【PTen】Add dot and matmul grad kernel in pten #38713
Merged
Conversation
This file contains hidden or 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! |
chenwhql reviewed Jan 6, 2022
| | ||
| // TODO(chenweihang): add debug flags later | ||
| // TODO(chenweihang): deal with complex cases later | ||
| if (framework::IsComplexType(kernel_type.data_type_)) { |
Contributor
There was a problem hiding this comment.
这里是否可以使用pten_kernel的data type
Contributor Author
There was a problem hiding this comment.
由于传入的KernelSignature和Kernel数据结构都不具有data_type信息，所以需要使用kernel_type的数据
… pten_matmul_grad
YuanRisheng reviewed Jan 6, 2022
Comment on lines +1893 to +1898
| if (current_vector_size > start_idx) { | ||
| pt_kernel_context_->SetOutputWithoutSetRange(start_idx, {nullptr}); | ||
| } else { | ||
| pt_kernel_context_->EmplaceBackOutputWithoutSetRange({nullptr}); | ||
| } | ||
| end_idx = start_idx + 1; |
Comment on lines 360 to 365
| } else { | ||
| kernel_ctx->SetOutputWithoutSetRange( | ||
| start_idx + offset, | ||
| experimental::MakePtenTensorBaseFromVar( | ||
| outs_vector[offset]->MutableVar(), out_def)); | ||
| } |
Comment on lines 374 to 381
| } else { | ||
| if (current_vector_size > start_idx) { | ||
| kernel_ctx->SetOutputWithoutSetRange(start_idx, {nullptr}); | ||
| } else { | ||
| kernel_ctx->EmplaceBackOutputWithoutSetRange( | ||
| experimental::MakePtenTensorBaseFromVar( | ||
| outs_vector[offset]->MutableVar(), out_def)); | ||
| kernel_ctx->EmplaceBackOutputWithoutSetRange({nullptr}); | ||
| } | ||
| kernel_ctx->AssignOutputRange(std::make_pair(start_idx, start_idx + 1), | ||
| i); |
Contributor
There was a problem hiding this comment.
这里建议将这段逻辑挪到开头，使用iter == outs.end判断执行后直接continue，这样可以优化代码结构，减少if else逻辑嵌套便于代码维护与理解
| paddle::platform::complex<float>, | ||
| paddle::platform::complex<double>) {} | ||
| | ||
| PT_REGISTER_CTX_KERNEL(matmul_grad_grad, |
Contributor
There was a problem hiding this comment.
这里建议命名与函数一致：matmul_double_grad，alias_name也如此
… pten_matmul_grad
raindrops2sea approved these changes Jan 11, 2022
Contributor
| 怀疑该PR导致了linear反向性能下降一倍：
linear_2的nvprof结果如下：
linear_2的nvprof结果如下： 新的linear反向计算多了一个 |
Contributor Author
| 收到，我排查一下 |
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
5 participants
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
Others
PR changes
Others
Describe
迁移dot和matmul反向一阶、二阶和三阶计算kernel到pten中。
为了完成PTen反向计算kernel与框架的适配，本PR中还包括了以下几项调整：
OpProto信息，与前向Op的处理有所不同，因此本PR中调整了相应的处理逻辑并为迁移的每个反向kernel对应的Op配置GetExpectedPtenKernelArgs，该解决方案后续有可能会替换。paddle::optional<const DenseTensor&>来包裹此类可能为空的输入变量。为此也在pten中增加了对paddle::optional<const DenseTensor&>输入类型的支持。DenseTensor& operator=(DenseTensor&& other)。