8000 Fix OpenMP failure on ARM by jbajic · Pull Request #21772 · arangodb/arangodb · GitHub
[go: up one dir, main page]
Skip to content

Fix OpenMP failure on ARM #21772

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.

Sign up for GitHub

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

Jump to bottom
Open
wants to merge 50 commits into
base: devel
Choose a base branch
from
Open

Fix OpenMP failure on ARM #21772

wants to merge 50 commits into from
+29 −48

Conversation

jbajic
Copy link
Contributor
@jbajic jbajic commented May 19, 2025 •
edited
Loading

Scope & Purpose

After Circle CI upgraded their machines to use cgroupv2, we have experienced test failures on ARM machines testing the vector index. This is an example of one of the stacktraces:

Assertion failure at kmp_affinity.cpp(4440): __kmp_avail_proc == __kmp_topology->get_num_hw_threads().
OMP: Error #13: Assertion failure at kmp_affinity.cpp(4440).
OMP: Hint Please submit a bug report with this message, compile and run commands used, and machine configuration info including native compiler and operating system versions. Faster response will be obtained by including all program sources. For information on submitting this issue, please see https://github.com/llvm/llvm-project/issues/.

There were attempts to disable the KMP_AFFINITY and limit the number of threads and cores used, but to no avail. Unfortunately, the LLVM OpenMP implementation issue still has not been resolved. It is unclear if this is an LLVM issue, or limitations of cgroupv2 and lxcfs to limit the resources in not properly propagated. Using the GCC OpenMP implementation fixes the failures.

  • 💩 Bugfix
  • 🍕 New feature
  • 🔥 Performance improvement
  • 🔨 Refactoring/simplification

Checklist

  • Tests
    • Regression tests
    • C++ Unit tests
    • integration tests
    • resilience tests
  • 📖 CHANGELOG entry made
  • 📚 documentation written (release notes, API changes, ...)
  • Backports
    • Backport for 3.12.0: (Please link PR)
    • Backport for 3.11: (Please link PR)
    • Backport for 3.10: (Please link PR)

Related Information

(Please reference tickets / specification / other PRs etc)

  • Docs PR:
  • Enterprise PR:
  • GitHub issue / Jira ticket:
  • Design document:

jbajic added 30 commits April 3, 2025 10:49
@jbajic
Reduce OMP threading
6e76df9
@jbajic
Disable kmp affinity
dcae2c5
@jbajic
Disable on openmp
9627409
@jbajic
Disable KMP_AFFINITY and LIBOMP_AFFINITY
bcab65f
@jbajic
Reduce omp number of threads
9699c84
@jbajic
Print cpu stats
8a6e620
@jbajic
Use KMP_DEVICE_THREAD_LIMIT
d5b8823
@jbajic
More verbose
bf0c5a1
@jbajic
Check restore env
794011e
@jbajic
Update env variables
d565324
@jbajic
Merge branch 'devel' into fix/arm-failure
f5cec0b
@jbajic
Enable tests
76ba626
@jbajic
Enable vector index tests
b884a99
@jbajic
Clear string
9d81206
@jbajic
Print omp variables
02fd177
@jbajic
Fix dependency
83d1504
@jbajic
Add flags only for arm
cc938a4
@jbajic
Fix arch var
d03933e
@jbajic
Refactor
1870ad1
@jbajic
Merge branch 'devel' into fix/arm-failure
8661867
@jbajic
Fix
8360527
@jbajic
Disable kmp
abfa5d5
@jbajic
Try testing libomp
0f0d8d1
@jbajic
Clone arangod for testing
4c941ae
@jbajic
Clone arango
13457e1
@jbajic
Fix path
14e555b
@jbajic
Fix
5741da6
@jbajic
Remove verbose
25468b4
@jbajic
Remove test changes
045c48f
@jbajic
Check env
141d9b5
@cla-bot cla-bot bot added the cla-signed label May 19, 2025
@jbajic jbajic force-pushed the bug-fix/fix-openmp-arm-failure branch 2 times, most recently from db65be7 to 7a060be Compare May 19, 2025 12:30
@jbajic jbajic force-pushed the bug-fix/fix-openmp-arm-failure branch from 7a060be to 74d1034 Compare May 19, 2025 12:45
@jbajic jbajic self-assigned this May 20, 2025
@jbajic jbajic changed the title Bug fix/fix openmp arm failure Fix OpenMP failure on ARM May 20, 2025
@jbajic jbajic marked this pull request as ready for review May 20, 2025 14:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees

@jbajic jbajic

Labels
cla-signed
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@jbajic
0