Explicit array bounds #458
Merged
Conversation
This file contains 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
matt-frey
added
bug
This reverts commit 7f4cbf6.
…c into fix-early-exit-in-nearest
…c into fix-early-exit-in-nearest
Fix MPI related issues
Read parcels in chunks
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Outdated suggestions cannot be applied.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.
There is a GCC issue that changes the bounds of allocatable arrays when calling subroutines with
dimension(..). See also https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97046 (and https://stackoverflow.com/questions/63824065/lbound-of-an-array-changes-after-call-to-mpi-gatherv-when-using-mpi-f08-module).As the latest MPI module
mpi_f08heavily usesdimension(..)in its routines, we get wrong bounds after MPI calls, although they are correct before. According to the link I posted, the issue affectsgcc/9.2.0andgcc/10.2.0. It worked on my laptop as I am usinggcc/9.3.0. One workaround is to specify the bounds explicitly in the MPI calls.This PR further changes
MPI_INTtoMPI_INTEGERand removes the calls toMPI_Win_syncas we get the errorWrong synchronization of RMA callson Archer2. These calls are intended to synchronise the private and public RMA memory. However, in a unified memory model, these are not necessary. We have put them in for safety reasons only.Plus changes in #459 and #460.