mesa/src/intel/compiler/test_opt_copy_propagation.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

116 lines
3.2 KiB
C++
Raw Normal View History

/*
* Copyright © 2016 Intel Corporation
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
* SPDX-License-Identifier: MIT
*/
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
#include "test_helpers.h"
#include "brw_builder.h"
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
class copy_propagation_test : public brw_shader_pass_test {};
TEST_F(copy_propagation_test, basic)
{
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_builder bld = make_shader();
brw_builder exp = make_shader();
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_reg v0 = vgrf(bld, exp, BRW_TYPE_F);
brw_reg v1 = vgrf(bld, exp, BRW_TYPE_F);
brw_reg v2 = vgrf(bld, exp, BRW_TYPE_F);
brw_reg v3 = vgrf(bld, exp, BRW_TYPE_F);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
bld.MOV(v0, v2);
bld.ADD(v1, v0, v3);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
/* Propagation makes ADD use v2 as source. */
EXPECT_PROGRESS(brw_opt_copy_propagation, bld);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
exp.MOV(v0, v2);
exp.ADD(v1, v2, v3);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
EXPECT_SHADERS_MATCH(bld, exp);
}
TEST_F(copy_propagation_test, maxmax_sat_imm)
{
static const struct {
enum brw_conditional_mod conditional_mod;
float immediate;
bool expected_result;
} test[] = {
/* conditional mod, imm, expected_result */
{ BRW_CONDITIONAL_GE , 0.1f, false },
{ BRW_CONDITIONAL_L , 0.1f, false },
{ BRW_CONDITIONAL_GE , 0.5f, false },
{ BRW_CONDITIONAL_L , 0.5f, false },
{ BRW_CONDITIONAL_GE , 0.9f, false },
{ BRW_CONDITIONAL_L , 0.9f, false },
{ BRW_CONDITIONAL_GE , -1.5f, false },
{ BRW_CONDITIONAL_L , -1.5f, false },
{ BRW_CONDITIONAL_GE , 1.5f, false },
{ BRW_CONDITIONAL_L , 1.5f, false },
{ BRW_CONDITIONAL_NONE, 0.5f, false },
{ BRW_CONDITIONAL_Z , 0.5f, false },
{ BRW_CONDITIONAL_NZ , 0.5f, false },
{ BRW_CONDITIONAL_G , 0.5f, false },
{ BRW_CONDITIONAL_LE , 0.5f, false },
{ BRW_CONDITIONAL_R , 0.5f, false },
{ BRW_CONDITIONAL_O , 0.5f, false },
{ BRW_CONDITIONAL_U , 0.5f, false },
};
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
for (unsigned i = 0; i < ARRAY_SIZE(test); i++) {
brw_builder bld = make_shader();
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
enum brw_conditional_mod cmod = test[i].conditional_mod;
brw_reg imm = brw_imm_f(test[i].immediate);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_reg v0 = bld.vgrf(BRW_TYPE_F);
brw_reg v1 = bld.vgrf(BRW_TYPE_F);
brw_reg v2 = bld.vgrf(BRW_TYPE_F);
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
bld.MOV(v0, v1)->saturate = true;
bld.SEL(v2, v0, imm)->conditional_mod = cmod;
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
EXPECT_PROGRESS_RESULT(test[i].expected_result,
brw_opt_copy_propagation, bld);
}
}
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
TEST_F(copy_propagation_test, mixed_integer_sign)
{
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_builder bld = make_shader();
brw_builder exp = make_shader();
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_reg v0 = vgrf(bld, exp, BRW_TYPE_UD);
brw_reg v1 = vgrf(bld, exp, BRW_TYPE_D);
brw_reg v2 = vgrf(bld, exp, BRW_TYPE_UD);
brw_reg v3 = vgrf(bld, exp, BRW_TYPE_UD);
brw_reg v4 = vgrf(bld, exp, BRW_TYPE_UD);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
bld.MOV(v1, v0);
bld.BFE(v2, v3, v4, retype(v1, BRW_TYPE_UD));
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
/* Propagation makes BFE use v0 as source. */
EXPECT_PROGRESS(brw_opt_copy_propagation, bld);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
exp.MOV(v1, v0);
exp.BFE(v2, v3, v4, v0);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
EXPECT_SHADERS_MATCH(bld, exp);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
}
TEST_F(copy_propagation_test, mixed_integer_sign_with_vector_imm)
{
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_builder bld = make_shader();
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
brw_reg v0 = bld.vgrf(BRW_TYPE_W);
brw_reg v1 = bld.vgrf(BRW_TYPE_UD);
brw_reg v2 = bld.vgrf(BRW_TYPE_UD);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
bld.MOV(v0, brw_imm_uv(0xffff));
bld.ADD(v1, v2, retype(v0, BRW_TYPE_UW));
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
