fix: correct DDP handling and backward gradients - #214
Open
chen2021673 wants to merge 1 commit into
Open
Conversation
- honor DDP reduction configuration and validate process group devices - fix elementwise and CUDA linear backward correctness - remove the unsupported distributed optimizer test - clean up related build diagnostics and test documentation
chen2021673
commented
Aug 31, 2026
| EXPECT_TRUE(state.contains("adam.v.first")); | ||
| } | ||
|
|
||
| TEST_P(OptimizerParameterNamesTest, DistributedOptimizerPropagatesNamesToShardOptimizer) { |
Contributor
Author
There was a problem hiding this comment.
当前测试体系不支持多进程，所以这个测试实际上被跳过了，没有真正运行
chen2021673
commented
Aug 31, 2026
|
|
||
| // Fast path: no broadcast — grad_a and grad_b are both just grad_output | ||
| if (a_dims_ == b_dims_) { | ||
| return {grad_output, grad_output}; |
Contributor
Author
There was a problem hiding this comment.
这里有问题，两个返回值指向同一个对象，梯度累积会出错。删掉。
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Fix several backend-independent correctness issues in DDP, autograd, and CUDA linear backward.
Changes
average_in_collectiveacross DDP reduction pathsExpbackward and avoid aliased gradients inAddbackward