Cambricon / mlu-ops

Efficient operation implementation based on the Cambricon Machine Learning Unit (MLU) .
MIT License
103 stars 102 forks source link

[Feature](mlu-ops): removing error throwing in mluops library. #1038

Closed DanieeelLiu closed 2 months ago

DanieeelLiu commented 5 months ago

Thanks for your contribution and we appreciate it a lot. :rocket::rocket:

1. Motivation

removing error throwing in mluops compiled library.

2. Modification

kernels/utils/cnnl_helper.cpp

Are new test cases added? If so, please post the corresponding generator-PR link here.

3. Test Report

If you want to know how to do operator testing, you can see GTest-User-Guide-zh.

3.1 Modification Details

3.1.1 Accuracy Acceptance Standard

For static threshold standard details, see: MLU-OPS™ Accuracy Acceptance Standard.

3.1.2 Operator Scheme checklist

3.2 Accuracy Test

3.2.1 Accuracy Test

If you have checked the following items, please tick the relevant box.

3.2.2 Parameter Check

Test Point-1: `CALL_CNNL return an error when parameter is nullptr.

2024-5-29 11:16:13] [CNNL] [Error]:[cnnlCreateTransposeDescriptor] Check failed: desc != NULL. 
[2024-5-29 11:16:13] [MLUOP] [Error]:CNNL_HELPER: Internal cnnl api call error accured.
[2024-5-29 11:16:13] [MLUOP] [Error]:"CNNL_STATUS_BAD_PARAM in cnnlCreateTransposeDescriptor(nullptr) at /projs/platform/liuduanhui/mluops-dev/mlu-ops/kernels/three_nn_forward/three_nn_forward.cpp line: 53"
[2024-5-29 11:16:13] [MLUOP] [Error]:"MLUOP_STATUS_BAD_PARAM in mluOpGetThreeNNForwardWorkspaceSize(handle_, known_desc, &workspace_size_)"

3.4 Summary Analysis

ci passed normally.