本文へ移動
cccskills
無料GitHub で公開

mechsys

Use when a task mentions MechSys, tdem, tlbmdem, DEM/Particle/Voronoi optimization, CUDA DEM tests, LBM-DEM, or the MechSys simulation library in this workspace.

日本語の概要は準備中です。原文の説明を表示しています。

Axtal/mechsysスター 19(GitHub 19 + cccskills 0)

実行スコア 未検証

インストール方法を見る

含まれるファイル（200）

  • SKILL.md5.0 KB
  • .agents/skills/mechsys/SKILL.md5.0 KB
  • .gitignore2.3 KB
  • AGENTS.md9.2 KB
  • authors.txt22 B
  • CMakeLists.txt22.8 KB
  • debian/changelog1.5 KB
  • debian/compat2 B
  • debian/control1.4 KB
  • debian/copyright1.4 KB
  • debian/rules909 B
  • lib/adlbm/Cell.h9.9 KB
  • lib/adlbm/Domain.h14.0 KB
  • lib/adlbm/Lattice.h6.9 KB
  • lib/datfiles.h13.0 KB
  • lib/dem.h1.9 KB
  • lib/dem/basic_functions.h3.3 KB
  • lib/dem/cylinder.h3.0 KB
  • lib/dem/dem.cuh39.0 KB
  • lib/dem/distance.h27.7 KB
  • lib/dem/domain.h163.6 KB
  • lib/dem/dompargen.h49.2 KB
  • lib/dem/edge.h2.2 KB
  • lib/dem/face.h5.6 KB
  • lib/dem/graph.h4.2 KB
  • lib/dem/gsd.h60.0 KB
  • lib/dem/interacton.h47.2 KB
  • lib/dem/particle.h49.9 KB
  • lib/dem/special_functions.h18.5 KB
  • lib/dem/torus.h2.3 KB
  • lib/dem/visualise.h11.8 KB
  • lib/dfn/Domain.h91.6 KB
  • lib/dfn/Fracture.h9.7 KB
  • lib/dfn/Graph_x.h3.2 KB
  • lib/dfn/Line.h960 B
  • lib/dfn/Loop.h11.9 KB
  • lib/dfn/Math.h15.4 KB
  • lib/dfn/Random_function.h10.3 KB
  • lib/doxymain.h10.1 KB
  • lib/draw.h3.1 KB
  • lib/emlbm/Cell.h8.2 KB
  • lib/emlbm/Domain.h20.9 KB
  • lib/emlbm/Lattice.h6.4 KB
  • lib/emlbm2/Cell.h8.0 KB
  • lib/emlbm2/Domain.h14.4 KB
  • lib/emlbm2/Lattice.h4.5 KB
  • lib/flbm/Domain.h101.6 KB
  • lib/flbm/lbm.cl17.6 KB
  • lib/flbm/lbm.cuh49.5 KB
  • lib/geomtype.h2.9 KB
  • lib/inpfile.h51.4 KB
  • lib/lbm/Cell.h12.2 KB
  • lib/lbm/Dem.h5.3 KB
  • lib/lbm/Domain.h170.5 KB
  • lib/lbm/Dompargen.h39.3 KB
  • lib/lbm/Interacton.h5.3 KB
  • lib/lbm/Lattice.h20.4 KB
  • lib/lbmdem/Domain.h50.0 KB
  • lib/lbmdem/lbmdem.cuh40.7 KB
  • lib/lbmmpm/Domain.h22.9 KB
  • lib/linalg/jacobirot.h7.1 KB
  • lib/linalg/laexpr.h43.2 KB
  • lib/linalg/matrix.h13.4 KB
  • lib/linalg/matvec.h89.9 KB
  • lib/linalg/mumps.h4.1 KB
  • lib/linalg/quaternion.h5.3 KB
  • lib/linalg/sparse_crmatrix.h15.5 KB
  • lib/linalg/sparse_matrix.h18.5 KB
