use elementwise to optimize gelu forward implementation on GPU #38188
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
Zjq9409
changed the title
AshburnLee
reviewed
Dec 17, 2021
paddle/fluid/operators/gelu_op.cu
Outdated
| std::vector<const framework::Tensor*> ins; | ||
| std::vector<framework::Tensor*> outs; | ||
| ins = {in}; | ||
| outs = {out}; |
Zjq9409
changed the title
ZzSean
reviewed
Dec 20, 2021
paddle/fluid/operators/gelu_op.cu
Outdated
| }; | ||
|
|
||
| template <typename T> | ||
| struct GeluNoApproximateFunctor { |
paddle/fluid/operators/gelu_op.cu
Outdated
| template <typename DeviceContext, typename T> | ||
| typename std::enable_if< | ||
| std::is_same<DeviceContext, platform::CUDADeviceContext>::value>::type | ||
| default_gelu_fw(const framework::ExecutionContext& ctx, |
There was a problem hiding this comment.
不需要重新写一个新的函数，直接特化一个CUDA版本的GeluKernel就可以了
paddle/fluid/operators/gelu_op.cu
Outdated
| using MT = typename details::MPTypeTrait<T>::Type; | ||
| inline HOSTDEVICE T operator()(T x) { | ||
| // this function is tanh approximation of gelu | ||
| MT mx = static_cast<MT>(x); |
There was a problem hiding this comment.
这里的命名可以参考activation_op.cu中的命名方式
Zjq9409
changed the title
Zjq9409
force-pushed
the
gelu_opt
branch
2 times, most recently
from
107f8ed
to
dac385f
Compare
ZzSean
reviewed
Dec 21, 2021
paddle/fluid/operators/gelu_op.cu
Outdated
| }; | ||
|
|
||
| template <typename DeviceContext, typename T> | ||
| class GeluCUDAKernel : public framework::OpKernel<T> { |
There was a problem hiding this comment.
特化的话不用改名字，还用GeluKernel就好，DeviceContext这个模版参数用CUDADeviceContext就可以
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Performance optimization
PR changes
OPs
Describe
使用elementwise优化gelu算子GPU前向计算，前向算子性能数据如下：
