Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
gaoqiong
composable_kernel
Commits
fa9da1a4
Commit
fa9da1a4
authored
Jun 19, 2023
by
Jun Liu
Browse files
Merge branch 'amd-develop' into amd-master
parents
4c105089
457308e3
Changes
1000
Hide whitespace changes
Inline
Side-by-side
Showing
20 changed files
with
39 additions
and
35 deletions
+39
-35
example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_xdl_bf16.cpp
example/09_convnd_fwd/convnd_fwd_xdl_bf16.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_xdl_fp16.cpp
example/09_convnd_fwd/convnd_fwd_xdl_fp16.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_xdl_fp32.cpp
example/09_convnd_fwd/convnd_fwd_xdl_fp32.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_xdl_fp64.cpp
example/09_convnd_fwd/convnd_fwd_xdl_fp64.cpp
+1
-1
example/09_convnd_fwd/convnd_fwd_xdl_int8.cpp
example/09_convnd_fwd/convnd_fwd_xdl_int8.cpp
+1
-1
example/09_convnd_fwd/run_convnd_fwd_dl_example.inc
example/09_convnd_fwd/run_convnd_fwd_dl_example.inc
+1
-1
example/09_convnd_fwd/run_convnd_fwd_example.inc
example/09_convnd_fwd/run_convnd_fwd_example.inc
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/CMakeLists.txt
...e/10_convnd_fwd_multiple_d_multiple_reduce/CMakeLists.txt
+20
-16
example/10_convnd_fwd_multiple_d_multiple_reduce/common.hpp
example/10_convnd_fwd_multiple_d_multiple_reduce/common.hpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_bf16.cpp
...wd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_bf16.cpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp16.cpp
...wd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp16.cpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp32.cpp
...wd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp32.cpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int4.cpp
...wd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int4.cpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int8.cpp
...wd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int8.cpp
+1
-1
example/10_convnd_fwd_multiple_d_multiple_reduce/run_convnd_fwd_max_example.inc
...multiple_d_multiple_reduce/run_convnd_fwd_max_example.inc
+1
-1
example/12_reduce/reduce_blockwise.cpp
example/12_reduce/reduce_blockwise.cpp
+1
-1
example/12_reduce/reduce_blockwise_impl.hpp
example/12_reduce/reduce_blockwise_impl.hpp
+1
-1
example/12_reduce/reduce_blockwise_two_call.cpp
example/12_reduce/reduce_blockwise_two_call.cpp
+1
-1
No files found.
Too many changes to show.
To preserve performance only
1000 of 1000+
files are displayed.
Plain diff
Email patch
example/09_convnd_fwd/convnd_fwd_dl_fp32.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_dl_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_dl_int8.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_dl_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_xdl_bf16.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_xdl_fp16.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_xdl_fp32.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_xdl_fp64.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_common.hpp"
...
...
example/09_convnd_fwd/convnd_fwd_xdl_int8.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "convnd_fwd_common.hpp"
...
...
example/09_convnd_fwd/run_convnd_fwd_dl_example.inc
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
...
...
example/09_convnd_fwd/run_convnd_fwd_example.inc
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/CMakeLists.txt
View file @
fa9da1a4
add_custom_target
(
example_convnd_fwd_reduce_xdl
)
add_example_executable
(
example_convnd_fwd_max_xdl_int8 convnd_fwd_max_xdl_int8.cpp
)
add_example_executable_no_testing
(
example_convnd_fwd_max_xdl_bf16 convnd_fwd_max_xdl_bf16.cpp
)
add_example_executable_no_testing
(
example_convnd_fwd_max_xdl_fp16 convnd_fwd_max_xdl_fp16.cpp
)
add_example_executable
(
example_convnd_fwd_max_xdl_fp32 convnd_fwd_max_xdl_fp32.cpp
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_int8
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_bf16
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_fp16
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_fp32
)
if
(
USE_BITINT_EXTENSION_INT4
)
add_example_executable
(
example_convnd_fwd_max_xdl_int4 convnd_fwd_max_xdl_int4.cpp
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_int4
)
endif
(
USE_BITINT_EXTENSION_INT4
)
list
(
APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942
)
set
(
target 0
)
foreach
(
gpu IN LISTS GPU_TARGETS
)
if
(
gpu IN_LIST gpu_list AND target EQUAL 0
)
add_custom_target
(
example_convnd_fwd_reduce_xdl
)
add_example_executable
(
example_convnd_fwd_max_xdl_int8 convnd_fwd_max_xdl_int8.cpp
)
add_example_executable_no_testing
(
example_convnd_fwd_max_xdl_bf16 convnd_fwd_max_xdl_bf16.cpp
)
add_example_executable_no_testing
(
example_convnd_fwd_max_xdl_fp16 convnd_fwd_max_xdl_fp16.cpp
)
add_example_executable
(
example_convnd_fwd_max_xdl_fp32 convnd_fwd_max_xdl_fp32.cpp
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_int8
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_bf16
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_fp16
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_fp32
)
if
(
USE_BITINT_EXTENSION_INT4
)
add_example_executable
(
example_convnd_fwd_max_xdl_int4 convnd_fwd_max_xdl_int4.cpp
)
add_dependencies
(
example_convnd_fwd_reduce_xdl example_convnd_fwd_max_xdl_int4
)
endif
(
USE_BITINT_EXTENSION_INT4
)
set
(
target 1
)
endif
()
endforeach
()
\ No newline at end of file
example/10_convnd_fwd_multiple_d_multiple_reduce/common.hpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include <algorithm>
#include <cassert>
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_bf16.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "common.hpp"
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp16.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "common.hpp"
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_fp32.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "common.hpp"
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int4.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#ifndef CK_EXPERIMENTAL_BIT_INT_EXTENSION_INT4
#error Should compile this file with ck::int4_t support
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/convnd_fwd_max_xdl_int8.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include "common.hpp"
...
...
example/10_convnd_fwd_multiple_d_multiple_reduce/run_convnd_fwd_max_example.inc
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
...
...
example/12_reduce/reduce_blockwise.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include <iostream>
#include <initializer_list>
...
...
example/12_reduce/reduce_blockwise_impl.hpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#pragma once
...
...
example/12_reduce/reduce_blockwise_two_call.cpp
View file @
fa9da1a4
// SPDX-License-Identifier: MIT
// Copyright (c) 2018-202
2
, Advanced Micro Devices, Inc. All rights reserved.
// Copyright (c) 2018-202
3
, Advanced Micro Devices, Inc. All rights reserved.
#include <iostream>
#include <numeric>
...
...
Prev
1
2
3
4
5
6
7
8
9
…
50
Next
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment