 +3976,7 @@ static uint32_t vkd3d_shader_dst_param_order(const struct vkd3d_shader_dst_param return order; }
-static uint32_t sm4_src_register_order(const struct sm4_src_register *src) +static uint32_t sm4_token_count_from_src_register(const struct vkd3d_shader_src_param *src) { uint32_t order = 1; if (src->reg.type == VKD3DSPR_IMMCONST) @@ -4001,9 +3995,9 @@ static void write_sm4_instruction(const struct tpf_writer *tpf, const struct sm4
size += instr->modifier_count; for (i = 0; i < instr->dst_count; ++i) - size += vkd3d_shader_dst_param_order(&instr->dsts[i]); + size += sm4_token_count_from_dst_register(&instr->dsts[i]); for (i = 0; i < instr->src_count; ++i) - size += sm4_src_register_order(&instr->srcs[i]); + size += sm4_token_count_from_src_register(&instr->srcs[i]); size += instr->idx_count; if (instr->byte_stride) ++size; @@ -5350,7 +5344,7 @@ static void write_sm4_gather(const struct tpf_writer *tpf, const struct hlsl_ir_ const struct hlsl_deref *resource, const struct hlsl_deref *sampler, const struct hlsl_ir_node *coords, DWORD swizzle, const struct hlsl_ir_node *texel_offset) { - struct sm4_src_register *src; + struct vkd3d_shader_src_param *src; struct sm4_instruction instr;
memset(&instr, 0, sizeof(instr));
That should cause compilation to fail, right? (And probably with something like an internal compiler error, because this is something vkd3d-shader itself would trigger.)
Right. We don't want incorrect shaders to compile. I will put ```c ERR("Unhandled register modifier %#x.\n", src->modifiers); vkd3d_unreachable(); ``` then.
The naming here seems a bit unfortunate. As far as I can tell, this function determines the TPF source token count, which isn't a property of the vsir source parameter as such. I.e., I'd expect this to be called something along the lines of "sm4_token_count_from_src_register()", although you could possibly make a case for something like "vsir_src_register_get_tpf_token_count()". (vkd3d_shader_dst_param_order() has that issue as well.)
"sm4_token_count_from_src_register()" sounds good to me, since other static functions in tpf.c start with "sm4_" too. I also renamed vkd3d_shader_dst_param_order().
That should cause compilation to fail, right? (And probably with something like an internal compiler error, because this is something vkd3d-shader itself would trigger.)
Right. We don't want incorrect shaders to compile. I will put
ERR("Unhandled register modifier %#x.\n", src->modifiers); vkd3d_unreachable();then.
I'll approve it, but I don't think that's terribly robust. Much like assert() and abort(), this escalates what would by definition be a bug in our code from causing compilation failure to causing program termination.
@@ -4001,9 +3995,9 @@ static void write_sm4_instruction(const struct tpf_writer *tpf, const struct sm4 size += instr->modifier_count; for (i = 0; i < instr->dst_count; ++i) - size += vkd3d_shader_dst_param_order(&instr->dsts[i]); + size += sm4_token_count_from_dst_register(&instr->dsts[i]); for (i = 0; i < instr->src_count; ++i) - size += sm4_src_register_order(&instr->srcs[i]); + size += sm4_token_count_from_src_register(&instr->srcs[i]); size += instr->idx_count; if (instr->byte_stride) ++size;
As an aside, note that we don't necessarily need to calculate the size in advance here; we could do something like this:
```c token_position = put_u32(buffer, 0); ... size = (buffer->size - token_position) / sizeof(uint32_t); token |= (size << VKD3D_SM4_INSTRUCTION_LENGTH_SHIFT); set_u32(buffer, token_position, token); ```
Might be a nice follow-up to this series.
This merge request was approved by Henri Verbeet.
On Tue Oct 3 15:06:23 2023 +0000, Henri Verbeet wrote:
That should cause compilation to fail, right? (And probably with
something like an internal compiler error, because this is something vkd3d-shader itself would trigger.)
Right. We don't want incorrect shaders to compile. I will put
ERR("Unhandled register modifier %#x.\n", src->modifiers); vkd3d_unreachable();then.
I'll approve it, but I don't think that's terribly robust. Much like assert() and abort(), this escalates what would by definition be a bug in our code from causing compilation failure to causing program termination.
@@ -4001,9 +3995,9 @@ static void write_sm4_instruction(const structtpf_writer *tpf, const struct sm4
size += instr->modifier_count; for (i = 0; i < instr->dst_count; ++i)
size += vkd3d_shader_dst_param_order(&instr->dsts[i]);
size += sm4_token_count_from_dst_register(&instr->dsts[i]);
size += sm4_src_register_order(&instr->srcs[i]);
size += sm4_token_count_from_src_register(&instr->srcs[i]);As an aside, note that we don't necessarily need to calculate the size in advance here; we could do something like this:
token_position = put_u32(buffer, 0); ... size = (buffer->size - token_position) / sizeof(uint32_t); token |= (size << VKD3D_SM4_INSTRUCTION_LENGTH_SHIFT); set_u32(buffer, token_position, token);Might be a nice follow-up to this series.
I see. Nice observation.
-
Francisco Casas -
Francisco Casas (＠fcasas)
-
Henri Verbeet (＠hverbeet)