intel/compiler: add a new pass to lower shading rate into HW format

Rework:
 * Jason: Modernize brw_nir_lower_shading_rate_output:
   1. Use nir_shader_instructions_pass()
   2. Use *_imm builder helpers.
   3. Use nir_intrinsic_base() instead of ->const_index[0]

v2: Also lower loads (Caio)

v3: Update stage check to trigger lowering (Caio)

v4: Assert on != MESH (Caio)

v5: Fixup instruction insertion (Caio)

Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Caio Oliveira <caio.oliveira@intel.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13739>
This commit is contained in:
Lionel Landwerlin
2021-02-17 13:47:36 +02:00
committed by Marge Bot
parent e227bb9fd5
commit cebf284ac1
5 changed files with 130 additions and 0 deletions

View File

@@ -98,6 +98,19 @@ gl_shader_stage_is_callable(gl_shader_stage stage)
stage == MESA_SHADER_CALLABLE;
}
static inline bool
gl_shader_stage_can_set_fragment_shading_rate(gl_shader_stage stage)
{
/* According to EXT_fragment_shading_rate :
*
* "This extension adds support for setting the fragment shading rate
* for a primitive in vertex, geometry, and mesh shading stages"
*/
return stage == MESA_SHADER_VERTEX ||
stage == MESA_SHADER_GEOMETRY ||
stage == MESA_SHADER_MESH;
}
/**
* Number of STATE_* values we need to address any GL state.
* Used to dimension arrays.

View File

@@ -1110,6 +1110,9 @@ brw_postprocess_nir(nir_shader *nir, const struct brw_compiler *compiler,
OPT(nir_lower_idiv, &options);
}
if (gl_shader_stage_can_set_fragment_shading_rate(nir->info.stage))
brw_nir_lower_shading_rate_output(nir);
brw_nir_optimize(nir, compiler, is_scalar, false);
if (is_scalar && nir_shader_has_local_variables(nir)) {

View File

@@ -120,6 +120,8 @@ bool brw_nir_lower_conversions(nir_shader *nir);
bool brw_nir_lower_scoped_barriers(nir_shader *nir);
bool brw_nir_lower_shading_rate_output(nir_shader *nir);
bool brw_nir_lower_storage_image(nir_shader *nir,
const struct intel_device_info *devinfo);

View File

@@ -0,0 +1,111 @@
/*
* Copyright (c) 2021 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.
*/
/*
* Lower the shading rate output from the bit field format described in the
* SPIRV spec :
*
* bit | name | description
* 0 | Vertical2Pixels | Fragment invocation covers 2 pixels vertically
* 1 | Vertical4Pixels | Fragment invocation covers 4 pixels vertically
* 2 | Horizontal2Pixels | Fragment invocation covers 2 pixels horizontally
* 3 | Horizontal4Pixels | Fragment invocation covers 4 pixels horizontally
*
* into a single dword composed of 2 fp16 to be stored in the dword 0 of the
* VUE header.
*
* When no horizontal/vertical bits are set, the size in pixel size in that
* dimension is assumed to be 1.
*
* According to the specification, the shading rate output can be read &
* written. A read after a write should report a different value if the
* implemention decides on different primitive shading rate for some reason.
* This is never the case in our implementation.
*/
#include "brw_nir.h"
#include "compiler/nir/nir_builder.h"
static bool
lower_shading_rate_output_instr(nir_builder *b, nir_instr *instr,
UNUSED void *_state)
{
if (instr->type != nir_instr_type_intrinsic)
return false;
nir_intrinsic_instr *intrin = nir_instr_as_intrinsic(instr);
if (intrin->intrinsic != nir_intrinsic_load_output &&
intrin->intrinsic != nir_intrinsic_store_output)
return false;
if (nir_intrinsic_base(intrin) != VARYING_SLOT_PRIMITIVE_SHADING_RATE)
return false;
b->cursor = intrin->intrinsic == nir_intrinsic_load_output ?
nir_after_instr(instr) : nir_before_instr(instr);
if (intrin->intrinsic == nir_intrinsic_store_output) {
assert(intrin->src[0].is_ssa);
nir_ssa_def *bit_field = intrin->src[0].ssa;
nir_ssa_def *fp16_x =
nir_i2f16(b,
nir_ishl(b, nir_imm_int(b, 1),
nir_ishr_imm(b, bit_field, 2)));
nir_ssa_def *fp16_y =
nir_i2f16(b,
nir_ishl(b, nir_imm_int(b, 1),
nir_iand_imm(b, bit_field, 0x3)));
nir_ssa_def *packed_fp16_xy = nir_pack_32_2x16_split(b, fp16_x, fp16_y);
nir_instr_rewrite_src(instr, &intrin->src[0],
nir_src_for_ssa(packed_fp16_xy));
} else {
assert(intrin->intrinsic == nir_intrinsic_load_output);
nir_ssa_def *packed_fp16_xy = &intrin->dest.ssa;
nir_ssa_def *u32_x =
nir_i2i32(b, nir_unpack_32_2x16_split_x(b, packed_fp16_xy));
nir_ssa_def *u32_y =
nir_i2i32(b, nir_unpack_32_2x16_split_y(b, packed_fp16_xy));
nir_ssa_def *bit_field =
nir_ior(b, nir_ishl_imm(b, nir_ushr_imm(b, u32_x, 1), 2),
nir_ushr_imm(b, u32_y, 1));
nir_ssa_def_rewrite_uses_after(packed_fp16_xy, bit_field,
bit_field->parent_instr);
}
return true;
}
bool
brw_nir_lower_shading_rate_output(nir_shader *nir)
{
/* TODO(mesh): Add Shading Rate support. */
assert(nir->info.stage != MESA_SHADER_MESH);
return nir_shader_instructions_pass(nir, lower_shading_rate_output_instr,
nir_metadata_block_index |
nir_metadata_dominance, NULL);
}

View File

@@ -90,6 +90,7 @@ libintel_compiler_files = files(
'brw_nir_lower_rt_intrinsics.c',
'brw_nir_lower_scoped_barriers.c',
'brw_nir_lower_shader_calls.c',
'brw_nir_lower_shading_rate_output.c',
'brw_nir_lower_storage_image.c',
'brw_nir_opt_peephole_ffma.c',
'brw_nir_rt.h',