  • lib/linalg/sparse_triplet.h14.0 KB
  • lib/linalg/superlu.h3.9 KB
  • lib/linalg/superlud.h6.6 KB
  • lib/linalg/umfpack.h7.9 KB
  • lib/linalg/vector.h11.6 KB
  • lib/matfile.h17.5 KB
  • lib/mesh/alphashape.h5.6 KB
  • lib/mesh/delaunay.h5.4 KB
  • lib/mesh/mesh.h79.8 KB
  • lib/mesh/paragrid3d.h14.2 KB
  • lib/mesh/structured.h51.4 KB
  • lib/mesh/unstructured.h49.3 KB
  • lib/mpm/Corner.h3.5 KB
  • lib/mpm/Domain.h76.8 KB
  • lib/mpm/mpm.cuh5.6 KB
  • lib/mpm/Node.h7.8 KB
  • lib/mpm/Particle.h11.8 KB
  • lib/mpm/shape_functions.h3.8 KB
  • lib/nn/Domain.h22.2 KB
  • lib/numerical/min.h9.8 KB
  • lib/numerical/montecarlo.h4.9 KB
  • lib/numerical/multipleroots.h6.5 KB
  • lib/numerical/numdiff.h3.6 KB
  • lib/numerical/odesolver.h12.1 KB
  • lib/numerical/quadrature.h9.1 KB
  • lib/numerical/root.h9.1 KB
  • lib/python/msys_ana.py3.0 KB
  • lib/python/msys_drawmesh.py30.6 KB
  • lib/python/msys_fcrits.py21.7 KB
  • lib/python/msys_fig.py16.2 KB
  • lib/python/msys_invs.py17.2 KB
  • lib/python/msys_linfit.py3.9 KB
  • lib/python/msys_matvec.py4.4 KB
  • lib/python/msys_plt.py19.3 KB
  • lib/python/msys_pyshell.py8.8 KB
  • lib/sph/Boundary_Condition.h2.8 KB
  • lib/sph/Domain.h129.3 KB
  • lib/sph/Dompargen.h4.9 KB
  • lib/sph/Functions.h12.0 KB
  • lib/sph/Particle.h43.4 KB
  • lib/util/array.h29.7 KB
  • lib/util/colors.h15.6 KB
  • lib/util/fatal.h4.0 KB
  • lib/util/maps.h23.9 KB
  • lib/util/numstreams.h5.0 KB
  • lib/util/stopwatch.h3.7 KB
  • lib/util/string.h6.9 KB
  • lib/util/tree.h7.7 KB
  • lib/util/util.h8.7 KB
  • lib/vtk/arrow.h7.1 KB
  • lib/vtk/arrows.h6.2 KB
  • lib/vtk/axes.h11.1 KB
  • lib/vtk/cube.h4.9 KB
  • lib/vtk/cylinder.h5.0 KB
  • lib/vtk/disk.h5.4 KB
  • lib/vtk/isosurf.h5.4 KB
  • lib/vtk/line.h3.2 KB
  • lib/vtk/plane.h6.0 KB
  • lib/vtk/sgrid.h13.6 KB
  • lib/vtk/sphere.h3.7 KB
  • lib/vtk/spheres.h9.9 KB
  • lib/vtk/text.h3.2 KB
  • lib/vtk/text2d.h2.9 KB
  • lib/vtk/triangulate.h4.5 KB
  • lib/vtk/ugrid.h4.6 KB
  • lib/vtk/win.h7.2 KB
  • lib/vtkcelltype.h7.8 KB
  • MECHSYS_HANDBOOK.md23.2 KB
  • MEMORY.md12.1 KB
  • Modules/FindBLITZ.cmake2.1 KB
  • Modules/FindCGAL.cmake2.2 KB
  • Modules/FindDEPS.cmake13.4 KB
  • Modules/FindDEPSFL.cmake14.3 KB
