This CL replaces the post-hoc implicit flex correction (`flexInterp_cgsolve`) with a **linearly-implicit effective metric** `M̃ = M + (h² + h·damping)·K` carried by the CG constraint solver itself. Contact/friction forces and implicit flex elasticity are now computed against one consistent metric, instead of the solver seeing `M` and a post-solve correction changing `qacc` behind its back.
Gate (unchanged semantics): `solver="CG"` + implicit/implicitfast integrator + pyramidal cones + flex stiffness present. Newton and PGS are untouched. `solver="CG"` remains the user-facing contract — the factorization is an implementation detail of the preconditioner.
### What's in the metric
- **mjData `efm_*`** (arena, efc-like lifetime/skip semantics; built in `mj_fwdPosition`, value-refreshed in `mj_fwdVelocity`): the per-step stiffness CSR `efm_B_*`, its reverse-Cholesky factor `efm_dofid` + `efm_L_*` (nested-dissection ordered, separators-first for the reverse factorization), and the smooth-force shift `efm_c = h·K·qvel`.
- **`mjd_flexStiff_assemble`** now assembles stretch (Gauss–Newton), standard dim-2 bending, and — via the cached corotated stiffness `d->flexelem_krot` — interp stiffness (all node bodies on simple sliders: point Jacobian is I₃, `flex_centered` not required; fixed nodes drop like pins) into one dof-level CSR. `mjd_effMulAdd`/`mjd_effSolve` apply the metric, with matrix-free operator fallbacks where assembly does not apply.
- **mjModel `efm0_*`** (`nefm0dof`/`nefm0L`): the constant part of the metric factor — currently the dim-2 bending factor, computed once in `mj_setConst` — so bending-only models pay zero per-step factorization cost. Naming mirrors mjData's `efm_*` with the standard `0`-suffix (reference/constant) idiom, and is deliberately not bending-specific: future constant contributors extend it without renames.
- The solver consumes the metric through pre-shifted `qfrc_smooth` and the metric products `Ma`/`Mv`/`Mgrad`; `qacc_smooth` becomes the unconstrained minimizer of the implicit dynamics, which makes the no-constraint shortcut and the warmstart choice consistent by construction.
- **`mj_inverse` adds `B·qacc − c`**, making inverse dynamics discrete-consistent with the gated forward dynamics — exact, since the gated path has no qDeriv term (new test `ForwardTest.GatedFlexInverseConsistency`).
### Performance
All numbers: ms/step over the same 2000-step window, models as shipped on each side (old code with the old model settings vs this CL with the new ones).
The new solver path activates on exactly two shipped models — the ponchos, the only flex models that need an implicit integrator (poncho on Euler degenerates to >200 ms/step). For them, this CL trades speed for consistency: the implicit bending solve now runs inside every solver iteration, where the contact solve can see the stiffness, instead of once after the solve. Solver iterations drop because the curvature is visible, but each iteration pays for the implicit solve:
| model | before | after | solver iters/step |
|---|---|---|---|
| poncho | 2.47 | 3.30 (1.33×) | 16.8 → 11.8 |
| poncho_edgeequality | 1.96 | 2.72 (1.39×) | 13.2 → 10.0 |
What that price buys: contact forces consistent with the implicit elasticity (previously the post-hoc correction changed `qacc` after the constraint solve), discrete-consistent inverse dynamics, and the removal of the post-hoc special case from the integration path. Raising poncho's timestep from 2 to 5 ms leaves its per-step cost nearly flat, so the consistency price can be recovered by taking fewer steps where accuracy allows.
Every other flex model was measured stable on Euler at its shipped timestep and switches to it (these models predate the post-hoc integrator; implicit was never load-bearing for them). They end up equal or faster than before: bunny_multicell 0.47 → 0.40, trampoline 0.28 → 0.25, plate 1.02 → 0.99, pancake 0.34 → 0.33.
Finally, the per-step factorization makes configurations practical that the old code could only integrate explicitly: implicit stretch elasticity (`elastic2d="stretch"`/`"both"`, dim-3 solids) and factorized interp stiffness. No before/after exists for these — stock has no implicit treatment of stretch at all.
### Behavior changes
- With the post-hoc correction deleted, interp/bending models running `solver="Newton"` (or elliptic cones, or islands) now integrate flex elasticity **explicitly** (previously: post-hoc implicit). Affects e.g. `gripper_trilinear` (stable, and faster, but different semantics). Follow-up options: Newton-side metric support, or a documented fallback.
