static void
build_color_shaders(struct nir_shader **out_vs,
struct nir_shader **out_fs,
uint32_t frag_output)
{
nir_builder vs_b;
nir_builder fs_b;
nir_builder_init_simple_shader(&vs_b, NULL, MESA_SHADER_VERTEX, NULL);
nir_builder_init_simple_shader(&fs_b, NULL, MESA_SHADER_FRAGMENT, NULL);
vs_b.shader->info.name = ralloc_strdup(vs_b.shader, "meta_clear_color_vs");
fs_b.shader->info.name = ralloc_strdup(fs_b.shader, "meta_clear_color_fs");
const struct glsl_type *position_type = glsl_vec4_type();
const struct glsl_type *color_type = glsl_vec4_type();
nir_variable *vs_out_pos =
nir_variable_create(vs_b.shader, nir_var_shader_out, position_type,
"gl_Position");
vs_out_pos->data.location = VARYING_SLOT_POS;
nir_intrinsic_instr *in_color_load = nir_intrinsic_instr_create(fs_b.shader, nir_intrinsic_load_push_constant);
nir_intrinsic_set_base(in_color_load, 0);
nir_intrinsic_set_range(in_color_load, 16);
in_color_load->src[0] = nir_src_for_ssa(nir_imm_int(&fs_b, 0));
in_color_load->num_components = 4;
nir_ssa_dest_init(&in_color_load->instr, &in_color_load->dest, 4, 32, "clear color");
nir_builder_instr_insert(&fs_b, &in_color_load->instr);
nir_variable *fs_out_color =
nir_variable_create(fs_b.shader, nir_var_shader_out, color_type,
"f_color");
fs_out_color->data.location = FRAG_RESULT_DATA0 + frag_output;
nir_store_var(&fs_b, fs_out_color, &in_color_load->dest.ssa, 0xf);
nir_ssa_def *outvec = radv_meta_gen_rect_vertices(&vs_b);
nir_store_var(&vs_b, vs_out_pos, outvec, 0xf);
const struct glsl_type *layer_type = glsl_int_type();
nir_variable *vs_out_layer =
nir_variable_create(vs_b.shader, nir_var_shader_out, layer_type,
"v_layer");
vs_out_layer->data.location = VARYING_SLOT_LAYER;
vs_out_layer->data.interpolation = INTERP_MODE_FLAT;
nir_ssa_def *inst_id = nir_load_system_value(&vs_b, nir_intrinsic_load_instance_id, 0);
nir_ssa_def *base_instance = nir_load_system_value(&vs_b, nir_intrinsic_load_base_instance, 0);
nir_ssa_def *layer_id = nir_iadd(&vs_b, inst_id, base_instance);
nir_store_var(&vs_b, vs_out_layer, layer_id, 0x1);
*out_vs = vs_b.shader;
*out_fs = fs_b.shader;
}
static nir_shader *
build_buffer_fill_shader(struct radv_device *dev)
{
nir_builder b;
nir_builder_init_simple_shader(&b, NULL, MESA_SHADER_COMPUTE, NULL);
b.shader->info->name = ralloc_strdup(b.shader, "meta_buffer_fill");
b.shader->info->cs.local_size[0] = 64;
b.shader->info->cs.local_size[1] = 1;
b.shader->info->cs.local_size[2] = 1;
nir_ssa_def *invoc_id = nir_load_system_value(&b, nir_intrinsic_load_local_invocation_id, 0);
nir_ssa_def *wg_id = nir_load_system_value(&b, nir_intrinsic_load_work_group_id, 0);
nir_ssa_def *block_size = nir_imm_ivec4(&b,
b.shader->info->cs.local_size[0],
b.shader->info->cs.local_size[1],
b.shader->info->cs.local_size[2], 0);
nir_ssa_def *global_id = nir_iadd(&b, nir_imul(&b, wg_id, block_size), invoc_id);
nir_ssa_def *offset = nir_imul(&b, global_id, nir_imm_int(&b, 16));
offset = nir_swizzle(&b, offset, (unsigned[]) {0, 0, 0, 0}, 1, false);
static nir_shader *
build_resolve_compute_shader(struct radv_device *dev, bool is_integer, int samples)
{
nir_builder b;
char name[64];
nir_if *outer_if = NULL;
const struct glsl_type *sampler_type = glsl_sampler_type(GLSL_SAMPLER_DIM_MS,
false,
false,
GLSL_TYPE_FLOAT);
const struct glsl_type *img_type = glsl_sampler_type(GLSL_SAMPLER_DIM_2D,
false,
false,
GLSL_TYPE_FLOAT);
snprintf(name, 64, "meta_resolve_cs-%d-%s", samples, is_integer ? "int" : "float");
nir_builder_init_simple_shader(&b, NULL, MESA_SHADER_COMPUTE, NULL);
