Skip ROCm test for MPZCH kernel #4271
Open
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
Summary: Pull Request resolved: pytorch#4214 X-link: facebookresearch/FBGEMM#1290 Opensource FBGEMM CUDA Kernel for MPZCH feature ### Major changes - Create a folder named `faster_hash` under the `fbgemm/fbgemmgpu/src` folder. - Copy the following files to the created folder from `fbsource/fbcode/caffe2/torch/fb/retrieval` - faster_hash.cpp - faster_hash.cu - common_utils.cuh - Revise the `faster_hash.cpp` - Change `namespace fb` to `namespace fbgemm_gpu`. - Comment out `using namespace torch::fb::turborec;` - Change `TORCH_LIBRARY_IMPL(fb, ...)` to `TORCH_LIBRARY_IMPL(fbgemm, ...)` - Fix namespace calling issue due to the namespace change. - Revise the `faster_hash.cu` - Change `namespace fb` to `namespace fbgemm_gpu`. - Change `TORCH_LIBRARY_IMPL(fb, ...)` to `TORCH_LIBRARY_IMPL(fbgemm, ...)` - Fix namespace calling issue due to the namespace change. - Revise the `common_utils.cuh` file - Change `namespace fb` to `namespace fbgemm_gpu`. - Add a BUCK file to compile the cpp and cuda library. - Copy the `faster_hash_test.py` file to the `fbgemm/fbgemm_gpu/test` folder. - Add a section in the BUCK file under the `test` folder for `python_unittest` of `faster_hash_test`. - In the `faster_hash_test.py` file - Load the `faster_hash` related libraries with `torch.ops.load` API. - Replace all the `torch.ops.fb` to `torch.ops.fbgemm`. - Following other test files to add `opensource` and `gpu availability` check. ### Questions - After refactorying, the API calls `torch.ops.create_zch_buffer`, `torch.ops.zero_collision_hash`, `torch.ops.fbgemm.zero_collision_hash`, and `torch.ops.fbgemm.create_zch_buffer` are all valid, while `torch.ops.create_zch_buffer` and `torch.ops.zero_collision_hash` may incur certain parameter mismatches. How to resolve this issue and disable the API calls without `fbgemm`? - How to integrate the refactoryed library into fbgemm so the test can call something like `from fbgemm_gpu import create_zch_buffer, zero_collision_hash`? Differential Revision: D75505020 Reviewed By: ionuthristodorescu
Differential Revision: D76045051
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
|
This pull request was exported from Phabricator. Differential Revision: D76045051 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Differential Revision: D76045051