Issues

Tension cutoff at yield_stress_min = 0 gives exactly zero viscosity and a singular operator (DIVERGED_LINEAR_SOLVE)

#759

· lmoresi opened

on Sep 18, 2026

yield_stress_min is not scaled like yield_stress: a dimensional floor changes the answer, a non-dimensional one does not

#758

· lmoresi opened

on Sep 18, 2026

Schur preconditioner scaling omits the solvent and DEVSS viscosities, and the pressure solve runs away at high Wi

#755

· lmoresi opened

on Sep 17, 2026

JIT C-source generation is non-deterministic across MPI ranks (flaky hard abort at np>1)

#752

· lmoresi opened

on Sep 16, 2026

Eulerian stress history blows up at a shear-wave reflection (Waters & King start-up)

#749

· lmoresi opened

on Sep 16, 2026

SNES_Scalar still assembles its tangent with derive_by_array: the #457 transpose survives in the one solver #493 did not convert

#747

· lmoresi opened

on Sep 16, 2026

After solve(), psi_star[0] holds the new stress on the Stokes family but the previous step's stress on NavierStokesSLCN

#742

· lmoresi opened

on Sep 14, 2026

Particle stress history (Lagrangian_Swarm) gives a NaN first residual on the viscoelastic cylinder benchmark

#738

· lmoresi opened

on Sep 14, 2026

Integration-point stress history grows an inlet instability when dt is halved (Wi 5 cylinder)

#737

· lmoresi opened

on Sep 13, 2026

The nodal semi-Lagrangian stress history carries ~50% too much stress at a no-slip wall: two independent transports converge together, it stays flat elsewhere

#735

· lmoresi opened

on Sep 12, 2026

Nitsche penalty scale regressed with mesh.cell_size(): the Zhong spherical-shell response went from 0.2% to 2.4-5.7% off

#734

· lmoresi opened

on Sep 11, 2026

CI runtime: shard the test batches across parallel jobs when the single job tightens again

#730

· lmoresi opened

on Sep 11, 2026

Navigation Menu

All issues

Tension cutoff at yield_stress_min = 0 gives exactly zero viscosity and a singular operator (DIVERGED_LINEAR_SOLVE)

yield_stress_min is not scaled like yield_stress: a dimensional floor changes the answer, a non-dimensional one does not

Schur preconditioner scaling omits the solvent and DEVSS viscosities, and the pressure solve runs away at high Wi

JIT C-source generation is non-deterministic across MPI ranks (flaky hard abort at np>1)

Eulerian stress history blows up at a shear-wave reflection (Waters & King start-up)

SNES_Scalar still assembles its tangent with derive_by_array: the #457 transpose survives in the one solver #493 did not convert

After solve(), psi_star[0] holds the new stress on the Stokes family but the previous step's stress on NavierStokesSLCN

Particle stress history (Lagrangian_Swarm) gives a NaN first residual on the viscoelastic cylinder benchmark

Integration-point stress history grows an inlet instability when dt is halved (Wi 5 cylinder)

The nodal semi-Lagrangian stress history carries ~50% too much stress at a no-slip wall: two independent transports converge together, it stays flat elsewhere

Nitsche penalty scale regressed with mesh.cell_size(): the Zhong spherical-shell response went from 0.2% to 2.4-5.7% off

CI runtime: shard the test batches across parallel jobs when the single job tightens again

Search results