Issues

• Debug builds with -Mbounds fail test_penta_omp_4 and test_ab_checkpoint_omp_1 on main

  bug
  Bug

  xcompact3d/x3d2#468 · Filter by author kaanolgu opened on Oct 27, 2026 · Version 0.1.0

• FFT Poisson 110 zeroes a valid mode, leaving divergence after projection

  bug
  Bug

  xcompact3d/x3d2#465 · Filter by author ia267 opened on Oct 23, 2026 · Version 0.1.0
  

• Channel smoke tests start with no noise and do not test kx, kz modes in the Poisson solve

  bug
  xcompact3d/x3d2#463 · Filter by author kaanolgu opened on Oct 23, 2026 · Version 0.1.0
  

• FFT Poisson solver gives wrong results with the 'bottom' y-stretching mode

  bug
  xcompact3d/x3d2#462 · Filter by author kaanolgu opened on Oct 23, 2026 · Version 0.1.0
  

• Distributed compact solver leaves div_u residual for non-periodic x

  xcompact3d/x3d2#461 · Filter by author ia267 opened on Oct 23, 2026
  

• memcheck gives false MISMATCH when other processes use the GPU

  bug
  Bug

  xcompact3d/x3d2#458 · Filter by author ia267 opened on Oct 22, 2026 · Version 0.1.0
  1

  

• Add IBM mask as an input parameter

  enhancement
  Task

  xcompact3d/x3d2#456 · Filter by author ia267 opened on Oct 21, 2026 · Version 0.1.0
  

• Channel (010) case multi GPU support [CUDA]

  corecuda
  Feature

  xcompact3d/x3d2#453 · Filter by author kaanolgu opened on Oct 21, 2026 · Version 0.1.0
  

• FFT Poisson solver: support the 111 case (non-periodic x, y and z)

  enhancement
  Feature

  xcompact3d/x3d2#449 · Filter by author ia267 opened on Oct 16, 2026 · Version 0.1.0
  

• ABL fringe region: shifted periodic and concurrent precursor

  enhancement
  xcompact3d/x3d2#441 · Filter by author ia267 opened on Oct 13, 2026

• Move the multi-rank state of the CUDA Poisson solver into derived types

  xcompact3d/x3d2#437 · Filter by author kaanolgu opened on Oct 9, 2026
  

• Add pipeline dependency/ordering to GitLab CI

  xcompact3d/x3d2#428 · Filter by author kaanolgu opened on Oct 1, 2026

• Avoid duplicate GitLab pipeline runs on merge

  xcompact3d/x3d2#427 · Filter by author kaanolgu opened on Oct 1, 2026

• Build the OpenMP FFT Poisson solver and its tests with WITH_MPI=OFF

  xcompact3d/x3d2#426 · Filter by author kaanolgu opened on Sep 30, 2026 · Version 0.1.0
  

• New GitLab Routines requires some tweaks

  xcompact3d/x3d2#423 · Filter by author kaanolgu opened on Sep 30, 2026
  1

  

• OMP backend: build the FFT Poisson solver in place instead of by function-result assignment

  enhancementomp
  Bug

  xcompact3d/x3d2#421 · Filter by author kaanolgu opened on Sep 30, 2026 · Version 0.1.0
  

• Group spectral kernel extents, offsets and a*/b* coefficients into derived types

  enhancement
  Task

  xcompact3d/x3d2#413 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0

• z_step_{fw,bw} for the z post-process

  enhancement
  Task

  xcompact3d/x3d2#411 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0

• Paired even/odd loops pair_{split,merge} helpers

  enhancement
  Task

  xcompact3d/x3d2#410 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0

• Shared gather-index helper for enforce/undo_periodicity_{x,xy}_omp

  enhancement
  Task

  xcompact3d/x3d2#409 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0

• [OMP-cpu] Cache-blocked transpose in forward & backward FFT

  enhancement
  Task

  xcompact3d/x3d2#408 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0
  

• Fuse the OpenMP 110 spectral passes

  ompperformance
  Feature

  xcompact3d/x3d2#407 · Filter by author kaanolgu opened on Sep 25, 2026 · Version 0.1.0
  

• GPU-aware I/O: further performance improvements

  cudaperformance
  xcompact3d/x3d2#402 · Filter by author ia267 opened on Sep 23, 2026
  

• Add cylinder 110 case CUDA/OMP for regression tests

  xcompact3d/x3d2#390 · Filter by author ia267 opened on Sep 22, 2026 · Version 0.1.0
  

• Channel (010) case y-stretching for multi CPU [OpenMP]

  xcompact3d/x3d2#388 · Filter by author kaanolgu opened on Sep 21, 2026 · Version 0.1.0