ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-20 04:49:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

CK-Tile Grouped GEMM refactor and post PR fixes (#1756)

Browse Source 
* Grouped gemm simple code refactor

* Offset invoker

* Invoke generic Run, and replace name of parrtitioner variable

* Tests fix type

* Removed namespaces

* Add template param to avoid implicit cast

* Remove generic function

* Constant value

* underline enum to int16_t

* Generalize partitioner function

* Remove whitespaces

* Rename function

* Using support

* Clang-format

* Clang-format

* Fn-partitioner description fn

* Typo

* Typo 2

* Better description

* Better description

* Refactor after review

* Use ctr instead of set fn

* Inovke ctr and typo

* Comments

* Remove unnecessary comment

* Review, remove modulo

[ROCm/composable_kernel commit: 3c93d3c444]
This commit is contained in:
Mateusz Ozga
2025-01-21 21:06:10 +01:00
committed by GitHub
parent c6a6e93628
commit b63e4bc4b8
17 changed files with 363 additions and 388 deletions
Download Patch File Download Diff File Expand all files Collapse all files

57
include/ck_tile/core/arch/arch.hpp
View File

@@ -1,5 +1,5 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
@@ -12,18 +12,37 @@
namespace ck_tile {
enum struct address_space_enum
template <typename, bool>
struct safe_underlying_type;
template <typename T>
struct safe_underlying_type<T, true>
{
generic,
using type = std::underlying_type_t<T>;
};
template <typename T>
struct safe_underlying_type<T, false>
{
using type = void;
};
template <typename T>
using safe_underlying_type_t = typename safe_underlying_type<T, std::is_enum<T>::value>::type;
enum struct address_space_enum : std::uint16_t
{
generic = 0,
global,
lds,
sgpr,
vgpr,
constant,
vgpr
};
enum struct memory_operation_enum
enum struct memory_operation_enum : std::uint16_t
{
set,
set = 0,
atomic_add,
atomic_max,
add
@@ -109,4 +128,30 @@ CK_TILE_DEVICE void s_nop(index_t cnt = 0)
#endif
}
#define CK_CONSTANT_ADDRESS_SPACE \
__attribute__((address_space( \
static_cast<safe_underlying_type_t<address_space_enum>>(address_space_enum::constant))))
template <typename T>
__device__ T* cast_pointer_to_generic_address_space(T CK_CONSTANT_ADDRESS_SPACE* p)
{
// cast a pointer in "Constant" address space (4) to "Generic" address space (0)
// only c-style pointer cast seems be able to be compiled
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wold-style-cast"
return (T*)(p); // NOLINT(old-style-cast)
#pragma clang diagnostic pop
}
template <typename T>
__host__ __device__ T CK_CONSTANT_ADDRESS_SPACE* cast_pointer_to_constant_address_space(T* p)
{
// cast a pointer in "Generic" address space (0) to "Constant" address space (4)
// only c-style pointer cast seems be able to be compiled;
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wold-style-cast"
return (T CK_CONSTANT_ADDRESS_SPACE*)p; // NOLINT(old-style-cast)
#pragma clang diagnostic pop
}
} // namespace ck_tile

37
include/ck_tile/core/utility/amd_address_space.hpp
View File

@@ -1,37 +0,0 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2024, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
#include "ck_tile/core/config.hpp"
// Address Space for AMDGCN
// https://llvm.org/docs/AMDGPUUsage.html#address-space
namespace ck_tile {
#define CK_CONSTANT_ADDRESS_SPACE __attribute__((address_space(4)))
template <typename T>
__device__ T* cast_pointer_to_generic_address_space(T CK_CONSTANT_ADDRESS_SPACE* p)
{
// cast a pointer in "Constant" address space (4) to "Generic" address space (0)
// only c-style pointer cast seems be able to be compiled
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wold-style-cast"
return (T*)p; // NOLINT(old-style-cast)
#pragma clang diagnostic pop
}
template <typename T>
__host__ __device__ T CK_CONSTANT_ADDRESS_SPACE* cast_pointer_to_constant_address_space(T* p)
{
// cast a pointer in "Generic" address space (0) to "Constant" address space (4)
// only c-style pointer cast seems be able to be compiled
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wold-style-cast"
return (T CK_CONSTANT_ADDRESS_SPACE*)p; // NOLINT(old-style-cast)
#pragma clang diagnostic pop
}
} // namespace ck_tile