DAOS-16722 client: to intercept PMPI_Init() in libpil4dfs

[wiliamhuang]

Intercept PMPI_Init() to avoid calling daos_init() if MPI_Init() is intercepted by other library (like darshan and mpip).

Features: pil4dfs

Required-githooks: true
Skipped-githooks: codespell

Before requesting gatekeeper:

• Two review approvals and any prior change requests have been resolved.
• Testing is complete and all tests passed or there is a reason documented in the PR why it should be force landed and forced-landing tag is set.
• Features: (or Test-tag*) commit pragma was used or there is a reason documented that there are no appropriate tags for this PR.
• Commit messages follows the guidelines outlined here.
• Any tests skipped by the ticket being addressed have been run and passed in the PR.

Gatekeeper:

• You are the appropriate gatekeeper to be landing the patch.
• The PR has 2 reviews by people familiar with the code, including appropriate owners.
• Githooks were used. If not, request that user install them and check copyright dates.
• Checkpatch issues are resolved. Pay particular attention to ones that will show up on future PRs.
• All builds have passed. Check non-required builds for any new compiler warnings.
• Sufficient testing is done. Check feature pragmas and test tags and that tests skipped for the ticket are run and now pass with the changes.
• If applicable, the PR has addressed any potential version compatibility issues.
• Check the target branch. If it is master branch, should the PR go to a feature branch? If it is a release branch, does it have merge approval in the JIRA ticket.
• Extra checks if forced landing is requested
  • Review comments are sufficiently resolved, particularly by prior reviewers that requested changes.
  • No new NLT or valgrind warnings. Check the classic view.
  • Quick-build or Quick-functional is not used.
• Fix the commit message upon landing. Check the standard here. Edit it to create a single commit. If necessary, ask submitter for a new summary.

[github-actions]

Ticket title is 'Hang in zeInit in pil4dfs interception library when preloading darshan'
Status is 'In Review'
Labels: 'ALCF,triaged,request_for_2.6.2'
https://daosio.atlassian.net/browse/DAOS-16722

[wiliamhuang]

All CI tests finished without issues.

[knard38]

For my understanding, we are expecting to get the reference on the original PMPI_Init function.
However, I do understand how we are sure of which function reference we will get, if there is several LD_PRELOAD on this last one. For example, if Darshan is redefining it, could it be possible that we get the reference on the darshan one.

[wiliamhuang]

The issue of deadlock is caused by calling daos_init() inside MPI runtime library's MPI_Init()/PMPI_Init(). We could avoid this issue as long as the PMPI_Init() from libpil4dfs is executed before the PMPI_Init() implemented in MPI runtime library. It does not not matter the PMPI_Init() from libpil4dfs or the PMPI_Init() from darshan is executed first.

[knard38]

OK, thanks for the explanation.

[mchaarawi]

let's hold off merging for now until i test on aurora since this is not extremelly time critical anymore.
i have access to darshan there so should be quick once i get an ECB.

[wiliamhuang]

let's hold off merging for now until i test on aurora since this is not extremelly time critical anymore. i have access to darshan there so should be quick once i get an ECB.

Thank you very much!