b.shader->info->name = ralloc_strdup(b.shader, name);
b.shader->info->cs.local_size[0] = 16;
b.shader->info->cs.local_size[1] = 16;
b.shader->info->cs.local_size[2] = 1;
nir_variable *input_img = nir_variable_create(b.shader, nir_var_uniform,
sampler_type, "s_tex");
input_img->data.descriptor_set = 0;
input_img->data.binding = 0;
nir_variable *output_img = nir_variable_create(b.shader, nir_var_uniform,
img_type, "out_img");
output_img->data.descriptor_set = 0;
output_img->data.binding = 1;
nir_ssa_def *invoc_id = nir_load_system_value(&b, nir_intrinsic_load_local_invocation_id, 0);
nir_ssa_def *wg_id = nir_load_system_value(&b, nir_intrinsic_load_work_group_id, 0);
nir_ssa_def *block_size = nir_imm_ivec4(&b,
b.shader->info->cs.local_size[0],
b.shader->info->cs.local_size[1],
b.shader->info->cs.local_size[2], 0);
nir_ssa_def *global_id = nir_iadd(&b, nir_imul(&b, wg_id, block_size), invoc_id);
nir_intrinsic_instr *src_offset = nir_intrinsic_instr_create(b.shader, nir_intrinsic_load_push_constant);
src_offset->src[0] = nir_src_for_ssa(nir_imm_int(&b, 0));
src_offset->num_components = 2;
nir_ssa_dest_init(&src_offset->instr, &src_offset->dest, 2, 32, "src_offset");
nir_builder_instr_insert(&b, &src_offset->instr);
nir_intrinsic_instr *dst_offset = nir_intrinsic_instr_create(b.shader, nir_intrinsic_load_push_constant);
dst_offset->src[0] = nir_src_for_ssa(nir_imm_int(&b, 8));
dst_offset->num_components = 2;
nir_ssa_dest_init(&dst_offset->instr, &dst_offset->dest, 2, 32, "dst_offset");
nir_builder_instr_insert(&b, &dst_offset->instr);
nir_ssa_def *img_coord = nir_iadd(&b, global_id, &src_offset->dest.ssa);
/* do a txf_ms on each sample */
nir_ssa_def *tmp;
nir_tex_instr *tex = nir_tex_instr_create(b.shader, 2);
tex->sampler_dim = GLSL_SAMPLER_DIM_MS;
tex->op = nir_texop_txf_ms;
tex->src[0].src_type = nir_tex_src_coord;
tex->src[0].src = nir_src_for_ssa(img_coord);
tex->src[1].src_type = nir_tex_src_ms_index;
tex->src[1].src = nir_src_for_ssa(nir_imm_int(&b, 0));
tex->dest_type = nir_type_float;
tex->is_array = false;
tex->coord_components = 2;
tex->texture = nir_deref_var_create(tex, input_img);
tex->sampler = NULL;
nir_ssa_dest_init(&tex->instr, &tex->dest, 4, 32, "tex");
nir_builder_instr_insert(&b, &tex->instr);
tmp = &tex->dest.ssa;
nir_variable *color =
nir_local_variable_create(b.impl, glsl_vec4_type(), "color");
if (!is_integer && samples > 1) {
nir_tex_instr *tex_all_same = nir_tex_instr_create(b.shader, 1);
tex_all_same->sampler_dim = GLSL_SAMPLER_DIM_MS;
tex_all_same->op = nir_texop_samples_identical;
tex_all_same->src[0].src_type = nir_tex_src_coord;
tex_all_same->src[0].src = nir_src_for_ssa(img_coord);
tex_all_same->dest_type = nir_type_float;
tex_all_same->is_array = false;
tex_all_same->coord_components = 2;
tex_all_same->texture = nir_deref_var_create(tex_all_same, input_img);
tex_all_same->sampler = NULL;
nir_ssa_dest_init(&tex_all_same->instr, &tex_all_same->dest, 1, 32, "tex");
nir_builder_instr_insert(&b, &tex_all_same->instr);
nir_ssa_def *all_same = nir_ine(&b, &tex_all_same->dest.ssa, nir_imm_int(&b, 0));
nir_if *if_stmt = nir_if_create(b.shader);
if_stmt->condition = nir_src_for_ssa(all_same);
nir_cf_node_insert(b.cursor, &if_stmt->cf_node);
b.cursor = nir_after_cf_list(&if_stmt->then_list);
for (int i = 1; i < samples; i++) {
nir_tex_instr *tex_add = nir_tex_instr_create(b.shader, 2);
tex_add->sampler_dim = GLSL_SAMPLER_DIM_MS;
tex_add->op = nir_texop_txf_ms;
tex_add->src[0].src_type = nir_tex_src_coord;