- With the gate on, `mj_forward` outputs are timestep-dependent for gated models (they answer the linearly-implicit discrete problem); `qacc_smooth` and `mj_inverse` change accordingly. Non-gated models are bit-identical (full suite green throughout).
### Validation
- 1737/1737 tests, including new: `FlexStretchDerivatives` (FD-validated GN operator), `FlexStiffAssemble`/`FlexStiffAssembleInterp` (CSR ≡ operators), `GatedFlexInverseConsistency` (fails pre-change), equivalence tests vs the old post-hoc treatment (bending matches to 2e-11).
- Fingerprint discipline throughout: bending-only models bit-exact across every refactor; permutation/kernel changes verified iteration-identical.
### Known follow-ups (not in this CL)
3×3-block sparse Cholesky kernel (the numeric factorization is index-bound; projected ~3× on the factor); mjModel persistence of the factor's symbolic pattern (rest-pose ND makes sizes compile-time); the general effective-metric mode (all solvers, all PSD-safe force classes, behind an enable flag).
PiperOrigin-RevId: 948561856
Change-Id: I8b8e32ebd0428042af71647d0470d10773bf6daf
Introduce a new header `gil.h` defining `MutexLockIfGilDisabled` to support thread-safety in both standard and free-threaded CPython builds.
Protect critical shared states and registries:
- Guard global Python callback pointers in `callbacks.cc` using a mutex. Move `gil_scoped_acquire` into local blocks around refcount modifications to prevent `longjmp` from bypassing destructors.
- Protect raw pointer maps in `structs_wrappers.cc` with static mutexes.
- Replace TOCTOU race in `mjcb_time` initialization with thread-safe `std::call_once`.
- Add synchronization to lazy indexer array cache initialization in `indexers.cc` and `indexer_xmacro.h`.
- Protect vector mutations in `StructListBase::PopulateUpTo` in `structs.h` with a mutex.
- Revert unnecessary atomic changes to threadpool counters.
- Declare free-threading compatibility by passing `pybind11::mod_gil_not_used()` to all extension modules.
Fixes#3259Fixes#3256Fixes#2978
PiperOrigin-RevId: 941101502
Change-Id: Iec4ce58afcbc75d4b0be6a9a21fc8a47854242e3
This allows calling `mujoco.mj_copyData(dest, model, src)` to copy data between existing `mjData` instances.
PiperOrigin-RevId: 855150139
Change-Id: If122cb7acf590846e25ee15f9770c3c087147169
This function allows the caller to extract a subset of components of a state previously obtained via `mj_getState` without having to first write it back into `mjData`.
PiperOrigin-RevId: 822165077
Change-Id: I6433261f4f5ec2e8d024bb7dee8f7dbf1e62666c
- `qLDiagSqrtInv` is only required for the dual solvers. It is now computed as-needed rather than unconditionally.
- `mj_solveM2` now requires a new input array `sqrtInvD` which contains the square root of the inverse diagonal D (formerly saved in `qLDiagSqrtInv`).
PiperOrigin-RevId: 710805133
Change-Id: I0622d6a8da3882916824e9c10bad9223c122c321
This function extracts the 3D rotation from an arbitrary 3x3 matrix by refining the input quaternion. It is based on the paper "A robust method to extract the rotational part of deformations" by Müller, Matthias, Jan Bender, Nuttapong Chentanez, and Miles Macklin.
PiperOrigin-RevId: 700006006
Change-Id: I77550993233dea9cdf68601762a3ae7ded749bdf
These functions names and argument ordering are more consistent with the rest of the API.
PiperOrigin-RevId: 643788290
Change-Id: I783eda8021b80b82098e23ed95669b102bb82508
Fixes#1719.
Also improve documentation of `mju_sigmoid`, was previously only documented in changelog.
PiperOrigin-RevId: 642637458
Change-Id: I811a0bf8f881c8b76c9a62d2faec8fdec4e7585b
Before this change, MjDataWrapper contained a custom data structure called MjDataMetadata, which contained the sizes of various arrays needed to create MjData. This was used for serialization and deserialization, as well as copying MjData instances.
This worked fine for native MuJoCo models, but as soon as plugins were used, there was information in the model that was needed and not available in MjDataMetadata. Since plugins are so general, keeping MjDataMetadata was untenable.
PiperOrigin-RevId: 578205355
Change-Id: I11b9ee797d1da5aaf1fecbbe170b3b0f1f1a3e6c
Alessio Quaglino