This patch moves the following data structures from vec4_visitor to
vec4_vs_visitor, since they contain VS-specific data:
- struct brw_vs_compile *c
- struct brw_vs_prog_data *prog_data
- src_reg *vp_temp_regs
- src_reg vp_addr_reg
Since brw_vs_compile and brw_vs_prog_data also contain vec4-generic
data, the following pointers are added to the base class, to allow it
to access the vec4-generic portions of these data structures:
- struct brw_vec4_compile *vec4_compile
- struct brw_vec4_prog_key *vec4_key
- struct brw_vec4_prog_data *vec4_prog_data
---
src/mesa/drivers/dri/i965/brw_vec4.cpp | 26 ++++-----
src/mesa/drivers/dri/i965/brw_vec4.h | 19 +++---
.../drivers/dri/i965/brw_vec4_reg_allocate.cpp | 12 ++--
src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp | 67 ++++++++++++----------
.../dri/i965/test_vec4_register_coalesce.cpp | 12 ++--
5 files changed, 72 insertions(+), 64 deletions(-)
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.cpp
b/src/mesa/drivers/dri/i965/brw_vec4.cpp
index 5f9372d..847adc7 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4.cpp
@@ -410,8 +410,8 @@ vec4_visitor::pack_uniform_registers()
/* Move the references to the data */
for (int j = 0; j < size; j++) {
- prog_data->base.param[dst * 4 + new_chan[src] + j] =
- prog_data->base.param[src * 4 + j];
+ vec4_prog_data->param[dst * 4 + new_chan[src] + j] =
+ vec4_prog_data->param[src * 4 + j];
}
this->uniform_vector_size[dst] += size;
@@ -562,16 +562,16 @@ vec4_visitor::move_push_constants_to_pull_constants()
pull_constant_loc[i / 4] = -1;
if (i >= max_uniform_components) {
- const float **values = &prog_data->base.param[i];
+ const float **values = &vec4_prog_data->param[i];
/* Try to find an existing copy of this uniform in the pull
* constants if it was part of an array access already.
*/
- for (unsigned int j = 0; j < prog_data->base.nr_pull_params; j += 4) {
+ for (unsigned int j = 0; j < vec4_prog_data->nr_pull_params; j += 4) {
int matches;
for (matches = 0; matches < 4; matches++) {
- if (prog_data->base.pull_param[j + matches] != values[matches])
+ if (vec4_prog_data->pull_param[j + matches] != values[matches])
break;
}
@@ -582,11 +582,11 @@ vec4_visitor::move_push_constants_to_pull_constants()
}
if (pull_constant_loc[i / 4] == -1) {
- assert(prog_data->base.nr_pull_params % 4 == 0);
- pull_constant_loc[i / 4] = prog_data->base.nr_pull_params / 4;
+ assert(vec4_prog_data->nr_pull_params % 4 == 0);
+ pull_constant_loc[i / 4] = vec4_prog_data->nr_pull_params / 4;
for (int j = 0; j < 4; j++) {
- prog_data->base.pull_param[prog_data->base.nr_pull_params++] =
values[j];
+ vec4_prog_data->pull_param[vec4_prog_data->nr_pull_params++] =
values[j];
}
}
}
@@ -646,7 +646,7 @@ vec4_visitor::opt_set_dependency_control()
cfg_t cfg(this);
- assert(prog_data->base.total_grf ||
+ assert(vec4_prog_data->total_grf ||
!"Must be called after register allocation");
for (int i = 0; i < cfg.num_blocks; i++) {
@@ -1259,7 +1259,7 @@ vec4_visitor::setup_uniforms(int reg)
for (unsigned int i = 0; i < 4; i++) {
unsigned int slot = this->uniforms * 4 + i;
static float zero = 0.0;
