leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 26 Packages Projects Releases Wiki Activity

CUTLASS 3.7 (#2045)

Browse Source 
* CUTLASS 3.7

* clean up changelog

---------

Co-authored-by: yuzhai <yuzhai@nvidia.com>
Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
This commit is contained in:
Yujia Zhai
2025-01-18 09:53:07 -05:00
committed by GitHub
co-authored by yuzhai Haicheng Wu
parent 902dff3663
commit b78588d163
2030 changed files with 8947 additions and 3475 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/57_hopper_grouped_gemm/57_hopper_grouped_gemm.cu
+5 -6
View File
@@ -1,5 +1,5 @@
/***************************************************************************************************
* Copyright (c) 2023 - 2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
* Copyright (c) 2023 - 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
* SPDX-License-Identifier: BSD-3-Clause
*
* Redistribution and use in source and binary forms, with or without
@@ -550,11 +550,10 @@ void initialize(const Options &options) {
template <typename GemmT>
typename GemmT::Arguments args_from_options(const Options &options, bool host_problem_shapes_available = true)
{
cutlass::KernelHardwareInfo hw_info;
// Change device_id to another value if you are running on a machine with multiple GPUs and wish
// to use a GPU other than that with device ID 0.
hw_info.device_id = 0;
hw_info.sm_count = cutlass::KernelHardwareInfo::query_device_multiprocessor_count(hw_info.device_id);
int device_id = 0;
cutlass::KernelHardwareInfo kernel_hw_info = cutlass::KernelHardwareInfo::make_kernel_hardware_info<Gemm::GemmKernel>(device_id);
typename GemmT::Arguments arguments;
decltype(arguments.epilogue.thread) fusion_args;
@@ -590,7 +589,7 @@ typename GemmT::Arguments args_from_options(const Options &options, bool host_pr
{options.groups, problem_sizes.get(), options.problem_sizes_host.data()},
{ptr_A.get(), stride_A.get(), ptr_B.get(), stride_B.get()},
{fusion_args, ptr_C.get(), stride_C.get(), ptr_D.get(), stride_D.get()},
hw_info
kernel_hw_info
};
}
else {
@@ -599,7 +598,7 @@ typename GemmT::Arguments args_from_options(const Options &options, bool host_pr
{options.groups, problem_sizes.get(), nullptr},
{ptr_A.get(), stride_A.get(), ptr_B.get(), stride_B.get()},
{fusion_args, ptr_C.get(), stride_C.get(), ptr_D.get(), stride_D.get()},
hw_info
kernel_hw_info
};
}
examples/57_hopper_grouped_gemm/CMakeLists.txt
+1 -1
View File
@@ -1,4 +1,4 @@
# Copyright (c) 2023 - 2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# Copyright (c) 2023 - 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: BSD-3-Clause
#
# Redistribution and use in source and binary forms, with or without