ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-16 10:59:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

Extend permute scale support up to 6D (#1168)

Browse Source 
* Extend permute scale support up to 6D

* Fixes

* Fixes

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

* Update profiler/README.md

Co-authored-by: Lisa <lisajdelaney@gmail.com>

---------

Co-authored-by: Lisa <lisajdelaney@gmail.com>

[ROCm/composable_kernel commit: 66736edb95]
This commit is contained in:
Bartłomiej Kocot
2024-02-20 18:56:54 +01:00
committed by GitHub
parent 4acafee7c2
commit a08052cf6a
15 changed files with 898 additions and 299 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
library/src/tensor_operation_instance/gpu/permute_scale/CMakeLists.txt
View File

@@ -1,2 +1,7 @@
add_instance_library(device_permute_scale_instance
device_permute_scale_instances.cpp)
device_permute_scale_1d_instances.cpp
device_permute_scale_2d_instances.cpp
device_permute_scale_3d_instances.cpp
device_permute_scale_4d_instances.cpp
device_permute_scale_5d_instances.cpp
device_permute_scale_6d_instances.cpp)

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_1d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_1d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 1>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<1>{});
}
void add_device_permute_scale_1d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 1>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<1>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_2d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_2d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 2>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<2>{});
}
void add_device_permute_scale_2d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 2>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<2>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_3d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_3d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 3>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<3>{});
}
void add_device_permute_scale_3d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 3>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<3>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_4d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_4d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<4>{});
}
void add_device_permute_scale_4d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<4>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_5d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_5d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 5>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<5>{});
}
void add_device_permute_scale_5d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 5>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<5>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

29
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_6d_instances.cpp Normal file
View File

@@ -0,0 +1,29 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/library/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
void add_device_permute_scale_6d_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 6>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances<6>{});
}
void add_device_permute_scale_6d_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 6>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances<6>{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck

56
library/src/tensor_operation_instance/gpu/permute_scale/device_permute_scale_instances.cpp
View File

@@ -1,56 +0,0 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
#include "ck/ck.hpp"
#include "ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp"
#include "ck/utility/data_type.hpp"
#include "ck/library/tensor_operation_instance/add_device_operation_instance.hpp"
namespace ck {
namespace tensor_operation {
namespace device {
namespace instance {
using F16 = ck::half_t;
using F32 = float;
using Pass = ck::tensor_operation::element_wise::PassThrough;
using UnaryOp = ck::tensor_operation::element_wise::UnarySquare;
using Scale = ck::tensor_operation::element_wise::Scale;
// clang-format off
using device_permute_scale_f16_instances =
std::tuple <
DeviceElementwiseImpl<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4, 1, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4, 8, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4, 4, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4, 2, ck::Sequence<1>, ck::Sequence<1>>
>;
using device_permute_scale_f32_instances = std::tuple<
DeviceElementwiseImpl<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4, 1, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4, 8, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4, 4, ck::Sequence<1>, ck::Sequence<1>>,
DeviceElementwiseImpl<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4, 2, ck::Sequence<1>, ck::Sequence<1>>
>;
// clang-format on
void add_device_permute_scale_f16_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F16>, ck::Tuple<F16>, Pass, UnaryOp, Scale, 4>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f16_instances{});
}
void add_device_permute_scale_f32_instances(
std::vector<std::unique_ptr<
DeviceElementwise<ck::Tuple<F32>, ck::Tuple<F32>, Pass, UnaryOp, Scale, 4>>>& instances)
{
add_device_operation_instances(instances, device_permute_scale_f32_instances{});
}
} // namespace instance
} // namespace device
} // namespace tensor_operation
} // namespace ck