ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-19 04:19:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit 'a1589a9667517ddc73048c05c6f3c859db99851d' into develop

Browse Source
This commit is contained in:
assistant-librarian[bot]
2025-08-19 09:12:25 +00:00
parent e9577d7ef3
commit 40d03acb47
16 changed files with 129 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
example/01_gemm/run_gemm_example.inc
View File

@@ -1,7 +1,8 @@
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-2024, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
#include "ck/library/utility/validation_common.hpp"
template <typename ProblemType>
bool run_gemm(const ProblemType& problem_size, const ExecutionConfig& config)
@@ -53,6 +54,17 @@ bool run_gemm(const ProblemType& problem_size, const ExecutionConfig& config)
StrideB = f_get_default_stride(K, N, StrideB, BLayout{});
StrideC = f_get_default_stride(M, N, StrideC, CLayout{});
try
{
ck::utils::validate_gemm_strides_abc<ALayout, BLayout, CLayout>(
M, N, K, StrideA, StrideB, StrideC);
}
catch(const std::runtime_error& e)
{
std::cerr << "Error: " << e.what() << std::endl;
return false;
}
Tensor<ADataType> a_m_k(f_host_tensor_descriptor(M, K, StrideA, ALayout{}));
Tensor<BDataType> b_k_n(f_host_tensor_descriptor(K, N, StrideB, BLayout{}));

2
example/01_gemm/run_gemm_example_v2.inc
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

7
example/ck_tile/17_grouped_gemm/grouped_gemm.cpp
View File

@@ -29,10 +29,6 @@ float grouped_gemm_tileloop(const ck_tile::stream_config& s,
void* kargs_ptr,
bool splitk)
{
constexpr bool kPadM = false;
constexpr bool kPadN = false;
constexpr bool kPadK = false;
constexpr ck_tile::index_t TileParitionerGroupNum = 8;
constexpr ck_tile::index_t TileParitionerM01 = 4;
@@ -44,7 +40,6 @@ float grouped_gemm_tileloop(const ck_tile::stream_config& s,
using TilePartitioner = ck_tile::
GemmSpatiallyLocalTilePartitioner<GemmShape, TileParitionerGroupNum, TileParitionerM01>;
using Traits = ck_tile::TileGemmTraits<kPadM, kPadN, kPadK, ALayout, BLayout, CLayout>;
using GemmUniversalTraits =
ck_tile::PersistentTileGemmUniversalTraits<GemmConfig::kPadM,
GemmConfig::kPadN,
@@ -53,8 +48,6 @@ float grouped_gemm_tileloop(const ck_tile::stream_config& s,
ALayout,
BLayout,
CLayout>;
using GemmPipelineProblem =
ck_tile::GemmPipelineProblem<ADataType, BDataType, AccDataType, GemmShape, Traits>;
float ave_time{0};