Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

modified code for using ROCm backened within the PyTorch framework #2198

Closed
wants to merge 6 commits into from
Closed

Conversation

xiabo123
Copy link
Contributor

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Related issue: #1949 #2035
This PR enables HIP as a backened for pytorch only.

Modification

  1. Added './mmcv/ops/csrc/pytorch' to the include directory as hipify in pytorch needed to hipify 'spconv_utils.h'.
  2. rocblas has ambiguities for 'is_complex' ultimately from the order of the include-files in some files. The workaround in this PR is to add '#include spconv_utils.h' before the other file to break this ambiguity. This may go against the style guideline, unfortunately. Albeit, this issue has been reported and will be addressed with future versions of ROCm.
  3. Added __shfl_down without mask in ./mmcv/ops/common/cuda/correlation_cuda.cuh since HIP does not have warp level granularity yet. The test in ./tests/test_ops/test_correlation.py passed.

BC-breaking (Optional)

N/A

Use cases (Optional)

Using MMCV with the pytorch framework on AMD GPUs.

Checklist

Before PR:

  • I have read and followed the workflow indicated in the CONTRIBUTING.md to create this PR.
  • Pre-commit or linting tools indicated in CONTRIBUTING.md are used to fix the potential lint issues.
  • Bug fixes are covered by unit tests, the case that causes the bug should be added in the unit tests.
  • New functionalities are covered by complete unit tests. If not, please add more unit test to ensure the correctness.
  • The documentation has been modified accordingly, including docstring or example tutorials.

After PR:

  • If the modification has potential influence on downstream or other related projects, this PR should be tested with some of those projects, like MMDet or MMCls.
  • CLA has been signed and all committers have signed the CLA in this PR.

@CLAassistant
Copy link

CLAassistant commented Aug 15, 2022

CLA assistant check
All committers have signed the CLA.

@zhouzaida
Copy link
Collaborator

closed by #1918

@zhouzaida zhouzaida closed this Sep 13, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants