Repository navigation

[gpu-1] End-to-end GPU test of VlasovAmpereOneSpecies on 1 and 2 ranks - #734

Merged

max-models merged 37 commits into

cuda-namespacesfrom

gpu-1-vlasov-ampere-e2e

Oct 13, 2026

Merged

[gpu-1] End-to-end GPU test of VlasovAmpereOneSpecies on 1 and 2 ranks#734
max-models merged 37 commits into
cuda-namespacesfrom
gpu-1-vlasov-ampere-e2e

Commits on Oct 7, 2026
- feectools: Kronecker solver on the device
  
  max-models
  and
  claude
  committed
  aeebf95
- feectools: stencil 3D device kernels with 6D views
  
  max-models
  and
  claude
  committed
  39f1984
- Move the feectools submodule to the head of feectools#96 (64b10fa)
  
  max-models
  and
  claude
  committed
  081768b
- feectools: inner products stay on the device
  
  max-models
  and
  claude
  committed
  c58ccc0
- MassMatrixPreconditioner on the CuPy backend
  
  max-models
  and
  claude
  committed
  1845bc6
- End-to-end GPU test for LinearVlasovAmpereOneSpecies
  
  max-models
  and
  claude
  committed
  afcbb90
- Merge feectools-device-kronecker-solve (#703) into feectools-stencil-6d-views
  
  max-models
  and
  claude
  committed
  ca84d59
- Merge feectools-stencil-6d-views (#706) into feectools-inner-on-device
  
  max-models
  and
  claude
  committed
  73e5c6b
Commits on Oct 8, 2026
- Update feectools to latest commit in inner-on-device
  max-models
  committed
  85bc754
- Merge feectools-inner-on-device into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  ce8f174
- Merge mass-preconditioner-cupy into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  aa4f52d
- Merge linear-vlasov-ampere-gpu-e2e into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  240e1d4
- Merge cuda-namespaces (#722) into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  560db4f
- End-to-end GPU test of VlasovAmpereOneSpecies on 1 and 2 ranks
  
  max-models
  and
  claude
  committed
  e40e4d7
Commits on Oct 9, 2026
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  61a0875
- Merge devel into linear-vlasov-ampere-gpu-e2e
  max-models
  committed
  10b8c51
- Merge devel (#731 Kronecker preconditioners, feectools 0.7.0) into mass-preconditioner-cupy
  
  max-models
  and
  claude
  committed
  e5f5b1d
- Merge mass-preconditioner-cupy into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  aead20c
- Merge linear-vlasov-ampere-gpu-e2e into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  59dc84b
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  23fa155
- Merge devel into linear-vlasov-ampere-gpu-e2e
  max-models
  committed
  42b7c85
- Merge devel into mass-preconditioner-cupy
  
  max-models
  and
  claude
  committed
  d6d99e6
Commits on Oct 10, 2026
- Merge mass-preconditioner-cupy into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  4f9ad4a
- Merge linear-vlasov-ampere-gpu-e2e into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  99d7bba
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  cb25e33
- Diagonal scaling: include band entries that wrap onto the diagonal
  
  max-models
  and
  claude
  committed
  4d4b51f
- VlasovAmpereCoupling: update the ghost regions of the mid-point field before the push
  
  max-models
  and
  claude
  committed
  a0f1cf3
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  788c1b0
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  ca09409
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  e7e968b
- Merge cuda-namespaces into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  c6092ba
- Merge fix-diagonal-scaling-1d (#741) into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  f864be0
Commits on Oct 12, 2026
- Merge branch 'cuda-namespaces' into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  4405320
- Merge branch 'cuda-namespaces' into gpu-1-vlasov-ampere-e2e
  max-models
  committed
  727083e
Commits on Oct 13, 2026
- draw_markers: send the markers to their processes also without sorting boxes (#764)
  
  spossann
  and
  claude
  committed
  1d2e28e
- Review fixes for #734: run the e2e tests in CI, keep the _core_diagonal contract
  
  spossann
  and
  claude
  committed
  d356ad9
- Merge branch 'draw-markers-mpi-sort' (#765) into gpu-1-vlasov-ampere-e2e
  
  spossann
  and
  claude
  committed
  1a02d0c

Commits

Commits on Oct 7, 2026

Commits on Oct 8, 2026

Commits on Oct 9, 2026

Commits on Oct 10, 2026

Commits on Oct 12, 2026

Commits on Oct 13, 2026