Fix: MPI communication errors due to inconsistent R-coordinates in sparse matrix generation#6233
Merged
mohanchen merged 8 commits intodeepmodeling:developfrom May 29, 2025
Conversation
dyzheng
previously requested changes
May 26, 2025
mohanchen
approved these changes
May 29, 2025
dyzheng
pushed a commit
that referenced
this pull request
Sep 30, 2025
…atch problem during sparse matrix generation) (#6555) * Fixed the bug in memory statistics * Fix: MPI communication errors due to inconsistent R-coordinates in sparse matrix generation (#6233) * Fix MPI_ERR_TRUNCATE error * Add MPI compilation macro * Temp debug info print * Move sync operation into get_R_range --------- Co-authored-by: Taoni Bao <baotaoni@pku.edu.cn>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Linked Issue
Fix #6229
Description
When generating sparse matrices using
get_R_range, different MPI processes could produce inconsistent sets of R-coordinates. This led toMPI_ERR_TRUNCATEerrors duringMPI_Allreduceoperations because the data sizes across processes mismatched.My solution is:
Synchronize R-coordinates globally:
• Added
sync_all_R_coorto aggregate R-coordinates from all processes viaMPI_Allgatherv, ensuring a consistentall_R_coorset across all ranks.Fix MPI buffer size handling:
• Corrected buffer size calculations in
MPI_Allgathervto account for 3 integers (x, y, z) per R-coordinate, resolvingMPI_ERR_TRUNCATE.