Skip to content

Bump Score-P to v10.1 & OPARI2 to v2.0.10 - #2665

Open
Thyre wants to merge 2 commits into
openhpc:4.xfrom
Thyre:scorep-10.1
Open

Bump Score-P to v10.1 & OPARI2 to v2.0.10#2665
Thyre wants to merge 2 commits into
openhpc:4.xfrom
Thyre:scorep-10.1

Conversation

@Thyre

@Thyre Thyre commented Aug 9, 2026

Copy link
Copy Markdown
Contributor

No description provided.

Copilot AI lite review requested due to automatic review settings August 9, 2026 18:02
@github-actions

github-actions Bot commented Aug 9, 2026 •
edited
Loading

Copy link
Copy Markdown

📦 Package Count Analysis Results

Environment: UBI 10 Container
Target Branch: 4.x
Version: OpenHPC 4.x
Overall Status: ✅ Analysis completed successfully

🏭 Factory Repositories

Status: ✅ Factory repositories analysis completed successfully

Release Base OS aarch64 x86_64 noarch Total
4.2 EL_10 54 95 12 161
4.2 openEuler_24.03 48 55 12 115
Total All 102 150 24 276

Analysis performed by OpenHPC Package Count CI
Commit: edf2de63
Workflow: 32129743496

@github-actions

github-actions Bot commented Aug 9, 2026 •
edited
Loading

Copy link
Copy Markdown

📊 Build Order Analysis Results

Environment: AlmaLinux 10 Container
Total Spec Files Analyzed: 89
Status: ✅ Build order analysis completed successfully

📋 Build Order Output

📦 Build Order (84 packages)

➕ Show remaining 69 packages

Analysis performed by OpenHPC Build Order CI
Commit: edf2de63
Workflow: 32129743575

Copilot AI reviewed Aug 9, 2026
View reviewed changes

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull request overview

Updates the OpenHPC packaging and validation for the Score-P performance tool stack by bumping Score-P and OPARI2 to newer upstream releases, and removing now-obsolete Score-P v10.0 workaround artifacts.

Changes:

  • Bump Score-P RPM spec from 10.0 to 10.1 and drop the Intel MPI linker workaround patch and related install-time symlink workaround.
  • Bump OPARI2 RPM spec from 2.0.9 to 2.0.10.
  • Update the Score-P test to validate the presence of the current license documentation file (README.LICENSES.md) instead of the older COPYING path.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File Description
tests/perf-tools/scorep/tests/test_module Updates license-file availability check for the new Score-P doc layout.
components/serial-libs/opari2/SPECS/opari2.spec Bumps OPARI2 packaged version to 2.0.10.
components/perf-tools/scorep/SPECS/scorep.spec Bumps Score-P to 10.1 and removes v10.0-only patch/symlink workarounds.
components/perf-tools/scorep/SOURCES/Score-P-10.0_fix-intel-mpi-linker-error.patch Removes obsolete v10.0 Intel MPI linker workaround patch.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Comment thread tests/perf-tools/scorep/tests/test_module Outdated
Copilot AI review requested due to automatic review settings August 9, 2026 18:05
Copilot AI reviewed Aug 9, 2026
View reviewed changes

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

@github-actions

github-actions Bot commented Aug 9, 2026 •
edited
Loading

Copy link
Copy Markdown

🚀 CCache Statistics

Configuration 🐧 RHEL 🦊 openEuler
🖥️ x86_64 / gnu15 🟡 44.27% 🟡 43.55%
🖥️ x86_64 / intel 🟡 46.29% -
💪 aarch64 / gnu15 🟡 33.59% 🟡 33.55%
📊 Detailed Statistics

openEuler (aarch64)

Cacheable calls:    2498 / 5615 (44.49%)
  Hits:              838 / 2498 (33.55%)
    Direct:          437 /  838 (52.15%)
    Preprocessed:    401 /  838 (47.85%)
  Misses:           1660 / 2498 (66.45%)
Uncacheable calls:  3117 / 5615 (55.51%)
Local storage:
  Cache size (GiB):  0.0 /  5.0 ( 0.76%)
  Hits:              838 / 2498 (33.55%)
  Misses:           1660 / 2498 (66.45%)

openEuler (x86_64)

