Optimize the computation of log_softmax #40612
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! |
Xreki
reviewed
Mar 16, 2022
| @@ -293,9 +288,10 @@ __global__ void WarpSoftmaxForward(T* softmax, | |||
| } | |||
|
|
|||
| // data src | |||
| AccT srcdata_raw[kBatchSize][kLoopsV][kVSize]; | |||
| AccT srcdata[kBatchSize][kLoopsV][kVSize]; | |||
| T src_tmp[kBatchSize][kLoopsV][kVSize]; | |||
There was a problem hiding this comment.
这样弄了3个寄存器数组，srcdata_raw、srcdata、src_tmp，含义从变量名很难区分。
|
|
||
| HOSTDEVICE explicit inline ExpSubFunctor(Tx y) : y((Tx)(y)) {} | ||
| struct ExpFunctor { | ||
| HOSTDEVICE explicit inline ExpFunctor() {} |
Xreki
approved these changes
Mar 16, 2022
liqitong-a
pushed a commit
to liqitong-a/Paddle
that referenced
this pull request
Mar 17, 2022
* Optimize the computation of log_softmax * modify the var name
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Outdated suggestions cannot be applied.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
Optimize the computation of log_softmax。
合入#38992后，部分case性能有回退，原因是在softmax的基础上实现log softmax，存在重复计算，优化完计算后，性能与之前持平