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
b798c73b
"tests/vscode:/vscode.git/clone" did not exist on "56a76082edbd24d52c7e346942c40104159a43f4"
Commit
b798c73b
authored
May 17, 2023
by
rocking
Browse files
Refine naming
parent
b31307e9
Changes
9
Hide whitespace changes
Inline
Side-by-side
Showing
9 changed files
with
18 additions
and
18 deletions
+18
-18
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f16_instance.cpp
.../gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f16_instance.cpp
+1
-1
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f32_instance.cpp
.../gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f32_instance.cpp
+1
-1
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f16_instance.cpp
...gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f16_instance.cpp
+1
-1
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f32_instance.cpp
...gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f32_instance.cpp
+1
-1
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f16_instance.cpp
.../gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f16_instance.cpp
+2
-2
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f32_instance.cpp
.../gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f32_instance.cpp
+2
-2
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f16_instance.cpp
...gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f16_instance.cpp
+2
-2
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f32_instance.cpp
...gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f32_instance.cpp
+2
-2
library/src/tensor_operation_instance/gpu/pool_fwd/pool_fwd_instance_common.hpp
...ration_instance/gpu/pool_fwd/pool_fwd_instance_common.hpp
+6
-6
No files found.
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f16_instance.cpp
View file @
b798c73b
...
...
@@ -14,7 +14,7 @@ void add_device_pool2d_fwd_nhwc_f16_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F16
,
F16
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool2d_fwd_nhwc_f32_instance.cpp
View file @
b798c73b
...
...
@@ -14,7 +14,7 @@ void add_device_pool2d_fwd_nhwc_f32_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F32
,
F32
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f16_instance.cpp
View file @
b798c73b
...
...
@@ -14,7 +14,7 @@ void add_device_pool3d_fwd_ndhwc_f16_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F16
,
F16
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_avg_pool3d_fwd_ndhwc_f32_instance.cpp
View file @
b798c73b
...
...
@@ -14,7 +14,7 @@ void add_device_pool3d_fwd_ndhwc_f32_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F32
,
F32
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f16_instance.cpp
View file @
b798c73b
...
...
@@ -14,14 +14,14 @@ void add_device_pool2d_fwd_nhwc_f16_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F16
,
F16
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
false
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
false
>
{});
}
void
add_device_pool2d_fwd_nhwc_index_f16_instances
(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F16
,
F16
,
I32
,
ReduceOpId
,
true
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
true
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
true
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool2d_fwd_nhwc_f32_instance.cpp
View file @
b798c73b
...
...
@@ -14,14 +14,14 @@ void add_device_pool2d_fwd_nhwc_f32_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F32
,
F32
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
void
add_device_pool2d_fwd_nhwc_index_f32_instances
(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
4
,
2
,
F32
,
F32
,
I32
,
ReduceOpId
,
true
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
true
>
{});
instances
,
device_pool2d_fwd_nhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
true
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f16_instance.cpp
View file @
b798c73b
...
...
@@ -14,14 +14,14 @@ void add_device_pool3d_fwd_ndhwc_f16_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F16
,
F16
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
false
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
false
>
{});
}
void
add_device_pool3d_fwd_ndhwc_index_f16_instances
(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F16
,
F16
,
I32
,
ReduceOpId
,
true
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
true
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F16
,
F16
,
I32
,
F16
,
ReduceOpId
,
true
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/device_max_pool3d_fwd_ndhwc_f32_instance.cpp
View file @
b798c73b
...
...
@@ -14,14 +14,14 @@ void add_device_pool3d_fwd_ndhwc_f32_instances(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F32
,
F32
,
I32
,
ReduceOpId
,
false
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
false
>
{});
}
void
add_device_pool3d_fwd_ndhwc_index_f32_instances
(
std
::
vector
<
std
::
unique_ptr
<
DevicePoolFwd
<
5
,
3
,
F32
,
F32
,
I32
,
ReduceOpId
,
true
>>>&
instances
)
{
add_device_operation_instances
(
instances
,
device_pool
ing
3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
true
>
{});
instances
,
device_pool3d_fwd_ndhwc_instances
<
F32
,
F32
,
I32
,
F32
,
ReduceOpId
,
true
>
{});
}
}
// namespace instance
...
...
library/src/tensor_operation_instance/gpu/pool_fwd/pool_fwd_instance_common.hpp
View file @
b798c73b
...
...
@@ -25,14 +25,14 @@ template <typename InDataType,
typename
AccDataType
,
ReduceTensorOp
ReduceOpId
,
bool
OutputIndex
>
using
device_pool
ing
2d_fwd_nhwc_instances
=
using
device_pool2d_fwd_nhwc_instances
=
// clang-format off
std
::
tuple
<
DevicePool2dFwd_Input_N_Hi_Wi_C_Output_N_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
1
,
1
,
1
>
,
DevicePool2dFwd_Input_N_Hi_Wi_C_Output_N_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
2
,
1
,
2
>
,
DevicePool2dFwd_Input_N_Hi_Wi_C_Output_N_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
4
,
1
,
4
>
// clang-format on
>
;
// clang-format on
>
;
template
<
typename
InDataType
,
typename
OutDataType
,
...
...
@@ -40,14 +40,14 @@ template <typename InDataType,
typename
AccDataType
,
ReduceTensorOp
ReduceOpId
,
bool
OutputIndex
>
using
device_pool
ing
3d_fwd_ndhwc_instances
=
using
device_pool3d_fwd_ndhwc_instances
=
// clang-format off
std
::
tuple
<
DevicePool3dFwd_Input_N_Di_Hi_Wi_C_Output_N_Do_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
1
,
1
,
1
>
,
DevicePool3dFwd_Input_N_Di_Hi_Wi_C_Output_N_Do_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
2
,
1
,
2
>
,
DevicePool3dFwd_Input_N_Di_Hi_Wi_C_Output_N_Do_Ho_Wo_C
<
InDataType
,
OutDataType
,
IndexDataType
,
AccDataType
,
ReduceOpId
,
OutputIndex
,
256
,
256
,
1
,
4
,
1
,
4
>
// clang-format on
>
;
// clang-format on
>
;
}
// namespace instance
}
// namespace device
...
...
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