brw: Simplify the test code for brw passes The key change is to use a builder to write the expected shader result and compare that. To make this less error prone, a few helper functions were added - a way to allocate VGRFs from both shaders in parallel, that way the same brw_reg can be used in both of them; - assertions that a pass will make progress or not, and proper output when the unexpected happens; - use a common brw_shader_pass_test class so to collect some of the helpers; - make some helpers work directly with builder. The idea is to improve the signal in tests, so that the disasm comments are not necessary anymore. For example ``` TEST_F(saturate_propagation_test, basic) { brw_reg dst1 = bld.vgrf(BRW_TYPE_F); brw_reg src0 = bld.vgrf(BRW_TYPE_F); brw_reg src1 = bld.vgrf(BRW_TYPE_F); brw_reg dst0 = bld.ADD(src0, src1); set_saturate(true, bld.MOV(dst1, dst0)); /* = Before = * * 0: add(16) dst0 src0 src1 * 1: mov.sat(16) dst1 dst0 * * = After = * 0: add.sat(16) dst0 src0 src1 * 1: mov(16) dst1 dst0 */ brw_calculate_cfg(*v); bblock_t *block0 = v->cfg->blocks[0]; EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_TRUE(saturate_propagation(v)); EXPECT_EQ(0, block0->start_ip); EXPECT_EQ(1, block0->end_ip); EXPECT_EQ(BRW_OPCODE_ADD, instruction(block0, 0)->opcode); EXPECT_TRUE(instruction(block0, 0)->saturate); EXPECT_EQ(BRW_OPCODE_MOV, instruction(block0, 1)->opcode); EXPECT_FALSE(instruction(block0, 1)->saturate); } ``` becomes ``` TEST_F(saturate_propagation_test, basic) { brw_builder bld = make_shader(MESA_SHADER_FRAGMENT, 16); brw_builder exp = make_shader(MESA_SHADER_FRAGMENT, 16); brw_reg dst0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg dst1 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src0 = vgrf(bld, exp, BRW_TYPE_F); brw_reg src1 = vgrf(bld, exp, BRW_TYPE_F); bld.ADD(dst0, src0, src1); bld.MOV(dst1, dst0)->saturate = true; EXPECT_PROGRESS(brw_opt_saturate_propagation, bld); exp.ADD(dst0, src0, src1)->saturate = true; exp.MOV(dst1, dst0); EXPECT_SHADERS_MATCH(bld, exp); } ``` Reviewed-by: Ian Romanick <ian.d.romanick@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/33936>
2025-03-05 16:58:32 -08:00
EXPECT_NO_PROGRESS(brw_opt_copy_propagation, bld);
intel/brw: Copy prop from raw integer moves with mismatched types The specific pattern from the unit test was observed in ray tracing trampoline shaders. v2: Refactor the is_raw_move tests out to a utility function. Suggested by Ken. v3: Fix a regression caused by being too picky about source modifiers. This was introduced somewhere between when I did initial shader-db runs an v2. v4: Fix typo in comment. Noticed by Caio. shader-db: All Intel platforms had similar results. (Meteor Lake shown) total instructions in shared programs: 19734086 -> 19733997 (<.01%) instructions in affected programs: 135388 -> 135299 (-0.07%) helped: 76 / HURT: 2 total cycles in shared programs: 916290451 -> 916264968 (<.01%) cycles in affected programs: 41046002 -> 41020519 (-0.06%) helped: 32 / HURT: 29 fossil-db: Meteor Lake, DG2, and Skylake had similar results. (Meteor Lake shown) Totals: Instrs: 151531355 -> 151513669 (-0.01%); split: -0.01%, +0.00% Cycle count: 17209372399 -> 17208178205 (-0.01%); split: -0.01%, +0.00% Max live registers: 32016490 -> 32016493 (+0.00%) Totals from 17361 (2.75% of 630198) affected shaders: Instrs: 2642048 -> 2624362 (-0.67%); split: -0.67%, +0.00% Cycle count: 79803066 -> 78608872 (-1.50%); split: -1.75%, +0.25% Max live registers: 421668 -> 421671 (+0.00%) Tiger Lake and Ice Lake had similar results. (Tiger Lake shown) Totals: Instrs: 149995644 -> 149977326 (-0.01%); split: -0.01%, +0.00% Cycle count: 15567293770 -> 15566524840 (-0.00%); split: -0.02%, +0.01% Spill count: 61241 -> 61238 (-0.00%) Fill count: 107304 -> 107301 (-0.00%) Max live registers: 31993109 -> 31993112 (+0.00%) Totals from 17813 (2.83% of 629912) affected shaders: Instrs: 3738236 -> 3719918 (-0.49%); split: -0.49%, +0.00% Cycle count: 4251157049 -> 4250388119 (-0.02%); split: -0.06%, +0.04% Spill count: 28268 -> 28265 (-0.01%) Fill count: 50377 -> 50374 (-0.01%) Max live registers: 470648 -> 470651 (+0.00%) Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Caio Oliveira <caio.oliveira@intel.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30251>
2024-07-16 16:04:38 -07:00
}