Repository navigation
pkgdb: discover hdf5 and mpi variant subdirectories under dpkg - #49
Merged
Merged
Conversation
This was referenced Jul 13, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
On Debian/Ubuntu, hdf5 and openmpi tuck their headers and libraries into variant subdirectories
that
_buildDpkgPackageDatabasedid not probe:libhdf5-dev, serial):hdf5.hin/usr/include/hdf5/serial/, libs in/usr/lib/<multiarch>/hdf5/serial/libopenmpi-dev):mpi.hin/usr/lib/<multiarch>/openmpi/include/, libs in.../openmpi/lib/The emitter wrote only
<pkg>.dir, sohdf5.incpath/mpi.incpathfell back to the bare$(dir)/include(/usr/include) and the compiles failed withhdf5.h/mpi.hnot found.Fix
Give hdf5 and mpi bespoke emitter branches (like python/numpy/pybind11) that read the real locations
from
dpkg -L:hdf5.handlibhdf5.so, emithdf5.incpath/hdf5.libpathvia_dpkgAnchor.Emit
hdf5.parallel= the flavor (the header's enclosing dir:serial/openmpi/mpich) sothe existing
findstring mpigate inmake/extern/hdf5/init.mmfoldsmpiintohdf5.dependenciesfor parallel builds. The transitive edge stays owned by the make engine.mpi.h(the one on an/include/path, not the fortran module) andlibmpi.so,emit
mpi.incpath/mpi.libpath.Also corrects the stale
hdf5.parallelmarker hint (was "on/off"; the value is really the flavorserial/openmpi/mpich) and the default (off→serial).Verification
Builds pyre from scratch against apt packages in an
ubuntu:24.04container; hdf5/mpi resolve andthe full test suite passes.
The embedded clone of this logic in pyre's
packages/merlin/shells/MM.pygets the same fix in acompanion pyre PR.