[0-size XPU Tensor] Fix bugs for 0-size tensor handling in XPU kernels #77557
+322
−6
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
|
你的PR提交成功，感谢你对开源项目的贡献! |
tjujingzong
changed the title
Contributor
Author
|
/re-run all-failed |
1 similar comment
Contributor
Author
|
/re-run all-failed |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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 Category
Operator Mechanism
PR Types
Bug fixes
Description
利用 PaddleAPITest 发现以下 XPU Kernel 在处理 0-size Tensor（例如某些维度为 0、或元素数为 0 的输入）时支持不足，可能出现异常访问、报错或行为不一致。为此：
本次涉及并修复/加固的 XPU kernel（算子）如下：
fused_layernorm（paddle/phi/kernels/fusion/xpu/fused_layernorm_kernel.cc）activation（paddle/phi/kernels/xpu/activation_kernel.cc）atan（paddle/phi/kernels/xpu/atan_kernel.cc）bce_loss_grad（paddle/phi/kernels/xpu/bce_loss_grad_kernel.cc）bitwise（paddle/phi/kernels/xpu/bitwise.cc）conv（paddle/phi/kernels/xpu/conv_kernel.cc）cumprod（paddle/phi/kernels/xpu/cumprod_kernel.cc）diag（paddle/phi/kernels/xpu/diag_kernel.cc）embedding_grad（paddle/phi/kernels/xpu/embedding_grad_kernel.cc）gather_nd_grad（paddle/phi/kernels/xpu/gather_nd_grad_kernel.cc）huber_loss（paddle/phi/kernels/xpu/huber_loss_kernel.cc）label_smooth（paddle/phi/kernels/xpu/label_smooth_kernel.cc）logical（paddle/phi/kernels/xpu/logical_kernel.cc）pad（paddle/phi/kernels/xpu/pad_kernel.cc）prelu_grad（paddle/phi/kernels/xpu/prelu_grad_kernel.cc）scatter_grad（paddle/phi/kernels/xpu/scatter_grad_kernel.cc）sign（paddle/phi/kernels/xpu/sign_kernel.cc）take_along_axis_grad（paddle/phi/kernels/xpu/take_along_axis_grad_kernel.cc）