Bug 1924231
| Summary: | PETSc requires specifc MPI version at compile time | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | david08741 |
| Component: | petsc | Assignee: | Antonio T. sagitter <trpost> |
| Status: | CLOSED RAWHIDE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | rawhide | CC: | trpost |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2021-02-06 23:33:00 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 1921129 | ||
Hi David. I have modified and rebuilt PETSc in Rawhide branch (https://src.fedoraproject.org/rpms/petsc/c/412a009d556300487759ba3613e45c86eadb3e8a?branch=rawhide); are those changes fine for your needs? Next step is to build PETSc-3.14.3 and push it in stable branch (f33) if needed. Thanks for the quick rebuild. Yes, that is great, thanks a lot. |
Description of problem: PETSc checks during the build that it is build against the same mpi version, against which PETSc was build. That means, whenever openmpi or mpich is update, petsc needs to be rebuild, or depending packages are FTBFS. I think it should be safe to remove the check, but I guess upstream wouldn't want to merge such a patch, and keeping patches around isn't nice. Otherwise an explicit dependency on the specific mpi version used to compile could be added to the devel sub-packages. That would ensure that it fails earlier, and you would get FTI notices if that happens. Version-Release number of selected component (if applicable): petsc-3.13.4-2.fc33.x86_64 How reproducible: always Steps to Reproduce: 1. Try to rebuild bout++ Actual results: /usr/include/mpich-i386/petsc/petscsys.h:213:6: error: #error "PETSc was configured with one MPICH mpi.h version but now appears to be compiling using a different MPICH mpi.h version" 213 | # error "PETSc was configured with one MPICH mpi.h version but now appears to be compiling using a different MPICH mpi.h version" | ^~~~~ Compiling rk4.cxx In file included from /usr/include/mpich-i386/petsc/petscbag.h:4, from /usr/include/mpich-i386/petsc/petsc.h:5, from ../../../../include/bout/petsclib.hxx:50, from imex-bdf2.hxx:47, from imex-bdf2.cxx:4: /usr/include/mpich-i386/petsc/petscsys.h:213:6: error: #error "PETSc was configured with one MPICH mpi.h version but now appears to be compiling using a different MPICH mpi.h version" 213 | # error "PETSc was configured with one MPICH mpi.h version but now appears to be compiling using a different MPICH mpi.h version" | ^~~~~ Expected results: Builds as expected. Additional info: I hoped the mass rebuild would fix this at least temporary (as a new openmpi was build around 14 days ago) but then two days ago mpich was updated, and I am back to not being able to rebuild bout++