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
CI: Specify libc and libstdcxx versions in conda environments #121556
Conversation
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/121556
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 1 PendingAs of commit 1a12269 with merge base 953c6c3 (): This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. [ghstack-poisoned]
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. [ghstack-poisoned]
.ci/docker/common/install_conda.sh
Outdated
# Also ensure sysroot is using a modern GLIBC to match system compilers | ||
as_jenkins conda create -n py_$ANACONDA_PYTHON_VERSION -y\ | ||
python="$ANACONDA_PYTHON_VERSION" \ | ||
sysroot_linux-64=2.17 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Since this script supports aarch64, this needs to change depend on uname -m
# libstdcxx from conda default channels are too old, we need GLIBCXX_3.4.30 | ||
# which is provided in libstdcxx 12 and up. | ||
conda_install libstdcxx-ng=12.3.0 -c conda-forge |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Problem with this statement, is that it adds conda-forge channel, which allows the rest of the packages to be installed from conda-forge going forward. But may be this is fine
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Merge failedReason: New commits were pushed while merging. Please rerun the merge command. Details for Dev Infra teamRaised by workflow job |
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
@pytorchbot cherry-pick --onto release/2.3 -c regression --fixes "glibc errors" |
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. Pull Request resolved: #121556 Approved by: https://github.com/isuruf, https://github.com/malfet (cherry picked from commit 7a53ded)
Cherry picking #121556The cherry pick PR is at #121929 and it is linked with issue glibc errors Details for Dev Infra teamRaised by workflow job |
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. Pull Request resolved: #121556 Approved by: https://github.com/isuruf, https://github.com/malfet
Without this we get mismatches between the GLIBC and GLIBCXX ABI used by conda packages vs pytorch. Pull Request resolved: #121556 Approved by: https://github.com/isuruf, https://github.com/malfet (cherry picked from commit 7a53ded) Co-authored-by: Peter Bell <peterbell10@live.co.uk>
Stack from ghstack (oldest at bottom):
Without this we get mismatches between the GLIBC and GLIBCXX ABI used
by conda packages vs pytorch.