-
-
Notifications
You must be signed in to change notification settings - Fork 438
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
lci pp: fix messages larger than INT_MAX #6566
base: master
Are you sure you want to change the base?
Conversation
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
For your information: The same problem exists with the MPI parcel port. |
Yeah. I will work on the MPI parcelport later. |
Coverage summary from CodacySee diff coverage on Codacy
Coverage variation details
Coverage variation is the difference between the coverage for the head and common ancestor commits of the pull request branch: Diff coverage details
Diff coverage is the percentage of lines that are covered by tests out of the coverable lines that the pull request added or modified: See your quality gate settings Change summary preferencesCodacy stopped sending the deprecated coverage status on June 5th, 2024. Learn more |
It will most likely exist in all parcelports... |
Since this problem most likely exists in all parcelports, would you mind moving the test to the |
@hkaiser I don't think the For example, this test tests.regressions.util.distributed.lci.zero_copy_parcels_1001_no_zero_copy_optimization The cmake configuration is
But the command line is still
The same happens for my |
Any additional (unparsed) argument passed to
should do the trick. |
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
25d15b6
to
a4b1932
Compare
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
Coverage summary from CodacySee diff coverage on Codacy
Coverage variation details
Coverage variation is the difference between the coverage for the head and common ancestor commits of the pull request branch: Diff coverage details
Diff coverage is the percentage of lines that are covered by tests out of the coverable lines that the pull request added or modified: See your quality gate settings Change summary preferencesCodacy stopped sending the deprecated coverage status on June 5th, 2024. Learn more |
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
b77b02e
to
a312910
Compare
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
a312910
to
032147f
Compare
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
@JiakunYan PP LCI now works here with the appropriate changes 👍 |
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
@hkaiser The new test failed clang-16-debug mystically but worked fine for all other configurations. Do you have an instinct why? From the test log, it seems the data transfer has been completed but the program was stuck in the finalization. Update: I failed to reproduce this issue on Rostam with my own build. |
Performance test reportHPX PerformanceComparison
Info
Comparison
Info
Comparison
Info
Explanation of Symbols
|
Coverage summary from CodacySee diff coverage on Codacy
Coverage variation details
Coverage variation is the difference between the coverage for the head and common ancestor commits of the pull request branch: Diff coverage details
Diff coverage is the percentage of lines that are covered by tests out of the coverable lines that the pull request added or modified: See your quality gate settings Change summary preferencesCodacy stopped sending the deprecated coverage status on June 5th, 2024. Learn more |
Thank @m-diers for reporting the bug. Test program provided by @m-diers
The LCI parcelport did not properly support messages larger than INT_MAX. This PR fixes the HPX side problem. The LCI side problem will be fixed by uiuc-hpc/lci#75 . I will update the default LCI autofetch tag soon.
This PR also adds a test for large messages. @hkaiser, let me know if I put the test program in the right directory.