diff options
Diffstat (limited to 'src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp')
| -rw-r--r-- | src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp | 59 |
1 files changed, 43 insertions, 16 deletions
diff --git a/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp index 3a39128c0a..959b3ab21d 100644 --- a/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp +++ b/src/gpu/cl/kernels/ClGemmMatrixMultiplyReshapedKernel.cpp @@ -32,6 +32,7 @@ #include "arm_compute/core/Utils.h" #include "arm_compute/core/Validate.h" #include "arm_compute/core/utils/misc/ShapeCalculator.h" +#include "src/core/AccessWindowStatic.h" #include "src/core/CL/CLUtils.h" #include "src/core/CL/CLValidate.h" #include "src/core/helpers/AutoConfiguration.h" @@ -41,6 +42,10 @@ #include "support/Cast.h" #include "support/StringSupport.h" +#include <cstddef> +#include <cstdint> +#include <tuple> + namespace arm_compute { namespace opencl @@ -126,11 +131,17 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *src0, ITens const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info, ElementsProcessed &num_elements_processed) { - ARM_COMPUTE_UNUSED(src0, src1, src2); unsigned int &num_elems_processed_per_iteration_x = num_elements_processed[0]; unsigned int &num_elems_processed_per_iteration_y = num_elements_processed[1]; bool reinterpret_output_as_3d = gemm_info.depth_output_gemm3d != 0; + Window win{}; + Window win_out{}; + bool window_changed = false; + + // dst tensor auto initialization if not yet initialized + auto_init_if_empty(*dst, src0->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info))); + TensorInfo tmp_info(*dst); if(reinterpret_output_as_3d) @@ -146,7 +157,21 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *src0, ITens num_elems_processed_per_iteration_x = rhs_info.n0; num_elems_processed_per_iteration_y = lhs_info.m0; - Window win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); + win = calculate_max_window(tmp_info, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); + win_out = calculate_max_window(*dst, Steps(num_elems_processed_per_iteration_x, num_elems_processed_per_iteration_y)); + + if(src2 != nullptr) + { + const int bias_processed_per_iteration_x = num_elems_processed_per_iteration_x; + + const int bias_processed_per_iteration_y = gemm_info.broadcast_bias ? 1 : num_elems_processed_per_iteration_y; + + AccessWindowStatic src2_access(src2, 0, 0, + ceil_to_multiple(src2->dimension(0), bias_processed_per_iteration_x), + ceil_to_multiple(src2->dimension(1), bias_processed_per_iteration_y)); + + window_changed = update_window_and_padding(win, src2_access); // window used by the execute_window_loop + } // Collapse along the Z direction // This collapse needs to be here in order to tune the Z dimension of LWS @@ -154,7 +179,8 @@ std::pair<Status, Window> validate_and_configure_window(ITensorInfo *src0, ITens const unsigned int dimension_to_collapse = std::min(static_cast<unsigned int>(dst->num_dimensions()), 2u); collapsed = win.collapse(win, dimension_to_collapse); - return std::make_pair(Status{}, collapsed); + Status err = (window_changed) ? ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Insufficient Padding!") : Status{}; + return std::make_pair(err, collapsed); } } // namespace @@ -164,17 +190,14 @@ ClGemmMatrixMultiplyReshapedKernel::ClGemmMatrixMultiplyReshapedKernel() } void ClGemmMatrixMultiplyReshapedKernel::configure(const CLCompileContext &compile_context, - const ITensorInfo *src0, const ITensorInfo *src1, const ITensorInfo *src2, ITensorInfo *dst, float alpha, float beta, + ITensorInfo *src0, ITensorInfo *src1, ITensorInfo *src2, ITensorInfo *dst, float alpha, float beta, const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info) { ARM_COMPUTE_ERROR_ON_NULLPTR(src0, src1, dst); ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(src0, src1, src2, dst, alpha, beta, lhs_info, rhs_info, gemm_info)); - // dst tensor auto initialization if not yet initialized - auto_init_if_empty(*dst, src0->clone()->set_tensor_shape(misc::shape_calculator::compute_mm_shape(*src0, *src1, gemm_info))); - - auto padding_info = get_padding_info({ src0, src1, src2, dst }); + auto padding_info = get_padding_info({ src0, dst }); _reinterpret_output_as_3d = gemm_info.depth_output_gemm3d != 0; _use_dummy_work_items = preferred_dummy_work_items_support(CLKernelLibrary::get().get_device()); _add_bias = src2 != nullptr; @@ -188,14 +211,7 @@ void ClGemmMatrixMultiplyReshapedKernel::configure(const CLCompileContext &compi ElementsProcessed num_elements_processed{}; // Configure kernel window - auto win_config = validate_and_configure_window(src0->clone().get(), - src1->clone().get(), - (src2 != nullptr) ? src2->clone().get() : nullptr, - dst->clone().get(), - lhs_info, - rhs_info, - gemm_info, - num_elements_processed); + auto win_config = validate_and_configure_window(src0, src1, src2, dst, lhs_info, rhs_info, gemm_info, num_elements_processed); ARM_COMPUTE_ERROR_THROW_ON(win_config.first); ICLKernel::configure_internal(win_config.second); @@ -288,7 +304,18 @@ Status ClGemmMatrixMultiplyReshapedKernel::validate(const ITensorInfo *src0, con const GEMMLHSMatrixInfo &lhs_info, const GEMMRHSMatrixInfo &rhs_info, const GEMMKernelInfo &gemm_info) { + ElementsProcessed num_elements_processed{}; ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(src0, src1, src2, dst, alpha, beta, lhs_info, rhs_info, gemm_info)); + ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(src0->clone().get(), + src1->clone().get(), + src2 != nullptr ? src2->clone().get() : nullptr, + dst->clone().get(), + lhs_info, + rhs_info, + gemm_info, + num_elements_processed) + .first); + return Status{}; } |