NVIDIA / NVIDIA/cuvs

[CI] NEIGHBORS_MG_TEST flaky

Open
#2,588 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug
Dominant language
Cuda
Stars
854
Forks
236
Avg merge
3d 3h
Merged PRs (30d)
62

Description

Reported here https://github.com/NVIDIA/cuvs/pull/2584#issuecomment-5623785637

[ RUN      ] AnnMGTest/AnnMGTestF_float.AnnMG/8
terminate called after throwing an instance of 'raft::cuda_error'
  what():  CUDA error encountered at: file=/tmp/conda-bld-output/bld/rattler-build_libcuvs-headers/host_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_/include/raft/util/cudart_utils.hpp line=139 function=void raft::copy(Type*, const Type*, size_t, rmm::_RMM_26_12::cuda_stream_view) [with Type = long int; size_t = long unsigned int]: call='cudaMemcpyAsync(dst, src, len * sizeof(Type), cudaMemcpyDefault, stream.get())', Reason=cudaErrorMisalignedAddress:misaligned address
Obtained 16 stack frames
#1 in /opt/conda/envs/test/bin/gtests/libcuvs/../../../lib/libcuvs.so(+0x6707f6) [0x7dc83781f7f6]
#2 in /opt/conda/envs/test/bin/gtests/libcuvs/../../../lib/libcuvs.so(+0x1c6d413) [0x7dc838e1c413]
#3 in /opt/conda/envs/test/bin/gtests/libcuvs/../../../lib/libcuvs.so(+0x1c6d69f) [0x7dc838e1c69f]
#4 in /opt/conda/envs/test/bin/gtests/libcuvs/../../../lib/libgomp.so.1: GOMP_parallel +0x46 [0x7dc82717ef36]
#5 in /opt/conda/envs/test/bin/gtests/libcuvs/../../../lib/libcuvs.so(+0x1c755d6) [0x7dc838e245d6]
#6 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x70796) [0x5eed6ff16796]
#7 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x9752c) [0x5eed6ff3d52c]
#8 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x97801) [0x5eed6ff3d801]
#9 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x97ba7) [0x5eed6ff3dba7]
#10 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x98026) [0x5eed6ff3e026]
#11 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x9eac3) [0x5eed6ff44ac3]
#12 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x9efb1) [0x5eed6ff44fb1]
#13 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x2eddf) [0x5eed6fed4ddf]
#14 in /usr/lib/x86_64-linux-gnu/libc.so.6(+0x2a601) [0x7dc826b5e601]
#15 in /usr/lib/x86_64-linux-gnu/libc.so.6: __libc_start_main +0x88 [0x7dc826b5e718]
#16 in ./../../..//bin/gtests/libcuvs/NEIGHBORS_MG_TEST(+0x2eece) [0x5eed6fed4ece]

CMake Error at run_gpu_test.cmake:26 (execute_process):
  execute_process failed command indexes:

    1: "Abnormal exit with child return code: Subprocess aborted"
....
The following tests FAILED:
	 32 - NEIGHBORS_MG_TEST (Failed)
/__w/cuvs/cuvs

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing NEIGHBORS_MG_TEST, especially AnnMGTest/AnnMG/8, using the failure output in the issue. Inspect run_gpu_test.cmake and the referenced stack trace to locate the misaligned CUDA copy. Done means the test no longer flakes or aborts with cudaErrorMisalignedAddress in CI.

Written by the indexing model from the issue text.

Assessment

Tech stack
cmake
Domain
ci-cd, testing-qa
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.