  • Modules/FindFLTK.cmake2.1 KB
  • Modules/FindGMSH.cmake4.2 KB
  • Modules/FindGOTOBLAS.cmake2.0 KB
  • Modules/FindGSL.cmake2.5 KB
  • Modules/FindHDF5.cmake3.1 KB
  • Modules/FindIGRAPH.cmake2.4 KB
  • Modules/FindLocLAPACK.cmake2.1 KB
  • Modules/FindMECHSYS.cmake2.4 KB
  • Modules/FindMTL.cmake2.0 KB
  • Modules/FindMUMPS.cmake2.3 KB
  • Modules/FindPARMETIS.cmake2.1 KB
  • Modules/FindPROC.cmake2.2 KB
  • Modules/FindSCALAPACK.cmake2.1 KB
  • Modules/FindSOPLEX.cmake2.2 KB
  • Modules/FindSPHASH.cmake2.0 KB
  • Modules/FindSUPERLU.cmake2.4 KB
  • Modules/FindSUPERLUD.cmake2.2 KB
  • Modules/FindSUPERLUMT.cmake2.2 KB
  • Modules/FindTENSORS.cmake2.0 KB
  • Modules/FindUMFPACK.cmake2.8 KB
  • Modules/FindVORO.cmake2.3 KB
  • Modules/FindVTK.cmake3.1 KB
  • Modules/mechsys-config.in3.3 KB
  • Modules/mechsyscc.in6.9 KB
  • msys_sandbox.zip3.4 KB
  • patches/blitz/do_patch.bash293 B
  • patches/blitz/funcs.h.diff33 B
  • patches/blitz/mathfunc.h.diff30 B
  • patches/blitz/range.h.diff166 B
  • patches/igraph/community_leiden.c.diff628 B
  • patches/igraph/do_patch.bash757 B
  • patches/igraph/README.md1.5 KB
  • patches/igraph/uninit.c.diff326 B
  • patches/mumps/do_patch.bash128 B
  • patches/mumps/Makefile.inc3.9 KB
  • patches/superlumt/do_patch.bash633 B
  • patches/superlumt/install_makefile.diff50 B
  • patches/superlumt/install.bash251 B
  • patches/superlumt/make.pthreads1.1 KB
  • patches/superlumt/pctest.bash1.6 KB
  • patches/superlumt/pdtest.bash1.6 KB
  • patches/superlumt/pstest.bash1.6 KB
  • patches/superlumt/pztest.bash1.6 KB
  • patches/superlumt/test.bash58 B
  • patches/superlumt/testing_makefile.diff492 B
  • patches/voro/cell0.3.hh.diff283 B
  • patches/voro/CMakeLists.txt85 B
  • patches/voro/container0.3.hh.diff51 B
  • patches/voro/do_patch.bash228 B
  • patches/voro/src_CMakeLists.txt29 B
  • README.txt815 B
  • scripts/additional_installations.bash8.9 KB
  • scripts/install_apt_deps.bash772 B
  • scripts/install_compile_deps.bash11.3 KB
  • scripts/install_flashlite.bash9.9 KB
  • scripts/old/01_create_mechsys_deb.sh1.3 KB
  • scripts/old/02_gen_repository.sh1.3 KB
  • scripts/old/autocrop_convert_eps.bash97 B
  • scripts/old/change_header.sh3.1 KB

