diff options
author | SiCong Li <sicong.li@arm.com> | 2023-02-22 17:24:09 +0000 |
---|---|---|
committer | SiCong Li <sicong.li@arm.com> | 2023-03-06 16:19:11 +0000 |
commit | 47f177e679874dc901888973c5fc237b756b38cb (patch) | |
tree | 130386717101d0c2440111cb288faa21df8ab151 /src/runtime/CL/tuners | |
parent | adfcacc8e39888a9a62e33c178041642d0a3047a (diff) | |
download | ComputeLibrary-47f177e679874dc901888973c5fc237b756b38cb.tar.gz |
Fix LWS search space used by CLTuner
* Ensure CLTuner uses the real GWS used by run(), instead of the
static GWS (which is usually changed at run time), by caching GWS in
each kernel
Note this is a somewhat inelegant workaround. The real issue stems
from the fact that execution window and scheduler are very much
coupled with our operator run() / run_op() method.
(Please see COMPMID-5934)
* Restrict LWS values to explore within GWS bound for exhaustive mode
* Refactor gws_from_window() to include all the information required
to calculate GWS
* Log lws search space used for tuning
* Fix ClDirectConv2dKernel config id
Resolves COMPMID-5892
Signed-off-by: SiCong Li <sicong.li@arm.com>
Change-Id: I420490d8b94d13ada2e44eb0a12078f883379334
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9193
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/CL/tuners')
-rw-r--r-- | src/runtime/CL/tuners/CLTuningParametersList.cpp | 27 |
1 files changed, 15 insertions, 12 deletions
diff --git a/src/runtime/CL/tuners/CLTuningParametersList.cpp b/src/runtime/CL/tuners/CLTuningParametersList.cpp index 6cb2212794..6f3e32491a 100644 --- a/src/runtime/CL/tuners/CLTuningParametersList.cpp +++ b/src/runtime/CL/tuners/CLTuningParametersList.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2019-2021 Arm Limited. + * Copyright (c) 2019-2021, 2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -31,7 +31,7 @@ constexpr unsigned int max_lws_supported_x{ 64u }; constexpr unsigned int max_lws_supported_y{ 32u }; constexpr unsigned int max_lws_supported_z{ 32u }; -/** Non instantiable base class for Tuning parameters combinations that use Index2Cooard mapping */ +/** Non instantiable base class for Tuning parameters combinations that use Index2Coord mapping */ class CLTuningParametersList : public ICLTuningParametersList { protected: @@ -162,10 +162,13 @@ CLTuningParams CLTuningParametersListExhaustive::operator[](size_t index) CLTuningParametersListExhaustive::CLTuningParametersListExhaustive(const cl::NDRange &gws, CLTuningInfo tuning_info) { - ARM_COMPUTE_UNUSED(gws); - search_space_shape[0] = max_lws_supported_x; - search_space_shape[1] = max_lws_supported_y; - search_space_shape[2] = max_lws_supported_z; + const auto lws_x_max = std::min(static_cast<unsigned int>(gws[0]), max_lws_supported_x); + const auto lws_y_max = std::min(static_cast<unsigned int>(gws[1]), max_lws_supported_y); + const auto lws_z_max = std::min(static_cast<unsigned int>(gws[2]), max_lws_supported_z); + + search_space_shape[0] = lws_x_max; + search_space_shape[1] = lws_y_max; + search_space_shape[2] = lws_z_max; search_space_shape[3] = 1; if(tuning_info.tune_wbsm) { @@ -183,9 +186,9 @@ CLTuningParams CLTuningParametersListNormal::operator[](size_t index) CLTuningParametersListNormal::CLTuningParametersListNormal(const cl::NDRange &gws, CLTuningInfo tuning_info) { - auto lws_x_max = std::min(static_cast<unsigned int>(gws[0]), max_lws_supported_x); - auto lws_y_max = std::min(static_cast<unsigned int>(gws[1]), max_lws_supported_y); - auto lws_z_max = std::min(static_cast<unsigned int>(gws[2]), max_lws_supported_z); + const auto lws_x_max = std::min(static_cast<unsigned int>(gws[0]), max_lws_supported_x); + const auto lws_y_max = std::min(static_cast<unsigned int>(gws[1]), max_lws_supported_y); + const auto lws_z_max = std::min(static_cast<unsigned int>(gws[2]), max_lws_supported_z); // Initialize the tuning parameters values to test _lws_x = {}; @@ -227,9 +230,9 @@ void CLTuningParametersListNormal::initialize_lws_values(std::vector<unsigned in CLTuningParametersListRapid::CLTuningParametersListRapid(const cl::NDRange &gws, CLTuningInfo tuning_info) { - auto lws_x_max = std::min(static_cast<unsigned int>(gws[0]), 8u); // Limit exploration to 1 - 8 - auto lws_y_max = std::min(static_cast<unsigned int>(gws[1]), 4u); // Limit exploration to 1 - 4 - auto lws_z_max = std::min(static_cast<unsigned int>(gws[2]), 4u); // Limit exploration to 1 - 4 + const auto lws_x_max = std::min(static_cast<unsigned int>(gws[0]), 8u); // Limit exploration to 1 - 8 + const auto lws_y_max = std::min(static_cast<unsigned int>(gws[1]), 4u); // Limit exploration to 1 - 4 + const auto lws_z_max = std::min(static_cast<unsigned int>(gws[2]), 4u); // Limit exploration to 1 - 4 // Initialize the LWS values to test _lws_x = {}; |