[BIG tensor]Fix the infinite loop issue in argsort. #74434
Merged
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 Outdated suggestions cannot be applied.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
Pcard-73145
在以下case argsort会导致hang住:
在 [PHI] Fixed argsort big tensor bug #72712 中实现了拆分为多个 cub kernel;但拆分计算batch时每次参与 sort 的元素不超过 2^30 个
在被sort的维度大于2^30时,(element_per_call / segment_size)为0,导致死循环
在被sort的维度大于2^31时,调用
DeviceSegmentedRadixSort::SortPairs依旧有问题,参考torch处理,对该情况添加报错信息。https://github.com/pytorch/pytorch/blob/aeb5321b6360c899808d3461789b3bbd6265756e/aten/src/ATen/native/cuda/Sort.cpp#L63