nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
/*
|
|
|
|
|
* Copyright © 2020 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 "gtest/gtest.h"
|
|
|
|
|
#include "nir.h"
|
|
|
|
|
#include "nir_builder.h"
|
|
|
|
|
#include "nir_phi_builder.h"
|
|
|
|
|
|
|
|
|
|
#define UNROLL_TEST_INSERT(_label, _type, _init, _limit, _step, \
|
|
|
|
|
_cond, _incr, _rev, _exp_res, \
|
|
|
|
|
_exp_instr_count, _exp_loop_count) \
|
|
|
|
|
TEST_F(nir_loop_unroll_test, _label) \
|
|
|
|
|
{ \
|
2023-08-12 16:17:15 -04:00
|
|
|
nir_def *init = nir_imm_##_type(&bld, _init); \
|
|
|
|
|
nir_def *limit = nir_imm_##_type(&bld, _limit); \
|
|
|
|
|
nir_def *step = nir_imm_##_type(&bld, _step); \
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
loop_unroll_test_helper(&bld, init, limit, step, \
|
|
|
|
|
&nir_##_cond, &nir_##_incr, _rev); \
|
|
|
|
|
EXPECT_##_exp_res(nir_opt_loop_unroll(bld.shader)); \
|
|
|
|
|
EXPECT_EQ(_exp_instr_count, count_instr(nir_op_##_incr)); \
|
|
|
|
|
EXPECT_EQ(_exp_loop_count, count_loops()); \
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
namespace {
|
|
|
|
|
|
|
|
|
|
class nir_loop_unroll_test : public ::testing::Test {
|
|
|
|
|
protected:
|
|
|
|
|
nir_loop_unroll_test()
|
|
|
|
|
{
|
|
|
|
|
glsl_type_singleton_init_or_ref();
|
|
|
|
|
static nir_shader_compiler_options options = { };
|
|
|
|
|
options.max_unroll_iterations = 32;
|
|
|
|
|
options.force_indirect_unrolling_sampler = false;
|
|
|
|
|
options.force_indirect_unrolling = nir_var_all;
|
|
|
|
|
bld = nir_builder_init_simple_shader(MESA_SHADER_VERTEX, &options,
|
|
|
|
|
"loop unrolling tests");
|
|
|
|
|
}
|
|
|
|
|
~nir_loop_unroll_test()
|
|
|
|
|
{
|
|
|
|
|
ralloc_free(bld.shader);
|
|
|
|
|
glsl_type_singleton_decref();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
int count_instr(nir_op op);
|
|
|
|
|
int count_loops(void);
|
|
|
|
|
|
|
|
|
|
nir_builder bld;
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
} /* namespace */
|
|
|
|
|
|
|
|
|
|
int
|
|
|
|
|
nir_loop_unroll_test::count_instr(nir_op op)
|
|
|
|
|
{
|
|
|
|
|
int count = 0;
|
|
|
|
|
nir_foreach_block(block, bld.impl) {
|
|
|
|
|
nir_foreach_instr(instr, block) {
|
|
|
|
|
if (instr->type != nir_instr_type_alu)
|
|
|
|
|
continue;
|
|
|
|
|
nir_alu_instr *alu_instr = nir_instr_as_alu(instr);
|
|
|
|
|
if (alu_instr->op == op)
|
|
|
|
|
count++;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return count;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
int
|
|
|
|
|
nir_loop_unroll_test::count_loops(void)
|
|
|
|
|
{
|
|
|
|
|
int count = 0;
|
|
|
|
|
foreach_list_typed(nir_cf_node, cf_node, node, &bld.impl->body) {
|
|
|
|
|
if (cf_node->type == nir_cf_node_loop)
|
|
|
|
|
count++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
return count;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
void
|
2023-08-12 16:17:15 -04:00
|
|
|
loop_unroll_test_helper(nir_builder *bld, nir_def *init,
|
|
|
|
|
nir_def *limit, nir_def *step,
|
|
|
|
|
nir_def* (*cond_instr)(nir_builder*,
|
|
|
|
|
nir_def*,
|
|
|
|
|
nir_def*),
|
|
|
|
|
nir_def* (*incr_instr)(nir_builder*,
|
|
|
|
|
nir_def*,
|
|
|
|
|
