Skip to content
Snippets Groups Projects
Select Git revision
  • 51cb767b44056e283091cf697ab5de354b39b3bc
  • master default protected
  • zoom_merge protected
  • darwin/gear_chemistry_fluxes
  • fstasys/VEP_Fm2
  • reyz/gear_preSN_feedback
  • fewer_gpart_comms
  • MAGMA2_matthieu
  • MHD_canvas protected
  • forcing_boundary_particles
  • melion/BalsaraKim
  • darwin/sink_mpi_physics
  • darwin/gear_mechanical_feedback
  • FS_VP_m2_allGrad
  • improve-snap-to-ic
  • karapiperis/Bcomoving_as_a2_Bphysical
  • split-space-split
  • reyz/debug
  • darwin/gear_preSN_fbk_merge
  • gas-splitting
  • stars_sidm_iact
  • v2025.10 protected
  • v2025.04 protected
  • v2025.01 protected
  • v1.0.0 protected
  • v0.9.0 protected
  • v0.8.5 protected
  • v0.8.4 protected
  • v0.8.3 protected
  • v0.8.2 protected
  • v0.8.1 protected
  • v0.8.0 protected
  • v0.7.0 protected
  • v0.6.0 protected
  • v0.5.0 protected
  • v0.4.0 protected
  • v0.3.0 protected
  • v0.2.0 protected
  • v0.1.0-pre protected
  • v0.1 protected
  • v0.0 protected
41 results

swiftsim

user avatar
Matthieu Schaller authored
Simplify, permit user control over affinity

This ensures we 'do the right thing' when the user imposes affinity through e.g. Intel MPI's I_MPI_PIN_DOMAIN or other mechanisms. Also no longer does works with a shuffled ```cpuid``` array, which will hopefully have less surprising failure modes, and means we don't have to handle hyperthreading explicitly. If we want to be clever in this way e.g. to maximise available cache, then we could replace libnuma with hwloc as discussed elsewhere.

Hopefully this gives us a reasonable affinity which is easily overridden using standard techniques.

See merge request !120
51cb767b
History
Name Last commit Last update