- prog_data->base.param[slot] = &zero;
+ vec4_prog_data->param[slot] = &zero;
}
this->uniforms++;
@@ -1268,9 +1268,9 @@ vec4_visitor::setup_uniforms(int reg)
reg += ALIGN(uniforms, 2) / 2;
}
- prog_data->base.nr_params = this->uniforms * 4;
+ vec4_prog_data->nr_params = this->uniforms * 4;
- prog_data->base.curb_read_length = reg - 1;
+ vec4_prog_data->curb_read_length = reg - 1;
return reg;
}
@@ -1403,7 +1403,7 @@ vec4_visitor::run()
}
base_ir = NULL;
- if (c->key.base.userclip_active && !c->key.base.uses_clip_distance)
+ if (vec4_key->userclip_active && !vec4_key->uses_clip_distance)
setup_uniform_clipplane_values();
emit_thread_end();
diff --git a/src/mesa/drivers/dri/i965/brw_vec4.h
b/src/mesa/drivers/dri/i965/brw_vec4.h
index 54109e4..a5aa996 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4.h
+++ b/src/mesa/drivers/dri/i965/brw_vec4.h
@@ -210,8 +210,10 @@ class vec4_visitor : public backend_visitor
{
public:
vec4_visitor(struct brw_context *brw,
- struct brw_vs_compile *c,
- struct brw_vs_prog_data *prog_data,
+ struct brw_vec4_compile *vec4_compile,
+ struct gl_program *p,
+ const struct brw_vec4_prog_key *vec4_key,
+ struct brw_vec4_prog_data *vec4_prog_data,
struct gl_shader_program *prog,
struct brw_shader *shader,
void *mem_ctx);
@@ -228,8 +230,9 @@ public:
}
struct gl_program *p;
- struct brw_vs_compile *c;
- struct brw_vs_prog_data *prog_data;
+ struct brw_vec4_compile *vec4_compile;
+ const struct brw_vec4_prog_key *vec4_key;
+ struct brw_vec4_prog_data *vec4_prog_data;
unsigned int sanity_param_count;
char *fail_msg;
@@ -251,9 +254,6 @@ public:
int *virtual_grf_use;
dst_reg userplane[MAX_CLIP_PLANES];
- src_reg *vp_temp_regs;
- src_reg vp_addr_reg;
-
/**
* This is the size to be used for an array with an element per
* reg_offset
@@ -507,6 +507,11 @@ private:
void setup_vp_regs();
dst_reg get_vp_dst_reg(const prog_dst_register &dst);
src_reg get_vp_src_reg(const prog_src_register &src);
+
+ struct brw_vs_compile * const c;
+ struct brw_vs_prog_data * const prog_data;
+ src_reg *vp_temp_regs;
+ src_reg vp_addr_reg;
};
/**
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_reg_allocate.cpp
b/src/mesa/drivers/dri/i965/brw_vec4_reg_allocate.cpp
index f9ebc3c..1fe7ab7 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_reg_allocate.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_reg_allocate.cpp
@@ -76,7 +76,7 @@ vec4_visitor::reg_allocate_trivial()
next += this->virtual_grf_sizes[i];
}
}
- prog_data->base.total_grf = next;
+ vec4_prog_data->total_grf = next;
foreach_iter(exec_list_iterator, iter, this->instructions) {
vec4_instruction *inst = (vec4_instruction *)iter.get();
@@ -87,9 +87,9 @@ vec4_visitor::reg_allocate_trivial()
assign(hw_reg_mapping, &inst->src[2]);
}
- if (prog_data->base.total_grf > max_grf) {
+ if (vec4_prog_data->total_grf > max_grf) {
fail("Ran out of regs on trivial allocator (%d/%d)\n",
- prog_data->base.total_grf, max_grf);
+ vec4_prog_data->total_grf, max_grf);
return false;
}
@@ -221,12 +221,12 @@ vec4_visitor::reg_allocate()
* regs in the register classes back down to real hardware reg
* numbers.
