Fix in stable_distribute and stable_distribute_inplace #25
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Outdated suggestions cannot be applied.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: Currently, the
mxx::stable_distribute*
functions may crash on a process with rank >= p if the input size on every process with rank >= p is zero and the global input size is less than the total number of processes. This is because the functions try to determine the index of the first element of the current partition using the result of themxx::exscan
which leads to a divide by zero error in the functionmxx::partition::blk_dist_buf::rank_of
.Solution: A process should try to compute its send counts only if the input size on the process is non-zero.
I have also added a test that was failing before the fix.