Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
TonyTonyFun
Paddle
提交
8f8a6848
P
Paddle
项目概览
TonyTonyFun
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
0
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
0
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
0
Issue
0
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
未验证
提交
8f8a6848
编写于
7月 04, 2022
作者:
L
Leo Chen
提交者:
GitHub
7月 04, 2022
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
unify cpu context (#44049)
上级
2b0c22ad
变更
2
隐藏空白更改
内联
并排
Showing
2 changed file
with
9 addition
and
10 deletion
+9
-10
paddle/fluid/eager/api/generated/eager_generated/backwards/scale_node.cc
...ger/api/generated/eager_generated/backwards/scale_node.cc
+9
-9
paddle/fluid/platform/device_context.h
paddle/fluid/platform/device_context.h
+0
-1
未找到文件。
paddle/fluid/eager/api/generated/eager_generated/backwards/scale_node.cc
浏览文件 @
8f8a6848
...
@@ -117,20 +117,20 @@ void ScaleAPI(const paddle::experimental::Tensor& x,
...
@@ -117,20 +117,20 @@ void ScaleAPI(const paddle::experimental::Tensor& x,
paddle
::
platform
::
DeviceContextPool
::
Instance
();
paddle
::
platform
::
DeviceContextPool
::
Instance
();
if
(
expected_kernel_place
==
paddle
::
platform
::
CPUPlace
())
{
if
(
expected_kernel_place
==
paddle
::
platform
::
CPUPlace
())
{
auto
*
dev_ctx
=
dynamic_cast
<
paddle
::
platform
::
CPUDeviceContext
*>
(
auto
*
dev_ctx
=
pool
.
Get
(
expected_kernel_place
));
dynamic_cast
<
phi
::
CPUContext
*>
(
pool
.
Get
(
expected_kernel_place
));
if
(
!
dev_ctx
)
{
if
(
!
dev_ctx
)
{
PADDLE_THROW
(
paddle
::
platform
::
errors
::
Fatal
(
PADDLE_THROW
(
paddle
::
platform
::
errors
::
Fatal
(
"Cannot convert device_context to
CPUDevice
Context."
"Cannot convert device_context to
phi::CPU
Context."
"This indicates backend mismatch."
"This indicates backend mismatch."
"Pleas double check your expected place"
));
"Pleas double check your expected place"
));
}
}
ScaleDeviceDispatch
<
p
addle
::
platform
::
CPUDevice
Context
>
(
*
dense_tensor
.
get
(),
ScaleDeviceDispatch
<
p
hi
::
CPU
Context
>
(
*
dense_tensor
.
get
(),
*
dev_ctx
,
*
dev_ctx
,
scale
,
scale
,
bias
,
bias
,
bias_after_scale
,
bias_after_scale
,
dense_out
.
get
());
dense_out
.
get
());
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
#if defined(PADDLE_WITH_CUDA) || defined(PADDLE_WITH_HIP)
}
else
if
(
expected_kernel_place
==
paddle
::
platform
::
CUDAPlace
())
{
}
else
if
(
expected_kernel_place
==
paddle
::
platform
::
CUDAPlace
())
{
...
...
paddle/fluid/platform/device_context.h
浏览文件 @
8f8a6848
...
@@ -133,7 +133,6 @@ constexpr DeviceType kIPU = DeviceType::IPU;
...
@@ -133,7 +133,6 @@ constexpr DeviceType kIPU = DeviceType::IPU;
constexpr
DeviceType
kMLU
=
DeviceType
::
MLU
;
constexpr
DeviceType
kMLU
=
DeviceType
::
MLU
;
using
DeviceContext
=
phi
::
DeviceContext
;
using
DeviceContext
=
phi
::
DeviceContext
;
using
CPUDeviceContext
=
phi
::
CPUContext
;
template
<
typename
Place
>
template
<
typename
Place
>
struct
DefaultDeviceContextType
;
struct
DefaultDeviceContextType
;
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录