*/
- prog_data->base.total_grf = first_assigned_grf;
+ vec4_prog_data->total_grf = first_assigned_grf;
for (int i = 0; i < virtual_grf_count; i++) {
int reg = ra_get_node_reg(g, i);
hw_reg_mapping[i] = first_assigned_grf + brw->vs.ra_reg_to_grf[reg];
- prog_data->base.total_grf = MAX2(prog_data->base.total_grf,
+ vec4_prog_data->total_grf = MAX2(vec4_prog_data->total_grf,
hw_reg_mapping[i] + virtual_grf_sizes[i]);
}
@@ -321,7 +321,7 @@ void
vec4_visitor::spill_reg(int spill_reg_nr)
{
assert(virtual_grf_sizes[spill_reg_nr] == 1);
- unsigned int spill_offset = c->base.last_scratch++;
+ unsigned int spill_offset = vec4_compile->last_scratch++;
/* Generate spill/unspill instructions for the objects being spilled. */
foreach_list(node, &this->instructions) {
diff --git a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
index 79fd6a8..18f17c8 100644
--- a/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
+++ b/src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp
@@ -605,12 +605,12 @@ vec4_visitor::setup_uniform_values(ir_variable *ir)
int i;
for (i = 0; i < uniform_vector_size[uniforms]; i++) {
- prog_data->base.param[uniforms * 4 + i] = &components->f;
+ vec4_prog_data->param[uniforms * 4 + i] = &components->f;
components++;
}
for (; i < 4; i++) {
static float zero = 0;
- prog_data->base.param[uniforms * 4 + i] = &zero;
+ vec4_prog_data->param[uniforms * 4 + i] = &zero;
}
uniforms++;
@@ -632,14 +632,14 @@ vec4_visitor::setup_uniform_clipplane_values()
*/
int compacted_clipplane_index = 0;
for (int i = 0; i < MAX_CLIP_PLANES; ++i) {
- if (!(c->key.base.userclip_planes_enabled_gen_4_5 & (1 << i)))
+ if (!(vec4_key->userclip_planes_enabled_gen_4_5 & (1 << i)))
continue;
this->uniform_vector_size[this->uniforms] = 4;
this->userplane[compacted_clipplane_index] = dst_reg(UNIFORM,
this->uniforms);
this->userplane[compacted_clipplane_index].type = BRW_REGISTER_TYPE_F;
for (int j = 0; j < 4; ++j) {
- prog_data->base.param[this->uniforms * 4 + j] = &clip_planes[i][j];
+ vec4_prog_data->param[this->uniforms * 4 + j] = &clip_planes[i][j];
}
++compacted_clipplane_index;
++this->uniforms;
@@ -648,12 +648,12 @@ vec4_visitor::setup_uniform_clipplane_values()
/* In Gen6 and later, we don't compact clip planes, because this
* simplifies the implementation of gl_ClipDistance.
*/
- for (int i = 0; i < c->key.base.nr_userclip_plane_consts; ++i) {
+ for (int i = 0; i < vec4_key->nr_userclip_plane_consts; ++i) {
this->uniform_vector_size[this->uniforms] = 4;
this->userplane[i] = dst_reg(UNIFORM, this->uniforms);
this->userplane[i].type = BRW_REGISTER_TYPE_F;
for (int j = 0; j < 4; ++j) {
- prog_data->base.param[this->uniforms * 4 + j] = &clip_planes[i][j];
+ vec4_prog_data->param[this->uniforms * 4 + j] = &clip_planes[i][j];
}
++this->uniforms;
}
@@ -689,7 +689,7 @@ vec4_visitor::setup_builtin_uniform_values(ir_variable *ir)
int swiz = GET_SWZ(slots[i].swizzle, j);
last_swiz = swiz;
- prog_data->base.param[this->uniforms * 4 + j] = &values[swiz];
+ vec4_prog_data->param[this->uniforms * 4 + j] = &values[swiz];
if (swiz <= last_swiz)
this->uniform_vector_size[this->uniforms]++;
}
@@ -2303,7 +2303,7 @@ vec4_visitor::visit(ir_texture *ir)
void
vec4_visitor::swizzle_result(ir_texture *ir, src_reg orig_val, int sampler)
{
- int s = c->key.base.tex.swizzles[sampler];
+ int s = vec4_key->tex.swizzles[sampler];
this->result = src_reg(this, ir->type);
dst_reg swizzled_result(this->result);
@@ -2417,8 +2417,8 @@ void
vec4_visitor::emit_psiz_and_flags(struct brw_reg reg)
{
if (intel->gen < 6 &&
