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
jerrrrry
infinicore
Commits
6b717b30
Commit
6b717b30
authored
Apr 09, 2025
by
PanZezhong
Browse files
issue/4/fix 修改info构建
parent
a8955429
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
4 additions
and
4 deletions
+4
-4
src/infiniop/ops/causal_softmax/cuda/causal_softmax_cuda.cu
src/infiniop/ops/causal_softmax/cuda/causal_softmax_cuda.cu
+4
-4
No files found.
src/infiniop/ops/causal_softmax/cuda/causal_softmax_cuda.cu
View file @
6b717b30
...
@@ -17,11 +17,11 @@ infiniStatus_t Descriptor::create(
...
@@ -17,11 +17,11 @@ infiniStatus_t Descriptor::create(
infiniopHandle_t
handle
,
infiniopHandle_t
handle
,
Descriptor
**
desc_ptr
,
Descriptor
**
desc_ptr
,
infiniopTensorDescriptor_t
y_desc
)
{
infiniopTensorDescriptor_t
y_desc
)
{
CausalSoftmaxInfo
info
;
auto
info
=
CausalSoftmaxInfo
::
create
(
y_desc
)
;
CHECK_
STATUS
(
createCausalSoftmaxInfo
(
&
info
,
y_desc
)
);
CHECK_
RESULT
(
info
);
*
desc_ptr
=
new
Descriptor
(
*
desc_ptr
=
new
Descriptor
(
new
Opaque
{
reinterpret_cast
<
device
::
cuda
::
Handle
*>
(
handle
)
->
internal
()},
new
Opaque
{
reinterpret_cast
<
device
::
cuda
::
Handle
*>
(
handle
)
->
internal
()},
info
,
0
,
handle
->
device
,
handle
->
device_id
);
info
.
take
()
,
0
,
handle
->
device
,
handle
->
device_id
);
return
INFINI_STATUS_SUCCESS
;
return
INFINI_STATUS_SUCCESS
;
}
}
...
@@ -42,7 +42,7 @@ infiniStatus_t launchKernel(void *data, infiniDtype_t dtype, size_t batch_size,
...
@@ -42,7 +42,7 @@ infiniStatus_t launchKernel(void *data, infiniDtype_t dtype, size_t batch_size,
infiniStatus_t
Descriptor
::
calculate
(
void
*
workspace
,
size_t
workspace_size
,
infiniStatus_t
Descriptor
::
calculate
(
void
*
workspace
,
size_t
workspace_size
,
void
*
data
,
void
*
data
,
void
*
stream_
)
{
void
*
stream_
)
const
{
cudaStream_t
stream
=
(
cudaStream_t
)
stream_
;
cudaStream_t
stream
=
(
cudaStream_t
)
stream_
;
if
(
_opaque
->
internal
->
maxThreadsPerBlock
()
==
CUDA_BLOCK_SIZE_1024
)
{
if
(
_opaque
->
internal
->
maxThreadsPerBlock
()
==
CUDA_BLOCK_SIZE_1024
)
{
CHECK_STATUS
(
launchKernel
<
CUDA_BLOCK_SIZE_1024
>
(
CHECK_STATUS
(
launchKernel
<
CUDA_BLOCK_SIZE_1024
>
(
...
...
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