mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-06-29 11:16:59 +00:00
717f2efef7
[CK] add composable kernel support on gfx1250 (#6978) ## Motivation Add composable kernel support on gfx1250. ## Technical Details <!-- Explain the changes along with any relevant GitHub links. --> ## Test Plan <!-- Explain any relevant testing done to verify this PR. --> ## Test Result <!-- Briefly summarize test outcomes. --> ## Submission Checklist - [ ] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests. --------- Co-authored-by: Qun Lin <qlin@amd.com> Co-authored-by: jialuo12_amdeng <jia.luo@amd.com> Co-authored-by: Andriy Roshchenko <andriy.roshchenko@amd.com> Co-authored-by: hsivasun_amdeng <haresh.sivasuntharampillai@amd.com>
341 lines
12 KiB
C++
341 lines
12 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2025, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#include "mx_wmma_op_test_common.hpp"
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x128_bf8_bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x128_BF8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x128_bf8_fp8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x128_BF8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x128_fp8_bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x128_FP8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x128_fp8_fp8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x128_FP8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x64_f8f8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x64_FP8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 64;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x64_f8bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x64_FP8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 64;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x64_bf8f8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x64_BF8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 64;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f16_16x16x64_bf8bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF16WMMA16x16x64_BF8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = ck::half_t;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 64;
|
|
using AccType = ck::half_t;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f32_16x16x128_bf8_bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF32WMMA16x16x128_BF8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = float;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = float;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f32_16x16x128_bf8_fp8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF32WMMA16x16x128_BF8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::bf8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = float;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = float;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f32_16x16x128_fp8_bf8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF32WMMA16x16x128_FP8_BF8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::bf8_t;
|
|
using CType = float;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = float;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|
|
|
|
// Unscaled WMMA: test wmma_f32_16x16x128_fp8_fp8_gfx125
|
|
TEST(MXWMMA_UNSCALED, MXF32WMMA16x16x128_FP8_FP8_GFX125)
|
|
{
|
|
using ALayout = ck::tensor_layout::gemm::RowMajor;
|
|
using BLayout = ck::tensor_layout::gemm::ColumnMajor;
|
|
using CLayout = ck::tensor_layout::gemm::RowMajor;
|
|
using AType = ck::f8_t;
|
|
using BType = ck::f8_t;
|
|
using CType = float;
|
|
|
|
constexpr int BLOCK_M = 16;
|
|
constexpr int BLOCK_N = 16;
|
|
constexpr int BLOCK_K = 128;
|
|
using AccType = float;
|
|
|
|
auto pass = run_mx_wmma_unscaled_test<ALayout,
|
|
BLayout,
|
|
CLayout,
|
|
AType,
|
|
BType,
|
|
CType,
|
|
AccType,
|
|
BLOCK_M,
|
|
BLOCK_N,
|
|
BLOCK_K>(common_init);
|
|
EXPECT_TRUE(pass);
|
|
}
|