nir_def*),
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
bool reverse)
|
|
|
|
|
{
|
|
|
|
|
nir_loop *loop = nir_push_loop(bld);
|
|
|
|
|
|
|
|
|
|
nir_block *top_block =
|
|
|
|
|
nir_cf_node_as_block(nir_cf_node_prev(&loop->cf_node));
|
|
|
|
|
nir_block *head_block = nir_loop_first_block(loop);
|
|
|
|
|
|
|
|
|
|
nir_phi_instr *phi = nir_phi_instr_create(bld->shader);
|
2023-08-14 11:56:00 -05:00
|
|
|
nir_def_init(&phi->instr, &phi->def, 1, 32);
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
|
2023-08-17 15:26:32 -05:00
|
|
|
nir_phi_instr_add_src(phi, top_block, init);
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
|
2023-08-12 16:17:15 -04:00
|
|
|
nir_def *cond = cond_instr(bld,
|
2023-08-14 11:56:00 -05:00
|
|
|
(reverse ? limit : &phi->def),
|
|
|
|
|
(reverse ? &phi->def : limit));
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
|
|
|
|
|
nir_if *nif = nir_push_if(bld, cond);
|
|
|
|
|
nir_jump(bld, nir_jump_break);
|
|
|
|
|
nir_pop_if(bld, nif);
|
|
|
|
|
|
2023-08-14 11:56:00 -05:00
|
|
|
nir_def *var = incr_instr(bld, &phi->def, step);
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
|
2023-08-17 15:26:32 -05:00
|
|
|
nir_phi_instr_add_src(phi, nir_cursor_current_block(bld->cursor), var);
|
nir/loop_analyze: Determine iteration counts for more kinds of loops
If loop iterator is incremented with something other than regular
addition, it would be more error prone to calculate the number of
iterations theoretically. What we can do instead, is try to emulate the
loop, and determine the number of iterations empirically.
These operations are covered:
- imul
- fmul
- ishl
- ishr
- ushr
Also add unit tests for loop unrollment.
Improves performance of Aztec Ruins (sixonix
gfxbench5.aztec_ruins_vk_high) by -1.28042% +/- 0.498555% (N=5) on Intel
Arc A770.
v2 (idr): Rebase on 3 years. :( Use nir_phi_instr_add_src in the test
cases.
v3 (idr): Use try_eval_const_alu in to evaluate loop termination
condition in get_iteration_empirical. Also restructure the loop
slightly. This fixed off by one iteration errors in "inverted" loop
tests (e.g., nir_loop_analyze_test.ushr_ieq_known_count_invert_31).
v4 (idr): Use try_eval_const_alu in to evaluate induction variable
update in get_iteration_empirical. This fixes non-commutative update
operations (e.g., shifts) when the induction varible is not the first
source. This fixes the unit test
nir_loop_analyze_test.ishl_rev_ieq_infinite_loop_unknown_count.
v5 (idr): Fix _type parameter for fadd and fadd_rev loop unroll
tests. Hopefully that fixes the failure on s390x. Temporarily disable
fmul. This works-around the revealed problem in
glsl-fs-loop-unroll-mul-fp64, and there were no shader-db or fossil-db
changes.
v6 (idr): Plumb max_unroll_iterations into get_iteration_empirical. I
was going to do this, but I forgot. Suggested by Tim.
v7 (idr): Disable fadd tests on s390x. They fail because S390 is weird.
Almost all of the shaders affected (OpenGL or Vulkan) are from gfxbench
or geekbench. A couple shaders in Deus Ex (OpenGL), Dirt Rally (OpenGL),
Octopath Traveler (Vulkan), and Rise of the Tomb Raider (Vulkan) are
helped.
