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
93e7f92a
Commit
93e7f92a
authored
Nov 18, 2022
by
Astha Rai
Browse files
removed debugging print statements
parent
16f02f76
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
4 additions
and
7 deletions
+4
-7
example/43_elementwise_permute_2d/elementwise_permute_4D_fp16_2d.cpp
...elementwise_permute_2d/elementwise_permute_4D_fp16_2d.cpp
+3
-3
include/ck/tensor_operation/gpu/device/device_elementwise_2d.hpp
.../ck/tensor_operation/gpu/device/device_elementwise_2d.hpp
+1
-4
No files found.
example/43_elementwise_permute_2d/elementwise_permute_4D_fp16_2d.cpp
View file @
93e7f92a
...
@@ -48,10 +48,10 @@ int main()
...
@@ -48,10 +48,10 @@ int main()
bool
do_verification
=
true
;
bool
do_verification
=
true
;
bool
time_kernel
=
true
;
bool
time_kernel
=
true
;
const
int
N
=
128
;
const
int
N
=
120
;
const
int
H
=
128
;
const
int
W
=
128
;
const
int
C
=
128
;
const
int
C
=
128
;
const
int
H
=
32
;
const
int
W
=
1024
;
/**const int N = 120;
/**const int N = 120;
const int H = 32;
const int H = 32;
...
...
include/ck/tensor_operation/gpu/device/device_elementwise_2d.hpp
View file @
93e7f92a
...
@@ -83,10 +83,7 @@ struct DeviceElementwise : public DeviceElementwiseBase<InDataTypeTuple,
...
@@ -83,10 +83,7 @@ struct DeviceElementwise : public DeviceElementwiseBase<InDataTypeTuple,
const
index_t
loop_step_n
=
num_threads_n
*
NPerThread
;
const
index_t
loop_step_n
=
num_threads_n
*
NPerThread
;
const
auto
pad_m
=
math
::
integer_least_multiple
(
m
,
loop_step_m
)
-
m
;
const
auto
pad_m
=
math
::
integer_least_multiple
(
m
,
loop_step_m
)
-
m
;
const
auto
pad_n
=
math
::
integer_least_multiple
(
n
,
loop_step_n
)
-
n
;
const
auto
pad_n
=
math
::
integer_least_multiple
(
n
,
loop_step_n
)
-
n
;
std
::
cout
<<
NumDim_m
<<
" m: "
<<
m
<<
" loop_step_m: "
<<
loop_step_m
<<
" pad_m: "
<<
pad_m
<<
std
::
endl
;
std
::
cout
<<
NumDim_n
<<
" n: "
<<
n
<<
" loop_step_n: "
<<
loop_step_n
<<
" pad_n: "
<<
pad_n
<<
std
::
endl
;
const
auto
desc_mn_pad
=
transform_tensor_descriptor
(
const
auto
desc_mn_pad
=
transform_tensor_descriptor
(
desc_mn
,
desc_mn
,
make_tuple
(
make_right_pad_transform
(
m
,
pad_m
),
make_right_pad_transform
(
n
,
pad_n
)),
make_tuple
(
make_right_pad_transform
(
m
,
pad_m
),
make_right_pad_transform
(
n
,
pad_n
)),
...
...
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