SKILL.md（原文）

インストールする前に、エージェントに与えられる指示の中身を確認できます。

MechSys

This is the visible, downloadable copy of the MechSys skill. DSH discovers .agents/skills/mechsys/SKILL.md; keep the two files in sync.

This is the project-local skill for the MechSys checkout. It is discovered automatically from .agents/skills/mechsys; do not install a separate plugin just to provide this static guidance.

The authoritative context is:

  1. AGENTS.md — automatic, concise agent rules.
  2. MECHSYS_HANDBOOK.md — architecture, build/run commands, exactness procedure, benchmarks, and known traps.
  3. MEMORY.md — distilled experience and a short verification checklist.
  4. The optimization reports:
    • CPU_DEM_OPTIMIZATION_CHANGES.md
    • CUDA_OPTIMIZATION_CHANGES.md
    • REDUCED_TEST_CU_03_PERFORMANCE.md

Read the relevant documents before making substantive changes, and update them when a durable lesson or corrected benchmark is learned.

Non-negotiable rules

  • Do not assume a given checkout or HEAD contains the optimized state. Check git status, the reports, and compare against a clean baseline created from the chosen commit.
  • CPU DEM exactness is checked at Nproc=1 by comparing HDF5 datasets, not whole-file MD5. Expect neq [] and matching printed energies.
  • Preserve candidate-list order and the original Distance arithmetic in the contact path. Do not sort the lists.
  • The CPU Verlet cutoff is Rsum + 4.0*ContactDispGlobal(). Do not lower it without a new safe derivation and an exact comparison.
  • Broadphase bounds must be conservative. If in doubt, remove a rejection rather than risk a missed contact.
  • CUDA test_cu_03/04 use atomicAdd; the baseline is non-deterministic. Do not claim bitwise equality for those GPU runs. Nthread=512 fails for the current large test; tested range is up to 384.
  • Keep the separate pTranslate + pRotate fallback. Making fused movement unconditional broke test_cu_04.
  • The outer solve loop owns Time/iter; publish them to CUDA with SetClock. Do not add independent clock increments in DEM or LBM substeps.
  • After changing lib/dem/*, build at least one CPU target, one CUDA target, and the relevant LBM-DEM target.

Rotation caches

Particle owns wrot = Rotation(w,Q) and Qc = Conjugate(Q). They are refreshed by CalcProps, InitializeVelocity, Rotate, UpdateContacts, and after a user Setup callback.

  • CUDA UploadParticle copies the host caches; it does not recompute them.
  • CUDA DnloadParticle copies the device caches back.
  • The built-in device movement kernels refresh the device caches after changing Q/w.
  • A custom GPU movement/rotation kernel that changes Q/w must refresh DPar[].wrot/DPar[].Qc itself.
  • Host code that assigns Q/w directly and bypasses the normal paths must call Particle::UpdateRotCache() before using the caches.

Build and run

From the repository root:

cd tdem
make -j8 test_periodic
make -j8 test_cu_01 test_cu_02 test_cu_03 test_cu_04
./test_periodic ../benchmarks/cpu_voronoi/tper_small 8

cd ../tlbmdem
make -j8 tlbmdem01 tlbmdem02
make -j8 tlbmdem_cu_01

test_periodic takes a FileKey prefix (the .inp suffix is appended). test_cu_03 takes Nproc, Verlet distance, and threads per block: ./test_cu_03 1 0.05 128. Full LBM-DEM and CUDA tests are long; use a temporary reduced copy for smoke tests and do not modify the checked-in test just to shorten it.

For one-off cases, use the configured standalone compiler: ./mechsyscc case.cu from the repository root (or the equivalent mechsyscc in the current installation).

FLBM / GPU fluid cases

  • lib/flbm/Domain.h is the fluid LBM; lib/flbm/lbm.cuh holds the CUDA kernels. Use D2Q9 with iVec3_t(nx,ny,1) for a 2-D GPU run.
  • The GPU streaming kernel resets BForce every step. A constant body force must be written back to dom.pBForce on the device in the Setup callback before collision; setting the host array once is not enough.
  • dom.Nthread controls the CUDA block size. The Nproc argument of Domain::Solve affects host packing/transfers on the GPU path.
  • CPU and GPU body-force conventions differ: CPU StreamSC adds 0.5*dt*BForce to the reported velocity, while the GPU fused kernel returns the raw second moment. The raw GPU moment carries a -0.5*nu*phi bias in permeability; use u_physical = u_raw + 0.5*dt*fx*phi.
  • Keep one-off FLBM cases outside the checkout; do not add them to tflbm/ unless they belong in the library test suite.

Reporting

Report measured wall times, separate solve time from total wall time, and explicitly distinguish what compiled, what ran, what matched a baseline, and what was faster. If a verification was not done, say so.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

コメント

ログインなしで書き込めます。使ってみた感想、質問、動かなかった環境などを気軽にどうぞ。

まだコメントはありません。最初のコメントを書いてみませんか。

送信するとすぐに公開されます。メールアドレスや電話番号などの個人情報は書かないでください。URL は 2 つまでです。

0 / 1000 文字

このフォームは reCAPTCHA で保護されており、Google のプライバシーポリシーと利用規約が適用されます。

このスキルの問題を報告する