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.

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

Fix validation workflow on aarch64 with conda 23.11.0 and GLIBC_2.25 #1645

Merged
merged 7 commits into from
Dec 16, 2023
Merged

Fix validation workflow on aarch64 with conda 23.11.0 and GLIBC_2.25 #1645

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
1a6a45b
Debug aarch64 clone
huydhn Dec 14, 2023
96abef8
Merge branch 'main' into debug-aarch64-clone
huydhn Dec 14, 2023
f329da5
Debug
huydhn Dec 15, 2023
0241cf6
Fix validation workflow with conda 23.11.0 and GLIBC_2.25
huydhn Dec 15, 2023
d6ad9cb
Gate the change on linux-aarch64 and keep the old LD_LIBRARY_PATH
huydhn Dec 15, 2023
fa7e4c3
Try to unset LD_LIBRARY_PATH in the workflow instead
huydhn Dec 15, 2023
07e5cd9
Fix copy/paste typo
huydhn Dec 15, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .github/scripts/validate_binaries.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,6 +39,12 @@ else
${PWD}/check_binary.sh
fi

# NB: The latest conda 23.11.0 pulls in some dependencies of conda-libmamba-solver that
# require GLIBC_2.25, which is not available in the current aarch64 image causing the
# suqsequence git command to fail. Basically, they don't work with CentOS 7 which AML 2
# is based on https://github.com/ContinuumIO/anaconda-issues/issues/12822
unset LD_LIBRARY_PATH

Copy link
Contributor

@atalman atalman Dec 15, 2023 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This logic probably better done here: https://github.com/pytorch/builder/blob/main/.github/workflows/validate-aarch64-linux-binaries.yml#L99

It will apply only for aarch64 builds

if [[ ${TARGET_OS} == 'windows' ]]; then
python ./test/smoke_test/smoke_test.py ${TEST_SUFFIX}
else
Expand Down
5 changes: 4 additions & 1 deletion test/smoke_test/smoke_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -258,7 +258,10 @@ def smoke_test_modules():
if module["repo"]:
if not os.path.exists(f"{cwd}/{module['repo_name']}"):
print(f"Path does not exist: {cwd}/{module['repo_name']}")
subprocess.check_output(f"git clone --depth 1 {module['repo']}", stderr=subprocess.STDOUT, shell=True)
try:
subprocess.check_output(f"git clone --depth 1 {module['repo']}", stderr=subprocess.STDOUT, shell=True)
except subprocess.CalledProcessError as exc:
raise RuntimeError(f"Cloning {module['repo']} FAIL: {exc.returncode} Output: {exc.output}") from exc
try:
smoke_test_command = f"python3 {module['smoke_test']}"
if target_os == 'windows':
Expand Down
Loading