- ((prog_data->base.vue_map.slots_valid & VARYING_BIT_PSIZ) ||
- c->key.base.userclip_active || brw->has_negative_rhw_bug)) {
+ ((vec4_prog_data->vue_map.slots_valid & VARYING_BIT_PSIZ) ||
+ vec4_key->userclip_active || brw->has_negative_rhw_bug)) {
dst_reg header1 = dst_reg(this, glsl_type::uvec4_type);
dst_reg header1_w = header1;
header1_w.writemask = WRITEMASK_W;
@@ -2426,7 +2426,7 @@ vec4_visitor::emit_psiz_and_flags(struct brw_reg reg)
emit(MOV(header1, 0u));
- if (prog_data->base.vue_map.slots_valid & VARYING_BIT_PSIZ) {
+ if (vec4_prog_data->vue_map.slots_valid & VARYING_BIT_PSIZ) {
src_reg psiz = src_reg(output_reg[VARYING_SLOT_PSIZ]);
current_annotation = "Point size";
@@ -2435,7 +2435,7 @@ vec4_visitor::emit_psiz_and_flags(struct brw_reg reg)
}
current_annotation = "Clipping flags";
- for (i = 0; i < c->key.base.nr_userclip_plane_consts; i++) {
+ for (i = 0; i < vec4_key->nr_userclip_plane_consts; i++) {
vec4_instruction *inst;
inst = emit(DP4(dst_null_f(), src_reg(output_reg[VARYING_SLOT_POS]),
@@ -2471,7 +2471,7 @@ vec4_visitor::emit_psiz_and_flags(struct brw_reg reg)
emit(MOV(retype(reg, BRW_REGISTER_TYPE_UD), 0u));
} else {
emit(MOV(retype(reg, BRW_REGISTER_TYPE_D), src_reg(0)));
- if (prog_data->base.vue_map.slots_valid & VARYING_BIT_PSIZ) {
+ if (vec4_prog_data->vue_map.slots_valid & VARYING_BIT_PSIZ) {
emit(MOV(brw_writemask(reg, WRITEMASK_W),
src_reg(output_reg[VARYING_SLOT_PSIZ])));
}
@@ -2502,11 +2502,11 @@ vec4_visitor::emit_clip_distances(struct brw_reg reg,
int offset)
* if the user wrote to it; otherwise we use gl_Position.
*/
gl_varying_slot clip_vertex = VARYING_SLOT_CLIP_VERTEX;
- if (!(prog_data->base.vue_map.slots_valid & VARYING_BIT_CLIP_VERTEX)) {
+ if (!(vec4_prog_data->vue_map.slots_valid & VARYING_BIT_CLIP_VERTEX)) {
clip_vertex = VARYING_SLOT_POS;
}
- for (int i = 0; i + offset < c->key.base.nr_userclip_plane_consts && i < 4;
+ for (int i = 0; i + offset < vec4_key->nr_userclip_plane_consts && i < 4;
++i) {
emit(DP4(dst_reg(brw_writemask(reg, 1 << i)),
src_reg(output_reg[clip_vertex]),
@@ -2527,7 +2527,7 @@ vec4_visitor::emit_generic_urb_slot(dst_reg reg, int
varying)
varying == VARYING_SLOT_COL1 ||
varying == VARYING_SLOT_BFC0 ||
varying == VARYING_SLOT_BFC1) &&
- c->key.base.clamp_vertex_color) {
+ vec4_key->clamp_vertex_color) {
inst->saturate = true;
}
}
@@ -2556,7 +2556,7 @@ vec4_visitor::emit_urb_slot(int mrf, int varying)
break;
case VARYING_SLOT_CLIP_DIST0:
case VARYING_SLOT_CLIP_DIST1:
- if (this->c->key.base.uses_clip_distance) {
+ if (this->vec4_key->uses_clip_distance) {
emit_generic_urb_slot(reg, varying);
} else {
current_annotation = "user clip distances";
@@ -2828,8 +2828,8 @@ vec4_visitor::move_grf_array_access_to_scratch()
if (inst->dst.file == GRF && inst->dst.reladdr &&
scratch_loc[inst->dst.reg] == -1) {
- scratch_loc[inst->dst.reg] = c->base.last_scratch;
- c->base.last_scratch += this->virtual_grf_sizes[inst->dst.reg];
+ scratch_loc[inst->dst.reg] = vec4_compile->last_scratch;
+ vec4_compile->last_scratch += this->virtual_grf_sizes[inst->dst.reg];
}
for (int i = 0 ; i < 3; i++) {
@@ -2837,8 +2837,8 @@ vec4_visitor::move_grf_array_access_to_scratch()
if (src->file == GRF && src->reladdr &&
scratch_loc[src->reg] == -1) {
- scratch_loc[src->reg] = c->base.last_scratch;
- c->base.last_scratch += this->virtual_grf_sizes[src->reg];
+ scratch_loc[src->reg] = vec4_compile->last_scratch;
+ vec4_compile->last_scratch += this->virtual_grf_sizes[src->reg];
}
}
}
@@ -2937,12 +2937,12 @@
vec4_visitor::move_uniform_array_access_to_pull_constants()
* add it.
