Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

update support new device docs #6963

Merged
merged 3 commits into from
Dec 25, 2017
Merged

update support new device docs #6963

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
44cad80
update docs
QiJune Dec 25, 2017
c882e35
follow comments
QiJune Dec 25, 2017
15db7ad
fix typo
QiJune Dec 25, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 9 additions & 7 deletions doc/design/support_new_device.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,13 +25,14 @@ There are mainly three parts that we have to consider while integrating a new de

### Place and DeviceContext

Please remind that device and computing library are not one-to-one corresponding. A device can have a lot of computing libraries and a computing library can also support several devices.

#### Place
Fluid uses class [Place](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L55) to represent different devices and computing libraries. There are inheritance relationships between different kinds of `Place`.
Fluid uses class [Place](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L55) to represent the device memory where data is locating. If we add another device, we have to add corresponding `DevicePlace`.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

where data is locating -> where data is located

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done


```
| CPUPlace --> MKLDNNPlace
Place --| CUDAPlace --> CUDNNPlace
| CPUPlace
Place --| CUDAPlace
| FPGAPlace
```

Expand All @@ -43,7 +44,7 @@ typedef boost::variant<CUDAPlace, CPUPlace, FPGAPlace> Place;

#### DeviceContext

Fluid uses class [DeviceContext](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L30) to manage the resources in different hardwares, such as CUDA stream in `CDUADeviceContext`. There are also inheritance relationships between different kinds of `DeviceContext`.
Fluid uses class [DeviceContext](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L30) to manage the resources in different libraries, such as CUDA stream in `CDUADeviceContext`. There are also inheritance relationships between different kinds of `DeviceContext`.


```
Expand Down Expand Up @@ -106,7 +107,7 @@ template <typename Place>
size_t Used(Place place);
```

To implementing these interfaces, we have to implement MemoryAllocator for different Devices
To implementing these interfaces, we have to implement MemoryAllocator for different Devices.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

implementing -> implement

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Done



#### Tensor
Expand Down Expand Up @@ -243,6 +244,7 @@ REGISTER_OP_CUDA_KERNEL(
Generally, we will impelement OpKernel for all Device/Library of an Operator. We can easily train a Convolutional Neural Network in GPU. However, some OpKernel is not sutibale on a specific Device. For example, crf operator can only run on CPU, whereas most other operators can run at GPU. To achieve high performance in such circumstance, we have to switch between different Device/Library.


We will discuss how to implement an efficient OpKernel switch policy.
For more details, please refer to following docs:

- TBD
- operator kernel type [doc](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/operator_kernel_type.md)
- switch kernel [doc](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md)