[Phi] Support cudnn kernel moving & move softmax kernels #39547
Merged
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! |
… pten/support_cudnn_kernel_moving
chenwhql
changed the title
…/chenwhql/Paddle into pten/support_cudnn_kernel_moving
…/chenwhql/Paddle into pten/support_cudnn_kernel_moving
… pten/support_cudnn_kernel_moving
chenwhql
changed the title
zyfncg
previously approved these changes
Feb 24, 2022
| public: | ||
| using Type = float; | ||
| }; | ||
|
|
There was a problem hiding this comment.
这里和 paddle/fluid/operators/amp/fp16_type_traits.h 重复了，后边可以考虑合并。
There was a problem hiding this comment.
嗯嗯，这是移过来的，后续不推荐使用fluid下的paddle/fluid/operators/amp/fp16_type_traits.h，可以删除这个文件
zyfncg
approved these changes
Feb 25, 2022
YuanRisheng
approved these changes
Feb 25, 2022
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Outdated suggestions cannot be applied.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
New features
PR changes
Others
Describe
[Phi] Support cudnn kernel moving
支持phi注册GPUDNN的kernel，迁移softmax op kernel作为验证
由于并没有单独的GPUDNNContext，这给目前的体系引入了一些问题，当我们using CUDNNContext = GPUContext时，如果gpu和cudnn的kernel函数命名是一样的，会导致实例化冲突，例如：
SoftmaxKernel<float, GPUContext>和SoftmaxKernel<float, GPUDNNContext>实质上是一个实例化，但它们的实现是不一样的，这导致编译时，GPUDNN的kernel实例化被覆盖了，实际上GPUDNN注册的是GPU的kernel，在运行时引入了计算结果错误。由于目前GPUDNNContext的设计并不明确，和现在的GPUContext本质上就是一样的，所以暂时通过更改函数名解决冲突，GPUDNN的kernel命名为SoftmaxGPUDNNKernel。
但这样引入的问题是，通过
SoftmaxKernel<T, Context>将无法调用到SoftmaxGPUDNNKernel，gpudnn的kernel自成一个小体系，但影响不大，cudnn kernel的数目极少（个位数）。这里后续需要确认下，SoftmaxKernel原先的cuda kernel是否还有必要保留，如果没有必要的话，将其删除，并将cudnn kernel移到gpu下会合适一些。TODO：