Cacheable calls:    3502 / 7623 (45.94%)
  Hits:             1525 / 3502 (43.55%)
    Direct:          711 / 1525 (46.62%)
    Preprocessed:    814 / 1525 (53.38%)
  Misses:           1977 / 3502 (56.45%)
Uncacheable calls:  4121 / 7623 (54.06%)
Local storage:
  Cache size (GiB):  0.0 /  5.0 ( 0.91%)
  Hits:             1525 / 3502 (43.55%)
  Misses:           1977 / 3502 (56.45%)

RHEL (aarch64/gnu15)

Cacheable calls:    2495 / 5585 (44.67%)
  Hits:              838 / 2495 (33.59%)
    Direct:          437 /  838 (52.15%)
    Preprocessed:    401 /  838 (47.85%)
  Misses:           1657 / 2495 (66.41%)
Uncacheable calls:  3090 / 5585 (55.33%)
Local storage:
  Cache size (GiB):  0.0 /  5.0 ( 0.76%)
  Hits:              838 / 2495 (33.59%)
  Misses:           1657 / 2495 (66.41%)

RHEL (x86_64/gnu15)

Cacheable calls:    3558 / 7710 (46.15%)
  Hits:             1575 / 3558 (44.27%)
    Direct:          758 / 1575 (48.13%)
    Preprocessed:    817 / 1575 (51.87%)
  Misses:           1983 / 3558 (55.73%)
Uncacheable calls:  4152 / 7710 (53.85%)
Local storage:
  Cache size (GiB):  0.0 /  5.0 ( 0.91%)
  Hits:             1575 / 3558 (44.27%)
  Misses:           1983 / 3558 (55.73%)

RHEL (x86_64/intel)

Cacheable calls:    3813 / 13467 (28.31%)
  Hits:             1765 /  3813 (46.29%)
    Direct:          825 /  1765 (46.74%)
    Preprocessed:    940 /  1765 (53.26%)
  Misses:           2048 /  3813 (53.71%)
Uncacheable calls:  9654 / 13467 (71.69%)
Local storage:
  Cache size (GiB):  0.0 /   5.0 ( 0.64%)
  Hits:             1765 /  3813 (46.29%)
  Misses:           2048 /  3813 (53.71%)

🤖 Generated from workflow run 32129743605

@Thyre

Thyre commented Aug 9, 2026

Copy link
Copy Markdown
Contributor Author

Some tests are failing with:

2026-08-09T18:31:59.4280201Z sh: line 1: /opt/ohpc/pub/libs/gnu15/opari2/2.0.9/bin/opari2-config: No such file or directory
2026-08-09T18:31:59.4280846Z main_1786300266_922612.input.c:2:10: fatal error: opari2/pomp2_lib.h: No such file or directory
2026-08-09T18:31:59.4281322Z     2 | #include <opari2/pomp2_lib.h>
2026-08-09T18:31:59.4281594Z       |          ^~~~~~~~~~~~~~~~~~~~
2026-08-09T18:31:59.4281850Z compilation terminated.

I'm guessing Score-P was built with the old OPARI2 in the CI, but then the new version of this PR was installed, causing the error?
I can bump the minimum required version to v2.0.10, but it's not really required.

@Thyre
Thyre marked this pull request as ready for review August 9, 2026 18:42
@github-actions

github-actions Bot commented Aug 9, 2026 •
edited
Loading

Copy link
Copy Markdown

Test Results

 53 files  + 24   53 suites  +24   6m 27s ⏱️ + 5m 36s
251 tests +192  245 ✅ +192  6 💤 ±0  0 ❌ ±0 
361 runs  +256  353 ✅ +256  8 💤 ±0  0 ❌ ±0 

Results for commit 04ae8ca. ± Comparison against base commit c4c064d.

♻️ This comment has been updated with latest results.

@Thyre

Thyre commented Aug 10, 2026

Copy link
Copy Markdown
Contributor Author

@adrianreber I probably need to split OPARI2 and Score-P into two PRs again, right?

@adrianreber

adrianreber commented Aug 10, 2026

Copy link
Copy Markdown
Member

@adrianreber I probably need to split OPARI2 and Score-P into two PRs again, right?

In theory that is no longer necessary. There is a step to calculate the right order. I can take a look later today.

