Skip to content

Commit

Permalink
Dem fixing resize containers (#1316)
Browse files Browse the repository at this point in the history
Description
Some std::vector<> in the DEM were not resize properly. The resize was being done before the exchange of ghost particles between processors, thus the max particle index was be too small. This was causing unexpected segmentation faults when cells had a lot of neighbouring cells. This would happen especially when dealing with periodic boundary condition.

Solution
The exchange of ghost particles must be done before the resize. This is a permanent fix.

Testing
A new test was added using the periodic hopper example. This test is currently crashing on master. It uses 2 processor, otherwise there is not ghost particles.

Documentation
N/A
  • Loading branch information
OGaboriault authored and mivaia committed Oct 15, 2024
1 parent 4a57a49 commit 631da7d
Show file tree
Hide file tree
Showing 5 changed files with 674 additions and 3 deletions.
6 changes: 6 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,12 @@
All notable changes to the Lethe project will be documented in this file.
The format is based on [Keep a Changelog](http://keepachangelog.com/).

## [Master] - 2024-10-11

### Fixed

- MAJOR Unexpected segmentation faults were occuring when using large simulation with aperiodic boundary condition. The cause of thuis bug was identify and fixed. It was related to the resizing of the force, torque and displacement vectors which was not considering the ghost particles when resized. [#1316](https://github.com/chaos-polymtl/lethe/pull/1316)

## [Master] - 2024-10-09

### Changed
Expand Down
Loading

0 comments on commit 631da7d

Please sign in to comment.