search

stfc / HartreeParticleDSL

MIT License
1 stars 1 forks source link

Particle migration doesn't work correctly. #71

Open LonelyCat124 opened 1 year ago

LonelyCat124 commented 1 year ago

There's an issue with the MPI implementation where something is not quite correct, and particles with 0 position seem to occur on every node.

LonelyCat124 commented 1 year ago

Ok, no rank is expecting to be sending a 0.0 value, but every node is receiving one. This is strange.

LonelyCat124 commented 1 year ago

It looks like its the sending/receving of position/velocity is wrong in the DSL code, it also needs * ...extent(2).