mirror of
https://gitlab.freedesktop.org/mesa/mesa.git
synced 2025-12-24 13:10:10 +01:00
Instead of always lowering everything, we add a threshold such that if the total indirected array size (AoA size) is above that threshold, it won't lower. It's assumed that the driver will sort things out somehow by, for instance, lowering to scratch. Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5909>
226 lines
8 KiB
C
226 lines
8 KiB
C
/*
|
|
* Copyright © 2016 Intel Corporation
|
|
*
|
|
* Permission is hereby granted, free of charge, to any person obtaining a
|
|
* copy of this software and associated documentation files (the "Software"),
|
|
* to deal in the Software without restriction, including without limitation
|
|
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
|
* and/or sell copies of the Software, and to permit persons to whom the
|
|
* Software is furnished to do so, subject to the following conditions:
|
|
*
|
|
* The above copyright notice and this permission notice (including the next
|
|
* paragraph) shall be included in all copies or substantial portions of the
|
|
* Software.
|
|
*
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
|
|
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
|
|
* IN THE SOFTWARE.
|
|
*/
|
|
|
|
#include "nir.h"
|
|
#include "nir_builder.h"
|
|
#include "nir_deref.h"
|
|
|
|
static void
|
|
emit_load_store_deref(nir_builder *b, nir_intrinsic_instr *orig_instr,
|
|
nir_deref_instr *parent,
|
|
nir_deref_instr **deref_arr,
|
|
nir_ssa_def **dest, nir_ssa_def *src);
|
|
|
|
static void
|
|
emit_indirect_load_store_deref(nir_builder *b, nir_intrinsic_instr *orig_instr,
|
|
nir_deref_instr *parent,
|
|
nir_deref_instr **deref_arr,
|
|
int start, int end,
|
|
nir_ssa_def **dest, nir_ssa_def *src)
|
|
{
|
|
assert(start < end);
|
|
if (start == end - 1) {
|
|
emit_load_store_deref(b, orig_instr,
|
|
nir_build_deref_array_imm(b, parent, start),
|
|
deref_arr + 1, dest, src);
|
|
} else {
|
|
int mid = start + (end - start) / 2;
|
|
|
|
nir_ssa_def *then_dest, *else_dest;
|
|
|
|
nir_deref_instr *deref = *deref_arr;
|
|
assert(deref->deref_type == nir_deref_type_array);
|
|
|
|
nir_push_if(b, nir_ilt(b, deref->arr.index.ssa, nir_imm_intN_t(b, mid, parent->dest.ssa.bit_size)));
|
|
emit_indirect_load_store_deref(b, orig_instr, parent, deref_arr,
|
|
start, mid, &then_dest, src);
|
|
nir_push_else(b, NULL);
|
|
emit_indirect_load_store_deref(b, orig_instr, parent, deref_arr,
|
|
mid, end, &else_dest, src);
|
|
nir_pop_if(b, NULL);
|
|
|
|
if (src == NULL)
|
|
*dest = nir_if_phi(b, then_dest, else_dest);
|
|
}
|
|
}
|
|
|
|
static void
|
|
emit_load_store_deref(nir_builder *b, nir_intrinsic_instr *orig_instr,
|
|
nir_deref_instr *parent,
|
|
nir_deref_instr **deref_arr,
|
|
nir_ssa_def **dest, nir_ssa_def *src)
|
|
{
|
|
for (; *deref_arr; deref_arr++) {
|
|
nir_deref_instr *deref = *deref_arr;
|
|
if (deref->deref_type == nir_deref_type_array &&
|
|
!nir_src_is_const(deref->arr.index)) {
|
|
int length = glsl_get_length(parent->type);
|
|
|
|
emit_indirect_load_store_deref(b, orig_instr, parent, deref_arr,
|
|
0, length, dest, src);
|
|
return;
|
|
}
|
|
|
|
parent = nir_build_deref_follower(b, parent, deref);
|
|
}
|
|
|
|
/* We reached the end of the deref chain. Emit the instruction */
|
|
assert(*deref_arr == NULL);
|
|
|
|
if (src == NULL) {
|
|
/* This is a load instruction */
|
|
nir_intrinsic_instr *load =
|
|
nir_intrinsic_instr_create(b->shader, orig_instr->intrinsic);
|
|
load->num_components = orig_instr->num_components;
|
|
|
|
load->src[0] = nir_src_for_ssa(&parent->dest.ssa);
|
|
|
|
/* Copy over any other sources. This is needed for interp_deref_at */
|
|
for (unsigned i = 1;
|
|
i < nir_intrinsic_infos[orig_instr->intrinsic].num_srcs; i++)
|
|
nir_src_copy(&load->src[i], &orig_instr->src[i], load);
|
|
|
|