tex_add->src[0].src = nir_src_for_ssa(img_coord);
tex_add->src[1].src_type = nir_tex_src_ms_index;
tex_add->src[1].src = nir_src_for_ssa(nir_imm_int(&b, i));
tex_add->dest_type = nir_type_float;
tex_add->is_array = false;
tex_add->coord_components = 2;
tex_add->texture = nir_deref_var_create(tex_add, input_img);
tex_add->sampler = NULL;
nir_ssa_dest_init(&tex_add->instr, &tex_add->dest, 4, 32, "tex");
nir_builder_instr_insert(&b, &tex_add->instr);
tmp = nir_fadd(&b, tmp, &tex_add->dest.ssa);
}
tmp = nir_fdiv(&b, tmp, nir_imm_float(&b, samples));
nir_store_var(&b, color, tmp, 0xf);
b.cursor = nir_after_cf_list(&if_stmt->else_list);
outer_if = if_stmt;
}
nir_store_var(&b, color, &tex->dest.ssa, 0xf);
if (outer_if)
b.cursor = nir_after_cf_node(&outer_if->cf_node);
nir_ssa_def *newv = nir_load_var(&b, color);
nir_ssa_def *coord = nir_iadd(&b, global_id, &dst_offset->dest.ssa);
nir_intrinsic_instr *store = nir_intrinsic_instr_create(b.shader, nir_intrinsic_image_store);
store->src[0] = nir_src_for_ssa(coord);
store->src[1] = nir_src_for_ssa(nir_ssa_undef(&b, 1, 32));
store->src[2] = nir_src_for_ssa(newv);
store->variables[0] = nir_deref_var_create(store, output_img);
nir_builder_instr_insert(&b, &store->instr);
return b.shader;
}
static nir_shader *
create_passthrough_tcs(void *mem_ctx, const struct brw_compiler *compiler,
const nir_shader_compiler_options *options,
const struct brw_tcs_prog_key *key)
{
nir_builder b;
nir_builder_init_simple_shader(&b, mem_ctx, MESA_SHADER_TESS_CTRL,
options);
nir_shader *nir = b.shader;
nir_variable *var;
nir_intrinsic_instr *load;
nir_intrinsic_instr *store;
nir_ssa_def *zero = nir_imm_int(&b, 0);
nir_ssa_def *invoc_id =
nir_load_system_value(&b, nir_intrinsic_load_invocation_id, 0);
nir->info->inputs_read = key->outputs_written;
nir->info->outputs_written = key->outputs_written;
nir->info->tcs.vertices_out = key->input_vertices;
nir->info->name = ralloc_strdup(nir, "passthrough");
nir->num_uniforms = 8 * sizeof(uint32_t);
var = nir_variable_create(nir, nir_var_uniform, glsl_vec4_type(), "hdr_0");
var->data.location = 0;
var = nir_variable_create(nir, nir_var_uniform, glsl_vec4_type(), "hdr_1");
var->data.location = 1;
/* Write the patch URB header. */
for (int i = 0; i <= 1; i++) {
load = nir_intrinsic_instr_create(nir, nir_intrinsic_load_uniform);
load->num_components = 4;
load->src[0] = nir_src_for_ssa(zero);
nir_ssa_dest_init(&load->instr, &load->dest, 4, 32, NULL);
nir_intrinsic_set_base(load, i * 4 * sizeof(uint32_t));
nir_builder_instr_insert(&b, &load->instr);
store = nir_intrinsic_instr_create(nir, nir_intrinsic_store_output);
store->num_components = 4;
store->src[0] = nir_src_for_ssa(&load->dest.ssa);
store->src[1] = nir_src_for_ssa(zero);
nir_intrinsic_set_base(store, VARYING_SLOT_TESS_LEVEL_INNER - i);
nir_intrinsic_set_write_mask(store, WRITEMASK_XYZW);
nir_builder_instr_insert(&b, &store->instr);
}
/* Copy inputs to outputs. */
uint64_t varyings = key->outputs_written;
while (varyings != 0) {
const int varying = ffsll(varyings) - 1;
load = nir_intrinsic_instr_create(nir,
nir_intrinsic_load_per_vertex_input);
load->num_components = 4;
load->src[0] = nir_src_for_ssa(invoc_id);
load->src[1] = nir_src_for_ssa(zero);
nir_ssa_dest_init(&load->instr, &load->dest, 4, 32, NULL);
nir_intrinsic_set_base(load, varying);
nir_builder_instr_insert(&b, &load->instr);
store = nir_intrinsic_instr_create(nir,
nir_intrinsic_store_per_vertex_output);
store->num_components = 4;
store->src[0] = nir_src_for_ssa(&load->dest.ssa);