The lost / gained shaders in OpenGL are an Aztec Ruins shader that goes
from SIMD16 to SIMD8. The spills / fills affected are in a single Aztec
Ruins (Vulkan) compute shader.
shader-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
total loops in shared programs: 5514 -> 5470 (-0.80%)
loops in affected programs: 62 -> 18 (-70.97%)
helped: 37 / HURT: 0
LOST: 2
GAINED: 2
Haswell and Broadwell had similar results. (Broadwell shown)
total loops in shared programs: 5346 -> 5298 (-0.90%)
loops in affected programs: 66 -> 18 (-72.73%)
helped: 39 / HURT: 0
fossil-db results:
Skylake, Ice Lake, and Tiger Lake had similar results. (Tiger Lake shown)
Instructions in all programs: 157374679 -> 157397421 (+0.0%)
Instructions hurt: 28
SENDs in all programs: 7463800 -> 7467639 (+0.1%)
SENDs hurt: 28
Loops in all programs: 38980 -> 38950 (-0.1%)
Loops helped: 28
Cycles in all programs: 7559486451 -> 7557455384 (-0.0%)
Cycles helped: 28
Spills in all programs: 11405 -> 11403 (-0.0%)
Spills helped: 1
Fills in all programs: 19578 -> 19588 (+0.1%)
Fills hurt: 1
Lost: 1
Signed-off-by: Yevhenii Kolesnikov <yevhenii.kolesnikov@globallogic.com>
Reviewed-by: Timothy Arceri <tarceri@itsqueeze.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/3445>
2020-01-17 13:01:01 +02:00
|
|
|
|
|
|
|
|
nir_pop_loop(bld, loop);
|
|
|
|
|
|
|
|
|
|
bld->cursor = nir_after_phis(head_block);
|
|
|
|
|
nir_builder_instr_insert(bld, &phi->instr);
|
|
|
|
|
|
|
|
|
|
nir_validate_shader(bld->shader, NULL);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
UNROLL_TEST_INSERT(iadd, int, 0, 24, 4,
|
|
|
|
|
ige, iadd, false, TRUE, 6, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(iadd_rev, int, 0, 24, 4,
|
|
|
|
|
ilt, iadd, true, TRUE, 7, 0)
|
|
|
|
|
#ifndef __s390x__
|
|
|
|
|
UNROLL_TEST_INSERT(fadd, float, 0.0, 24.0, 4.0,
|
|
|
|
|
fge, fadd, false, TRUE, 6, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(fadd_rev, float, 0.0, 24.0, 4.0,
|
|
|
|
|
flt, fadd, true, TRUE, 7, 0)
|
|
|
|
|
#endif
|
|
|
|
|
UNROLL_TEST_INSERT(imul, int, 1, 81, 3,
|
|
|
|
|
ige, imul, false, TRUE, 4, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(imul_rev, int, 1, 81, 3,
|
|
|
|
|
ilt, imul, true, TRUE, 5, 0)
|
|
|
|
|
#if 0 /* Disable tests until support is re-enabled in loop_analyze. */
|
|
|
|
|
UNROLL_TEST_INSERT(fmul, float, 1.5, 81.0, 3.0,
|
|
|
|
|
fge, fmul, false, TRUE, 4, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(fmul_rev, float, 1.0, 81.0, 3.0,
|
|
|
|
|
flt, fmul, true, TRUE, 5, 0)
|
|
|
|
|
#endif
|
|
|
|
|
UNROLL_TEST_INSERT(ishl, int, 1, 128, 1,
|
|
|
|
|
ige, ishl, false, TRUE, 7, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(ishl_rev, int, 1, 128, 1,
|
|
|
|
|
ilt, ishl, true, TRUE, 8, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(ishr, int, 64, 4, 1,
|
|
|
|
|
ilt, ishr, false, TRUE, 5, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(ishr_rev, int, 64, 4, 1,
|
|
|
|
|
ige, ishr, true, TRUE, 4, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(ushr, int, 64, 4, 1,
|
|
|
|
|
ilt, ushr, false, TRUE, 5, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(ushr_rev, int, 64, 4, 1,
|
|
|
|
|
ige, ushr, true, TRUE, 4, 0)
|
|
|
|
|
|
|
|
|
|
UNROLL_TEST_INSERT(lshl_neg, int, 0xf0f0f0f0, 0, 1,
|
|
|
|
|
ige, ishl, false, TRUE, 4, 0)
|
|
|
|
|
UNROLL_TEST_INSERT(lshl_neg_rev, int, 0xf0f0f0f0, 0, 1,
|
|
|
|
|
ilt, ishl, true, TRUE, 4, 0)
|