nir_ssa_dest_init(&load->instr, &load->dest,
|
|
orig_instr->dest.ssa.num_components,
|
|
orig_instr->dest.ssa.bit_size, NULL);
|
|
nir_builder_instr_insert(b, &load->instr);
|
|
*dest = &load->dest.ssa;
|
|
} else {
|
|
assert(orig_instr->intrinsic == nir_intrinsic_store_deref);
|
|
nir_store_deref(b, parent, src, nir_intrinsic_write_mask(orig_instr));
|
|
}
|
|
}
|
|
|
|
static bool
|
|
lower_indirect_derefs_block(nir_block *block, nir_builder *b,
|
|
nir_variable_mode modes,
|
|
uint32_t max_lower_array_len)
|
|
{
|
|
bool progress = false;
|
|
|
|
nir_foreach_instr_safe(instr, block) {
|
|
if (instr->type != nir_instr_type_intrinsic)
|
|
continue;
|
|
|
|
nir_intrinsic_instr *intrin = nir_instr_as_intrinsic(instr);
|
|
if (intrin->intrinsic != nir_intrinsic_load_deref &&
|
|
intrin->intrinsic != nir_intrinsic_interp_deref_at_centroid &&
|
|
intrin->intrinsic != nir_intrinsic_interp_deref_at_sample &&
|
|
intrin->intrinsic != nir_intrinsic_interp_deref_at_offset &&
|
|
intrin->intrinsic != nir_intrinsic_interp_deref_at_vertex &&
|
|
intrin->intrinsic != nir_intrinsic_store_deref)
|
|
continue;
|
|
|
|
nir_deref_instr *deref = nir_src_as_deref(intrin->src[0]);
|
|
|
|
/* Walk the deref chain back to the base and look for indirects */
|
|
uint32_t indirect_array_len = 1;
|
|
bool has_indirect = false;
|
|
nir_deref_instr *base = deref;
|
|
while (base && base->deref_type != nir_deref_type_var) {
|
|
nir_deref_instr *parent = nir_deref_instr_parent(base);
|
|
if (base->deref_type == nir_deref_type_array &&
|
|
!nir_src_is_const(base->arr.index)) {
|
|
indirect_array_len *= glsl_get_length(parent->type);
|
|
has_indirect = true;
|
|
}
|
|
|
|
base = parent;
|
|
}
|
|
|
|
if (!has_indirect || !base || indirect_array_len > max_lower_array_len)
|
|
continue;
|
|
|
|
/* Only lower variables whose mode is in the mask, or compact
|
|
* array variables. (We can't handle indirects on tightly packed
|
|
* scalar arrays, so we need to lower them regardless.)
|
|
*/
|
|
if (!(modes & base->var->data.mode) && !base->var->data.compact)
|
|
continue;
|
|
|
|
b->cursor = nir_instr_remove(&intrin->instr);
|
|
|
|
nir_deref_path path;
|
|
nir_deref_path_init(&path, deref, NULL);
|
|
assert(path.path[0] == base);
|
|
|
|
if (intrin->intrinsic == nir_intrinsic_store_deref) {
|
|
assert(intrin->src[1].is_ssa);
|
|
emit_load_store_deref(b, intrin, base, &path.path[1],
|
|
NULL, intrin->src[1].ssa);
|
|
} else {
|
|
nir_ssa_def *result;
|
|
emit_load_store_deref(b, intrin, base, &path.path[1],
|
|
&result, NULL);
|
|
nir_ssa_def_rewrite_uses(&intrin->dest.ssa, nir_src_for_ssa(result));
|
|
}
|
|
|
|
nir_deref_path_finish(&path);
|
|
|
|
progress = true;
|
|
}
|
|
|
|
return progress;
|
|
}
|
|
|
|
static bool
|
|
lower_indirects_impl(nir_function_impl *impl, nir_variable_mode modes,
|
|
uint32_t max_lower_array_len)
|
|
{
|
|
nir_builder builder;
|
|
nir_builder_init(&builder, impl);
|
|
bool progress = false;
|
|
|
|
nir_foreach_block_safe(block, impl) {
|
|
progress |= lower_indirect_derefs_block(block, &builder, modes,
|
|
max_lower_array_len);
|
|
}
|
|
|
|
if (progress)
|
|
nir_metadata_preserve(impl, nir_metadata_none);
|
|
else
|
|
nir_metadata_preserve(impl, nir_metadata_all);
|
|
|
|
return progress;
|
|
}
|
|
|
|
/** Lowers indirect variable loads/stores to direct loads/stores.
|
|
*
|
|
* The pass works by replacing any indirect load or store with an if-ladder
|
|
* that does a binary search on the array index.
|
|
*/
|
|
bool
|
|
nir_lower_indirect_derefs(nir_shader *shader, nir_variable_mode modes,
|
|
uint32_t max_lower_array_len)
|
|
{
|
|
bool progress = false;
|
|
|
|
nir_foreach_function(function, shader) {
|
|
if (function->impl) {
|
|
progress = lower_indirects_impl(function->impl, modes,
|
|
max_lower_array_len) || progress;
|
|
}
|
|
}
|
|
|
|
return progress;
|
|
}
|