*/
if (pull_constant_loc[uniform] == -1) {
- const float **values = &prog_data->base.param[uniform * 4];
+ const float **values = &vec4_prog_data->param[uniform * 4];
- pull_constant_loc[uniform] = prog_data->base.nr_pull_params / 4;
+ pull_constant_loc[uniform] = vec4_prog_data->nr_pull_params / 4;
for (int j = 0; j < uniform_size[uniform] * 4; j++) {
- prog_data->base.pull_param[prog_data->base.nr_pull_params++]
+ vec4_prog_data->pull_param[vec4_prog_data->nr_pull_params++]
= values[j];
}
}
@@ -2984,13 +2984,14 @@ vec4_visitor::resolve_ud_negate(src_reg *reg)
}
vec4_visitor::vec4_visitor(struct brw_context *brw,
- struct brw_vs_compile *c,
- struct brw_vs_prog_data *prog_data,
+ struct brw_vec4_compile *vec4_compile,
+ struct gl_program *p,
+ const struct brw_vec4_prog_key *vec4_key,
+ struct brw_vec4_prog_data *vec4_prog_data,
struct gl_shader_program *prog,
struct brw_shader *shader,
void *mem_ctx)
{
- this->c = c;
this->brw = brw;
this->intel = &brw->intel;
this->ctx = &intel->ctx;
@@ -3004,9 +3005,10 @@ vec4_visitor::vec4_visitor(struct brw_context *brw,
this->current_annotation = NULL;
memset(this->output_reg_annotation, 0, sizeof(this->output_reg_annotation));
- this->c = c;
- this->p = &c->vp->program.Base;
- this->prog_data = prog_data;
+ this->vec4_compile = vec4_compile;
+ this->p = p;
+ this->vec4_key = vec4_key;
+ this->vec4_prog_data = vec4_prog_data;
this->variable_ht = hash_table_ctor(0,
hash_table_pointer_hash,
@@ -3038,7 +3040,10 @@ vec4_vs_visitor::vec4_vs_visitor(struct brw_context *brw,
struct gl_shader_program *prog,
struct brw_shader *shader,
void *mem_ctx)
- : vec4_visitor(brw, c, prog_data, prog, shader, mem_ctx)
+ : vec4_visitor(brw, &c->base, &c->vp->program.Base, &c->key.base,
+ &prog_data->base, prog, shader, mem_ctx),
+ c(c),
+ prog_data(prog_data)
{
}
diff --git a/src/mesa/drivers/dri/i965/test_vec4_register_coalesce.cpp
b/src/mesa/drivers/dri/i965/test_vec4_register_coalesce.cpp
index 8408e9f..6d2bbf8 100644
--- a/src/mesa/drivers/dri/i965/test_vec4_register_coalesce.cpp
+++ b/src/mesa/drivers/dri/i965/test_vec4_register_coalesce.cpp
@@ -38,7 +38,7 @@ public:
struct intel_context *intel;
struct gl_context *ctx;
struct gl_shader_program *shader_prog;
- struct brw_vs_compile *c;
+ struct brw_vertex_program *vp;
vec4_visitor *v;
};
@@ -47,9 +47,8 @@ class register_coalesce_vec4_visitor : public vec4_visitor
{
public:
register_coalesce_vec4_visitor(struct brw_context *brw,
- struct brw_vs_compile *c,
struct gl_shader_program *prog)
- : vec4_visitor(brw, c, NULL, prog, NULL, NULL)
+ : vec4_visitor(brw, NULL, NULL, NULL, NULL, prog, NULL, NULL)
{
}
@@ -89,14 +88,13 @@ void register_coalesce_test::SetUp()
intel = &brw->intel;
ctx = &intel->ctx;
- c = ralloc(NULL, struct brw_vs_compile);
- c->vp = ralloc(NULL, struct brw_vertex_program);
+ vp = ralloc(NULL, struct brw_vertex_program);
shader_prog = ralloc(NULL, struct gl_shader_program);
- v = new register_coalesce_vec4_visitor(brw, c, shader_prog);
+ v = new register_coalesce_vec4_visitor(brw, shader_prog);
- _mesa_init_vertex_program(ctx, &c->vp->program, GL_VERTEX_SHADER, 0);
+ _mesa_init_vertex_program(ctx, &vp->program, GL_VERTEX_SHADER, 0);
intel->gen = 4;
}
--
1.8.2
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