store->src[1] = nir_src_for_ssa(invoc_id);
store->src[2] = nir_src_for_ssa(zero);
nir_intrinsic_set_base(store, varying);
nir_intrinsic_set_write_mask(store, WRITEMASK_XYZW);
nir_builder_instr_insert(&b, &store->instr);
varyings &= ~BITFIELD64_BIT(varying);
}
nir_validate_shader(nir);
nir = brw_preprocess_nir(compiler, nir);
return nir;
}
static nir_shader *
build_nir_itob_compute_shader(struct radv_device *dev)
{
nir_builder b;
const struct glsl_type *sampler_type = glsl_sampler_type(GLSL_SAMPLER_DIM_2D,
false,
false,
GLSL_TYPE_FLOAT);
const struct glsl_type *img_type = glsl_sampler_type(GLSL_SAMPLER_DIM_BUF,
false,
false,
GLSL_TYPE_FLOAT);
nir_builder_init_simple_shader(&b, NULL, MESA_SHADER_COMPUTE, NULL);
b.shader->info->name = ralloc_strdup(b.shader, "meta_itob_cs");
b.shader->info->cs.local_size[0] = 16;
b.shader->info->cs.local_size[1] = 16;
b.shader->info->cs.local_size[2] = 1;
nir_variable *input_img = nir_variable_create(b.shader, nir_var_uniform,
sampler_type, "s_tex");
input_img->data.descriptor_set = 0;
input_img->data.binding = 0;
nir_variable *output_img = nir_variable_create(b.shader, nir_var_uniform,
img_type, "out_img");
output_img->data.descriptor_set = 0;
output_img->data.binding = 1;
nir_ssa_def *invoc_id = nir_load_system_value(&b, nir_intrinsic_load_local_invocation_id, 0);
nir_ssa_def *wg_id = nir_load_system_value(&b, nir_intrinsic_load_work_group_id, 0);
nir_ssa_def *block_size = nir_imm_ivec4(&b,
b.shader->info->cs.local_size[0],
b.shader->info->cs.local_size[1],
b.shader->info->cs.local_size[2], 0);
nir_ssa_def *global_id = nir_iadd(&b, nir_imul(&b, wg_id, block_size), invoc_id);
nir_intrinsic_instr *offset = nir_intrinsic_instr_create(b.shader, nir_intrinsic_load_push_constant);
offset->src[0] = nir_src_for_ssa(nir_imm_int(&b, 0));
offset->num_components = 2;
nir_ssa_dest_init(&offset->instr, &offset->dest, 2, 32, "offset");
nir_builder_instr_insert(&b, &offset->instr);
nir_intrinsic_instr *stride = nir_intrinsic_instr_create(b.shader, nir_intrinsic_load_push_constant);
stride->src[0] = nir_src_for_ssa(nir_imm_int(&b, 8));
stride->num_components = 1;
nir_ssa_dest_init(&stride->instr, &stride->dest, 1, 32, "stride");
nir_builder_instr_insert(&b, &stride->instr);
nir_ssa_def *img_coord = nir_iadd(&b, global_id, &offset->dest.ssa);
nir_tex_instr *tex = nir_tex_instr_create(b.shader, 2);
tex->sampler_dim = GLSL_SAMPLER_DIM_2D;
tex->op = nir_texop_txf;
tex->src[0].src_type = nir_tex_src_coord;
tex->src[0].src = nir_src_for_ssa(img_coord);
tex->src[1].src_type = nir_tex_src_lod;
tex->src[1].src = nir_src_for_ssa(nir_imm_int(&b, 0));
tex->dest_type = nir_type_float;
tex->is_array = false;
tex->coord_components = 2;
tex->texture = nir_deref_var_create(tex, input_img);
tex->sampler = NULL;
nir_ssa_dest_init(&tex->instr, &tex->dest, 4, 32, "tex");
nir_builder_instr_insert(&b, &tex->instr);
nir_ssa_def *pos_x = nir_channel(&b, global_id, 0);
nir_ssa_def *pos_y = nir_channel(&b, global_id, 1);
nir_ssa_def *tmp = nir_imul(&b, pos_y, &stride->dest.ssa);
tmp = nir_iadd(&b, tmp, pos_x);
nir_ssa_def *coord = nir_vec4(&b, tmp, tmp, tmp, tmp);
nir_ssa_def *outval = &tex->dest.ssa;
nir_intrinsic_instr *store = nir_intrinsic_instr_create(b.shader, nir_intrinsic_image_store);
store->src[0] = nir_src_for_ssa(coord);
store->src[1] = nir_src_for_ssa(nir_ssa_undef(&b, 1, 32));
store->src[2] = nir_src_for_ssa(outval);
store->variables[0] = nir_deref_var_create(store, output_img);
nir_builder_instr_insert(&b, &store->instr);
return b.shader;
}