diff options
| author | Gunes Bayir <gunes.bayir@arm.com> | 2023-01-03 21:26:09 +0000 |
|---|---|---|
| committer | Gunes Bayir <gunes.bayir@arm.com> | 2023-01-06 13:45:22 +0000 |
| commit | 3a1e1256946028cb13f31521caec9f08235c7332 (patch) | |
| tree | cc63b85bc30ecdc81d7218d7adf05d27a96e6411 /src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp | |
| parent | b3077fbaee868579f9a41888fef1f71286d6757c (diff) | |
| download | ComputeLibrary-3a1e1256946028cb13f31521caec9f08235c7332.tar.gz | |
Handle Intermediate tensors within the sketch
- Intermediate tensor info objects are not created by the user anymore. They're returned from create_op and reused. This will prevent allocation of the intermediate tensors in case of possible interface misuse.
- Sketch object handles intermediate tensor info pointers inside its implementation class via a unique pointer vector
- Conv2d operator is migrated into the new interface
Resolves: COMPMID-5776
Change-Id: I9422e3681eef4f2d2922f6d0a5d7786380837c6d
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8906
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp')
| -rw-r--r-- | src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp | 14 |
1 files changed, 7 insertions, 7 deletions
diff --git a/src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp b/src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp index 60c2281433..cd5487c10b 100644 --- a/src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp +++ b/src/dynamic_fusion/sketch/gpu/operators/GpuOutput.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2022 Arm Limited. + * Copyright (c) 2022-2023 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -24,12 +24,13 @@ #include "arm_compute/dynamic_fusion/sketch/gpu/operators/GpuOutput.h" -#include "src/core/helpers/AutoConfiguration.h" #include "src/common/utils/Log.h" +#include "src/core/helpers/AutoConfiguration.h" #include "src/dynamic_fusion/sketch/ArgumentPack.h" #include "src/dynamic_fusion/sketch/gpu/GpuWorkloadSketchImpl.h" #include "src/dynamic_fusion/sketch/gpu/components/cl/ClComponentStore.h" +#include "src/dynamic_fusion/utils/Utils.h" namespace arm_compute { @@ -65,7 +66,7 @@ Status GpuOutput::validate_op(const GpuWorkloadSketch &sketch, { ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(src, dst); ARM_COMPUTE_RETURN_ERROR_ON(!src->has_valid_id()); - ARM_COMPUTE_RETURN_ERROR_ON(!dst->has_valid_id()); + ARM_COMPUTE_RETURN_ERROR_ON(!is_user_tensor(dst)); // Initialize the destination tensor info. TensorInfo dst_to_validate = *dst; @@ -79,12 +80,11 @@ Status GpuOutput::validate_op(const GpuWorkloadSketch &sketch, tensors.add_const_tensor(ACL_SRC_0, src); tensors.add_const_tensor(ACL_DST_0, &dst_to_validate); - const auto group = sketch.implementation().operator_group(); - const auto op = group.new_operator(operator_type, tensors); + const auto group = sketch.implementation().operator_group(); + const auto op = group.new_operator(operator_type, tensors); const auto success = group.try_add_operator(op, true); ARM_COMPUTE_RETURN_ERROR_ON_MSG(!success, "This operator cannot be fused into the workload."); - ARM_COMPUTE_UNUSED(success); const auto status = is_supported_op(*sketch.gpu_context(), src, dst); return status; @@ -101,7 +101,7 @@ void GpuOutput::create_op(GpuWorkloadSketch &sketch, auto_init_if_empty(*dst, *src); // Translate into components and add to component graph - auto &comp_graph = sketch.implementation().component_graph(); + auto &comp_graph = sketch.implementation().component_graph(); const auto sketch_ctx = sketch.implementation().context(); if(sketch_ctx->gpu_language() == GpuLanguage::OpenCL) |