@adrianreber

adrianreber commented Aug 10, 2026

Copy link
Copy Markdown
Member

Build order analysis failed on openEuler:

2026-08-09 18:11:02,018 About to run command misc/build_order.sh
2026-08-09 18:11:09,436 misc/build_order.sh failed, using original spec file order
2026-08-09 18:11:09,437 --> Building RPM from spec file components/perf-tools/scorep/SPECS/scorep.spec
2026-08-09 18:11:09,437 About to run command misc/get_source.sh scorep.spec
2026-08-09 18:11:12,992 About to run command misc/build_srpm.sh components/perf-tools/scorep/SPECS/scorep.spec gnu15 openmpi5
2026-08-09 18:11:13,462 /github/home/rpmbuild/SRPMS/scorep-gnu15-openmpi5-ohpc-10.1-1.src.rpm
2026-08-09 18:11:13,462 About to run command dnf -y builddep /github/home/rpmbuild/SRPMS/scorep-gnu15-openmpi5-ohpc-10.1-1.src.rpm
'lustre-client-ohpc-kmp'
'lustre-client-ohpc-tests-kmp'
Traceback (most recent call last):
  File "/__w/ohpc/ohpc/misc/build_order.py", line 110, in <module>
    v[i] = spec_dict[v[i]]
           ~~~~~~~~~^^^^^^
./components/perf-tools/scorep/SPECS/scorep.spec
Source42: OHPC_macros

This needs to be fixed. It works on RHEL.

On RHEL based clones the test failures is because of mvapich tests are failing. They work sometimes but sometimes they don't.

You can add a commit that removes the lines 145-147 in tests/ci/setup_slurm_and_run_tests.sh. This should fix the failures on RHEL and clones.

If you are motivated you can also take a look at the build order analysis error on openEuler. As we do not build the lustre package for a long time you could probably just skip any lustre based build order analysis. There is already some code to ignore errors. Let me know if you do not want to deal with this then I can do it.

@Thyre

Thyre commented Aug 11, 2026

Copy link
Copy Markdown
Contributor Author

Build order analysis failed on openEuler:

2026-08-09 18:11:02,018 About to run command misc/build_order.sh
2026-08-09 18:11:09,436 misc/build_order.sh failed, using original spec file order
2026-08-09 18:11:09,437 --> Building RPM from spec file components/perf-tools/scorep/SPECS/scorep.spec
2026-08-09 18:11:09,437 About to run command misc/get_source.sh scorep.spec
2026-08-09 18:11:12,992 About to run command misc/build_srpm.sh components/perf-tools/scorep/SPECS/scorep.spec gnu15 openmpi5
2026-08-09 18:11:13,462 /github/home/rpmbuild/SRPMS/scorep-gnu15-openmpi5-ohpc-10.1-1.src.rpm
2026-08-09 18:11:13,462 About to run command dnf -y builddep /github/home/rpmbuild/SRPMS/scorep-gnu15-openmpi5-ohpc-10.1-1.src.rpm
'lustre-client-ohpc-kmp'
'lustre-client-ohpc-tests-kmp'
Traceback (most recent call last):
  File "/__w/ohpc/ohpc/misc/build_order.py", line 110, in <module>
    v[i] = spec_dict[v[i]]
           ~~~~~~~~~^^^^^^
./components/perf-tools/scorep/SPECS/scorep.spec
Source42: OHPC_macros

This needs to be fixed. It works on RHEL.

On RHEL based clones the test failures is because of mvapich tests are failing. They work sometimes but sometimes they don't.

You can add a commit that removes the lines 145-147 in tests/ci/setup_slurm_and_run_tests.sh. This should fix the failures on RHEL and clones.

If you are motivated you can also take a look at the build order analysis error on openEuler. As we do not build the lustre package for a long time you could probably just skip any lustre based build order analysis. There is already some code to ignore errors. Let me know if you do not want to deal with this then I can do it.

Thanks for the pointer. I probably won't have the time to look into why the build order analysis fails on openEuler, due to courses and upcoming vacation.
Should this still be an issue after my vacation, I may be able to take a look then.

@adrianreber

adrianreber commented Aug 11, 2026

Copy link
Copy Markdown
Member

Let's ignore the openeuler CI failure for now. It works locally for me in a container. Not sure why it fails in CI.

Thyre added 2 commits August 18, 2026 13:02
@Thyre
Bump Score-P to v10.1
a311849 
Signed-off-by: Jan Andre Reuter <jan@zyten.de>
@Thyre
Bump OPARI2 to v2.0.10
04ae8ca 
Signed-off-by: Jan Andre Reuter <jan@zyten.de>
Copilot AI review requested due to automatic review settings August 18, 2026 11:02
Copilot AI reviewed Aug 18, 2026
View reviewed changes

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

Copilot code review Copilot
Copilot review effort
Applies to this pull request for everyone.Learn more about Copilot code review.
Copilot left review comments

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@Thyre @adrianreber